Skip to main content

Full text of "Organic seminar abstracts"

See other formats






Return this book on or before the 
Latest Date stamped below. 

University of Illinois Library 


Ll(,l — 1141 


Reactions Involving the Azomethlne Linkage of Pyridine 

R. C. Fuson, Septemberl? . ... 1 

Recent Investigations of Calabash Curare 

H. R. Snyder, September 1? 4 

The Mechanism of Indole Formation from Phenacylarylamines 

T. R. Govindachari, September 24 7 

Displacement Reactions 

E. R. Alexander, September 24 11 

Hydrofuranol Derivatives of the Hexitols 

John Lynde Anderson, October 1 ............. 14 

l-Cyano-l,3-butadiene and Its Reactions 

Bruce Englu nd, October 1 18 

Sodium Hydride in Organic Chemistry 

Melvin I. Kohan, October 8 22 

Synthesis of Amidlnes 

John B. Campbell, October 8 25 

Strecker Degradation 

A. S. Nagarkatti, October 15 28 

Recent Advances in Thiophene Chemistry 

P. D. Caesar, October 15 31 

Reactions of Dlhydrophyran and Related Compounds 

Sidney Baldwin, October 22 35 

A General Theory of Neighboring Groups and Reactivity in 
Nucleophilic Replacement Reactions 

R. W. Meikle, October 22 38 

The Structure and Synthesis of Ketoyobyrine 

William E. Goode, October 22 42 

The Addition of Amines to Acetylenic Ketones 

Carl S. Hornberger, October 29 45 

Recent Studies of the Jacobsen Reaction 

K. H. Takemura, October 29 48 

Syntheses of 5-Aminothiazoles 

Alex Kotch, October 29 51 

Digitized by the Internet Archive 

in 2012 with funding from 

University of Illinois Urbana-Champaign 


The Total Synthesis of (+)-Egtrone 

Janet B. Peterson, November 5 54 

By Products in the Reformat sky Reaction 

C. W. Fairbanks, November 5 57 

Studies on the Structure of Pyrldoxal Phosphate 

S. E. Neuhausen, November 5 60 


Aaron B. Herrick, November 12 63 

The Synthesis of dl- Lysine 

Robert £• Carnahan, November 12 ...... 66 

The Rearrangement of Alpha Haloke tones 

/Y,#. A. DeWalt, Jr., November 12 69 

Formation of Carbon- Silicon Bonds 

H. W. Hill, Jr., November 19 72 

The Reaction of Mono61efins with Maleic Anhydride, Sulfur 
Trioxlde, Formaldehyde and Azodicarboxylic Ester 

H. DeWald, November 19 75 

New Organic Insecticides 

George I. Poos, November 19 78 

A New Synthesis of Arylacetic Acids 

Harold J. Watson, December 3 83 

Recent Syntheses of Spiranes 

William R. Miller, December 3 85 

Recent Studies of the Stobbe Condensation 

H. Rosenberg, December 10 88 

Scope of the Arbuzov Reaction 

Clarie Blue stein, December 10 ....... . 91 

Punched Cards for the Individual Organic Chemist 

Karl F. Heumann, December 10 95 


Edward F. Elslager, December 17 98 

The Oxo Process 

Edward F. Riener, December 17 102 

The Behavior of Organic Acids and Esters in Sulfuric Acid 

R. G. Bannister, December 17 105 

Enolization Tendency in Aromatic {3-Dlketones 

Alfred S. Spriggs, January 7 108 


The Structure of Meldrum f s Acid 

Richard H. Tennyson, January 7 Ill 

Rearrangement of Some Peroxide Esters 

Jean V. Crawford, January 7 114 

The Use of Higher Diazohydro carbons in the Arndt-Eistert 

Synthesis Paul M. Mader, January 14 11? 

The Significance of the Transition State in Aromatic 

George R. Coraor, January 14 120 


When an excess of G-rignard reagent is heated with pyridine, 
alkyl- and arylpyridines are produced. With phenylmagnesium bromide 
the 2-phenylpyridirre which is formed is accompanied by lesser 
amounts of 2,6-diphenylpyridine. This result indicates that the 
primary addition compound loses KgBrH or its equivalent to yield 
2-phenylpyridine which in turn can react to form the diphenyl- 

Lithium alky Is and aryls react in a similar way but more 
smoothly and at lower temperatures. When phenyllithium is heated 
with pyridine for eight hours in toluene a ^0-50$ yield of 2- 
phenylpyridine is obtained (2). . 

Quinoline has been shown to behave in a surprising manner, 
yielding a diphenylquinoline as well as the expected 2-phenyl- 
quinoline (3). The diphenyl compound (m.p. S6-S7 ) could be pre- 
pared also by the action of phenyllithium on 2-phenylquinoline, 
and it was considered likely that addition of the reagent to 2- 
phenylquinoline had occurred in the l,k manner. The new base was 
not 2, ^-diphenylquinoline (I), however; this compound is known and 
melts at 112°. It differed also from 2 (o-biphenyl) quinoline (II), 
a compound which might be formed by a 1,"?- addition involving the 
phenyl radical. 

C fi H 




The compound was finally shown to be 2,2-diphenyl-l,2-dihydro- 
quinoline (III), arising from 1,2 addition to 2-phenylpyridine. 

This structure was confirmed by the observation that the re- 
action of 2-phenylquinoline with p-tolyllithium gave the same 
product as the condensation of 2-p-tolylquinoline with phenyl- 
lithium, namely, 2-pheny*l-2-p-tolyl-l,2-dihydroquinoline (IV), 

GE. 3 G Q^i^hl 

W" CeH 


C K H 

6 a S 

\V N C 7 H 7 

^6 ri 5Ll 





A few years ago a new type of reaction involving the azo- 
me thine group was discovered by Ernmert and Asendorf (4-) . It had 
long been known that when the sodium derivative of pyridine is 
treated with water 2 i 2'-dipyridyl (V) and ^^'-dipyridyl (VI) are 
formed, the corresponding tetrahydro compounds being intermediates 

+ 2Na 










H a 



There are good reasons for believing that these reactions in- 
volve a free radical which might be written with the unpaired 
electron in the 2 or the H- position . 


— > 




Ernmert and Asendorf conceived the idea that it might be pos- 
sible to cause such radicals to unite with ketyls from ketones, 
producing compounds analogous to pinacols. They were able to pre- 
pare these compounds by treating a mixture of pyridine and a ketone 
with magnesium in the presence of mercuric chloride. From acetone, 
for example, they obtained 2- (2-hydroxyisopropyl) -pyridine (VII) . 





?H 3 

CH 3 



QH 3 

GH 3 



CH 3 



This remarkable synthesis has been modified by Bachman arid 
M cue el (5) who obtained yields of 2S% based on the magnesium. 

-3- ** 

i > 

Similar results were obtained "by Emrnert and Asendorf with methyl 
ethyl ketone, acetophenone and benzophenone. The new reaction was 
applied successfully also to a- and p-picolines. 


1. Bergstrom, Chem. Revs., 3^5, 77 (19W . 

2. Evans and Allen, Org. Syntheses, IS, "JO. 

3. Gilman and Gainer, J. Am. Chem. Soc . , 69, ^77 (19^7). 

4. Emrnert and Asendorf, Ber., J2B, llgg (1939). 

5. Bachman and Micucci, J. Am. Chem. Soc, 22, 23SI (19^3). 

Reported by R. C. Fuson 
September 17, 19 fe 



Since the last Seminar report (l) on the alkaloids of curare 
there have been four papers from the laboratories of Karrer 
(2,3,4) and Wieland (5) dealing with structure studies and with 
the isolation of new alkaloids from calabash curare. Earlier 
studies (l) in Wieland 1 s laboratory had established the presence 
of the following substances in various specimens of calabash curare 


C-Curarine-I C 30 H S1 N 3 C1 

C-Curarine-II C 2G H ss NgGl 

C-Curarine-III CsoH^NsCl 

C-Toxiferine-II C 30 H 33 N 3 C1 

C-Dihydrotoxiferine-I C 30 H 33 N 3 C1 
C-Isodihydrotoxif erine-I C 30 H 33 N 3 C1 

♦The prefix C, an abbreviation for calabash, denotes the 
source of the alkaloid. 

Other related substances, toxiferine-I (C 3oH 33 0N 3 Cl) , toxiferine- 
Ila (C 30 H 3S 3 N 3 Cl) and toxif erine-IIb, along with toxif erine-II, 
had been isolated in Wieland' s laboratory from the bark of 
Strychnos toxif era , a vine known to be employed by some of the 
Indians in the preparation of curare. 

In the recent studies, Karrer and Schmid fractionated 200 g. 
of dry calabash curare and obtained eight new alkaloids: "alkaloid 
A", C 30 H S3 ON 3 C1; "alkaloid B", C 3 ©H 38 N 3 C1; C-calebassine, 
C 3 o H 25°N 3 Cl_, C-toxiferine-I, C 30 H 33 0N 3 Cl, C-calebassinin, 
OigH 33 3 N 3 Cl; C-fluorocurine, C 30 H 33 3 N 3 C1; C-alkalold-UB, 
Ci 9 H 33 3 N 3 Cl(?) ; and C-alkaloid-X. C-Curarine-I is the principal 
alkaloid isolated both by Karrer and by Wieland. C-Toxif erine-I 
is said to be the most active alkaloid known, the limiting toxic 
dose for the dog being about 0.01 mg./kg. (subcutaneous injection). 
Karrer and Schmid believe C-toxiferine-I to be identical with the 
toxiferine-I of Wieland. 

The most extensive structure studies have been carried out on 
C-curarine-I and C-dihydrotoxif erine-I. From the earlier work it 
was believed that one of the nitrogen atoms of C-curarine-I was 
a quaternary ammonium nitrogen. Pyrolysis (300°) at 10"" 4 mm. 
converts the substance to nor- C-curarine-I and methyl chloride. 
The methiodide of nor-C-curarine-I can be converted to the 
chloride and picrate, both of which are identical x«rith the corres- 
ponding salts of C-curarine-I. Thus the quaternary nitrogen atom 
is attached to at least one methyl group. Treatment of C-curarine- 
I with strong bases leads to a dimeric ether (C 40 H 43 0N 4 ) . This 
reaction appears to be that of a quaternary quinoline or iso- 
quinoline nucleus, the quaternary hydroxide changing to a pseudo- 
base and thence to an ether. Comparison of the pK value of nor- 
C-curarine-I with values for various aromatic and hydroaromatic 


I I U - 

v-\/n,h 3 0H 






/^*V X N NCH 3 

1 ! 

quinoline and isoquinoline compounds suggests that a tetrahydroiso- 
quinoline system is present. 

The second nitrogen atom of C-curarine-I is neutral. In zinc 
dust distillations the unmistakable odor of indole is evident. 

Heating of C-dihydrotoxif erine-I with sulfur or zinc produces 
isoquinoline (i) , and the reaction with zinc yields in addition 
substances believed to be skatole (II) and p-ethylin&ole (ill). 










♦ - CH s CH s 



It thus appears likely that this alkaloid contains the ring system 
sho^n in structure IV. 



B I 0"\ 




Cleavage of ring C along either path indicated by the dotted lines 
and dehydroge nation would give rise to isoquinoline and skatole or 


p-e thyl indole . Wieland has postulated that C-dihydrotoxif erine-I 
contains the nucleus shown in structure IV. Karrer independently 
suggested the presence of the tetrahydro-^-carboline system 
(rings A, B, andC in structure IV) in C-toxif erine-I on the basis 
of similarities in color reactions of the alkaloid and known tetra- 
hydro-p-carboline derivatives. The formula of the C-dihydro- 
toxif erine-I corresponds to a methochloride of a dihydro derivative 
of IV. The substance IV has been synthesized (6) in connection 
with studies of yohimbine, but its physiological action has not yet 
been reported. 

Bibl iography 

Jones, Organic Seminar, University of Illinois, April 25, 1947. 
Karrer and Schmid, Helv. Chim. Acta, 29, 1855 (1946). 
Karrer and Schmid, Helv. Chim. Acta, _30_,. 1162 (1947). 
Schmid and Karrer, Helv. Chin. Acta. 30, 2080 (1947). 
Wieland, Witkop and Bahr, Ann., 558,. 144 . (1947). 
Clemo and Swan, J. Chem, Soc, 1946, 617. 

Reported by H. R, Snyder 
September 17, 1948 


Mohlau discovered that phenacylaniline was converted to 2- 
phenyl indole, when exposed to air, or heated with phosphorus 

pentachloride or aniline, 
mained obscure, in spite of 

The mechanism of this reaction has re- 
much investigation 8 and controversy. 

Fischer and Schmidt proposed that phenacylaniline in its 

enol form (la) cyclised to 3-phenyl 
to 2-phenyl indole and offered some 

indole, which then isomer! sed 
experimental evidence. 


V ^>N-Ph 








p \\. NH— \C-H 



According to Bischler , the phenacylaniline in its enol form 
combined with another molecule of aniline to give the diamine (IV), 
which then lost the initial aniline residue to give 2-phenyl indole. 
As proof, phenacylbromide when boiled with para toluidine was shown 
to give 2-phenyl- 5- methyl- indole. 

Bischler 1 s mechanism was generally accepted until it was 
observed 5 ' e that phenacyl aniline on boiling with aniline gave the 
triamine (V) and no diamine of the type (IV) could be isolated. 
Also, phenacylaniline on boiling with o- toluidine was converted 
to phenacyl -o- toluidine and aniline and gave 2-phenyl- 7-methyl- 
indole only in the presence of acid* 

The clue to the mechanism of formation of 2- aryl- indoles was 
obtained from an observation made independently by Julian et al r 
and Stevens and Mcgeoch 8 . o:-Bromo-propiophenone gave 6n treatment 
with aniline not only the expected a-anilino-propiophenone (VII), 
but also the isomeric a-anilino-benzyl ketone (VI II)' . The isomers 
were interconvertible and on boiling with aniline and hydrochloric 
acid gave 2-phenyl- 3-methyl- indole , 

CH 3 CHBr 

C 6 H 5 ~C=0 



■CH^NH.Ph C 6 H 5 -CH-NH'Ph 

C 6 H 5 0=0 


+ CH 3 -C=0 

Ph-NH 3 




C 6 H 5 -CH-B; 
CH 3 — C=0 

CeH B -C-NHPh 

|l Ph.NH 

-CH 3 
'J-CqHs " CH 3 -C-NB'Ph 

C 6 H 5 -C-NHPh 


CH 3 -C-0H 

r..; •,;■ { '■ 



Julian et al concluded that the i some rizat ion of VII to VIII 
may be the first step, followed by enolization, replacement by an 
aniline residue and final cyelization through loss of an aniline 
molecule. They showed that in the conversion of desylaniline to 
2, 3-diphenyl- indole, the intermediate desylanilineanil that should 
be formed according to the above mechanism could be isolated. The 
fact that ct-anilino-propiomesitylene could not be cyclized to 
the corresponding indole was considered further proof that aniline 
addition was an essential step, since in this particular case steric 
hindrance may prevent aniline addition. 


Recent essential findings relevant to the mechanism of 
indolization of phenacylarylamines follow: 

1. A phenacylamine of the type Ph. NH. CHR. COR* where R and R' 
are both aryl groups may have considerable stability in the pure 
state. When heated to moderate temperature in the presence of small 
quantities of acid, it may readily isomerize to the phena.cyl 
compound PhNH.CHR' . • COR and at higher temperatures in the presence 

of acids may then indolise; the indole obtained from either of the 
isomeric phenacylamine s will therefore be 2-R , -3-R- indole formed 
by the cylization of the second and more stable isomer. 

2. If an N-alkyl grout) is inserted in the Phenacylamine s, the 
two resulting isomers Ph.NR" .CHR. COR T and Ph.NR" .CHR* . COR do not 
undergo detectable interconversion and under vigorous conditions 
cyclize directly giving different indoles. 

3. The pure dry hydrobromides of phenacylamines do not yield 
indoles on heating but decompose. 

4. In the case of phenacyl primary amines, aniline hydrobro- 
mide was a more effective catalyst than aniline hydrochloride or N- 
ethyl aniline hydrobromide. 

m 9 

The mechanism postulated by Brown and Mann has been modified 
slightly and is presented below with the reactions of the isomeric 
1-phenyl-p-methyiphenacylaniline and 1-p-tolylphenacylaniline as 
example . 

Ph-(?H-CO-Tol + Ph CH— -CO-Tol Tol-CK-CO-Ph „+ Tol-CH-CO-Ph 



H Jf OH 

Ph Xl'a 


Ph XVIa 



H (±> 




Ph-CH u- 

*8? an 

it* OH 

Ph-CH C— Tol 





H @ ^ OH 



vir OH 




© W 



« r.":i.\Y 



<y\ c-Toi + 

I I + H 

VsNH^ Ph 



A plausible mechanism for the conversion of XI to XVI would 

Tol-CH-C— Ph - 
© NH 

_► Tol-d— C—Ph 





OH <5> 
Tol-6 -#— Ph 



Tol-CO— CH—Ph 




The mechanism clarifies the following experimentally observed facts: 

1. The pure dry phenacylamine hydrobromides do not give 
indoles, because only the cations XIa and XVIa will be ore sent and 
these are clearly inactive k 

2. Aniline hydrochloride the salt of a stronger acid and N- 
ethyl aniline hydrobromide the salt of a stronger base are less 
dissociated and are therefore less effective than aniline hydro- 
bromide as catalysts in this reaction, which depends on proton 

3. With phenacylalkylaniline s, proton addition will fpvour 
the preponderance of the inactive cations of the type XIa and XVIa 
over the active carbonium ions of the type XII and XVII, because 

of the strong basic character of alkylaniline s. Under the vigorous 
conditions required, direct cyclisation precedes isomerization, 
there being very little formation of the isomeric indoles. 


1. Mo'Vlau, Ber., 14, 173 fl&S-l)-; .15, 2480 (1882); 18, 165 (1885). 

2. Wolff, ibid., 21, 124 (1888); 22.428 (1889); Monlau, 25, 2485 
(1892); Hell and Cohen, 37,866"XL904) ; Strain, J. Am.Tihem. Soc. 

J5J., 269 (1929) . 

3. Fischer and Schmidt, Ber, 21, 1071, 1811 (1,888) . 

4. Bischler, ibid., 25, 2868 Tl892) . 

5. Crowther, Mann and Purdie, J. Chem. Soc, 58 (1943). 



6. Verkade and Janet sky, Rec trav. chim., 62, 763, 775 (1943); 
Verkade et al, ibid., 64, 129, 139, 289T1945); 65, 193, 691, 

7. Julian et al, J. Am. Chen. Soc . , 67, 1203 (1945). Julian, 
Abstracts, Fall Meeting of A.C.S., 1933. 

8. McG-eoch and Stevens, J. Chem. Soc, 1032 (1935). 

9» Brown and Mann, J. Chem. Soc, 847, 858 (1948). 

T. R. Gov inda char i 
September 24, 1948 



Probably the most important type of reaction in organic 
chemistry is a displacement reaction of the general form. 

A + BC . -* AB +0 

Depending upon the electronic structure of A there are two 
different processes- whereby this transformation may be effected* 

;A + ;B : C; -> ;AtS J + C: (Electrophilic Attack) 
'.'£'. + *B : C- — > iX*Bl + :C'. (Mucleophilic Attack) 

Examples of Reactions Initiated by Electrophilic Attack: 

l) Carbonium ion reactions 2) Aromatic substitution reactions 

Examples of Reactions Initiated by Nucleophillc Attack ; 

1) liS.' + ;. IVX; -+ HOR + :'^P 

2) R 3 N: + •■.. R\X: -» [R 4 N] + [:X/J^ 

3) CH 3 CdCH: + R -*X; -» CH 3 COCKR + [ : X: ] 


4) (EtOOC)sOT: + CH S :N(CH 3 ) 3 -> (EtOOC) 3 CH-CH 3 + :N(CH 3 ) 3 

1 } 

Routes leading to substitution at carbon : 

In many instances of substitution at a carbon atom it is 
found that if a homologous series is arranged in the order of in- 
creasing chain branching the total rate of reaction passes through 
a minimum: 


Relative rates of reaction with hydroxide ion 

CH 3 Br 2140 

CH 3 CH 3 Br 171 

(CH 3 ),CHBr 5 

(0H 3 ) 3 CBr 1010 

With the first members of the series the rate is proportional to 
the concentration of hydroxide ion and to the concentration of the 
alkyl halide. In the region of the minimum no simple mathematics 
obtains, and with the highly branched members of the series the 
reaction is completely independent of hydroxide ion. 

The most widely accepted explanation for the phenomenon is to 
assume that substitution at a carbon atom can occur by two paths: 



..<£> .. ,. .. .. ■•& 

A) ;A; + ;C - X: -> :A;-.C" + :X; 

rate = k[:A: ] [;C - X] 
Substitution at carbon Nucleus - Bimolecular, kinetically 2nd order 

slow .. . .(£) 

B) ;C -.-X;. - — * -'C G> + :X; 

I ..A: (fast) 
:C.:A-' + :X: 

rate= k [ :c - X] 

Substitution - at carbon _Nucleus - kinetically 1st order . '. si 
Principal Evidence for £L_1 and SJ2,: 

S^2. - l) Kinetics 2) Always accompanied by V/alden Inversion 

Si. - l) Verification of expected kinetic effects upon addition 
-JL of an inert salt. 2) Always accompanied by extensive 

racemir.ation. 3) Hydrolysis of certain compounds is 
rigorously independent of hydroxide ion. (a-phenylethyl 
chloride 6 and t -butyl chloride 7 ). 

Principal objection? to SmI Mechanism: 

l) Some inversion frequently accompanies extensive race- 
mization. 2) The solvated cprbonium ion must have a 
separate momentary planar configuration before it re- 
acts T -rith the solvent shell to form a protonated 
alcohol molecule: 

C(H a O) (CH 3 ) 3 C<£>] -» (CH 3 ) 3 C-0H+ (n-l) H 2 

Interpretation of the S^l Process as a Termolecular Reaction : 

Since S^l reactions usually have been carried out in the 
presence of a large excess of solvating molecules, it has not been 
possible to determine the kinetic order with respect to the 
solvent. Recently, however, Swain has studied a number of solvo- 
lytic reactions in an inert solvent (benzene) with relatively 
small amounts of added tertiary amines, methanol, and phenol. In 
this system it has been found possible to determine the kinetic 
erder of any of the components in the reaction mixture 8 ' 9 * 
Surprisingly enough the reaction was found to be strictly third 
order in all cases. The following equations summarize some 
experiments which seem to be particularly pertinent: 

1) j6 J CCl + CH 3 «H + pyridine -* # 3 COCF 3 t pyridine. HC1 

rate= 1%. [j6 3 CCl] [CH 3 OH] (indep. of pyridine) 


2) jb 3 CCl + CHg^H + phenol 

$ 3 COCH 3 + HC1 + phenol 
(no phenol consumed) 

rate = k 2 [# 9 CCl] [CH 3 OH] [0 OH] (k 3 seven times k x ) 

3) CH-Br + CH 3 OH + pyridine 

ns¥-0H s + Br^ 

rate = k 3 [CH 3 Br] [CH 3 OH] [pyridine] 

4) CH 3 Br + CH 3 0H + phenol 

CHgOCHs + HBr + phenol 

rate = k 4 [CH 3 Br] [CH 3 OH] [j60H] 

In the light of these data 5 , Swain believes that neither a 
simple "push" (S^-2) nor a simple "pull" (%l) is sufficient to 
effect a displacement reaction. He suggests that all displacements 
actually proceed by a termolecular process in which reaction is a 
result of simultaneous nucleophilic and electrophilic attack. For 
the solvolysis of triphenylmethyl chloride, (a reaction which has 
been regarded traditionally as an example of Sjjl), the process 
may be outlined as follows: 

*.n« -C CI H-i 


CH 3 ^ p CH 3 

:0: + JJ-OI + H-o -► 

H^ jl ffy> k 

naclebphilic electrophilic 

reagent reagent 




transition complex 

CH 3 v 

spj — M 

H -"• 4> J> 




PH 3 



solvated carbonium ion 


hydrogen bonded chloride ion 



H^e /6/6 

tH© ♦ 


-:o - 

protonated ether molecule 

According to this point of view phenol enters the rate eauatlon in 
2) and 4) without undergoing reaction because it acts exclusively 
as an aid toward removing the chloride ion by hydrogen bonding. In 
equation 3) methanol is a better acceptor than pyridine. Failure 
of hydroxide ion to accelerate the hydrolysis of certain halldes 
in aqueous solution may be interpreted as due to steric hindrance 
or to the fact that water is already quite adequate to effect the 
"push-pull" operation of the termolecular process. Although hydro- 
xide ion might be expected to perform this operation even more 
efficiently, Swain has pointed out that the amount of hydroxide ion 

-:?£ , 



which can be added to a reaction mixture is small in comparison 
to the concentration of water molecules present. Consequently 
the effect of hydroxyl ion might not be detected. 

General References ? 

1. Hammett, "Physical Organic Chemistry", pp. 131-183. 

2. Hughes, Trans. Faraday Soc, _37, 603 (1941). 

3. Dostrovsky, Hughes, and Ingold, J. Chem. Soc, 1946 , 173. 

4. Hughes, ibid., 1946 , 968 

5. Evans, Trans. Faraday Soc, _42,719 (1946). 

Hydrolysis of Halldes : 

6'. Ward, J. Chem. Soc. , 1927, 446. 

7. Swain and Ross, J. Am. Chem. Soc, jS3, 658 (1946). 

Termoleeulafr S^l Processes 

8. Swain, ibid., 70, 1119 (1948). 

9. Swain, Mechanisms Symposium, Colby Junior College, New London, 
New Hampshire, July 1948. 

E. R. Alexander 
September 24, 1948 



Anhydro he 
and seven membe 
those derivativ 
containing five 
dianhydro hex it 
types are 1,4- a 
various anhydro 
the se compounds 

xitols containing rings of three, four, five, six, 
rs are known. This seminar however is limited to 
es of mannitol (I), sorbitol (II), and iditol (ill) 

membe red anhydro rings. Both monoanhydro and 
ols have been prepared; representatives of these 
nhydromanitol or 1,4-mannitan (IV) and 1,4,3,6- 
ol or isomannide (V) . In the older literature, 

derivatives were described, but the structures of 

were not proved (1,2,3,4). 

CH 3 OH 

CH 3 0H 

CH 3 0H 















CH 3 OH 


CH 3 OH 


CH 3 OH 




QH 3 — 1 





HC — 



CH 3 OH 

CH 3 
— CH 
HC - 


-CH 3 

Methods of preparation of the monoanhydro hexitols : 

1. By heating the hexitol. For example, 1,4-mannitan (IV) 
results from mannitol on dry distillation (4) . 

2. By heating the hexitol in the oresence of acid. Thus 
1,4-sorbitan is obtained when sorbitol is heated in vacuo in the 
presence of sulfuric acid (5). 

3. By reduction of the corresponding anhydro sugar. When 
3, 6-anhydromannose is reduced with sodium-amalgam, 3,6-mannitan, 
identical with (IV), is obtained (6). Likewise 3,6-sorbitan has 
been prepared by the sodium-amalgam reduction (7) and the catalytic 
reduction (8) of 3, 6-anhydroglucose . 

4. By deamination. G-lucamine ( 1-aminosorbitol) is converted 
easily into 1,4-sorbitan by the action of nitrous acid (9). 

5. By heati 
conversion of 1,6 
mannitol" and "2, 
acid has been rep 
shown to be 2, 5-a 
inversion at C 3 ( 
to be 1,4-anhydro 
of such a Walden 
ditosylsorbltol t 
sodium hydroxide 

ng certain derivatives with acid or base. The 
-dibenzoylmannitol into "2, 5-anhydro-l, 6-dibenzoyl- 
4-anhydro-l,6-dibenzoylmannitol" on heating with 
orted (10). However the former product has been 
nhydro-l,6-dibcnzoylsorbitol caused by a V/alden 
11) while the latter product has been demonstrated 
-2 or 3,6-dibenzoylmannitol (12). Another example 
inversion was noted in the conversion of 1,6- 
o 1-tosy 1-2, 5-anhydro-L- iditol by alcoholic 



Methods of preparation of the dianhydro hexitols: 

1. By heating the hexitol or certain derivatives, 
has been obtained from mannitol by dry distillation (4) 
heating 1,6-dichloromannitol -in vacuo (14). 

and by 

2. By heating the hexitol or monoanhydrohexitol with acid. 
Mannitol, sorbitol, and iditol can be dehydrated to isomannide (V) , 
isosorbide (l, 4. 3,6-dianhydrosorbitol) , and isoidide (1,4,3,6- 
dianhydro iditol) respectively by the action of heat and hydro- 
chloric acid (4,8,14,15,16) or heat and sulfuric acid (12,3-5,17,18). 
Isomannide and isosorbide are obtained by heating 1,4-mannitan and 
1,4- or 3,6-sorbitan with acid. 

Less frequently used methods 

of synthesis of the dianhydro hexitols 

3. Raney nickel dehydrogenation and subsequent hydrogenation 
Partially converts isosorbide and isomannide into 1, 4,3,6-dianhydro- 
L- iditol (18). 

4. In addition 1,4-mannitan (IV) may be converted into iso- 
mannide by selective tosylation to give 6-tosyl-l, 4-anhydromannitol 
(Vl) which is then acetylated to 6-tosy 1-2, 3, 5-triacety 1-1, 4-anhy- 
dromannitol (VII) . By treating (VIl) with methanolic sodium 
methoxide, isomannide (v) is formed (14). Similarly isosorbide 
may be synthesized from 1,4- or 3,6-sorbitan (8). 

5. By heating in tctrachloroe thane with a trace of 
p-toluene sulfonic acid, 1,6-dibenzoylsorbitol and 1,6-dibenzoyl- 
mannitol are converted in part into the corresponding 2,5-dibenzcyl- 
isosorbide and 2, 5-dibenzoylisomannidc (10,17). 

CH 3 - 




pCH 3 C 6 H 4 50 3 Cl HOOH 

-> i 

pyridine HO — i 


CH 3 - 

CH 3 C00GH I 

(CH 3 C0) 3 CHgCOOGH I NaOMo 

HC -* 




CH 3 OTs 





I 1 


1 °< 




• CH S 


Properties of the anhydro hexitols : 

1. Ring stability. The hydrofuranol derivatives of the 
hexitols possess ring structures which are extremely stable to base 
being unattached by several hours' heating wi th sodium methoxide. 



They are also stable to mild, dilute acid but are cleaved by strong 
acid. Thus isomannide, heated with fuming hydrochloric acid, is 
converted to 1, 6-diohloromannitol (14), while ring scission of 
isosorbide with hydrochloric acid leads to both 1,6-dichloro sorbitol 
and 6-chloro-l,4-anhydro sorbitol (19). The ring system is stable 
to oxidizing agents such as nitric acid at 100°; the dimethyl 
derivative of isomannide can be recovered unchanged after several 
hours' heating with this reagent (14). 

2. Reactivity of the free hydroxyl groups. The primary 
hydroxyl groups present in the 1,4- and 3,6-anhydro hexitols react 
preferentially in the formation of , say, 6-tosyl-l, 4-mannitan 
(see equation IV - VII above). The secondary hydroxyls of the 
dianhydro hexitols undergo a variety of reactions, such as ether- 
ification, esterification, and replacement (15, 2<\ 21, 22). 

Proof of the structure of isomannide : 

1. According to Hockett, Fletcher, Sheffield, G-oepp, and Soltzberg: 

(a) 1, 4-anhydro-2 or 3,6-dibenzoylmannitol is converted ^n 
heating with acid into dibenzoyldianhydromannitol which can be 
partially hydrolyzed to 1, 4-mannitan with barium hydroxide (10) — 
proves the existence of the 1,4- or 3,6- ring in isomannide. 

(b) Of the 27 possible dianhydro structures, only six possess 
the 1,4- or 3,6- ring, These are: 1,2,3,6-, 1,4,2,3-, 1,4,2,5-, 
1,4,2,6-, 1,4,3,5-, and 1,4,3,6. Since isomannide does not react 
with lead tetrarctate, 1,2,3,6- and 1,4,2,3- are eliminated. Like- 
wise the absence of primary hydroxyls was shown by experiments with 
triphenylmethyl chloride which eliminates the 1,4,2,5- and 1,4,3,5- 
dianhydro structures, 

(c) The 1,5,3,6- structure was synthesized (23) and was not 
identical with isomannide. Therefore the structure must be 1,4,3,6- 
dianhydromannitol . 

2. According to Wiggins (14) : 

(a) Isomannide undergoes ring scission to 1, 6-dichloromannitol 
with fuming hydrochloric acid (pressure). Therefore carbons 1 and 

6 are involved in the rings. 

(b) Lead tetraacetate is without action, indicating no 
adjacent hydroxyls. 

(c) Isomannide treated with thionyl chloride and pyridine 
yields 2,5-dichlorodianhydromannitol. 

(d) Thus the rings must be either 1,4,3,6- or 1,3,4,6-. 


(3) Mannitan can be converted into Isomannide using only 
neutral or alkaline reagents and therefore isomannide contains 
the 1,4,3,6- ring structure. 


1. Berthelot, Ann, chim. (3), 47, 297 (1856). 

2. Vignon, Ann. chim., phye. (577 %> 4& 9 (1871). 

3. Bouchardat, ibid, (5), 6, 101 Tl875). 

4. Fauconnier, Oompt. rendu., 95, 991 (1882); ibid, 100 , 914 (1885) 

Bull. soc. chim., 41, 119*^1884). 

5. Soltzberg, Goepp, ancTFreudenberg, J.A.C.S. , _68, 919, (1946). 

6. Valentin, Collection Czechoslav. Chem. Communications, 8, 35 

(1936)-- (CA J30, 3409 (1936)), 

7. Fischer and Zach, Ber. , 45, 2068 (1912). 

8. Montgomery and Wiggins, J. Chem. Soc. 1946, 390. 

9. Bashford and Wiggins, ibid, 1948 , 299. 

10. Brigl and Grtiner, Ber., 66, 1945 (1933); ibid, 67. 1582 (1934). 

11. Hockett, Zief, and Goepp, J.A.C.S,, 68, 935 (1946), 

12. Hockett, Fletcher, Sheffield, G-oepp, and Soltzberg, ibid, 68 , 

930 (1946). 

13. Vargha and Puskas, Ber., 76> 85 9 (1945). 

14. Wiggins, J. Chem. Soc. 1945, 4. 

15. Montgomery and Wiggins, ibid, 1947 , 433. 

16. Wiggins, ibid, 1947 , 1403. 

17. Hockett. Fletcher, Sheffield, and Goepp, J.A.C.S., J38, 927 


18. Fletcher and Goepp, ibid, 68, 939 (1946). 

19. Montgomery and Wiggins, J. Chem, Soc. 1948 ;, 237. 

20. Gregory and Wiggins, ibid, 1947 , 1405. 

£1. Haworth, Gregory, and Wiggins, ibid, 1946, 488. 

22. Montgomery and Wiggins, ibid,, 1946 , 393. 

23. Hockett and Sheffield, J. A. C. S. , 68, 937 (1946), 

Reported by John Lynde Anderson 
October 1, 1948 



! .-' 



"V .' '■ "- 



I Preparation 

Coffman first reported the preparation of l-cyano-l,3~ 
butadiene , obtained by the action of sodium cyanide on l-chloro-2,5*- 
butadiene (l). He believed that the reaction proceeded in two stepe 
as shown, although the intermediate was not Isolated. 

CH S =C=CHCH 3 C1 + NaCN -3 [pH 3 =C=CIHCK 3 CN] -> CH 3 =CHCH=CKCN 

Coffman established the structure of his new compound by 
cold alkaline hydrolysis to (3-vinylacryllc acid, reduction to n- 
amyl amine, and alkaline permanganate oxidation to oxalic acid. 

Of several patented preparative methods the most general are 
the pyrolysis of esters of crotonaldehyde cyanohydrin (2-4) or of 
the diesters of acetaldol cyanohydrin (5,6). Others are given 

OH 3 =CHCH=CHX (in aqueous + NaCN/ fln \ -> CH 3 =CHCK=CHCN (7) 

&=C1 or Br emulsion) vaq,; 

CH 2 =CHCSCH + HCN -* CH 3 =CHCH=CHCN (8) 

CuCl catalyst 


dehydration catalyst 
NH 3 -» CH 3 =CHCH=CHCN (9) 

400 -500 C 

The most convenient method of preparation (60^ yields) is the 
pyrolysis of the benzoate of crotonaldehyde cyanohydrin (10). 

NaCN,C s H 5 C0Cl QC0C 6 H 5 575° 


-10° + 

C 6 H 5 C00H 

II cis-trans Isomers 

Slight variations in the boiling point and index of 
refraction of products from different runs provided the first 
evidence for the existence of cis and trans forms of l-cyano-1,3- 
butadlene (10). Likewise, variations were noted in reaction rates 
of different samples in copolymerizatlons with butadiene. Frac- 
tionation of the product of pyrolysis produced two isomeric fractions, 
one boiling at 49.5°/31,5 mm, the other at 53,0°/ 31. 5 mm. 

Copolymerization studies with butadiene showed that the 
higher boiling fraction reacted at a faster rate than did the low 
boiling fraction. Also, of the two rubber-like polymers so 


■. r> 


obtained, only the one produced with the high boiling cyano- 
butadiene had a strong odor, previously noted in copolymers made 
with the mixture of cis - and trans 1-cyanobutadiene. 

Ill Diels-Alder Reactions 

To explain the presence in the rubber-like polymer of a by- 
product which was volatile enough to have a strong odor, it was 
suggested that a Diels-Alder reaction may hrve occurred between the 
more reactive high boiling 1-cyanobutadiene and butadiene (10). 
Tests with liquid butadiene permitted the isolation of a Diels-Alder 
adduct with the high boiling fraction, but none wag obtained with 
the low boiling fraction. Similarly, the high boiling fraction 
formed an adduct with maleic anhydride, whereas the low boiling 
fraction did not. 

Reasoning by analogy to els - and trans -piperylene, provisional 
assignment of configuration of the isomeric 1-cyano-l, 3-butadienes 
has been made. It has been shown that only trans- plperylene gives 
Diels-Alder adducts with maleic anhydride or acrylonitrile (II). 
Thus the more reactive higher boiling 1-cyano-l, 3-butadiene is 
assigned the trans configuration. The reduced reactivity of the 
cis -form is explained on the basis of steric hindrance. The shape 
of the cis- molecule is such as to prevent the dienophlle from close 
approach to the ends of the diene system. 


II > 

CH-Cv —* no reaction 




HCT N H 11 ^0 

! + CH-C' -» 

HC ^0 

^CH 3 


Aromatization and subsequent hydrolysis of the cyano group 
from the butadiene and 1-cyano-l, 3-butadiene adduct produced 2-vinyl 
benzamide (12). Thus the adduct must be a 2-cyano-l-vinylcyclo- 
hexene. The position of the cyclohexene double bond cannot be 
assigned on the basis of the above facts. 


* ...» : ** 

■■■ .'■> 

--'.)„ . 

• u : •»,'•;. 


IV Reactions with Active Methylene Compounds 

The value of 1-cyano-l, 3-butadiene ps a preparative inter- 
mediate has been further demonstrated (13,14). A five carbon chain 
may be readily introduced into organic compounds possessing 
sufficiently active hydrogen atoms by a reaction resembling cyano- 
ethylation. In general, the reaction may be represented by the 
following, where the H in RH is an active hydrogen. 


All additions studied were catalyzed by a 38$ aqueous solution of 
trie thy lme thy lammonium hydroxide. Addition was 1,4 across the con- 
jugated diene system in all cases studied. 

CH 3 

1) 1-cyano-l, 3-butadiene + 2-nitropropane —> CH 3 -C-CH 3 CH=CHCH 3 CN (13) 

N0 3 

2) » + nitroethane -» CH 3 -C(CK 3 CH=CHCH 3 CN) 3 (13) 

N0 8 

3) " + nitromethane -* 3 N=C(CH 2 CH=CHCK 3 CN) 3 (13) 

S X NC 2 

4) " + nitrocyclohexane — » < V 

X ./\jH 3 CH=CHCH 3 CN (13) 

5) » + ethyl malonate -* (EtO^C) 3 C(CH 3 CH=CHCH 2 CN) 3 (14) 

Et0 3 C 

6) » + ethyl acetoacetate -> ' )3(CH s CH=CHCH 3 CN) 3 (l4) 

GH 3 C0 

Et0 3 G 
V) " + ethyl cyanoacetate -> C(CH 3 CH=CHCH 3 CN) 3 (14) 


To illustrate oossible applications of the reaction, adducts 
from nitro alkanes were converted to saturated nitro cyanides, 
nltro acids, amino acids, amino cyanides, and diamines, and un- 
saturated amino cyanides and amino acids by appropriate operations 
on the functional groups present (13). 

The adducts 5) to 7) were unstable and lost one molecule of 
1-cyano-l, 3-butadiene when heated (14). However, the Ms- adducts 
could be hydrogenated to give 5-substituted-l,9-dicyano nonanes. 
The mono-adducts can also be modified as in the preceeding paragraphs 

, I- ... ., 

" *3 ' <~. ; ■, • '• ■ 



It was reported that the methylene groups in benzylcyanide, 
acetophenone , and desoxyben 7 oin were insufficiently active to add 
across 1-cyano-l, 3-butadiene (14). 


1. Coffman, J. Am. Chem. Soc, J57, 1981 (1935). * 

2. Treppenhauer, D. R. Patent 673,427; C. A. J53, 4272 (1939). 

3. Gudgeon, Can. Patent 398,840; C. A. 35, 7423 1 (1941). 3 

4. G-udgeon and I.C.I. , Ltd., Brit. Patent 520,272; O.A. 36, 499 

(1942). ; 

5. Gudgeon and Hill, U.S. Patent 2,264,025, C.A. j56, 1622 (1942). 

6. Gudgeon and Hill, Brit. Patent 515,737; O.A. 35, 5916 (l94l). 

7. Carter and Johnson, U. S. Patent 2,276,156; OTA. J36, 4524 3 (1942) 

8. Kurtz and Schwartz, U. S. Patent 2,322,696; C.A. 38, 118 7 (1944). 

9. Hanford, U. S. Patent, 2,334,192; C.A. 38, 2668 7 Cl944). 
10. Snyder, Stewart, and Meyers, J. Am. Chem. Soc, in press. 

UL. Frank, Emmick, and Johnson, J. Am. Chem. Soc., _69, 2313 (1947). 

12. Synder and Poos, J, Am. Chem. Soc, forthcoming publication. 

13. Charlish, Davies, and Rose, J. Chem. Soc, Feb. 1948, p. 227. 

14. Charlish, Davies, and Rose, J. Chem. Soc, Feb. 1948, p. 232. 

Reported by Bruce Englund 
October 1, 1948 



Sodium hydride has recently become available in large 
quantities; preliminary investigations have demonstrated a useful- 
ness that partially overlaps that of metallic sodium or alkali 
alkoxides but is also unique in some important respects. 

I. Properties of pure NaH (l) . 

1. Grey to white, crystalline, free flowing pox^der insoluble 
in inert solvents. o 

2. Infusible (Dissociates into Na and H s at 400-430 G) . 

3. Decomposes readily in damp air. NaH + H 3 — * NaOH + H 3 . 

4. Ionizable. NaH -» Na" 1 " + H (2) . o 

5. Ignition point (in pure dry oxygen) > 230 C; I. P. of Na 
(dry air) = 120° C. 

II. The handling of NaH. Traces of Na may be removed from the 
commercial material by treatment with liquid ammonia (l). For some 
purposes a finer grained material is required; this is readily 
obtained by adding ceramic spheres as an abrasive to the reaction 
flask (3). Waste material may be destroyed in much the same manner 
as Na. 

III. Applications to organic chemistry. 

1. Catalytic reduction of aromatic hydrocarbons (8 references 
in (1)). 

T, p T = 250-300 C 

a. Naphthalene — > Tetralin only 

NaH p = 500-1000 p.s.i. 

b. Other similar polynuclear 

condensed systems contain- T,p Partial reduction 
ing bonds of relatively — » products 
high olefinic character NaH 

In the case of naphthalene, by using NaH in excess of the amount 
required for conversion of any sulfur present to sodium sulfide, it 
is possible to effect hydrogenation and desulfurization simultane- 
ously. It is stated that NaH should be generally useful in this 
manner, but evidence of its use with other compounds is not presented 

2. Polymerization catalyst. NaH has been employed success- 
fully in the polymerization of butadiene (4), crotonaldehyde and 
others, but it has not demonstrated any special value to justify 
its use in preference to other catalysts. 

3. Catalyst for the cyanoethylation reaction (5). 

4. Preparation of sodio derivatives of active H compounds. 

a. CH=CH + NaH liq.' NH 3 CH=CNa + H 3 (l) 

— > 

This reaction produces no acetylene reduction compounds as by- 
products unlike the related reaction with Na. 

> V . ■ 

r T '- 

.Toy.. -■ t 

-; . ': '■-■• - 

: i?fi' 

i ■ ) 



tr. alcohol - + 

b. CH 3 (COOR) 3 + NaH -> [CH(C00R) 3 ] Na + H 3 (l) 

This is the best way in whicb to prepare sodio malonic ester 
free of ethoxide ion} the reaction goes to completion rapidly. 

c. Alkoxides are easily prepared by dropping the alcohol 
on a suspension of the NaH in -&a inert solvent; the reaction is 
not prolonged by deactivating surface effects such as' are noted 
with Na. Certain alcohols susceptible to Na reduction are convert- 
ible to their alkoxides only by using NaH. Ex., furfuryl alcohol 
or eleostearyl alcohol CH 3 (CH 3 ) 3 (CH=CH) 3 (CH 3 ) 7 C00H (l) . 

5. Action on carbonyl compounds. 

a. Introduction. Swamer and Hauser (6) proved both of the 
following prototypical carbonyl reactions to occur with NaH: 

(I) - + 
-CHC- + NaH -» C-OC-3 Na + H 3 

(II) -CC- +NaH -> -0CH- 

The latter type reaction (II) has been observed only with 
aldehydes and ketones containing no oc H-atom. Thus, benzophenone 
on treatment with NaH in boiling xylene and subsequent hydrolysis 
yields benzhydrol. Benzyl benzoate in 92^ yield is obtained from 
benzaldehyde using 0.05 equiv. NaH probably by first forming Na 
benzylate which then acts in the usual manner (7) . Methyl benzoate 
is stable to NaH in boiling xylene . 

The same investigators (6) attempted to prepare ketone anions, 
but they found that either self- condensation or no reaction at all 

b. Ester-ester condensations. In general, the use of 
NaH holds these advantages: 

(1) Simpler equipment and less time for completeness 
of reaction are required than with an alkoxide . 

(2) The preparation of a particular ester does not 
necessitate the use of the corresponding alkoxide. 

(3) NaH may be employed at higher temperatures than 
Na without producing competitive reactions like 
acyloin formation. 

Esters up to and including the Ci 8 acid ester are self- 
condensed in better than 90^ yield, and it is interesting to note 
they were all cleaved in excellent yield to the corresponding 
ketones (8) . Ex. , 

2CH 3 (CH 3 ) 1S C00CH 3 +2NaH -» CH 3 (CH 3 ) 15 CHC00CH 3 -* [CH 8 (CH a ) 18 ] a C0 

C0(CH 3 ) ls CH 3 


; i 




Early work failed to establish that the hydride and not the 
alkoxide formed by reaction with traces of alcohol present was the 
actual catalyst. This was demonstrated by the successful self-con- 
densation of ethyl isovalerate using NaH (6) and the failure to 
effect this reaction with sodium e thy late (9). 

As yet only one ineffectual attempt to condense an ester with 
only one cc H-atom (ethylisobutyrate) has been reported (6). 

Succino succinic ester has been prepared with NaH in a 
Dieckmann type cyclization (3). 

c. Ketone- ester condensations. Sodium hydride has demon- 
strated a particular utility with reactions involving high molecular 
weight compounds. Thus, the following condensation* are best 
carried out using NaH. 

CH 3 (CH 2 ) 8 COOCH 3 + CH 3 COCH 


CH 3 (CH 3 ) 8 C0CH 2 C0CH 3 (l) 


/^V \-COCH 



C 6 H 5 C00CH 3 + 



; -COG 6 H 5 


Sodium hydride has been shown to work very well in the Stobbe 
type condensation involving a ketone plus succinic ester (10). It 
is also recommended in the synthesis of carbethoxy and ethoxalyl 
derivatives by the reaction of ketones with ethyl carbonate and 
ethyl oxalate (11) . 






Hansley and Carlisle, Chem. Eng. News, 23, 1332 (1945) (Review 

article with 35 references). 

Bardwell, J. Am. Chem. Soc, 44, 2499 (1922). 

Green and LaForge, ibid., 70, 2287 (1948). 

Schirmacher and Van Zutphe*nT U.S. P. 1,838,234 (Dec. 29, 1931); 

also, B. P. 315,356 (1929). 
Brusori, U.S. P. 2,227,510 (June 23, 1942). 

Swamer and Hauser, J. Am. Chem. Soc. 68, 2647 (1946). 

Hammett, "Physical Organic Chemistry", p. 352 

Hansley, U.S.P. 2,218,026 (Oct. 15, 1940) . 

Roberts and McElvain, J. Am. Chem. Soc, 59, 2007 (1937). 

10. Daub and Johnson, ibid., 70, 418 (1948). "~~ 

11. Soloway and LaForge, ibidTy 69, 2677 (1947). 

Reported by Melvin I. Kohan 
October 8, 1948 


Amidines are of interest as medicinal s - for example, some 
have recently been found effective against typhus infections. 
Amidines are also used to synthesize a large number of heterocyclic 
comoounds, particularly pyrimidines. 

Several common methods, all of which possess peculiar dis- 
advantages, have been used for the synthesis of amidines (l). They 
may be represented as follows: 

I. HC1 NH NH 3 NH 

RON + ROH -* R-C-OK ' HC1 -* R-C-NH 3 • HC1 

p PC1 5 CI RNH 3 NR 

II. R-C-NHR ~> R-U=NR -» R-C-NHR 


RON + RNH 3 -► R-C-NHR -> R-d-NHR 

Recently, several new methods have been developed for the 
synthesis of amidines. Amidines or mono- N- substituted amidines 
result in good yield from the reaction of alkyl and aryl nitriles 
with ammonium or primary amine salts of sulfonic acids (2). 

+ 180-300° NH 

IV. RCN + [R'S0 3 ][R"NH 3 ] -* R-d-NHR" • R f S0 3 H 

Amidines may also be prepared by merely heating together 
carboxylic acids and sulfonamides (3). The following overall 
equation applies: 

heat NH 

V. RCOOH + 2R*S0 3 NH 3 -> R-b-NH 3 « R r S0 3 H + R T S0 3 H 

It has been proposed that the reaction occurs in five steps: 

1. RCOOH + R' S0 3 NH 3 ^ RC0NH 3 + R' S0 3 H 

2. RC0NH 3 + R'S0 3 H + R T S0 3 NH 3 $1 RC0NHS0 S R' + R T S0,NH 4 


3. RC0NHS0 3 R ! ^ H-fr-OSO-jR 1 

NH _> 

4. R-C-0S0 3 R f *- RCN + R' S0 3 H 

-> IfH 

5. RCN + R' S0 3 NH 4 <— R-C-NH 3 f R' S0 3 H 

The series of steps has been substantiated by the isolation of 
several of the intermediate compounds in good yield. 



Carboxylic acids react with N- substituted sulfonamides to yield 
N,N'-disubstituted amidines (4). 

VI. RCOOH + 2R' SOgNHR' ' -> R-C-NHR" ■ R'g0 3 H + R f S0 3 H 

A series of reaction steps, partially identical to those given for 
reaction V, were proposed to explain the reaction. 

This type of reaction is made more versatile by initially 
preparing the mixed imides which occur above as intermediates. The 
mixed imides are prepared from acid chlorides and N- substituted 
sulfonamides in the presence of base. They may then be reacted 
with ammonium or primary or secondary amine salts of sulfonic acids 
to yield mono-, di- or tri-N- substituted amidines. 

VII. RC0C1 + R 1 SO s NHR" -> RCONR"S0 3 R* 

VIII. RCONR'SOgR' + [R , S0 3 ][NK 3 R'"R ] ->R-C-NR ! "R .R» S0 3 H + R» S0 3 H 

This general method ot synthesis of amidines is very useful 
since amidines of all degrees of substitution can be prepared and in 
which all of the N-substituents can be different if desired. 

Amidines may also be prepared by heating nitriles with ammonium 
thiocyanate or substituted ammonium thiocyanates (5). 


IX. RON -*o R-C-NR'R" (R 1 and R» ' may be H) 


The yields are sometimes quite high but the reaction conditions are 
quite critical. 

Amidines result in fair yield from the decomposition of the 
complexes formed from alkyl or aryl nitriles and aminomagne sium 
halides (6). 

X. R'R"NH '+ C 3 H s MgBr~» R f R"N-MgBr + C 3 H 6 

NMgBr H 3 NH 

XI. R ! R"N-MgBr + RON -> R-C-NR'R" -> R-C-NR f R» 

This reaction fails with the halomagnesium derivatives of primary 
amines and of diary lamines. 


* -f 


2-Substituted-4,5-dihydroglyoxalines, of recent interest as 
medicinals, can be prepared in excellent yields by heating nitriles 
with a sulfonic acid salt of ethylene diamine (7). 

+ 230° NH + 

XII. RON + [R'S03][NH 3 -CH 3 CH 2 -NH 3 ] -» [R-C-NH-CH. 3 CH 2 -NH 3 KH , SOg 

* ^N— CH 3 
R 1 S0 3 NH 4 + R-C J 

N NH~-CH 3 

Bases containing two dihydroglyoxaline nuclei are readily prepared 
from dinitriles. Yields from the reaction are so good that it can 
be used for the identification of nitriles. Tetrahydropyrimidines 
are conveniently prepared from nitriles and the salt of trimethylene 
diamine , 

N,N-Disubstituted amidines may be prepared from nitriles and 
amines, using Friedel-Craf t type catalysts (8). 

A1C1 3 NH 

XIII. RON + R'R"NH -> R-d-NR r R" 


1. Higgins, Org. Seminar, U. of Illinois, Feb. 28, 1947. 

2. Oxley and Short, J. Chem. Soc, 1946 , 147. 

3. ©xley, Partridge, Robson and Short, ibid . , 1946 , 763. 

4. Oxl ey and Short, lb id . , 1947, 382. 

5. Partridge and Short, ibid . , 1947 , 390. 

6. Hullin, Miller and Short, ibid., 1947 , 394. 

7. Oxley and Short, ibid ., 1947 , 497. 

8. Oxley, Partridge and Short, fold . , 1947 , 1110. 

Reported by John B. Campbell 
October 8, 1948 

i I v- "V* 

* * 



In 1862 Strecker (l) observed for the first tine, that 
alloxan reacted with alanine to give acetaldehyde and carbon 
dioxide. The interaction of a-amino acids with carbonyl compounds 
in aqueous solution or in suspension to give aldehydes and ketones 
with one carbon atom less, has been termed the "Strecker Degrad- 

Various workers in this field have studied the degradation 
of a-amino acids and many substances such as dehydroascorbic acid, 
alloxan, and 2-methyl-l, 4-naphthaquinone were found to be effective. 
A detailed investigation on the pcope and limitations was made by 
Schftnberg and co-workers (2) at the Fouad University in Egypt. 
Their results may be summarized as follows. 

1. Nature o f a- Amino Acids :-The two hydrogens on the nitrogen 
atom must be unsubstituted. However the hydrogens on the co-carbon 
atom may be substituted; thus cc-aminoisobutyric acid yields acetone 
when treated with p-benzoquinone (4), or with methyl-glyoxal (5), 
Proline, an a-secondary amino acid does not undergo degradation 
with ninhydrin (3). 

2. Final State of the Amino group of the g- Amino Acid Sub- 
jected to Strecker Degradation ;-The character of the reaction 
products depends upon the nature of the carbonyl compound employed. 

(a) The amino group may be eliminated as ammonia. 





+ R-CH-C0 2 H 


V Vs-CQ 




Jt3-0H + RCHO + NH 3 + CO. 

(b) The amino group may become linked to the carbonyl com- 
pound which affects the degradation converting it into an amino 
compound of similar structure (Transamination). Thus alanine is 
formed when a-aminophenylacetic acid is subjected to degradation 
by the action of pyruvic acid. 

NH 3 


CHO + CH 3 -CH-C00H + C0 3 
NH 3 

(c) The amino group may enter into combination with the 
carbonyl compound used as the degrading agent producing a nitro- 
genous compound of a complicated character. Thus triketoindane , 
when used in the degradation, is transformed into a violet-blue 
imino compound (3) . 


C-ONH 4 
\ /CO 

N ccr N cc 

C 6 H 4v ^C— N— C v \C e H, 

3. Nature of Carbonyl Compounds which bring ^bout the Strecker 
Degradation : -The degradation is not effected by mono cprbonyl com- 
pounds. Experiments with various carbonyl comoounds have shown (2) 
that only those containing the group -CO- (CH=CH) n -CO- are effective. 
However, there are exceptions such as oarabanic acid, s-bibenzoyl- 
ethylene and dibenzoylstilloene which are ineffective in aqueous 
media probably due to their inability to form Schiff's base, which 
is an essential step in the degradation. 

4. Reaction Mechanism of Strecker Degradation :-G-rassmann and 
Arnim (3) suggested the following scheme for the degradation. 

/CO^ CO. H 

R-CH-C0 3 H + C 6 H 4 CO -» R-C-C0 3 H + C 6 H 4 X A 

NH 3 ^CCT ' NH X C0 ^OK 

R-C-C0 3 H H 3 RCHO + NH 3 + C0 3 

NH ~» 

The triketoindane functions only as a dehydrogenating agent. 
However the degradation has been brought about by such weak dehy- 
drogenating agents as anthraquinone and not by the stronger ones 
such as azobenzene or methylene blue. 

Schonberg and co-workers (2) advance another mechanism. 

R H 

-CO R -H 3 -C--=N-C-CO a H -C-N=C-H H 3 -C-NH 3 

I + H 3 N-g-C0 3 H -» J H fr* I! -> || + RCHO 

-CO H -CO -C-OH -C-OH 

This scheme explains the necessity of the two hydrogen atoms 
on the nitrogen atom, as well an the formation of aldehydes from 
non-acidic substances. Benzaldehyde is formed from benzylamine by 
the action of alloxan or isatin. 

-CO -CO -COH H 3 -COH 

+ PhCH 3 NH 3 ~> | ~» II -» |l + PhOHO 

-CO -C=N-CH 3 Fh -C-N=CHPh -C-NH 3 

NH 3 C0 2 H (5o 3 H C0 3 H C0 3 H 



Schttnberg's scheme appears to be more generalized than that 
suggested by Herbst (6) for the interaction of a ketonic acid and 
an a-amino-acid. 

-H 2 
RC0-C0 3 H + R'qH-C0 3 H -» RC=N-CHR ! -+ R^H-N^R' 

+H 3 

-» R-CH-COOH + R'CHO + C0 3 

NH 3 

This involves first the condensation of the carbonyl group of 
the ketonic acid with the amino group, followed by the migration 
of a hydrogen from the a-carbon of the amino acid to the ^-carbon 
of the ketonic acid. Subsequently decarboxylation takes place, an 
aldehyde is split off and a new amino acid is obtained. 


1. Strecker, Annalen, 123, 363 (1862). 

2. A. Schonberg, R. Moubasher and Akila Mostafa (Mrs. Said), 
J. Chem. Soc, 1948, 176. 

3. G-rassmann and Arnirn, Annalen, 509, 288 (1934). 

4. Langenbeck, Ber. , 61, 942 (192877 

5. Neuberg and Kobel, Bio chem, Z., 188 , 197 (1927). 

6. Herbst and Engel, J. Biol. Chem., 107, 505 (1934). 

Reported by A. S. Nagarkatti 
October 15, 1948. 


'.-.Or r ' 



Since the Seminar report (l) and Steinkopf's treatise (2) on 
the chemistry of thiophene, a commercial process for the prepara- 
tion of this compound has been developed (3,4) which has made it 
available in large quantities for the first time. The resultant 
renewal of interest in thiophene has been marked both in the fields 
of organic and biological chemistry. In the former, at least one 
important point has been established. Thiophene is more properly 
considered an anolog of phenol rather than benzene (5;. This dis- 
cussion will be limited to four types of organic reactions which 
best illustrate this fact. 


The alkylation of thiophene with olefins was not reported 
until the last two years. Since then a wide selection of olefins 
has been used as alkylating agents in the presence of such catalysts 

Activated silica-alumina type clays (6) 
Phosphoric acid on kieselguhr (?) 
Sulfuric acid of 70-96$ concentration (8) 
Boron 'fluoride complexes (8) 
Anhydrous aluminum chloride and others (8) 

The tardy discovery of this reaction can only be explained by the 
assumption that prior attempts failed because of the selection of 
alkylating conditions based on the old benzene- thiophene analogy. 

The following general precepts may be found useful in the 
sale ctrUm of optimum conditions for the alkylation of thiophene. 

1. Alkylation with reactive olefins, Buch as isobutylene, is 
catalyzed best by sulfuric acid of 70-80$ concentration, boron 
fluoride ether complex, or other mild catalysts at temperatures of 
the order of 70-80°. Alkylation with the less reactive straight- 
chain olefins, such as 1-octene or 1-hexadecene, requires active 
catalysts, such as concentrated sulfuric acid (combined with the 
olefin first) or boron fluoride water complex. 

2. Sulfuric acid or dihydroxyfluoboric acid generally give 
products rich in monoalkylthiophene; boron fluoride complexes give 
products rich in dialkylthiophenes. 

3. The preparation of the a- isomers is favored by mild re- 
action conditions and short reaction times. Considerable quantitie,< 
of the p-isomers occur at elevated temperatures in the presence of 
strong catalysts. This was noted by Appleby and coworkers (7), 

who used phosphoric acid on kieselguhr as the catalyst at 270°. 


Although thiophene like benzene can be acylated with acyl- 
halides in the presence of mole equivalents of metal halides (9) y 

•- .v. 



more economic methods for the synthesis of low molecular weight 
thienylketones have recently been reported (10-14). It was found 
that thiophene can be acylated with acid anhydrides in the presence 
°^ catalytic amounts of anhydrous zinc chloride, iodine, or hydr- 
iodic acid; or by passing the reagents over activated clays or 
ortho-phosphoric acid. 

Aluminum chloride and stannic chloride in catalytic amounts 
do not promote acylation of thiophene and higher concentrations 
of zinc chloride tend to decrease the yield. It is reasonable to 
assume that the zinc chloride does not form the complex usually 
associated with acylation reactions catalyzed by metal halides. 
Benzene, phenol, and resorcinol do not acylate this way. 

The mechanism by which traces of iodine or hydriodic acid 
catalyze the acylation has not been elucidated, Iodic anhydride, 
bromine, hydrochloric and hydrobromic acids fail as catalysts. 

Since acyl halides proved to be less efficient acylating 
agents with these catalysts, these methods are limited practically 
to the use of available anhydrides. However, a good method for the 
preparation of higher molecular weight thienylketones was reported 
(15; in which molecular equivalents of phosphorus pentoxide were 
used to promote the acylation of thiophene with organic acids. 
Benzene was used as the solvent since it is completely inert in 
this reaction. The yields, in general, increase with increasing 
molecular weight of the acid employed. With oleic acid, a 42fc 
yield of an acylated, unidentified dimer is obtained in addition 
a bbft yield of 2- (^ 9, 10-octadecenoyl)- thiophene (J.) . Strangely, 
23fo yield of 2,5-didecanoylthiophene (il) is obtained with decanoic 
acid in addition to a 42% yield of the mono-derivative. This does 
not occur with the lower molecular weight derivatives. 2,5-Diaceto 
thiophene has been isolated but in less than 5^ yields. 


/ \-d-(CH 3 ) 7 CH=CH(CH 2 ) 7 CH : 

q s p 

CH 3 (CH 3 ) e 3-/ \-8(CH a ) B CH 
I I 



In recent studies (16-18) thiophene and its homologs were 
found to undergo tran smetallation with sodium amalgam and an alkyl 
or aryl halide as shown in reaction I. Subsequent treatment of the 
thienyl sodium with carbon dioxide or ethylene oxide offers an 
excellent method for the preparation of alkylthienyl carboxylic 
acids or ethanols, respectively. 

\ + Na(Hg) + R'X 




(CH 3 ^0 




84/ y 
the f 
II wa 


The products obtained in the metalation of 2-chlorothJ or>hene 
found to be dependent on the solvent employed. In benzene, an 
ield of thienyl sodium is obtained; in anisole or butyl ether 

is no metalation of the thiophene. Of particular interest is 
act that high yields of 5- chloro- 2- thienyl sodium are obtained 

diethyl ether as the solvent. The mechanism shown in reaction 
s suggested to account for the product obtained. Neither 
urn nor potassium behave in this ma nner. 

4 — } 

S H 


1/2H 3 + XV \ 


Aminomethylatlon (Mannich ) 



X- / \ Na 


Recently it has been found (19-21) that thiophene and its 
homologs possess hydrogens of sufficient reactivity to undergo a 
type of Mannich reaction in the presence of formaldehyde and ammoniurr 
chloride. The products after neutralization consist of a mixture 
of primary (III) and secondary (IV) amines, together with a con- 
siderable quantity of sub-resinous amines of unknown structure. No 
tertiary amines were isolated unless one of the ct-positions on the 
thiophene was blocked (v ). 
■ SL / .8. \ /' s 


-CH 3 NH 3 (40?) 


\-CH 3 \ 3 NH (20fc) /r/ \-CH 


Analogous products are obtained with hydroxy lamine hydrochlo- 
ride. Depending on the conditions us ed substantial yields of 
2- thenylhydroxy lamine (VI), di- (2- the nyl) -hydroxy lamine (VII), or 
di(5-hydroxymethyi-2- thenyl)-hydroxylamine (VIIl) can be isolated. 

/ \-CH 3 NH0H (80/) 


NOH (50?0 

/ \CH 

a NOH 


Thiophene does not enter into the Mannich reactions with alkyl- 
amine hydrochlldes, probably because of side reaction. 

.. I ,.> 

■ J 




1. Trumbull, Organic Seminar, Univ. of Illinois, May 8, 1946. 

2. Steinkopf, "Die Chemie des Thiophens," Dresden and Leipzig, 1941. 

3. Rasmussen, Hansford and Sachanen, Ind. Eng. Chem._38, 376 (1946). 

4. Rasmussen and Ray, Chem. Inds. 60, 583,620 (1947). 

5. Caesar and Sachanen, Ind. ^ng. Chem. _40, 922 (1948). 

6. Kutz and Corson, J. Am. Chem. Soc, 68, 1477 (1946). 

7.. Appleby and coworkers, J. Am. Chem. Soc, .70, 1552 (1948). 

8. Caesar, Paper accepted for publication in J. Am. Chem. Soc, 

9. Campaigne and Diedrich, J. Am. Chem. Soc, J70, 391 (1948) . 

10. Hartough and Kosak, ibid. 68, 2639 (1946). 

11. Hartough and Rosak, ibid. _69, 1012, 1014 (1947). 

12. Hartough and Kosak, ibid. £9, 3093 (1947). 

13. Hartough and Kosak, ibid, 70, 867 (1948). 

14. Heid and Levine, J. Org. Chem. 13, 409 (1948). 

15. Hartough and Kosak, J. Am. Chem. Soc, 69, 3098 (1947). 

16. Schick and Hartough, ibid. 70, 286 (194FJ. 

17. Schick and Hartough, ibid. 70, 1645 (1948). 

18. Schick and Hartough, ibid, 70, 1646 (1948). 

19. Lukasiewicz and Murray, ibid, .68, 1389 (1946). 

20. Hartough, Lukasiewicz and Murray, ibid. .70, 1146 (1948). 

21. Hartough, ibid. 69, 1355 (1947). 

Reported by P. D. Caesar 
October 15, 1948 



I P re p aration: In 1933 Paul (2) first prepared 2,3-dihydro-l 4- 

pyran (or 3, 4-dihydro-l, 2-pyran) in 44? yield by passing tetra- 
hydrofurfuryl alcohol over A1 S 3 at 370-80°; A thorough study o? 

righ'ares-fof (l°2). haS imPr ° Ved ^ yi6ldS f~™ ^pf (SHo^s 

Paul (2) proved the structure of dihydropyran as follows: 

opnnoi'n. The position of the oxygen was verified by the reaction 
sequence • 

H 3 2HBr 

dihydropyran -+ tetrahydropyran ~- 1 ; 5 dibromopentane 




Any of the following structures would give 2- methyl- N-phenylpy rro- 

tt nu „ TT „ lidine under similar treatment. 

H s CH 3 CH 3 CH 

II or I I 
CH a C=CH 3 CH 3 C-CH 3 


,^^ Q 2 ' *. TW ° ? tructural isomers are possible for dihydropyran with 

d » ^l^t ° f th . e d ° Uble bond ' Partial hydrolysis with 
A? -? 3 %\ ^ tGm P era ture to an aldehyde and the instability 

dfhv^n d i lb S° mide addUCt are results consistent only with the 34- 
svn?h^T^?"?r an ^ trUCturee Recen tly Paul and Tchelitcheff (io) 
synthesized the other isomer; 5, 6-dihydro-l, 2-pyran and found their 
chemical reactivities distinctly different. ' 

11 Rlng Nuclear ReaotWg (Reactions of Double Bond) 

A. Hydrogg nation : Dihydropyran may be easily reduced catalyt- 
ically (10,14) to tetrahydro -pyran. 

B. Hal^geji^tion: Dihydropyan may be halogenated with Br 3 

LX'^ ° r Wlth C h in CCl 4 (9) to thecorresponding 2,3-dihalo- 
tetrahyd ropy ran. The relatively unstable dibromide upon 
lntp^ at1 ^ deh ^ r °halogenates to 1, 5-epoxy-2-bromo-l~ 
peniene. The analagous chloro compound is obtained by 
distilling the stable dichlorotetrahydropyran with diethyl- 

C Addition of HBr; Dry hydrogen bromide adds to dihydropyran 

resimiies on standing. 
D. ,%-Alkyl o r Aryl Derivatives via Grignard Reage nts! The 

addition of R24gX to 2-bromotetrahydropyran may be effected 
at low temperatures (6,8). Further treatment of the result- 
ing product with HBr under suitable conditions produces the 
corresponding 1, 5-dibromoparaf fin. Under similar conditions 
nVLn ^° m ° f? d d , ichlor °-tetrahydropyrans also undergo 
alkylation with slightly lower yields. Only the 2-halogen 
of the ring is replaced* b 



E. 2-Alkoxy Derivatives : In the presence of a trace of acid, 
alcohols and phenols add readily to dihydropyran, to pro- 
duce the acetals, 2-alkyoxy or aryloxy tetrahydropyrans 
(15). Dihalo tetrahydropyrans, when treated with an alcohol 
in the presence of the corresponding sodium alkoxide, yield 
the ?-alkoxy-3-halo tetrahydropyrans (9, 14). 

Ill Ring Opening Reactions (involving Oatom) 

A. . Hydrolysis : Dihydropyran is hydrolyzed to 5-hydroxypentanal 
with dilute HC1 (3,12,13,17). The 5-hydroxypentanal exists 
in solution almost completely in the form of the cyclic 
hemiacetal, 2- hydroxy tetrahydropyran (12) . Reduction of 5- 
hydroxypentanal by a variety of methods (3,12,13) produces 
1, 5-pentanediol in excellent yields. 

Amino alcohols are produced by reductive amination of 
5-hydroxypentanal with liquid ammonia or the appropriate 
amine using Raney nickel and hydrogen under pressure (13) 
The side chain of the antimalarial SN 13,276 [8- (5-isopropyl 
aminoamylamino)-6-methoxyquinoline] was synthesized using a 
similar method (l). 

B. 2,4-Pentadienal: Woods and Sanders (14) dehydrohalogenated 
2-ethoxy-3-bromotetrahydropyran with alcoholic KOH to form 
2 ethoxy-ZNs 3 - dihydropyran. The latter upon acid hydrolysis 
yield ed a polymeric material rather than the desired 5- 
hydroxy- A 3 -pentenal. The 2, of 
this aldehyde, however, could be isolated by hydrollzing 
in the reagent as a solvent. Steam distillation of the 
acid solution obtained from the H 3 P0 4 hydrolysis of 2- 
ethoxy A 3 dihydropyran produced a compound which proved to 
be 2,4 pentadienal (14,16). This product undergoes Diels- 
Alder addition reactions (16) either as the diene or 




^CH 3 




II >o 


reflux /CH^ 
-> CH v CH- 6^ 
in toluene || 
CH CH-C'' 
x CH 3 *0 




CH 3 
' ^C-CH 3 

^C-CH 3 
CH 3 

H 2 C=CH-CH X 3 ^C-CH 3 
sealed tube g II 

-* 0=C-CH C-CH 3 
150° ^CH^ 

ft I- 



2,4-Pentadienal + C fl ,H B MgBr 

The reaction of phenyl magnesium bromide with 2, 4-pentadien&l 
did not product the alcohol expected by 1£2 addition (16). The 
alcohol produced is unstable in air, and cinnamaldehyde can be 
isolated from the decomposed mixture. Reduction of the alcohol 
with Raney nickel and hydrogen gave 5-phenyl-l-pentanol. The 
reaction is presumed to proceed as follows: 

CH 3 =CH-CH=CHCH0 CsH5 il g r CH 3 =CHCH=CHC (OH) C 6 H 5 -» tar + C e H e CH=CHCK 

I Double Ally lie 

Ni * shift of (OH) 

C 6 H s (CH 3 ) 4 GH 2 OH «— C 6 H 5 CH=CH-0H=CHCH 3 OH 

H 3 

Woods and Schwartzman (18) produced 1, 3,5-hexatriene by the follow- 
ing reactions: 

CH 3 MgBr A1 2 3 

CH 3 =0H-CH=CHCHO -> CH 3 =CH-CH=CH-CH(0H) CH 3 -» 

CH 8 = CH- CH= CH- CH= CH 3 

The cis isomer is the open chain analog of benzene. 

IV Miscellaneous Reactions 

A* Pyrolysis : Pyrolysis of 3,4-dihydro-l, 2-pyran ih presence of 
equal parts of Al 2 3 and Si0 3 as catalysts (10, 2o) cleaves it into 
ethylene and acrolein. Pyrolysis of 5, 6-dihydro-l, 2-pyran produces 
formaldehyde and butadiene (10) . 

B. Dihydropyran, when passed over A1 3 3 at 400 in a stream of 
H 3 S (19), is converted in &0% yield to dihydrothiapyran. 


1. Drake et al. J. Am. Chem. Soc. , 68, 1529 (1946). 

2. Paul, Bull. soc. chim, (4) _53, 1489 (1933); 3. ibid., (5) 1, 

971 (1934); 4, ibid., (5) 1, 1397 (1934); 5. ibid-, (5) 2, 
745 (1935); 6. ibid., (5) 2, 311 (1935); 7. Oompt. rend., 198, 
375 (1934); 8. ibid., 198, 1246 (1934); 9. ibid., 218, 122 (194< 

10. Paul and Tchelitcheff ibid., J224, 1722 (1947). 

11. Sawyer and Andrus "Org. Syn." J23, 25 (1943). 

12. Schniepp and G-eller, J. Am. Chem. Soc, j38, 1646 (1946). 

13. Woods et al., ibid., 68, 2111 (1946);. 14. ibid., 68, 2483 (1946); 
15. ibid., 69, 2246 (1947); 16. ibid., 69, 2926 (1947); 17. 
"Org. Syn. 27, 43 (1947); 18. Abstracts, Chicago Meeting, A.C.S. 
April (1948TT 

19. Yuriev, Dubrovina, Tregubov, J. den. Chem. U.S.S.R., 16, 850, 

20. Wilson Nature, 157, 846 (1946). 

Reported by Sidney Baldwin 
October 22, 1948 


Two mechanisms for nucleophilic replacement reactions at a 
saturated carbon atom are currently recognized (1,2), One is the 
now familiar (3) bimolecular, S N 2, substitution with complete 
Walden inversion. 

The second mechanism has been termed unimolecular, 3^1 (1,2,4). 
It seems to consist of at least two steps, the most probable rate- 
determining step being an ionization. 

Ionization to an ion-pair, solvated in a way characteristic 
of ions, may be thought to be the rate-determining step in the @L1 
mechanism. Solvation of the ions makes this step feasible:, there- 
fore, the rate varies with the arrangement of solvent molecules 
around what is to be the ion-oair. Solvent molecules must be in- 
cluded in the transition state, without, however, drawing bonds 
between the solvent molecules and the carbonium ion (5). If the 
carbonium ion is very reactive it will react preferentially with 
a molecule in the solvation cluster to give inversion as the major 
steric result (l). If the reaction of the carbonium ion takes 
place after dissociation of the ion-pair, complete racemization 
is the steric result (4). 

To understand the rates and steric results of nucleophilic 
replacement reactions of the most complex compounds it is necessary 
to demonstrate and understand the effects of substituent groups 
other than their supply or withdrawal of electrons to the seat of 
substitution by induction and resonance (l)„ One of the most 
interesting effects is that of participation of a group on a 
neighboring carbon atom in a replacement process at a carbon atom. 
Thus, a replacement reaction might really consist of two steps, 
the first one an intramolecular S 2 reaction, the second the open- 
ing of a ring. Two inversions or^apparent retention will be the 
steric result. This is symbolized (6a) below, Y. and Z indicating 
the leaving and entering groups, respectively. 

A >A 

-Y A .+ 

(a) >&•£—$% -+ ytjr^vs -* >V"- £o 

(11) (iv) 

Participation of neighboring groups in displacement reactions 
has long been known with such groups as 0~ (from OH) and MH 3 , 
prior ring closure (7) to isolable oxide or imine occurring on 
attempted displacement of halide in a halohydrin or aminohallde. 

/ H 
(B) >Ctj—C< + OH" -> >C- — C< + HOH 

P % fast P 



/0 X 
Cp C„ + CI 

.' .• * : .' "• '.": :• > •:.'' , 

■■■/■\ ?'• . . ; ; ' 

'") i 

;n ;• /•■• 



In these cases the groups are those which also take part in known 
bimolecular, Sm2, displacements (2) symbolized in equations (C) and 






ROR' + X 





— > 


(RNH 2 R» ) + 

Of equal interest is the participation in nucleophillc displacement 
reactions by such neighboring groups as OAc, Br and OCH 3 . Winstein 
and co-workers (6) have accumulated a great deal of data which seems 
to indicate that these groups, when bonded to a neighboring carbon 
atom, do participate in the reaction by the formation of an inter- 
mediate of type (II), (see equation A). 

In this connection, rate measurements lead to an understanding 
of the rate-determining ionization step. This may be a one- stage 
ring closure to the cyclic intermediate (il) with Walden inversion 
(W.I.) at CL, an ionization to the substituted carbonium ion (ill), 
or both (6a;. 

(W.I.) A 


^ _^s + - 

(E) >d rt -*„ ?c r~ G a 

(I) (ii) 

(F) f, A 


k c g + 

>C—- C* -» >d! C< 

p ; a -y p a 

(I) (III) 

There are 
(III), is 
steric re 
ant inversion. 

Winstein and co-workers (6) observed the expected Insensivity 
of solvolysis rate to changes in structure, solvent and departing 
groups in the si reaction through intermediate (III). However, in 
dealing with some compounds which were shown to have a first order 
rate constant, the solvolysis rate was found to vary widely with 
structure, and this was attributed to the operation of both 
mechanisms, (E) and (F). 

These trends are exactly those predicted by a qualitative 
theory which recognizes that alpha substitution stabilizes (II) 
and, to a greater extent, (ill), and that beta substitution stabi- 
lizes (il). Thus, beta substitution increases k&and k^/k 

IT. »■ 

i '. ■ ) . 



vnile alpha substitution increases both k A and k n and decreases 

*a /k ~ A 

An alternative mechanism was considered in which mechanism 
(F) was the sole one, the open carbonium ion always being formed 
in the rate-determining step. In this interpretation, deviation! 
of the solvolysis rate from the calculated values, k /k h , would 
be due to polarization of the neighboring groups, an8 to steric 
effects Guch as envisioned by Brown (8) and termed w B"-strain. 
However, a steric effect of the kind postulated appears to be too 
small to be of sufficient importance in calculating deviations from 
the calculated solvolysis rate. In addition, polarization of a 
neighboring group does not predict the contrast between CI on the 
one hand and Br and I on the other. With CI as the neighboring 
group, the solvolysis rate is not sensitive to structure and the 
reaction proceeds by mechanism (F); with Br and I it is, the re- 
action proceding predominantly by mechanism (e). 

The qualitative theory explains the trends observed in rates 
of closure of the ethylene oxide ring from variously substituted 
ehtylene chlorohydrins . Similarly, it is in accord with the 
favorable effect of alpha methyl substitution for closure of the 
ethylene imine (9) and the beta -lactone ring. 

St ero chemistry and Products : — To control steric results of 
mucleophilic displacement reactions it is often necessary to 
assess the relative tendencies for bimolecular displacement with 
inversion, or for unimolecular type displacement which csn lead 
to various steric results of which, perhaps, the most interesting 
is retention of configuration in the presence of a suitable neigh- 
boring group. For this purpose, a knowledge of .the rates in re- 
actions of the unimolecular type of the substituted compounds is 
essential, and toward this end Einstein's rate work (6) so far 
reported is useful. 

Correlating the present work on rates of unimolecular sol- 
volysis with previous (6e,f,g,h) x^ork on the stereochemistry and 
products of such reactions, it becomes clear that most of the pre- 
vious work has dealt with cases in which the rate-determining re- 
action step was the type (e) . Inversion of configuration to form 
intermediate (il), followed by a second inversion thereby convert- 
ing (il) to product (see reaction A), accounts for the clean-cut 
retention of configuration. 

For the situation where the rate work indicates the ionization 
is at least partly by mechanism (F), there is little information 
on products and stereochemistry. In the cases where the rate- 
determing step is (F), the rate constants refer to the rate of 
formation of open carbonium ion (ill) and do not yield information 
on the reaction paths (ill) follows. The ion (III) may close to 
(II), it may coordinate with reagent or solvent (Z) to give 
product (IV), or it may rearrange either to new ion (V) or in 
other ways, to mention some of the possibilities (6m). 

, " - 

, • r; 

' ': r>f 

'as- t .»•• 


• -. . <r\y\ 

t v - 'i v " 

*.- '■■■ . \ hi 











; C a < 




Also, the sterochemical results may be controlled by restriction 

of rotaticn around the C Q C bond. 

p cc 





Bateman, Church, Hughes, Ingold and Taher, J. Chem. Soc, 979 


Hammett, "Physical Organic Chemistry," McGraw-Hill, N.Y., N.Y., 

1940, Chapters V and VI. 

Olson, J. Chem. Phys., 1, 418 (1933). 

Beste and Hammett, J. Am. Chem. Soc, 62, 2481 (1940) . 

Winstein, J. Am. Chem. Soc, 61, 1635 71939) . 

Winstein and Co-workers, J. Am. Chem. Soc, j34, 2780 (1942). 














, 64 

, 2787 ( 

, 64 

, 2791 ( 

, 64 

, 2792 ( 

, 6l' 

, 2796 ( 

, 65' 

, 613 ( 

, 65 

, 2196 ( 

, 6s; 

, 119 ( 

, 70 

, 812 i 

, 70' 

, 816 ( 

, 70* 

, 821 ( 

, 70' 

, 828 ( 

, 70' 

, 838 ( 

, To' 

, 841 ( 

Braun and Weissbach, 



3052 1930) . 
Brown, Bartholomay and Taylor, J. Am. Chem. Soc, 66 , 
C. H. Young, Organic Seminar Abstract, Univ. of Illinois 
March 1948. 


, 19 

Reported by R. w. 
October 22, 1948 




The selenium dehydrogenation of yohimbine (I) gives two bases, 
yobyrlne (II), and tetrahydroisoyobyrine fill), and ketoyobyrine, 
a neutral substance having the formula C 30 H 1S 0N 3 . The structures 
of yobyrine and tetrahydroisoyobyrine, and of yohimbine itself, 
have been established beyond question c (1,2) 




H a C. 


W N 




Scholz (3), in 1935, proposed the structure (TV), for ketoyo- 
byrine. The facts that ketoyobyrine is optically inactive, that 
it is the product of a drastic dehydrogenation, and in particular 
that it has no basic properties, are incompatible with that formula. 
Also, ketoyobyrine is cleaved smoothly by amyl alcoholic potassium 
hydroxide to morharmane and hemellitylic acid. 

This cleavage was used as a basis for formula (V), proposed 
by Witkop (2) in 1943. However, this formula also can not be re- 
conciled with the basic character of ketoyobyrine. In addition. 
Raymond-Hamet (4) has made Uitkop's formula doubtful on spectro- 
scopic grounds. 






This year, almost simultaneously, Schlittcr and Spei 
and Woodward and '-itkop (6) proposed a new structure (Vl) 
yobyrine. Both groups deduced the formula from that of y 
on the basis of the following considerations: 
is heated with selenium, lass of the hydroxyl 
drat ion may be followed to some extent by the 
ring E; (b) the resulting intermediate Cvil) , 
should be subject to ready reduction r.leavage 
to give (VITl); (o) by rotation through 180° about the 0*1 
bond, (VIII) is in a position to undergo lactamlzation to 
fd) selenium may effect the further dehydrogenation of th 
isoquinolone (IX) to (VI). 

(a) when y 
group throu 
as a benzyl 
between N«4 

tel (5) 

for keto- 
gh dehy- 
tion of 

and C-21 

e dihydro- 






1 4 





The cleavage of ketoyobyrine by amyl alcoholic potassium hydrox- 
ide may be readily explained on the basis of this newly proposed 
According to Woodward and Witkip, the opening of the 
is followed by the migration of the^' 14 double bond 
three prototropic shifts. The resulting dihydropyridine 
then suffers loss of the side chain, giving norharmane 

structure . 
amide link 
to^ 5 ' 6 by 

and °,, -3-dimethylbenzoic acid, 


14 3 

i I I 

C*- T 3 -C=N 

14 3 

I I 

3 5 

5 6 

Schlitter and Speitel (5) and Julian, et. al. (7) have reported, 
independently, the synthesis of ketoyobyrine .""' The methods of 
synthesis were almost identical. 



-CH s CH 2 NH a 


< / /j>~CH 2 COOH 1^0° 


lv >-C00H 3 hoiTs 

CH P N : 

3 iJ 2 



H 3 C 


C0 a CHg 




Comparison of the ultraviolet absorption spectrum of synthetic 
ketoyobyrine with that of the product of natural origin showed the 
two to be identical. 


1. Scholz, Helv. Chim. Acta., 18, 923 (1935). 

2. Witkop, Ann., 554, 83 (1943TT cf. Clemo and Swan, J. Chem. Soo. , 

617 (1946); Julian, et al., J. Am. Chem. Soc, 2!2> 180 (1948). 

3. Scholz, Helv. Chim. Acta., 16, 1343 (1933). 

4. Raymond- Hamet, Compt, rend., 226 , 137 (1948), 

5. Schlitter and Speitel, Helv, Chim. Acta., 31, 1199 (1948). 

6. Woodward and ^itkop, J. Am. Chem. Soc, 70, 2409 (1948). 

7. Julian, Karpel, Magnani, and Meyer, J. Am. Chem. Soc,, 70, 2834 

Reported by William E. G-oode 
October 22, 1948 



With the recent interest in the chemistry of acetylene com- 
pounds, much is being done with additions to the ethynylcarbonyl 
system, -C=C~CC-, (1-8). . 

In general, the laboratory preparation of acetylenic ketones 
has utilized the following types of reactions. 

1. R-C=C-Na + R ? -C0C1 -» R-G=0-GO-^ (1,9,10,11,12) 

2. R-C^C-C=C-R' + H 3 ->R-@H 3 -C=C-C0«R ! (13) 

A1C1 3 , v 

3* R-C^C-CO-Cl + R'H -> R-C=C-C0-R' (8) 

4. R-C^C-MgX + R'-COg-R" -* R-C=C-C0-R' + R n 0MgX (8,15) 

5. &r~C£C-C0 3 Et + ArMgX -> Ar-C=C-CVAr (8) 

Recently, the work of Bowden, Jones, and co-workers (2) and of 
Liang (15) upon the oxidation of the corresponding carbinols has 
made readily available the acetylenic ketones. . The most satisfact- 
ory method of oxidation consists of the addition of Cr0 3 in dilute 
H 3 S0 4 to the carbinol in acetone solution. With proper adjustment 
of the concentration of the reagents, the reaction mixture will 
separate in1>o two layers of which the upper one consists mainly of 
the carbonyl in acetone. In this way, the carbonyl is protected 
from further oxidation. By this method, use is made of the con- 
venient action of acetylene d-rignard reagents upon aldehydes and 
ketones to give the ethynylcarbinol compounds (16,17,18,19). 

R-C=C-MgX + R'-CHO -> R-C=C-CH0H-R T -» R-C^C-CO-R' 

Chauvelier (6) has found a practiral way to prepare symmetrical 
diacetylenic ketones by the reaction of acetylene G-rignard reagents 
upon ethyl formate. 

R-C=C-MgX + HC0 3 Et -* R-C=C- CH0H-C=C-R ->• R-C=C-C0-C=C-R 

The reaction of ammonia and primary and secondary amines with 
ethynyl carbonyl compounds gives products in which the addition 
takes place across the triple bond to give a beta substituted 
ethylene carbonyl compound (1,2,6,11,16). 

R-C=C-C0-Rt + R»NH 3 -* R-C(R»NH) = CH-CO-R r 

With compounds of the ethynyl-ethylene carbonyl type, the addition 
is across the triple bond. 

R-C=C-CO-CH=CH-R , + R»MH 3 -> R-C(R"NH) = CH-CO-CH=CH-R i 

By the addition of diethyl amine to a vinylog of an acetylenic ke- 
tone the expected product (2) was obtained. 

+■ I 

- v 

; ... J r> ';.. 




CH 3 -CO-CH=CH-C=CH + (C 3 H e ) 3 NH -> CH 3 -CO-CH=CH-OH=CH-N(C 3 H s ) 3 

This product is unstable and was characterized merely by 
absorption spectrum. Other nitrogen compounds, hydrazine sulfate, 
hydroxy amine hydrochloride, guanidine nitrate, and ammonium sul- 
fate add to benzoyl acetylene to give respectively 3-phenyl- 
pyrazole (70$), 5-phenylisooxazole (9C^). 2-amino-4-phenylpyrimidine 
(2df), and 5-benzoyl-2-phenylpyridine (3). 

The addition of ammonia and primary and secondary amines to 
symmetrical diacetylene ketones proceeds as might be expected to 
give mono-addition products which have been well characterized 


R-C=C-C0-c£C-E + R'NH 3 -> R-C (R'MH) = CH-CO-0=C-E 

These products exist in two isomeric forms both of which have been 
isolated and interconverted. In all probability, these are cis- 
trans geometric isomers. To date, the addition of two moles of 
amine has not been reported. 

When the addition products of primary amines are heated above 
their pelting points or boiled with xylene, a rearrangement takes 
place and the corresponding tri-substituted lutidones 
in good yield. n Cl- 

are produced 

R-C(R , NH) = CH-CO-C>G-R 



R l 




The addition products of secondary amines do not undergo this re- 
arrangement. The existance of a resonate zwitterion accounts for 
the lack of color found a 

Hydrolysis of the -amino addition products (primary or second- 
ary) with water and acid produces beta diketones from acetylenic 
ketones and gamma pyrones from di-a&etylenic ketones by way of an 
unstable intermediate (6,20) acetylenic diketone. 

I. R-0(R , NH)=C-C0-R" -> 
II. R~0(R'NH) = C-CC-C=C-R 

R-CC-CH 3 -C0-R" 

-► R-C0-CHr,-G0-C=C-R 





This intermediate is converted by heating almost explosively 
the pyrone. (20) . 

When aniline addition products of bis-ethynyl ketone are 
treated to effect cyclization to the lutidone, only 50% of this 
product is obtained. From this reaction mixture, a red compound 
has been isolated and from its absorption spectrum and degradation 
to benzoic acid, phthalic acid, benzanilid, and carbon dioxide, it 
appears to have the following structure » 

■ '• '.' ' ' ' 

9k Y 


or its isomer 







Andre, Ann. chem., (8) _29, 557 (1913) . 

Bowden, Braude, Jones, and Weedon, J. Che eft. Soc . ; 1946, 39 and 


Bowden and Jones, J. Chem. Soc.., 1942 , 953. 

Chauvelier, J., Cornpt. rend., 224, 476 (1947). 

Chauvelier, J., Compt. rend.^ 212, 

Chauvelier, J., Ann. Chem., 


Moore, A. 
Barat, J. 

793 (1941). 
'(127"3, 391 (1948). 
Jones, and Laeey, J. Chem. Soc, 1946 , 27. 
C, Org. Seminar Abstracts, Univ. of 111., 

net. 18, 

Ind. Chem. Soc, 7, 851 (1930) C.A., 25, 2145 

416 (1945). 


Campbell, and Eby, J. Am. Chem. Soc, 60, 2882(3938) 
Moureu and Delange, Bull. Soc Chim., _31, 1329 (1904) 
Nightingale and TiTadsworth, J. Am. Soc, 67 , 
G-rignard and Tcheoufaki, Rec trav. Chim., .48, 901 
Tiffeneau and Deux, Compt. rend., 214, 892 (1942). 
Liang, Bull. Soc Chim., (4) 53, 44TT1933). 
Dupont, Bull. Soc Chim,, (4) 1_5, 604 (1914). 
Froming and Hennion, J. Am. Chem. Soc, 62, 653 (1940). 
Heilbron, Johnson, Jones, and Spinks, J. Chem. Soc, 1942 , 731. 
Jones and McComble, J. Chem. Soc, 1942 , 737. 
Chauvelier, J., Compt. rend., 226 , 927 (1948). 

Reported by Carl S. Hornberger 
October 29, 1948 

-. t 
' .7 

. o; 

t „", 




The name Jacob sen Reaction is given to those reactions in- 
volving the migration of an alkyl group or a halogen atom of the 
sulfonic acid derived from a polyalkylbenzene, a halogenated poly- 
alkylbenzene , or a polyhslogenated benzene, in the presence of con- 
centrated sulfuric acid (l). 

The reaction was first discovered by Herzig who, in 1831, 
noted the rearrangement of a polyhalogenated benzenesulf onic acid. 
It is named however after Jacobsen who was the first to observe the 
rearrangement of polyalkylbenzene sulfonic acids in 1886 (2). 

Studies of Cyclic Systems 

Hydrindene (l) and tetralin (il), and their derivatives, may 
be considered as ortho-dialkylbenzenes, and might be expected to 
undergo rearrangements under the conditions of the Jacobsen Re- 
action (3). 




A rearrangement of this type was observed by Schroter and G-tttzky 
(4) who in preparing octahydroanthracene- 9- sulfonic acid found that 
under prolonged heating and high temperatures, this compound was 
transformed into the octahydrophenanthrene sulfonic acid, 

Arnold and Barnes (3) subjected s-hydrindaoene (ill), 5,6,7.8-- 
tetrahydrobenz (f ) indan (IV), 5-ethyl-6-methylhydrindene (v) , 6,7- 
diethyltetralin (VI) , and octahydroanthracene to the conditions of 
the Jacobsen Reaction. 

H 3 S0 4 

tar + starting 

H 3 S0 4 


' ■ 

/ ■; 

' • 

,'-•'■ ... . . ' I 

'O'v'i !■"■' 'i ti.r 

:f -j (• ;* p 


H 3 S0 4 


NNR R = CPU or C 3 H 




Reaction Mechanism 

H 3 S0 4 

K 3 S0 4 


From the results of the above reactions, and other work which 
indicated that the steric effect of a methylene group present in 
various groupings, increase in the order, 

5-membered ring s 6-memebered ring/ CH 3 , 
(hydrindene) \ (tetralin) ^ 

Arnold and Barnes postulated the following regarding the mechanism 
of this reaction (3): 

1. The reaction proceeds by an initial sulfonation, replace- 
ment of an alkyl group by a second sulfonic acid group, and sub- 
sequent replacement of the first sulfonic acid group by the alkyl 

2. The reaction is possible only if the first 
sufficiently hindered by ortho-substituents such as 
the contribution of structure (A) to resonance 

SO 3 H- group is 
to decrease 






S0 3 H 









3. This reduction of resonance permits the entrance of a 
second S0 3 H-group, preferentially meta to the first, with the re- 
placement of an alkyl cation, or the opening of a saturated ring to 
form an intermediate chain with a cat ionic terminal carbon atom. 

4. The alkyl cation, or the intermediate chain, replaces 
preferentially the most hindered sulfonic acid group. 

. . , n 


Further Studies of Cyclic Systems 

Smith and Lo (5) have reported further studies of 6 ; 7-dialkyl- 
tetralins. The 6, 7-dimethyltetralin gave the expected 5,6-dimethyl 
tetralin. 6-Isopropyl-7-ethyltetralin and 6, 7-di-n-propyltetralin 
yielded small amounts of unidentified oils. 6~Isopropyl-7-methyl- 
tetralin on treatment with sulfuric acid lost the isopropyl group 
giving 6-methyltetralin. 




CH 3 

CHp CH p CH . 

H 2 S0 4 liquid 
—» hydrocarbon 

6- n-propy 1-7- methyl tetralin 

Oxid . 







There are two ways that (VIl) can undergo rearrangement to give 
the liquid hydrocarbon which was identified as the 5-n-propyl-6- 

(a) The 6-n-propyl-group may migrate to the 8- (or 5-) position 

(b) The tetramethylene ring may open and close ortho to the 
propyl group. 

According to the mechanism theory of Arnold and Barnes (.3), 
the product from (a) should be 5-isopropyl-6-methyltetralin since 
during the course of the reaction a free n-propyl cation would be 
existent, and this would rearrange to the more stable isopropyl 
cation. Smith and Lo conclude therefore that the reaction proceeds 
according to (b). 


1. Smith "Organic Reactions", Vol. I, John Wiley and Sons, Inc., 

N.Y., 1942, p. 371, 

2. Moyle .and Smith, J. Org. Chem,, 2, 112 (1937). 

3. Arnold and Barnes, J. Am. Chem. Soc. , 66, 960 (1944). 

4. Schroter and G-atzky, Ber., J30, 2035 (1927), 

5. Smith and Lo, J. Am. Chem. Soc, 70, 2209 (1948). 

Reported by K. H. Takemura 
October 29, 1948 

r . • 

... •■ ■ •;". 

t .. <• * 

l „..., 


In view of the use of the well-known sulf athlazole (the sulf- 
anilamide derivative of 2-aminothiazole) as an important sulfa drug, 
interest arose in the synthesis of the 4- and 5-aminothiazoles and 
their derivatives. Although 2-aminothiazole was synthesized as* 
early as 1889, the isomeric 4-aminothiazole and 5-aminothiazole are 
not known. Recently, however, derivatives of these compounds have 
been prepared by a variety of methods (1-8). 

3 N SH 4 


2 CH CH 5 H 3 N-C!-C X5-NH 3 

^3/ X S X 

1 (I) 

Thiaaole 5-Amino-2-thioamidothiazole 

A. From " chrysean ". Walla ch (l) obtained chrysean in 15-20^ 
yield by passing hydrogen sulfide into a saturated solution of 
potassium cyanide. Later it was shown (2,3) to be 5-amino-2- 
thioamidothiazole, (i) . The chrysean molecule is fragile, and the 
5-aminothiazole cannot be prepared by the nitrile, carboxylic acid, 
and decarboxylation route. 

Hellsing (2) prepared 5-aminothiazole-2-nitrile from chrysean 
by treatment with lead or silver salts, but attempts to hydrolyze 
this to the acid usually result in rupture of the ring, especially 
under alkaline conditions. Cautions acid hydrolysis has given 
5-amino-thiazole- 2- carboxylic acid in poor yield. On the other 
hand, in neutrpl solution with excess of O60 a the nitrile gives 5- 
aminothlazole-2-amide in very good yield. Acylation of the amino 
group permits normal degradation of the thioamide group, but the 
products are all so resistant to hydrolysis that the acyl group can- 
not be removed. 

Arnold and Scaife (3) prepared 5- (p-acetamidobenzenesulf on™ 
amido) chiazole-2- thioamide, (II j, from chrysean and p-acetamido- 
benzenesulf onyl chloride. 5- (p-Aminobenzenesulf onamld o )thiazole, 
(III), was obtained from the 2- thioamide, (II ), by converting this 
into the 2-nitrile with lead carbonate and boiling the solution 
with sodium hydroxide; deacetylat ion, hydrolysis of the nitrile 
group, and decarboxylation then took place. It was found that 
against streptococcal infections in mice, (III) was active but has 
no advantage over established sulfonamide drugs, whereas (II) is 

n 1 

?H N -CH 

( II II 

H 3 N~C-C JC-NH-S0 3 -C 6 H 4 -NHAc CH C-NH-S0 3 -C s H 4 -NH 3 

(II) (III) 


Y n ": 



Ui'Vry. r : - *■• 

/ . 

. 0- r 'i 
» .- • r.> 

:'". >' .'. 

••■ v I , n> 

:. :j; ;.r---re:;rr,., { ?-J^XflT3 • 



B. From thlaz ole- 5- carboxyl ic esters by the Gurtius reaction 
(4,5). 'Starting with ethyl 2, 4-dimethylthiazole-5-carboxyla te, 
which was prepared from thioacetamide and ethyl oc-chloroacetoacetate 
the corresponding 5-amino compound was obtained by the Curtius re- 
action. The ethereal solution of the azide obtained from the ester 
was added to a mixture of acetic acid and acetic anhydride. After 
decomposition of the azide and neutralization with Na 3 C0 3 , the 
acetamino compound separated in 99f; yield. Hydrolysis with HC1 
yielded the amino compound in 90^ yield. 

The following esters, (a-c), gave the corresponding 5- 
acet amino compounds. 

(a) Ethyl 4-methylthiazole-5-carboxylate . 

This ester was Prepared from thloformamide and ethyl a- 
chloro (or bromo) acetoacetate. 

(b) Ethyl 2-chloro-4-methylthiazole~5--carboxylPte . 

The reaction of ammonium thiocarbamate and ethyl a-chloro- 
acetoacetate gave ethyl 2-hydroxy-4-methylthiazole- 5-carboxyl- 
ate. Refluxing with P0C1 3 gave (b), the corresponding 2- 
chloro compound. 

(c) Ethyl 2- amino- 4-methylthiazole-5-carboxy late . 

This compound was prepared from thiourea and ethyl a- 
chloroacetoacetate . 

6" From 5-acetylthiazole s by the Beckmann rearrangement . 
4-Methyl-5-acetylthiazole, prepared from thiof ormamide and 
chloroacetylacetone, and 2,4-dimethyl-5-acetylthiazole , prepared 
from thioacetamide and chloroacetylacetone, were converted to the 
corresponding 5-acetamino compounds in 30^ and \of. yields, respect- 
ively, by treatment of their oximes with PC1 B . Treatment of their 
oximes with acetic anhydride and hydrogen chloride gave the acetate? 
of the oximes. (4) 

D. From the reduction of 5-nitrothlazole s (4). Nitration 
of 2-acetaminothiazole with concentrated sulfuric acid and fuming 
nitric acid yielded 2-acetamino-5-nitrothiazole, but reduction 
failed to give the 5-amino compound. Nitration of 2, 4-dimethyl- 
thiazole, prepared from thioacetamide and chloroacetone, gave 2,4- 
dimethyl-5-nitrothiazole. This nitro compound could be reduced 
with iron to the amino compound which on acetylation gave 2,4- 

£•. From the reduction of 5-azothiazoles (4) . 2- Hydroxy- 4~ 
methylthiazole, prepared from ammonium thiocarbamate and chloro- 
acetone, underwent coupling with diazotized p-toluidine to yield 
an azo dye which on reduction with sodium hydro sulfite furnished 2- 
hydroxy-4-methyl-5-aminothlazole . This method has limited applic- 
ability, for 2,4-dimethylthiazole or 2-amino-4-methylthiazole did 
not couple with diazotized p-toluidine. 

F. From amlnoa cetohitrile s and dithloacid derivatives . In 
connection with the study of penicillin, Heilbron (6) examined the 
reaction between sodium or methyl dithiophenylacctate and ethyl- 

■J ,■ :f 

T •-■ 


t -f, 



aminooyanoacetate (prepared by reducing ethyl nitrosooyanoaoetate 
with amalgamated aluminum). The product was at first thought to be 
acyclic, but subsequent investigation showed that it was 5-amino- 
4-carbethoxy-2-benzylthiazole, (IV) . Similarly, aminoacetonitrile 


Ph-CH 3 -5-SNa(Me) + H 3 N-CH-COOEt -* | | 

ON Ph-CH 3 -C v £-NH 8 


and sodium di thlopheny lace tat e afforded an excellent yield of 5- 

By employing sodium dithioformate and ethyl aminooyanoacetate, 
5-amino~4-carbethoxythiazole was obtained. In the same way a- 
aminobenzyl cyanide and sodium d ithioformate afforded 5-amino-4- 

This appears to be the most general synthesis of the 5- amino- 
thiazoles. It is noteworthy also that these ring syntheses take 
place at room temperature, several of them in aqueous neutral 

G-. From et-amino-nitrile s and carbon disulfide (7) . The re- 
action between carbon disulfide and a-aminobenzy 1 cyanide was easily 
effected at room temperature to yield 5-amino-2-mercapto-4-phenyl- 
thiazole. When this was treated with alkali and Raney nickel in 
hot ethanol, removal of one atom of sulfur proceeded spontaneously 
to give the known 5-amino-4-phenyl-thiazole . 

Ethyl aminocyanoacetate and carbon disulfide gave 5- amino- 2- 
mercapto-4-carbethoxythiazole. TMs was confirmed again by de- 
sulf urization with Raney nickel to a known thiazole. 


1. Wallach, 0., Ber., 7, 902, (1874). 

2. Hellsing, G., Ber., 32, 1497 (1899); Ber., 33, 1774 (1900) Ber., 

36, 3546 (1903). 

3. Arnold and Soaife, J. Chem. Son., 103 (1944). 

4. G-anapathi and Venkataraman, Proc. Indian Acad. Soi., 22A, 343 


5. Jensen and Hansen. Dansk. Tids. Farm., 17, 189-93 (1943); 

C.A. 2058 3 (1945), 

6. Cook, Heilbron, Levy, J. Chen. ^oc. , 1594 (1947). 

7. Cook, Heilbron, Levy, ibid., 1598 (1947). 

8. Cook, Heilbron, Levy, ibid., 201 (1948). 

Reported by Alex Kotch 
October 29, 1948 



The total synthesis of the natural estrogenic hormone (+)- 
estrone was reported at the beginning of this year in a brief com- 
munication by Anner and Miescher (l) . Several investigators have 
attempted to synthesize the hormone but have succeeded only in ob- 
taining mixtures of sterioisomers or structural isomers (2). As a 
result of extensive research in the stereochemistry and synthesis 
of the marrianolic and doisynolic acids, Miescher and coworkers 
have been able to limit the number of stereoisomers in epch of the 
intermediate steps in the synthesis and have therefore been able to 
obtain the correct one of the sixteen possible stereoisomers of 

"Natural" (+)-marrianolic acid (ila) has been obtained most 
conveniently from (+)-estrone (la) via the estrone benzyl ether 
(lb) (3). 

CH 3 



Ni,H 3 



CH s C00R 




R=CH 3 $ 

(Ila), R=H 
(lib), R=CH 3 

Under such mild conditions as these it is unlikely that the con- 
figuration at any of the asymmetric carbon atoms is altered. 

"Natural" (+)-doisynolic acid (ill) is the only product ob- 
tained from (+)-estrone by fusion with potassium hydroxide at 
275° (5). Since fusion of estriol with iDotassium hydroxide at 
275° yields a single (+)-marrianolic acid (4) identical with (Ila) , 
it is probable that the (+)-doi syndic acid (III) corresponds 
stereochemical^ to (+)-estrone (5). Finally, (+)-marrianolic acid 
as the 7-methyl ether dimethyl ester (lib) has been converted into 
(+)-doisynolic acid (ill) in the following reaction series (6,7): 



CH fl 

CH 3 


1. Conversion to acid 

CH 2 C00H 2. Pd-C 

A CH 3 

r M- COOCH. 


•■' ■ I 

a i • .... < 

'• i 

■ ■! . '• 



1. NaxClyool, N 3 H 4; 19Q 

— > 

2. Pyridine-HCl 




a H 

3 n 5 


The key to the total synthesis of (+)~estrone was the keto 
ester IV which both Robinson (8) and Bachmann (9) had obtained only 
as a liquid mixture of racemates. Robinson (10) had little success 
in his attempt to synthesize estrone from this mixture. Anner and 
Miescher (11) were able to separate the mixture into three crystal- 
line, racemic substances (IVA,B and C) by fractional crystallization 
With these compounds as starting materials, they were able to ac- 
complish the total synthesis of five out of the eight possible 
racemic doisynolic acids as follows; 
CH 3 



CH 9 




S- C00CH3 


(IV A, B, C) 

I C 2 H 5 MgI 

— C~ OH 





^— C 2 H 5 


cn 3 


H 3 , Pt 



(IIIA) "a" and"p" 
(lIIC),"o"and "|3" 

The racemic (III A), (+)-"p"-7-methyl doisynolic acid, did net 
depress the melting point of the "natural" (+)-7-methyl doisynolic 
acid and, similarly, the (III B),"cc" racemate corresponded to ( 

j - i - 

7-methyl lumidoisynolic acid (differs from the "natural" doisynolic 
acid only in configuration at C s ) . The (III A) , "a" racemic 7-methy. 
doisynolic acid has been related to 14-iso estrone. 

The synthesis of (+)-estrone 
since it was believed to have the 
of the three asymmetric centers. 

.'. .OOOCH3 

= + 

CH 3 0-^ 

started with the keto ester (IV A) 
requisite configuration at each 

mirror image 

(IV A) 



/ N...COOCH, 

1. Z»,BrCH 3 C00CK 3 

-J ©. -k 3 c 

3. +2H 


+ mirror image 

Af 3 
r v GOOCH 


((+)ll b) no melting point 

lowering on admixture 
with (II b) 

I Arndt Eistert on 2-monoester 
(+)-7-methyl homomarrianolic acid 

1. Cyclization (Dieckmann) 
^ 2. Dftme thy la t ion 

+ mirror 
....CH 2 C00CH 3 image 


similar to those 
of (( + )II Id) 


+ mirror image HO-4 v ; 

+ mirror 

(±) estrone 

(+)-14-isc estrone 

resolution via 

L-menthoxy acetates 
(+)-estrone (la) 



This is also a total synthesis of "a" estradiol since 
estradoil is a reduction product of (+)~estrone. In a similar 
manner /\ 8 > 9 -monodehydroisoestrone has been synthesized from ( + )- 
"a"-7-methyl monodehydromarrianolic acid which has, in turn, be"en 
obtained by total synthesis (12,13,14). 




Anner and Miescher, Experentia 4, 25 (1948). 

Bailey, Organic Seminar Abstracts, May 8, 1946; Miller, Organic 
Seminar Abstracts, October 17, 1947. 
7' H e r and Mie scher, Helv. Chim. Acta, 28, 156 (1945). 
4. Marrian and Haslewood, J. Biol. Chem.TIOl, 753 (1933). 
5 8 Shoppee- Annual Reports, 44, 170 (1947": — 

6. Soffer, Soffer and SherTT-'J. Am. Chem. Soc. . 67. 1435 (1945) . 

7. Heer and Miescher, Helv. Chim. Acta., 29, 18§5"^1946) . 

8. Robinson and Walker, J. Ohem. Soc, 747" { 1936). 

1§: Rob°§SlSn'anS S ^!£erf J^SIS? ^?; i§3 te)? 0C - ^'^ (l942[ 

^\ A gPrS^t!a ie §f|?6 ¥ (ilh) Chim ' Acta -' 59* 1«8 (l947 > ; 
12,13,14. Heer and Mleecher, Helv . Chim. Acta ., 31 , 219 (1948): ibid, 

229 (1948); ibid., 1289 (1948). ~ 

Reported by Janet B. Peterson 
November 5, 1948 

.■ '■ 



The Reformatsky reaction produces several by products, some 
of which may be expected and others rather unexpected. The ones 
usually expected (6) are the product of coupling 

(1) 2BrCH 3 C00C 3 H 5 + Zn -> | 

CH 3 ~C00C 3 H B 

CH s -C00C 3 H 5 

and a T-bromo-|3-ketoester (3) produced by the addition of one mole 
of the zinc halide to a mole of the bromoester. 

(2) 2CH 2 C00C 3 H 5 + Zn 

BrCH 3 -(}-CH 3 COOC 3 H 5 

OC 3 H 5 

C 3 H 5 OZnBr 

BrCH 3 CCK 3 C00C 3 H 5 

Aliphatic aldehydes or ketones sometimes undergo aldolization under 
the influence of the zinc salts forming the usual condensation 

Upon treating 1,4- dibromo-1, 4-dibenzoyl butane (l) with zinc 
they obtained two cyclopentane derivatives, A and B, and the reduced 
dibenzoyl butane instead of the expected cyclobutane derivative. 
The proposed course of the intramolecular Reformatsky reaction is 
as follows: 

<?H 3 - 

C,0C- 6 H 5 

,C0C 6 H 5 


CH 3 - 

-CH (DZnX 


-CH X C S H 5 

CH 3 - 



COC s H s 



C0C 6 H 5 
CK 3 -CHZnX 
CH 3 - CHZnX 
, COC 6 H 5 
H 3 



-CK yDZnX 




C 6 H 5 CO(CH 3 ) 4 COC s H« 

CH 3 -CH V 

VC 6 H 5 

CH 3 -CH CeH 5 
COC 6 H B 

CH 3 -CH 3 OH 

CH 3 -CH CeH5 


H a O 

CK 3 -CH-— Q 

CH 3 -CH/ ^ 6 H 5 
COC 6 H 5 

CH 3 -CH / ' 

C0C 6 H 5 


C0C 6 H s 



When 5, 8-dimethyl-l-tetralone (3) was treated with zinc and 
(a) ethyl bromoacetate (b) ethyl a-bromoproplonate and (c) ethyl 
a-bromoisobutyrate the principal by products were the reduced 



ester and the corresponding p-ketoester, R 3 CHCOCR 3 COOC 3 H 5 . By 
using an excess of the bromoester and zinc to compensate for these 
by products the yields of the desired products increased from 25$ 
to 85$ with ethyl cc-bromoacetate and from 17$ to 82$ with ethyl 
a-bromopropionate. No Reformat sky oroduct was obtained with ethyl 
a-bromoisobutyrate. In each case about the same amount of reduced 
ester was obtained, 10-15$. In addition a considerable amount of 
the corresponding p-ketoeeter was obtained; 15$ of ethyl aceto- 
acetate, 35$ of ethyl oc-propionyl propionate and 69$ of ethyl a- 
isobutrylisobutyrate. They were unable to detect any of the 
halogenated £-ketoester or coupled product. The proposed reaction 
path for the formation of the p-ketoester is as follows: 

(3) [R 3 CC00Et]ZnBr + R 3 CBrCOCEt -> 

OEt -EtOZnBr 

R 8 CBrC-CR 3 C00Et -> R s CBrC0CR 3 C00Et 

OZnBr , 

H 3 
R 3 CFC0CR 3 C00Et £-— [R 3 CC0CR 3 C00Et]ZnBr 

The experimental .support for this reaction path is the fact that, 
even in the absence of carbonyl compounds, ethyl a-bromopropionate 
reacts vigorously with zinc to give 16$ of the reduced ester and 
39$ of the ethyl o-propicnylpropionate . Ethyl a-bromoisobutyrate 
gave traces of the reduced ester and a 65$ yield of ethyl cc- 
isobutrybutryate when treated in like manner. 

The occurrance of the reduced ester has been explained in two 
ways, neither of which is satisfactory in itself. The first explan- 
ation (5) postulates the enolization of the ketone which reacts with 
the organomet^lic intermediate analogously to the formation of 
methane in a Zerwitinoff determination: 

(4) [RCHC00Et]ZnBr + RC0CH 3 R -> [RCOCHR]ZnBr + RCH 3 C00Et 

This postulate was confirmed by treating acetomesitylene, which is 
known to react with other orgpnometalics (4) by enolization, with 
zinc and methyl bromoacetate. It was possible to distill methyl 
acetate from the reaction mixture before hydrolysis showing that 
the reduced ester did not arise from a reaction of the bromozinc 
intermediate with water. 

Since appreciable amounts of the reduced ester are obtained 
when no ketone is present in the reaction mixture it was necessary 
to postulate another mechanism. The one proposed involves ^n acid- 
base type reaction involving the acidic hydrogen of the bromoester. 


[RCHC00Et]ZnBr + RCHBrCOOEt -» RCH,CC0Et + [RCBrC00Et]ZnBr 




This type of mechanism is supported by the fact that no reduced 
ester is produced in the self condensation of ethyl a-bromoiso- 
butyrate, a bromo ester which contains no cc hydrogen. 


1. Fuson and Farlow, J. Am. Chem. Soc, 56, 1953 (1934). 

2. Hann and Lapworth; Proc Chem. Soc, 19, 189 (1903). 

3. . Hussey and Newman, J. Am. Chem. Soc, 70, 3024 (1948). 

4. .Newman, ibid, 62, 870 (1940) . 

5. .Newman, ibid, 64, 2130 (1942). 

6. Shriner, "The Reformatsky Reaction" in "Organic Reactions", 

Vol. I, John Wiley and Sons, Inc., New York, N. Y. 1942. 

Reported by C. W. Fairbanks 
November 5, 1948 


I. - 



Enzymes are organic catalysts produced by living organisms. 
Thus far, every enzyme isolated has been found to be a protein. 
Most usually, enzymes are named and classified in terms of the re- 
action or reactions which they : catalyze and by ':..\e lr behavior to- 
ward substrates (the substances acted upon). 

An enzyme may be considered as a combination of a protein 
carrier (apoenzyme) and a coenzyme, an organic, heat-stable molecule 
which is a specific activator. 

G-unsalus et al. (1,2) ha 
of Streptococcus fecalls and 
tyrosine and Cther amino acid 
(i), but is markedly stimulat 
adenosine triphosphate or if 
with chemical phosphorylating 
phosphorylation used were; (a 
followed by silver dihydrogen 
phoric acid in the cold; . (c) 
pyridoxal hydrochloride with 
oxy chloride. 

ve shown that th 

Escherichia coli 

s is slightly st 

ed if supplied w 

the pyridoxal is 

agents (3,4,5). 

) treatment with 

phosphate; Ob) 

treatment of an 

sodium hydroxide 

e ability of suspension 
to decarboxylate 

imulated by pyridoxal 

ith pyridoxal and 
treated previously 

The methods of 
thionyl chloride, 
treatment with phos- 
aqueous solution of 
and phosphorus 

The last method gave a b% yield compared to the trace yields 
of the preceding two. Analytical data on the amorphous barium 
salts indicated that the coenzyme was a pyridoxal phosphate. 






Since the method of preparation (6) does not reveal the 
position of the phosphate group, other studies (7) were undertaken 
to secure structural evidence. The following results were reported: 

1. When pyridoxine (il) is phosphorylated ( 
and then oxidized with potassium permanganat 
which convert pyridoxine to pyridoxal (i), a 
with coenzyme activity. The authors deduce 
aldehyde group is free in the phosphorylated 
further substantiated by phosphorylating pyr 
and treating the resultant product, which po 
of coenzyme activity, with nitrous acid to 1 
group.. The resulting solution possessed def 
activity. This eliminates structure (IV). 

no coenzyme activity) 
e under conditions 

product is obtained 
from this that the 

pyridoxal. This was 
idoxal oxime (ill) 
ssessed only traces 
iberate the aldehyde 
inite coenzyme 

■ • - I ••■ . • ■••■ 

• ' ...■■• I '_ 

'. J 

. : : ' r " 

• • ";.'.''■•'' 

- •■ 

i ■'• f"\ 



-CH 3 





CH 3 - 

OP0 3 H 





2. The following evidence indicates that the phenolic hydroxy 
group of pyridoxal is free: (a) The cyclic ethyl acetal of pyri- 
doxal (v) was converted into the oxime of the phosphate (VII) by 
phosphorylation, followed by the action of hydroxy lamine. The 
oxime (VIl) differs from the oxime of the phosphorylated pyridoxal. 
(b) If the phenolic hydroxyl were phosphorylated, the absorption 
maximum at 2900 A which is present in acid solution should be main- 
tained under alkaline conditions, as is the case with pyridoxin-3- 
methylether (VIIl). This does not occur. 

PC 2 H 5 0C 3 H 5 

OH — CH— 

P0C1 3 X { 



CH *V 

6H ; 

in~pyridine no 9 VO^\-bu 9 NH 3 0H H0 3 ?0.<^\-CH 3 0H 

Then water 

CH •=— . . 







CH 3 0H 


H0CH 3 /^X-0CH 3 
<s i-CH 3 


3. Coupling of 2,4-dichloroquinone chlorimide (8) in the 6-position 
of pyridoxal readily takes place, but does not occur in phosphory- 
lated pyridoxal. CHO CHO 

H0CH 3 ,^\-0H 

-NCI + 







Karrer and Viscontini (9) reported that the synthetic cyclic 
ethyl acetal of pyridoxal- 3-phosphate (VI) possessed 
activity. These results could not be duplicated by G-unsalus and 
Umbreit (10) ; due possibly to the difference in the enzymes used in 
the subsequent testing. 

Karrer verified his oostul&ted structure for acetal pyri&oxal- 
3-phosphate (Vl) by its failure to couple with 2,4-dichloroquinone 

•■>' * .•.-} 

■r ;.. 



ahlorimide, -"Since this is a positive test for p-un substituted 
phenols, he concludes the phenolic hydroxyl must have been modified 
by esterif ication to the phosphate. 

Karrer (ll) claimed that the loss of the absorption maximum 
at 2900 A was to be expected, since compound VI existed in acid 
solution as a cation and in alkaline solution as an anion. 

He showed that (Vl) or the hydrolyzed product, pyridoxal-3- 
phosphate was at least as active a coenzyme of 1- tyrosine- 

decarboxylase as a mixture of pyridoxal and adenosine triphosphate . 
Attempts to phosphorylate the primary hydroxyl group of pyridoxal 
or of its various derivatives were unsuccessful. 

A new synthesis of pyridoxamine was reported; 
CH 3 0H CH=N0H 

H0CH 3 -7^\0H NaHCOg NaOAc H0CH 3 

H, KMn0 4 H0NH 3 'HC1 


CH 3 NH 3 
H0CH 3 ^ \-0H 

CH 3 Pt0 3 


J OH. 


1. G-unsalus, I. C. , and Bellamy, W. D., J. Biol. Ohem. 155 , 357 

2. Umbreit, W. W. , and G-unsalus, I. C, ibid., 159 , 333 (1945). 

3. G-unsalus, I. C, and Bellamy, W. D., ibid., 155, 557 (1944). 

4. G-unsalus, I. 0., Bellamy, W. D., and Umbreit, W. W. , ibid., 
155, 685 (1944). 

5. Umbreit, W. ¥. , Bellamy, W. p., and G-unsalus, I. C., Arch. 
Biochem. *, 185 (1945). 

6. G-unsalus, I. C, and Umbreit, W. W., Abstracts Am. Chem. Soc. , 
110th meeting, Chicago, 34B (1946). 

7. Heyl, D., Hai ris, S. A., and Folkers, K. , ibid., 35B (1946). 

8. Scudi, J. V,, J. Biol. Chem., 139, 707 (1941) . 

9. Karrer, P., and Viscontini, M., Helv. chim. acta, _30, 52 
(1947); 30, 524 (1947) . 

10. G-unsalus, J. C. and Umbreit, W. W. , J. Biol. Chem. JL70, 415 

11. Karrer, P., Viscontini, M. , and Forster, 0., Helv. chim. acta, 
31, 1004 (1948). 

Reported lay S. E. Neuhausen 
November 5, 1948 




I INTRODUCTION: The concept of hypercon jugation (or no-bond re- 
sonance) was first proposed by Baker and Nathan (l) in 1935. 
Since that time a large body of physical and chemical evidence 
supporting the theory has accumulated, and it is further sub- 
stantiated by quantum mechanical calculations. These facts 
have been reviewed recently (2, 3) and will not be presented 
here. This seminar will be limited to a qualitative discussion 
of carbon-hydrogen hypercon jugation and a presentation of the 
recently proposed carbon-carbon hypercon Jugation, 


A» General Expre ssion of the C oncepts Hyper con jugation occur s, 
"to a greater or lesser extent, whenever a saturated carbon 
atom (holding one or more hydrogen atoms) is attached to an 
unsaturated carbon atom. In general it is the tendency of 
the electron pair of one of the C-H bonds of the saturated 
group to drift toward the unsaturated atom. In the extreme 
state the C-H bond is broken and there is an accompanying 
polarization of the double bond (see sections B and C below), 

B« Anal ogy jtq Ca rbonyl- type Res onance; The acidic properties 
of "carbonyl compound's having'a hydrogen atom attached to 
the adjoining carbon atom can be attributed to resonance 
stabilization of the negative ion. 

-C= v -C= -C^O -C- : O 

(3 k 

t , 

-9- H -C:° + H -C: } <3 


Hypercon jugation merely extends this idea to olefinic un- 
saturation. The position of the equilibrium (i) is again 
further to the right than might be expected, because of the 
resonance structures (II and III), which stabilize the 
negative ion. 

T c- -g-c- 

1 11 in 

It is of importance to note that the formation of the proton 
does not necessarily precede the electronic shift, and for 
this reason structures of the following type may occasionally 
contribute appreciably to the resting state of the molecule; 

H-C-H H-C K\J H-C-H 'iFc-H 

H H H @ H 



0. Order of Electron- re lea sing Tendencle s of Alkyl Groups: 

Assuming that only the alpha hydrogen atoms can participate 
in hypercon jugation, the order of electron-releasing ten- 
dency of alkyl groups attached to an unsaturated system is: 

CH 3 > CH 3 CH 3 y (CH 3 ) 3 CH > (CH 3 ) 3 C 

The order for the inductive effect is the reverse of this. 

D. Example g of Carbon- Hydrogen Hyper con j ugat ion : 

1. Addition of HC1 to a 2-pentene: 

Since there are more C-H bonds in the methyl group 
than in the methylene group adjacent to the double 
bond in this molecule, hypercon jugation oredicts the 
observed preferential formation of 2- chloropentane 
whereas the Markownikoff rule is not applicable because 
each unsaturated atom holds one hydrogen atom. 

2. Bromination of Toluene: 

Toluene is brominated pbout four times faster than t- 
butyl benzene. This is the expected result on the 
basis of the order of electron-releasing tendencies 
(section C) . 

CARBON-CARBON HYP ERGON JURATION: Berliner and Bondhus have re- 
cently suggested (4, 5) that a carbon-carbon single bond of a 
saturated group attached to an aromatic nucleus can participate 
in hypercon jugation in much the same way as a carbon- hydrogen 
bond, although to a lesser degree. 

A. The Necessity for Carbon- Carbon Hyperconjugation: When 
benzene and t-butyl benzene compete for an insufficient 
amount of bromine, the ratio of the rates of bromination 
is overwhelmingly in favor of the t-butyl benzene (115:1 
at 45°C). The inductive effect can exolain the direction 
of this result but probably not the extent, nor the strong 
ortho-para orienting influence in t-butyl benzene. Carbon- 
hydrogen hyperconjugation offers no assistance, because t- 
butyl benzene has no C-H bonds alpha to the unsaturated 
system. Since no satisfactory explanation based on a re- 
cognized theory has been advanced, Berliner believes that 
resonance structures of the following type must be utilized 
to account for the considerable electron- re leasing cha^acte: 
of the t-butyl group in t-butyl benzene: 

@CH 3 
CH 3 — C — CH 3 


Cn<a— C CHq 


CH 3 C— CH 3 

©CH : 

CHo— C~ CHq 




Extension of 0-0 hyper conjugation to other alkyl groups 
gives an order of electron release identical with the order 
of the inductive effect: methyl <^ ethyl <^i-propyl S t-butyl. 

B. Justification of Oarbon-Carbon Hyper con juration: 

1. Participation of C-C single bonds in hyperconjugation 
is allowed by quantum mechanics. 

2. The origin and physical significance of the inductive 
effect is obscure. Therefore, it is advantageous 

to consider all electron release by alkyl groups as 
a resonance effect. 

3. Physical constants indicate that increasing contribu- 
tions to resonance are obtained as the number of carbon- 
carbon single bonds available for hyper conjugation is 

a. Resonance energies of alkyl benzenes: methyl 

( i-propyl < t-butyl 

b. Dipole moments of alkyl benzenes: methyl 

('ethyl ^i-propyl < t-butyl 

c. Molecular exaltations of alkyl benzenes: methyl 

^ ethyl /i-propyl ' k > t-butyl 

IV CONCLUSION: The facts suggest that there are two opposing 

orders of electron release by alkyl groups attached to an un- 
saturated system. One of these orders can be explained by 
carbon-hydrogen hyperconjugation. Either the inductive effect 
or carbon-carbon hyperconjugation can be used to account for the 
opposing effect. Neither can be accepted without reservation, 
since the concept of carbon-carbon hyperconjugation has not been 
adequately tested, and the inductive effect is ill-defined 
with respect to its origin and physical nature. 


1. Baker and Nathan, J. Chem. Soc, 1840 (1935). 

2. Deasy, Chem. Rev,, _36, 145 (1945). 

3. Saunders, Organic Seminar, Univ. of 111., 1946. 

4. Berliner and Bondhus, <J. Am. Chem. Soc, 68, 2355 (1946). 

5. Berliner and Bondhus, ibid., 70, 854 (19487. 

Reported by Aaron B. Herrick 
November 12, 1948 


The 1-form of lysine (a, £-diaminocaproic acid) is one of the 
essential amino acids. 

Previous to this year four methods (1,2,-3,4) for the synthesis 
of dl- lysine had been reported. 

The first method was that of Fischer and Weigert (l). 

C00C s H 5 
NaCH + C1CH 3 0H 3 CH 3 CN -> CN(CH 3 ) S CH (COOG s H B ) 3 

C00C 3 H B 

CN(CH 3 ) 3 CH(COOC 3 H B ) 3 + C 3 H 5 0N0 -> CN(CH 3 ) 3 g COOC 3 H 5 


GN(CH 3 ) 3 qC00C 3 H 5 Ha, C 3 H 5 OH H + 

NOH -> -> NH 3 (CH 3 ) 4 .CHNH 3 COOH 

Stirensen's (2) synthesis was similar in principle, but in- 
volved the reduction and hydrolysis of ^-cyanopropylphthalimido- 
malonic ester in the final step. 

The method of v. Braun (3) provided the first practical 
synthetic procedure. 

^ CH 3 -CH 3 ^CH S _CH^ 

CH 3 NH C 6 H 5 C0C1 CH 3 NCOC 6 H 5 PC1 B C 6 H 5 C0NH(CH a ) s Cl 

^ CHg-OH/ -* ^CHg-CHs" -> 

-> C 6 H 5 CONH(CH 3 ) 5 0N aq« KOH C 6 H 5 C0NH(CH 3 ) e COOK P, Br 3 

C e H 5 CONH(CH 3 ) 4 CHBrCOOH NH 4 OH H + , H 3 NH 3 (CH 3 ) 4 CHNH 3 COOH 

— > — » 

dl- Lysine has been synthesized from acrolein (4), but this 
method has not proved practical. 

CH 3 =CHCHO dry HC1 01CH 3 CH 3 CH(OC S H 5 ) 3 CN(CH 3 ) 3 CH(OC 3 H B ) 3 

— > — > 

C 3 H B OH 

Na C G H 5 C0C1 H + ,H 3 

-» NH 3 (CH 3 ) 3 CH(0C 3 H 5 ) 3 -> -+ C s H 5 C0NH(CH 3 ) 3 CHO 

C 3 H 5 0H 

C 5 H S N sol'n 
G 6 H 5 CONH(CH 3 ) 3 0KO + 0H s (0OOH) 3 . -+ C s H s CONH(CH 3 ) 3 CH=CHC00H 

C B H n N 

H 3 , cat. C 6 H 5 CONH(CH 3 ) 3 CH 3 CH 3 COOH -» continue according to 

—> v. Braun T s method 



The current laboratory synthesis is that of v.Braun (o) as 
improved by Eck and Marvel (b) and later by G-alat (6)* 

^.-CH S -CH S \ H 3 S0 4 ^CK 8 -CK 3 -C=0 H 3 

CH 3 C-NOH -* CH 3 I -* 

\ GHg-CHg^ ^ CH 3 ~CH 3 -NH 

NH 3 (CH 3 ) 5 GOOH C S H 5 C0C1 S0 3 Cl 3 C 6 H 5 CONH (CH 3 ) 4 CHC1C00H NH 3 

hydrolysi s 

~> NH 3 (CH 3 ) 4 CHNH 3 COOH 

This year three new methods have appeared. One of these 
permits the introduction of C which is desirable for metabolic 

One method involves the Curtius degradation of diethyl-a- 
cyanopimelate (7). 

C 3 H 5 OOCCH 3 CH 3 CH 3 0H 3 Br + CNCH 3 COOC 3 H B -* pH0H s 0H s 0H s CH 3 GOOC s H s 

COOC 3 H 5 

H 3 NNH 3 dihydrazide HONO diazide C 3 H B OH CHCH 3 OH 3 CH 3 CH 3 CNHCOOC 3 H 5 
-* -> -> I)JHCOOC 3 H 5 

HOI, HCOOH CHCH 3 (3H 3 CH 3 CH 3 BH 3 
-> NH 3 

This method does not seem to be of any great preparative 
significance since the yield is low (10$) and no other advantages 
are apparent. 

The cleavage of dihydropyran to 5-hydroxypentanal (8) provides 
the basis for an excellent preparative method (9), 

CH 3 CH 3 

/ \ / \ 

HC1 CH 3 CH 3 NaHS0 3 KCH CH 3 CH 3 

-> I I -> -> I I 

I y J CH 3 CHO 0H 3 CHCN 

x o^ I I 6h 


HO(CH 3 ) 4 CHNH 3 COOH 
hydrolysis 5} 
(NH 4 ) 3 C0 3 ^CO-NH / 

-» HO(GH 3 ) 4 CH | ; ^CO-NH 

^NH-CO ^i Br(CH 3 ) 4 CH | NH 3 

^NH-CO -* 

NH 3 (CH 3 ) 4 CH j 

x NH-CO 

Br(CH 3 ) 4 CH \ 



hydro lysis NH 3 (CH 3 ) 4 CHNH 3 COOH 
— > 






'N(CH 8 ) 4 0H 



This method affords a 40^ yield as compared to 23$ obtained by the 
modified method of v. Braun, Both methods use readily available 
materials and employ reactions which proceed with a minimum of 

14 f 

For the introduction of C into the lysine molecule, 2) -chloro- 
a-acetamidovaleric acid was considered a useful intermediate (10) 
since it could be resolved before the introduction of the radio- 
active carbon. 

C1CH 3 CH 3 CH 3 CHC00H KC *N C 1 4 NCH 3 CH 3 CH 3 CHC00H 


However, all attempts to prepare this intermediate failed. The in- 
troduction of C 14 was achieved by a modified Fischer- Weigert syn- 
thesis using T-chlorobutyronitrile containing a radioactive nit rile 
carbon atom. 


Cl(CH 3 ),Br + KC N 

C1(CH 3 ) S C 4 N 


1. Fischer and Weigert, Ber. , _35, 3772 (1902) . 

Sorensen, Compt. rend, trav . lab. Carlsberg, 6, 1 (1903). 

v. Braun, Ber. , 42, 839 (1909). 

Sugasawa, J. Pharm. Soc. Japan, 550, 1044 (1927) (C.A., _22, 

1572 (1928)). 
Eck and Marvel, J. Biol. Chen., 106 , 387 (1934); Org. Syntheses 

Coll. Vol. II pg. 76. 
G-alat, J. Am. Chem. Soc, 69, 86 (1947). 
G-agnon and Boivin, Can. J. Research, 26B , 503 (1948). 

8. Baldwin, Org, Seminar Abstracts, Univ. of 111., Oct. 22, 1948. 

9. G-audry, Can. J. Research, _26B, 387 (1948). 

10. Olynyk, Camp, Griffith, Woi-slowski, and Helmkamp, J. Org. Chem. 
13, 465 (1948). 




Reported by Robert E. Carnahan 
November 12, 1948 


When alpha haloke tones are treated with a strong base^ a 
carbon chain rearrangement often occurs. 

S*~*. NaOH 
R CCH 3 X GO (CH 3 \ -> HC(CH 3 ) 3 C0 3 Na 

. / R-.0H 3 AC0CH 3 X 


Treatment of the same haloke tone with alkoxides produces the 
methyl or ethyl esters (Route A) and occasionally hydroxy acetals 
corresponding to the original ketone (Route b) . 

The following examples have been observed; (l) C S H 5 CHC1C0CH 3 
i^hen treated with KOH or Na0CH 3 in ether gives potassium or methyl 
hydro cinnamate; 


(2) C S H B CHC1C0CH 3 -» 6 H 5 0H 3 CH 3 COOCH 3 + C 6 H 5 CK0H-C(0CH 3 ) 3 CH 3 ; 

CH 3 0H 

NaOCH 3 

(3) C 6 H 5 CH 3 C0CH 3 C1 *-* C 6 H 5 CH 3 CH 3 COOCH 3 


NaOC 3 H 5 

(4) CH 3 -CBrCH 3 COCH 3 -> CH 3 COHCH 3 C(OC 3 H 5 ) 3 CH 3 + (CH a ) 3 CC0 3 C 3 H 5 

C 3 H s OH 

A number of mechanisms (3) (4) (5) (6) have been advanced for 
this transformation but none of them is completely general. One 
mechanism of interest is the following: 

Ha OR 

Route A -CHX-CC-CH 3 - 

HOR + 


-CH- CO- QH- + 

£< Na 

K_.--0 Na 

NaX + 



H— C CH 




<3 & 

RO. Na 

+ RO 
Na \ 





-,'6; OR 

H-C ~C-H 



\5=0 + 

/ <??Na 










? ; 





-CH 3 -CH~ 









CH 3 - 

The carbonion ion (II) undergoes immediate intramolecular alkyl- 
ation forming the highly strained cyolopropanone derivative (ill).. 
Addition of the sodium alkoxide to the carbonyl group follows to 
give the alkoxide adduct (IV) . This adduct cleaves at ©or @> as 
indicated, to give the intermediates (V) or (Vl) which then abstract 
a proton from the alcohol which was formed (i-Il) to produce the 

Route B. Formation of the hydroxyacetals: " 



-CH 6-CHg- 




. (IX) 

\ OR 

-CH— -C-CH 3 - 
X O y 


NaBr + 

-£JH-$-CH 8 
•0: OR 





-CHOHC(OR) 3 CH 3 - + NaOR 

The following statements offer evidence for Routes A and B. 

Route A. (l) This mechanism satisfactorily explains the fact that 
C 6 H 5 CHC1C0CH 3 and its isomer C S H 5 CH 3 C0GH 3 C1 give the same ester. 

Route B. (1) The rearward attack by the alkoxide ion on the 
ethylene oxide intermediate (X) is reterded by using a larger or 
secondary alkoxide ion. 

The location of the halogen atom or either side of the cc 

carbon atom of the ketone is inmate rial. 

able to form a 

hydrogen on one of the carbon atoms. 

inactive and does not rearrange. 

or a 

However, the ketone must bt 
a-Chloroisobutrophenone is 

three membered ring, which requires at least one 

?'■•'( r\ 




The above mechanism does have some difficulties; (l) It fails 
to explain why cc-haloacetone does not undergo rearrangement but 
gives the normal metathesis product when treated with sodium 
methoxide and methanol. Experimental evidence indicates that a 
branched chain is necessary in order for cc-halo alphatic ketones to 
undergo the above rearrangement; (S) The point of cleavage of the 
proposed cyclopropanone intermediate does not follow a definite 
pattern. For example, isopropyl propyl ketone and isobutyl propyl 
ketone cleave between the tertiary carbon atom and the carbonyl 
group in the cyclic derivative from the former compound and between 
the secondary carbon atom and the carbonyl group in the cyclic 
derivative from the latter compound. The corresponding esters of 
isopropyl but y rate and dimethylbutyl acetate are produced in 
excellent yields; (3) Routes A and 3 do not explain the evidence 
that the ester rearrangement is favored when solid alkoxide and 
ether are used, nor do they explain the fact that the hydroxy acetal 
predominates when alkoxide and alcohol are used as the reagents. 


1. Richards, G-. , Compt. rend 200, 1944 (1935). 

2. Bergmann and Miekeley, Ber. J54, 802 (1931). 

3. McPhee, W. D. and Klingsberg, E., J. Am. Chem. Soc. , J36, 1132 


4. Aston, J. G-. and G-reenburg, R. H., ibid _62, 2590 (1940). 

5. Aston, J. G-. , Clark, J. T., Burgress, K. A., and G-reenburg, R.B 

ibid 64, 300 (1942). 

6. Favorskii, J. Russ. Phys. Chem. Soc, 26, 559 (1894), 

J. Prakt. Chem. J51, 533 (1894). 

7. Felley, D., Organic Seminar, 1948, May 7th, University of 


8. Aston, J. G-. , and Sacks, A., Abstracts of 109th Meeting A.C.S. 

April, 1946. 

9. Wagner, R. B., and Moore, J. A., Abstracts of 114th Meeting 

A.C.S. , September, 1948. 

Reported by H. A. DeWalt, Jr* 
November 12, 1948 



It is the purpose of this seminar to discuss the preparation 
of simple organo silicon compounds by methods which involve the 
formation of carbon-silicon bonds. 

Use of Organozinc Compound s* The earliest method of forming 
carbon- silicon bonds involved the use of zinc alky Is as indicated 
by the equation: 



sealed tube 

SIR* + 22nCl 


where R is either alkyl or aryl. Since such a substitution proceeds 
in a stepwise manner, a mixture of substitution products containing 
from one to four groups is obtained. However by controlling the 
molar quantities of reagents, it is Possible to make one of the 
products predominate. Obvious disadvantages of the method are the 
sealed tube conditions, the preparation and handling of the highly 
flammable and toxic zinc alkyls and the separation of products. 

Use of Organosodium Compounds . It will be noted that this 
method is essentially that of the Wurtz reaction. 


+ 4RC1 + 8Na 




The method finds its g eatest use in the preparation of tetra- 
alkyl- and tetraaryl-silanes. 

Use of G-rignard Reagents . This represents , the most universal 
method of preparing simple organosilicon compounds. 



SiCl 4 
SiHCl 3 

R 3 SiCl 
R 3 SiH 


It is difficult to prepare tetraalkyl- or tetraaryl-silanes 
by this method. If an excess of the G-rignard reagent is used, 
the trialkyl or triaryl product usually predominates. Tetra- 
phenylsilane may be prepared by this method only if the reaction 
mixture is heated to 160-180° for 3-4 hours. However if R is a 
group such as isopropyl, only two groups may be substituted. 

Recently G-rignard reagents containing silicon have been 
employed as indicated by the following equations: 

(<CH 3 ) 3 SiCH 3 MgCl + (CH 3 ) 3 SiCl 
(CH 3 ) 3 SiCH 3 MgCl + (CH 3 ) s 3iCl 3 

[(CH 3 ) 3 Si] 3 CH 2 (5) 

* [(CH 3 ) 3 SiCH 3 ] 3 Si(CH 3 ) 3 (6) 



If silicon tetraf luoride is used, the main Droduct is R 3 SiF 
along with smaller amounts of R 4 3i. No mono- or di- substituted 
products are formed. 

Use of Organolithium Compounds . It has been found advantage- 
ous to use organolithium coumpounds when the corresponding G-rignard 

reagents give low yields or fail. 

SiCl 4 + 4RLi 


+ 4LiCl 


In the case of the simple alkyl- or aryl-lithium compounds, 
the above reaction goes quite readily. However introduction of 
the fourth R group is slow and sometimes impossible with some 
sterically hindered lithium compounds. It should also be noted 
that trialkyl- and triaryl-silanes react with organolithium 

R,SiH + RLi 

R 4 Si 



The following table indicates the products obtained with 
various lithium derivatives: 

Starting Material 

SiCl 4 

SiCl 4 

SiHCl 3 
_i-Pr 3 SiH 
_i-Pr 3 SiH 
J.-Pr 3 SiH 

SiCl 4 

SiGl 4 

SiCl 4 







CeH 5 


R 4 Si 

R 3 SiCl 

R 3 SiH 

No reaction 

i-Pr 3 SiCH B 
No reaction 

R 4 Si 

R g SiCl 2 

R 4 Si 

% Yield 





Addition of SiOl A and SlH01 a to Unsaturated Compounds. 
Silicon tetrachloride can be made to add to ethylene or other 
olefinlc hydrocarbons under conditions of high temperature and 
high pressures in the presence of A1C1 3 . 

CH jj— OH • 


HC=CH + SiCl 

ClCH 3 CH 3 SiCl; 
ClCH=CHSiCl 3 


The addition of trichlorosilane to olefins in the presence 
of peroxides is quite general. 

RCH=CH 3 + SiHCl 3 peroxide RCH ? CH P SiCl 


Silicon tetrachloride will not add in an analogous manner. 

•. < 



Direct Method. 

direct uaion of 
with metallic silicon in the presence of 
a good method for preparing dialkyl- and 
Both this method and the G-rignard method 






alkyl or aryl halides 
finely divided copper is 
are being used commercial- 


If R is phenyl, 

finely divided silver has been found to be more 


1. Aid rich, Organic 

2. G-ilman and Clark 

3. G-ilman and Clark 

4. G-oodwin, Baldwin 

5. Rochow, Burkhard 

6. Whitmore, et al , 

7. Whitmore, et al. 

8. Whitmore, et al. 

9. Whitmore, et al. 
10. Whitmore, et al . 
11.. Whitmore, et al. 

Seminar Abstracts, Jan. 23, 1948. 

J. Am. Chem. Soc. , _68, 1675 (1946). 
, ibid., _69, 1499 (1947) . 
and McGregor, ibid., 69, 2247 (1947). 

Booth and Hartt, Chem. Revs., 41, 99 (1947), 

J. Am. Chem. Soc, 68, 1380 (1946). 

ibid., _7J5, 484 (19481. 

ibid., 69, 980 (1947). 

ibid., 69, > 188 (1947). 

ibid., 70, 2872 (1948). 

ibid., 70, 2876 (1948). 

Reported by H. W. Hill, Jr. 
November 19, 1948. 



Monoolefins have been shown to react with maleicanhydride, 
sulfur trioxide, formaldehyde, and azodicarboxylic ester to form 
1-1 adducts in which the original olefinic bond has migrated to an 
adjacent position. B-Pinene (i), for example, reacts with formal- 
dehyde at 180° to give "nopol" (II) in almost quantitative yields. (l 

,9 s • 

CH 2 =0 

— > 



As a result of various patent claims describing the addition 
of hydrocarbons with isolated double bonds to maleic anhydride, 
Alder (S) undertook a systematic study of the behavior of simple 
olefins toward this reagent. At elevated temperatures* 200° and 
under pressure, ethylene gave no simple addition product with 
maleic anhydride. However, propene, 2-butene, isobutylene, 
n-hexylene, n-heptylene, cyclopentene and cyclohexene reacted to 
give 1-1 adducts. In general, the yields increased with increasing 
molecular weight of the olefin. A few representative reactions are 
shown. Propene reacts to give allyl succinic anhydride and iso- 
butylene gives 2-methylallyl succinic anhydride. 

CH 3 =CH-CH 


•I > 





OH a - C 


OH a =0- OH. 




CH-C sX 

CHp=C-CH P -CH-0 


CH 3 I /0 

CH 3 -C V 

Alder regarded this reaction as a typical substitution in the 
allyl position, the allyl H atom migrating to saturate the maleic 
residue. He termed it a "substitution addition" reaction, However 
allyl-benzene reacted with maleic anhydride to give 3-phenylallyl 
succinic anhydride. 



"\.-CH 3 -CH=CH 9 



ii y 



^ - CH= CH- CH 3 - CH— C^ 

CH 3 -C 

Such a product is not in accord with the postulated substitution 
at the allyl position. 

Azodicarboxylic ester can replace maleic anhydride in these 
reactions. The reaction follows the same course as with maleic 
anhydride, with the advantage that it can usually be effected at 
room temperature. (2) 

The scope of the reaction was extended somewhat b; 
who treated several monofilef inic esters, methyl undecyl 
methyl oleate, with maleic anhydride at 200-250° to ob' 

The scope of the reaction was extended somewhat by Ross (3) 

'lenate and 
obtain good 
yields of the simple 1-1 adducts. With methyl oleate, an isomeric 
mixture is formed by the attachment of the maleic residue to C 9 or 
lo and the remaining double bond shifting to the C^-Cn or C 9 -C B 
positions respectively of the octadecanoic acid chain. From these 
examples, it is apparent that maleic anhydride will react readily 
whether the ethylenic linkage is terminal or toward the center of 
the chain. 

The essential similarity in the reactions of maleic anhydride 
with monodlefins and conjugated dienes is worthy of note. The 
reaction with monottlef ins however usually requires a temperature 
of 200° or more. 

Concomitant with the introduction of dioxane sulfotrioxide 
as a new sulfating or sulfonating agent, Suter et al (4,5,6,?) 
have treated a series of monottlefins with this reagent to obtain, 
in many cases, unsaturated sulfonic acids as the major products. 
Several of these reaction products have been tabulated below. The 
reactions were run in ethylene chloride with temperatures ranging 
from 0-20° G. 


Principal Products 


JH 3 
CHa-C CH 3 

<?H 3 QH 3 
CH 3 =C-CH 3 SC 3 H CH 3 -C-CH 3 S0 3 H 


CH 3 
C 9 H B -CH 3 -C=CH 3 

C 6 H 5 -CH=C-CH 3 S0 3 H •C fl H 5 CH 3 -C-CH 3 S0 3 H 


?H 3 
CgHg-C— CH 3 

<?H 3 S0 3 H m 3 
C 6 H 5 -C=CHS0 3 H C 6 H 5 -C— CH 3 S0 3 H 


C 6 H B ~CH=CH-CH 3 

CeH 5 CH=§-Cg 3 


C S H 5 -CH 3 -CH=CH 3 

c 6 h 5 ch=ch-ch 3 s0 3 h c s h 5 ch 3 -ch-ch 3 s0 3 h 



.( :>:!'( ■■ 

- *. . . 



Arnold and Dowdall ( 
tained by the reaction of 
paraformaldehyde, maleio 
case, the reaction is ace 
double bond into the six 
formation of these adduct 
plex, which is formed by 
the terminal carbon atom 
necessarily followed by a 

8) have characterized the products ob- 
methylenecylohexane and the reagents 
anhydride, and sulfur trioxide. In each 
ompanied by a shift of the exo cyclic 
mernbered ring. These men regard the 
s as occurring via transient cyclic com- 
a simultaneous attack of the reagent at 
of the olefin and an ct-methylenlc group, 
shift of the double bond. 
..S0 3 
CH^ J$ CH 3 50 3 H 






The unusual reactivity shown by the isobutylene type olefins might 
be attributed partly to hyperconjugation. 

R-CH 3 , 

'C=CH 3 


H-CH 8 

/J-CH 3 







(1943); C.A., 37, 

Bain, J. Am. Chem. Soc, 68, 638 (1946). 
Alder, Posher and Schmitz, Ber., 76B , 27 
4700 (1943) . 

Ross, Gebhart and Gerecht, J. Am. Chem. 3oc, j38, 1373 (1946). 
Suter, Keifer and Evans, J. Am. Chem. Soc, 60, 538 (1938). 
Suter, Malkemus and Archer, J. Am. Chem. Soc, 63, 1594 (1941) 
Suter and Truce, J. Am. Chem. Soc, 66, 1105 (1944). 
Bordwell, Suter and Weber, J. Am. Chem. Soc, 67, 827 (1945). 
Arnold and McDowdall, J. Am. Chem. Soc, 70* 2590 (1948). 

Reported by H. DeWald 
November 19, 1948 



The success achieved in combating insect pests with DDT and 
its analogs has stimulated a search for other organic compounds 
with practical insecticidal properties. This report is concerned 
mainly with the few of many insecticides tested which have shown 
sufficient promise to be of economic importance. 


A. BENZENE HEXACHLORIDE (l, 2, 3, 4, 5, 6-hexachlorocyclo- 
hexane). Although benzene hexachloride has beentajovn for a long 
time (l), its insecticidal properties were not discovered until 
1941-1942 (2)« It was developed in England during World War II 
and found to be toxic to a wide variety of insects (l). 

The principle process used in making benzene hexachloride 
consists of adding gaseous chlorine to benzene in the presence of 
light (3). Technical benzene hexachloride contains five stereoiso- 
mers (1,5) (a, p, jT) £T and f ) but only the gamma isomer has 
appreciable insecticidal activity. Normally only 10-12$ active 
isomer is obtained and so processes for concentrating the gamma 
isomer by extraction of the technical product x^ith cyclone xane, 
trichloroethylene, chloroform, toluene, xylene, etc. are used (2,4). 
It has recently been reported (6) that the chlorination of benzene 
in methylene chloride solution under the influence of a peroxide 
catalyst produces 18$ gamma isomer. The configuration of only the 
beta isomer has been established with certainty (1,7,8). 

A serious deterrent to the more widespread use of benzene 
hexachloride is the objectionable musty odor of the technical 
material. This odor is due to an impurity which can be only 
partially removed by a variety of treatments (2, 2a, 9, 10, ll). 

The problem of finding analytical methods for gamma benzene 
hexachloride was difficult due to the chemical similarity of the 
stereoisomers. Differences in the rates of dehydrochlorina tion 
of the isomers is the basis of one method (8, 12). Other useful 
methods for analysis are based on infrared absorption spectra (13), 
cryoscopic measurements (14), partition chromatography (15, 16) 
and polar ogra^hic methods (17). 

B. CHLORDANE. Chlordane is a chlorinated hydrocarbon, 
Cio^s^lp, which was found to be toxic to a variety of insects by 
Kearns fl8) in 1945. Its action is similar to DDT and benzene 
hexachloride and it shows much promise in controlling some insect 
species (19). 

The active constituent of chlordane is 1,2.4,5,6,7,8- 
octachloro-4,7-methano-3a,4, 7,7a-tetrahydroindane (l) which is made 
(2b) by adding chlorine to one of the double bonds of the Diels- 
Alder adduct formed from perchlorocyclopentadiene and cyclopenta- 

* '■■' 

'. % : \ 





\S H ^y^ H 

CI Hg 


C. CHLORINATED TERPENES. Chlorination of camphene (2.) 
to a chlorine content of 67-9$ produces a' material with the 
empirical formula Ci Hi Cl 8 which has been found toxic to a con- 
siderable number of household and agricultural insect pests. 

It has been shown (20) that cls-1, 8-dichloroparamenthane 
and, to a lesser extent, bornyl chloride ea&Sftit insectioidal 
activity. These compounds are made by treating a-or p-pinene with 
hydrogen chloride. 


A. TETRAETHYL PYROPHOSPHATE. The Germans developed a 
substitute for nicotine in aphid control during the last war which 
they termed Bladan (21, 22,23). The active principle was thought 
to be hexaethyl tetraphosphate, a product obtained by the reaction 
of triethyl orthophosohate with phosphorous oxyohloride (24, 25) 
or phosphorous pentoxide (26) at 150°. It has been shown (27, 28) 
however that the insecticidal principle is tetraethyl pyrophosphate 
(il) which is produced in these reactions to the extent of about 
15-8$ along with ethyl metaphosphate* Approximately 40$ tetraethyl 
pyrophosphate is produced by increasing the proportion of triethyl 
orthophosphate (2). Toy (29) has described the preparation of pure 
tetraethyl pyrophosphate in good yield by the controlled hydrolysis 
of diethyl chlorophosohate in pyridine. 

2(Et0) 3 P0Cl + H.,0 + 

2C 5 H 5 N --> (EtO) 3 P0 3 P0(0Et) 


Tetraethyl pyrophosphate is a highly toxic compound to 
both insects and warm-blooded animals, but hydrolyzcs rapidly to 
non- toxic products (27). 

B. PARATHION. Another insecticide developed in Germany 
during the war is 0, O-diethyl-0-p-nitrophenyl thiophosphate (III) 
which was designated E-605 by the Germans (30, 31, 32, 33) and was 
given the name Parathion in this country. 

Parathion is synthesized by the following sequence (33). 

130° NaOEt £-N0oC 6 H 4 0Na 

PC1 3 + S -> PSC1 3 -* (ET0) 3 PSC1 A 

2 hr EtOH C 6 H 5 C1, 130 

(EtO) 3 P0/ 




; : •:; 

-3- 80 

C. OTHER PHOSPHATES. A number of compounds closely re- 
lated to parathion were prepared and tested by G-erman chemists 
(51, 33). 

A series of 46 organic phosphates and phosphites have 
been tested by Ludvik and Decker (34) against various aphids« 
These workers found that some of the pyrophosphates, triphosphates 
and tetrapyrophosp hates tested were superior to nicotine as 


A * BIS (p-CHLOROPHENOXY) METHANE. This compound has been 
shown to be a highly effective miticide (35). It is prepared by 
treating p_-chlorophenol with an equimolar quantity of sodium in a 
solvent such as absolute ethanol and subsequently treating the 
phenolate dispersion with methylene chloride (36). 

B. l,l-BIS(p-CHLOROPHENYL) ETHANOL. Also termed di(p_- 
chlorophenyl) methyl carbinol (DMC), I,l- bis (p-chlorophenyl) ethanol 
is another very effective miticide. This compound is made from 
4,4 f ~dichlorobenz.ophenone and methyl magnesium bromide (2d). 

C. PYRETHRIN SYNERGISTS. The pyrethrins, which are the 
active alkaloids of pyrethrum, possess a more rapid paralytic 
effect on insects than any synthetic organic insecticide known. 
Pyrethrum is less toxic to warm-blooded animals than the synthetic 
insecticides, but relatively expensive. It has been found that 
compounds with a methylene dioxyphenyl grouping increase the 
toxicity of the pyrethrins (2, 37, 38). Several of the more 
important of these activators or synergists are listed below. 

1. PIPERONYL BUTOXIDE (39, 41) contains 80$ of cc~[ 2- (2-butoxy- 
ethoxy)ethoxy]-4, 5~methylenedioxy-2-propyltoluene (ivj , 

j, ^CH S CH 3 CH 3 

oh; u xP _^ oh 3 och 3 ch 3 ock 3Ch 3 oc 4 h 9 IVf 

2. PIPERONYL OYCLONENE (39, 40) (V) is obtained by condensing 
ethyl acetoacetate with cy clohexyl-3, 4-methylenedioxy styrylketone . 



^<^" "^>CH=CHOG 6 Hi 3 + OH 3 OOOH 3 C0 3 C S H 6 -> CH-C=C) 5 

CH 3 V/ / V>CH ,CH 30$ 


/0 > . / CH 2 C=0 V. 

CH 2 N <^ ^>CH CH 50^ 



3. PROPYL IS0ME(2c, 42) (Vl) is prepared by a Diels- Alder 
reaction between n-propyl maleate and igosafrole. 

H 3 H 

/ 0-V / N\-CH=0HCH 3 CHC0 3 C 3 H 7 

CH ; 















+ CHC0 3 C 3 H 7 


N-ch s 

.-C0 3 C 3 H 7 

H C 3 C 3H 7 


Slade, Chem. Ind . , 40, 314 (1945) . 

Haller and Busbey, Abstracts, A.C.3. meeting, Sept. 1948, 4J; 

Ind. Eng. Chem., In Press. 

Burrage et al, Brit, patent 592,677 (Sept. 27, 1947). 

Hyman, Mexican patent 45,398 (Mar. 19, 1947). 

Synerholm, U. S. patent 2,431,845 (Dec. 2, 1947). 

Ruthruf f et al, U. S. patent 2,430,586. 

Hardie, U. S. patent 2,218,148; 0. A,, 35, 1071 (1941). 

Cooke et al, Brit, patent 586,439; C. A. 41, 7641 (1947); Hay 

et al, Brit, patent 586,442; C. A. 41, 76"4T (1947). 

Van der Linden, Ber. , 45, 236 (1912TT 

Kauer, Britton and Alquist, Abstracts A. C. S. meeting, Sept. 

1948, 65L. 

Dickinson and Bilic'ie, J. Am. Chem. Soc, J50, 764 (l928).„ 

Cristol, J. Am. Chem. Soc, _69, 338 (1947). 

Burrage et al, Brit, patent 586,468; C. A. 41, 7643 (1947). 

Webster et al, Brit, patent 586,434; C. A. 41, 7642 (1947). 

Brit, patent 586,464; C. A« 41, 

Ind. Ltd, 

Gray an&Tm p. Chem . 

7642 (1947). 

LaClair, Anal. Chem., 

Daasch, Ind. Eng. Chem., Anal. Ed. 

Bowen and Pogorelskin, Anal. Chem. 

Ramsey and Patterson, Assoc, Off. Agr. 

20, 241 (1948). 


> 19> 

779 (1947). 
346 (1948) . 
Chem., 29, 337 (1946). 

38, 661 (1945); 


Aepli et al, Anal. Chem., _20,^ 610 (1948). 

Dragt, Anal, Chem,, 20 , 

Kearns e_t al, J. Econ. Ent. 

C. A., 40, 2917 (1946). 

Bussart, Soap and Sanit. Chem., 

Desalbres and Labotut, Chim. et 

42, 2719 (1948). 

Hall, U. S. Dept. Commerce, OTS, 

Kilgore, Soap and Sanit. Chem., 

Smadel and Curtis, U. S. Dept. Commerce, ' OTS, PB 240 (1945). 

Schrader, G-er. patent 720,577; C. A., 37, 2388 (1943). 

Schrader, U. 3. patent 2,336,302; G.A., 38, 2966 (1944). 

Woodstock, U. S. patent 2,402,703; C. A., 40, 5444 (1946). 

Hall and Jacob son, Ind. Eng. Chem., 40, 694 (1948). 

Harris, Agr. Chem., 2 (10) , 27 (194777 

Toy, Abstracts A. C. S. meeting, Sept. 1948. 25L. 

Martin and Shaw, BIOS Final Rpt, 1095 (1946). (U. S. Dept. 

Commerce, OTS, PB-L78244). 




(8), 126 (1948). 
, _58, 443 (1947); 

252 (1945) . 

(12), 138 (1945). 


• - ' - < 

f ■ 
» * 



31. Mumford and Perren, BIOS Final Rpt. 714 (no date). 
(U. 3. Dept. Commerce, OTS, PB-L87923R) . 

32. Tanner et al, BIOS Final Rpt. 1480 (1946). (U. S. Dept. 
Commerce, OTS, PB-81638). 

33. Thurston, FIAT Final Rpt. 949 (1946). (U. 3. Dept. Commerce, 
OTS, PB-60890). 

34. Ludvik and Decker, J. Econ. Ent., 40 , 97 (1947). 

35. Jepson, J. Econ. Ent., 39, 813 (1946); C. A., 41, 3575 (1947). 

36. Moyle, U. S. patent 2,330,234; C. A., 38. 1316~~T1944) . 

37. Eagleson, Soap and Sanit. Chem., 18 (12), 125 (1942) . 

38. HalLeret al, J. Org. Chem., 7, 183~~Tl942) . 

39. Wachs, Science, 105, 530 (1947). 

40. Hedenburg and Wachs, J. Am. Chem. , SO c. , 70, 2216 (1948). 

41. McAlister et al, J. Econ. Ent., 40, 906T1947). 

42. Synerholm and Hartzell, Contrib. Boyce Thompson Inst., 14 (2), 
79 (1945); C. A., 40, 669 (1946). 

Reported by G-eorge I. Poos 
November 19, 1948 


• r 




This paper describes the use of potassium permanganate in free 
radical reactions producing arylacetic acids. 

G-riehl (l) has prepared cc-naphthylacetic acid by adding 
powdered potassium permanganate slowly to an excess of naphthalene 
in boiling acetic anhydride. Eighty percent of the naphthalene was 
recovered, and the yield of a-naphthylacetic acid was 66^ of the 
naphthalene consumed. No by-products were isolated. Hydrogen 
peroxide and diacetyl peroxide were used also, but the yields seemed 
better with permanganate. 

G-riehl termed the reaction an oxidative dehydrogenation and 
represented it by the following equation: 

CH 3 C0 3 H 

+ (CH 3 CO) B + [0] 

CH-CO a H 

In view of recent studies of free radical reactions (2,3,4) 
and the concept that certain oxidations involving metallic oxidiz- 
ing agents proceed via free radicals (5), it seems reasonable 
that the course of this reaction could be represented by a series 
of steps involving free radicals, such as the following* 

(1) 4(CH 3 C0) 3 + KMn0 4 - 

(2) CH 3 C00. -4 .CH 3 + C0 3 



•CH 3 

• GH « 

+ (CH 3 C0) 2 


5CH,C00* + CH 3 C0 2 K + (CH,C0 3 ) 2 Mn 


•CH 3 C0 3 C0CH s 



•CH 3 C0 3 C0CH 3 

CK 3 C0 3 C0CH 3 




CH 2 C0 3 COCH 3 

+ H< 

: i .. 



Step (l) is more easily explained if traces of acetic acid are 
assumed to be present. It is well known that the permanganate ion 
is unstable toward reduction by water in acid solution; the reaction 
is regarded as proceeding through the hydroxy 1 radical, the per- 
manganate ion being reduced stepwise to manganous ion (6). By 
analogy, with acetic acid, step (l) may be written as 

2CH 3 C0 S H + [0] -> 2CH 3 C0O + [H 2 0] 

[H 3 0] + (CH 3 C0) 3 -> 2CH 3 C0 3 H 

with acetic acid being reformed as it is oxidized. 

Uses of the Reaction . This synthesis of a-naphthylacetic acid, 
being a decided improvement over the best previous method (7), 
is of possible commercial interest, since a-naphthylacetic acid 
has considerable importance as a plant-growth regulator. In a 
similar manner, G-riehl produced p-hydroxy-cc-naphthylacetic acid 
from |3-naphthol, ]>-phenylphenylacetic acid from biphenyl, and o- 
methoxyphenylacetic acid "from anisole. Aralkanes reacted in a 
different manner, producing dimers of the sort encountered by 
Kharasch (3) by means of decomposition of diacetyl proxide in 
alkylbenzenes. For example, ji-propylbenzene, treated with acetic 
anhydride and permanganate, gave 3,4-diphenylhexane. It is inter- 
esting that permanganate with acetic anhydride alone gave a 40^f 
yield of succinic anhydride. Kha'ranoh (?) obtained b 50^ yield of 
ac?cU niC aCld b,V allowln S diacetyl peroxide to decompose in acetic 

The use of permanganate to form arylacetic acids is obviously 
limited to compounds without groups easily oxidized. Also, the 
reaction must be carried out using insufficient permanganate, 
since the products are subject to further oxidation by permanganate 
Its advantage is twofold; the reagents are cheap and the handling 
of peroxides is eliminated. 


1. G-riehl, W. , Chem. Ber., J30, 410 (1947), 

2. Kharasch, M., et al,, J. Am. Chem. Soc. , _65, 15 (1943). 

3. Kharasch, M. , et al., J. Org. Chem., 1_0, 401 (1945). 

4. Waters, W. A.,~ lr The Chemistry of Free Radicals", Clarendon 
Press, Oxford (1946). 

5. Waters, W. A., Ann. Reports, 42, 130 (1945). 

6. Latimer, W. M. , "Oxidation Potentials", Prentice-Hall, Inc. 

7. 3hmuk, A., and G-useva, A., J. Applied Chem. (U. S.8.R.), 14 , 
1031 (1941); C. A., 39, 4069 (1945) . 

Reported by Harold J. W a tson 
December 3, 1948 


' t 



Spiro compounds are composed of two or more rings, two of 
which have a single atom in common. Splranes, then, are saturated 
spiro hydrocarbons. These compounds are named by adding the prefix 
"spiro" to the name of the normal aliphatic hydrocarbon of the same 
number of members. 

Relatively few spiranes have been prepared. The following 
synthesis, taken from the last seminar on this subject (l), is 
typical of the syntheses used up to 1941: 


HON ' / 


S CH s C0 3 C s H B 


NaOC s H 5 

Cl(CH 3 ). 3 C0 3 C 2 Hc 

HC0 3 C 3 N s 

v JCCN(CH 3 ) 3 C0 3 C 3 H, 

(1) H 3 S0 4 
-C0 2 

C0 3 C 2 He 

GH (CH 3 ) 3 C0 3 C 3 H 5 
C0 3 C 3 H 5 


(2) esterify \< 


C0 3 C 2 H 5 

00 3 H 


H 3 S0 4 

C0 3 C 3 H 5 



HC1 y 

(2) heat X. 
Ca(0H) 3 

•— — — t 

In 1941 Marvel and Brooks (2) reported the synthesis of spiro 
[4.5] decane by the following series of reactions' 
HO (CH 3 ) 3 CH=CH 3 

+ BrMg(CH 3 ) 3 CH=CH 3 

trace I 3 


r /^S-(CH 3 ) 3 CH=CH 3 84$ H 

3 oU 4 



H 3 
— > 


(exact position 
of double- bond 
not determined) 

.«'. ' ••'• \ ■■ 

■ . . 




5-Methyloyclopentanone gave the analogous 3-methyl spirane . The 
spirane I gave no reaction with bromine and on dehydrogenation over 
platinum or palladium on charcoal gave 33-5$ naphthalene. The 3- 
methylderivative on similar treatment gave 31$ 2-methylnaphthalene . 
Neither spirane could be dehydrogenated with selenium. 

Spiro [4,4] nonane has been synthesized by Zelinskil and 
Elagina (3) in the following manner: 




Al(Hg) x 


'CH 3 CH 3 CH 3 COOH 

1:1 H 3 S0 4 
— » 

Ba(OH) 2 



Platinized C 

160 0_> 





NH 4 V0 3 
HN0 3 



The final reduction took place in 68.7$ yield to give a colorless, 
mobile liquid of terpene-like ©dor. On hydrogenation over platin- 
ized charcoal (4) mixtures of isomeric nonane s and cyclopentane 
homologs were obtained. In a carbon dioxide atmosphere, treatment 
with platinized charcoal at 305-10° gave o-ethyltoluene. It is 
hypothesized that bicyelo[4. 3.0] nonane is an intermediate in this 

Perhaps the most interesting spirane currently being investi- 

gated is spiropentane, CH 3 CH 

CHo CH. 

The obvious method of prepar- 

ing this simplest spirane, by tre 
bromide with zinc, has been belie 
product (5). However in 1944 Mur 
means of Raman spectra, that the 
aqueous methanol contained an une 
acetamide as the solvent, togethe 
iodide and sodium carbonate, cond 
ment (7), increased the yield of 
investigation (8) indicated that 
spiropentane. It has now been sh 
carried out successfully in ethan 

atment of pentaerythrityl tetra- 
ved not to give the desired 
ray and Stevenson (6) showed, by 
oroduct of this reaction in 
xpected component . Use of molten 
r with the addition of sodium 
itions unfavorable for rearrange- 
this component to 40$. Further 
this compound was the desired 
own (9) that the reaction may be 
ol solution. 

Hydrogenation of spiropentane (10) gives a mixture of neo- 
pentane, dimethylcyclopropane and 2-methylbutane. ^o ethylcyclo- 
propane or n- propane was isolated, 

A recent patent states that 5 to 25$ spiropentane in gasoline 
hydrocarbons gives an aviation fuel of improved performance (ll) . 


, •• • '. 

:• 1R 



Recently a new synthesis of other spiranes, making use of the 
reaction utilized for the preparation of spiropentane, has been 
devised (12). The preparation of spiro[2. 5]octane will illustrate 
this method: 


CH 3 , . 0H 3 OH 

KOH XL ./ CH 9 OH 





H 3 ^ CH 3 OH PEr 9 y . CH 3 Br Zn 

Ni N ./ CH 3 0H X S" CH 3 Br C 3 H 5 0H \. 


4-Methylspiro[2.5]octane has also been synthesized, starting with 
orotonaldehyde rather than acrolein. Hydrogenation of III gives 
1,1-dimethylcyclohexane, the 4-methyl analog giving 1,1, 2- 1 rime thy 1- 
cyclohexane. From these reactions and the results of the hydro- 
genation of spirooentane (10), the generalization has been made 
(12) that cleavage of these compounds seems to occur exclusively 
at the bond opposite the gem - substituted carbon atom. 


1. Wearn, Organic Seminar Abstracts, University of Illinois, 
14 May 1941. 

2. Marvel and Brooks, J. Am. Chem. Soc, _63, 2630 (,1941). 

3. Zelinskii and Elagina, Compt. rend. acad. sol. UR3S _49, 268 

(1945) [0. A, 40, 6058 s (1946)]. 

4. Zelinskii and Elagina, ibid., _52, 227 (1946) [C.A. 41, 3769b 

5. Whitmore, "Organic Chemistry", D. VanNo strand Cc.Inc, New York 
1937. p. 632. 

6. Murray and Stevenson, J. Am. Chem. Soc, 66, 314, 812 (1944). 

7. Hass, McBee, Hinds and G-luesenkamo, Ind. Ehg. Chem. 28, 1178 

8. Donohue, Humphrey and Schomaker, J. Am. Chern. Soc, 67, 332 


9. Slabey, Nat'l. Advisory Comm. Aeronautics, Tech. Note No. 1023 

(1946) [C.A. 40, 3729 3 (1946)]; J. Am. Chem. Soc. 68, 1335 

10. Slabey, J. Am. Chem. Soc. _69, 475 (1947) . 

11. McCulloch, US Pat 2,411,582 (26 Nov. 46) [C.A. 41, 1091c (1947); 

12. Shortridge, Craig, G-reenlee, Derfer and Boord, J. Am. Chem. 
Soc, 70, 946 (1948). 

Reported by William R. Miller 
December 3, 1948 

,• ->•: 

■ r > 

A > ■■' { '■■■ 




The term Stobbe Condensation is applied to those alkoxide- 
catatyzed reactions between ketones and diethyl succinate which 
result in the formation of dibasic unsaturated acids. 


OR OPU- ) 

%H 3 






3 n B 



3 n 5 



£=pCH 3 COOH 


R 1 ' COOCUH, 

'3 n 5 


Since the previous report (l) on the Stobbe Condensation, 
several papers have been published dealing with both the modifica- 
tion and extension of the reaction (3,4,5,6). 

Modified Stobbe Condensation 

In 1945, Johnson and coworkers (2) introduced the use of 
potassium _t~butoxide in Jfc-but#lalcohol in place of classical ssdium 
ethoxide for effecting the condensation. This modified procedure 
has since been extended to the reaction of a number of different 
type ketones (3,4,5). In almost all cases, potassium _t-butoxide 
was found to be a far superior agent for the reaction by giving 
both higher yields and Purer products during shorter reaction times 

Decarbethoxylation Reaction 

Decomposition of the product of the Stobbe Condensation is 
generally brought about by an acid- catalyzed decarbethoxylation 
reaction to give the lactone and/ or the unsaturated acid. Recently 
it has been shown that the ~fi -lactones and unsaturated acids thus 
produced are interconvertible in a true acid-catalysed^f-lacto- 
enoic tautomerism (3). 

^C=(JCH 3 C00H 
R 1 CC0C 3 H 5 



/ \ 



R" •— C~' 





In the polycyclic series it was noted that the decarboxylation 
proceeded more rapidly than the tautomerism, making it possible to 
stop the process before equilibrium was reached. When this was 
done, the lactone was always found in higher proportion than at 
equilibrium, suggesting that the lactone is the precursor of the 
unsaturated acid. 

When the Stobbe Condensation was effected x^ith cy clohexanone 
and the resulting half-ester (or its alkaline hydrolysis products) 
was treated with a strong mineral acid, a high yield of the para- 
conic acid was obtained (5). 





CH 2 COOC 2 H 5 

cH 3 cmc 3 H s 

KOC(CH 3 ) 

3/ 3 

,/\ CH 3 COOH H+ 
-CHCOOC 3 H 5 



tCgH-CH 3 




This evidence (V— >Vl), together with the conclusion that the 
^-lactones are precursors of the unsaturated acids, offers support 
to the hypothesis that paraconic acids are intermediates in the 
decarboxylation reaction. The following mechanism may then be 
postulated for the decarboxylation of the itaconic acid: 

^c=cm* 3 cooH-4 



CCHCH 3 C=r 
R ,x COOH 






^CCH 3 CH 3 C=0 






Synth etic Appli ca t ions 

When the acid-catalyzed decarbethoxylation of the half-ester 
isd by the Stobbe Condensation) is followed by reduction of 

(torrid by the Stobb 
the resulting product, 
Pr^pi on j c fict-d residue 

a method is afforded for introducing a 

at the site of the carbonyl group of a ketone 






.CCH P CHpC=0 



This chain -lengthening process, followed by cyclization, has 
been used in the preparation of a number of synthetic intermediates. 
It has found greatest application in the preparation of fused 5- 
memberad ring polycyclic compounds. The synthetic scheme has re- 
cently been extended to cyclization of the half-esters derived 
from -cyclone xan one and cyclopentanone to give the bicyclic ketones 

Stobbe Condensation 

CH 3 C00H ZnDl 3 
-CHC00C 3 H 5 HOAc,Ac 3 




l|-CHCOOC 3 H 5 




. • 

■ • 



• • 

■ , • - - -. -. . •' ; 


F * 1 


- " ■ • 

.-■} , 




Johnson has applied the process to methyl p-tolyl ketone and 
has utilized the resulting acid in a new, improved synthesis of 
cadalene (4)* The steps in the synthesis are as follows: 






(1) Stobbe Condensation 

(2) Decarbethoxylation 





CCH 3 CH 3 C=0 

CH 3 


— > 





i-CCH 3 CH 3 COONa cat, 
CH 3 


X Q 

\ (1) (CH 3 ) 3 CHMgBr 

(2) Dehydrogenation 



HCH 3 CH 3 C0 3 H 

H 3 

CH(CH £ ) 


B ibliography 

1. Katz, Organic Seminars, University of Illinois, Nov. 14, 1945, 

2. Johnson, Goldman and Schneider, Jo Am. Chem. Soc, _67, 1357 
(1945) . 

3. Johnson, Peterson and Schneider, ibid*, 69, 74 (1947). 

4. Johnson and Jones, ibid., _69, 792 (1947). 

5. Johnson, Davis, Hunt and Stork, ibid., 22.) 3 ° 21 (1948). 

6. Plattner and Buohi, Helv. Chim. Acta, g9_, 1608 (1946). 

Reported by H. Rosenberg 
December 10, 1948 

' . 

',,.-, • ' - 

t ■■ t 

■■'\ «vjXvH .■;.•*? 



The Arbuzov reaction is a very general method for the pre- 
paration of phosphonic acid esters. Recent work has described 
its wide applicability and has outlined its limitations. Arbuzov 
(1) found that phosphite esters could be isomerized to phosphonic 
acid esters by heating with an alkyl hallde, as shown in equation I 

(R0) 3 P + R'X ^> R»_i4oR) s + RX I 

The procedure is sometimes modified, as in equation II, by using 
sodium dialkyl phosphites, which are less expensive than the tri- 
alkyl phosphites and usually permit milder reaction conditions. 

NaOP(OR) 3 + R'X £ R'P^(OR) 3 + NaX II 

1. Monohalogen compounds 

Alkyl halides :- Primary bromides and iodides give best results: 
secondary halides do not react. Tertiary halides of the type 
Ar 3 CBr give excellent yields (2). 

Ethane and methane phosphonates are formed in 9d% yields. 
Though butyl and amyl bromides react much more slowly with tri- 
al^yl phosphites and give poor yields, hexyl and larger bromides 
work more smoothly (3). With further increase in chain length more 
drastic conditions and longer reaction oeriods are required (4). 
However, the entire series has been obtained in yields of 80# by 
use of sodium dibutyl phosphite under mild conditions (5). 

Aralkyl halides :- A variety of substituted benzyl chlorides 
have been converted to phosphonates, both by use of trie thy 1 
phosphite (6) and sodium dibutyl phosphite (7). The chloromethyl 
^r^ P i attach S d Jo the -ring is active, so that the yields by both 
methods are 70-90<£. Kosolapoff ( Q ) has also prepared dibutyl o- 
thienylme thane phosphonate similarly. 

The chlorine atom in 9-chloroacridine was found to b e active 
though to undergo this reaction with trlethyl phosphite. However, 
when sodium dibutyl phosohite was used, a quantitative yield 
of acndone was obtained (9). 

_ Alkyl halides with ot her functional groups present: - Halogen 
substituted carboxylic esters yield phosphonocarboxylic esters in 
limited cases. 

(R0) 3 P0Na + XCH 3 C0 3 Et -» (R©) 3 P-CH 3 C0 3 Et 

(in S ii C inf U l f e sults have been obtained with cc-haloacetic esters 
mC i I* ' P-iodopropionic ester (13), and bromomalonic ester. 
tin ^f ter . r 5; le:Lde ^ its Phosphonate only with trialkyl phosphites 
up, 12;. ^When sodium dialkyl phosphites are used, self couoling of 
the organic ester is an important side reaction. In several in- 
stances, the coupling products along with disproportionate pro- 
ducts are the only organic compounds formed (ll). 

- 2 - 92 

a- Nit rob ro mo compounds do not give the expected a-nltro- 
phosphonates with triethyl phosphite. Instead, ethyl phosphate 
is formed; the nitrobromo compound is decomposed (14). 

Acyl halldes yield a-ketophosphonates (15). 

/P /P P 

(R0) 3 P+ CH 3 0-G1 -> CH 3 C-P-(OR) 3 

2. Polyhalogen compounds 

Methylene halldes : -In general it is possible to obtain two 
principal products from a reaction with methylene halldes, depend- 
ing upon the relative amounts of reactants (16,.17,18,3) i 

Q P 

(RO) 3 P-(CH s ) n -P(OR) 3 and (RO) 3 P- (CH 3 ) n Br, n=l,2,3 

Carbon tetrachloride :- When excess carbon tetrachloride is 
used just one of the chlorine atoms is reactive (9).- 

CC1 4 + (R0) 3 P -> C1 3 C-P-(0R) 3 

3. Ethylene oxide (19,20,21,22) 

The phosphites formed from ethylene oxide and phosphorus tri- 
chloride can undergo intramolecular Arbuzov reactions, 

PC1 3 + H 3 0-CH a -> C1CH 3 CH 3 0PC1 3 -> C1GH 3 CH 3 0P (OR) 3 
^0 y pyr.- ^ a 

^-- ^ Z P 

(R0) 3 PC1 + H 3 0-0H 3 ^-^" C1CH 3 CH 3 P (0R) 3 

However, when arylchlorophosphites, (ArO) 3 PGl, are treated with 
ethylene oxide, the products, ClCH 3 CH 3 0P(0Ar) 3 , do not isomerize 
as expected in a normal Arbuzov reaction; instead ethylene diphos- 
phonic esters are formed.- 

Excess ethylene oxide on PC1 3 or PBr 9 gives the corresponding 
trihaloethyl esters which are difficult to obtain pure since they 
lsomerize on distillation. 

4. Glycols (23,24) 

Cyclic phosphites can be prepared by action of PC1 3 on a glycol 
in presence of a tertiary base* The 5 and 6 membered ring phos- 
phites are stable and are formed in poor yields; those with 7 and 8 
membered rings are formed in good yields and polymerize easily. In 
isomerization, the ring may or may not be opened. This seems to 
depend en the substitution on the ring. 

(a) Pyridine p _ ^ dpy | R| Er OCH^E 

PC1 3 + CH 3 0H jL-tfbM. CH 3 CH 3 0-P-0 -► CH 3 CH 3 0P-0 -* R'-P=0 

6h 3 oh ? ®t % 6l ROH OR ^ OR 

m £jt 3 o 

i,-j m 



CH 3 OMe „ , , ,, r , R»Br 

(b) PC1 3 + 9HOH -> MeOCH 3 CH 2 CH 3 OP-0 -* MeOCH 3 C^ 3 CH 3 QP-0 -> 

CH 2 OH CI OR & 

! ? 

MeOCH 3 CH 3 CH 3 Qp--R' 

5 * Organometalllc reagents OR 

Oacodyl phosphonic esters have been prepared from ethyl alkyl 
arsenous iodides and sodium diethyl phosphite ("35 ). The P-As bond 
in these esters is much weaker than the corresponding P-C bond; on 
attempted hydrolysis to the phosphonic acid with 15^ HC1 at 150° the 
esters decompose. Analagous tin derivatives have been prepared us- 
ing both dialkyl and trialkyl tin halides on trialkyl phosphites 
(36,27), In these esters the P-Sn bond is quite weak since it is 
cleaved by dilute HC1 at room temperature. This preparation has 
also been tried with lead al'-yl halides, but only disproportionation 
products result (S?). 

Mechanism of the Reaction 

The Arbuzov reaction is generally considered to procede through 
ddition intermediate, with subsequent splitting out of alkyl 

an addition 

halide. For example: 

P-OEt + Mel 



EtO Me 

Eto. .0 

K V EtI 

EtO X x Me 

Arbuzov (l), who postulated this mechanism, found evidence 
for such an intermediate by preparing crystals of [CH 3 P(0C G H 5 ) 3 ] I; 
more recently Kamai and Belorossova T25) obtained a quantitative 
yield of [EtBuAsP(0Et) 3 ]l, crystals, m.p, 182°. 

In the intramolecular Arbuzov reaction, the intermediate 
is said to be a cyclic one (19). 

r — n 

[ (CLCH 2 CH 3 0) 3 P-0CH 3 CH 2 3 


1. A, Arbuzov, Chem. Zentr., _77 II, 1639 (1906). 

2. B. Arbuzov and Nikonorov, J. G-en. Chem. (U.S.S.R.), 17, 2139 
(1947); C. A., 42, 4546b (1948). 

3. Ford-Moore and Williams, J. Chem, Soc, 1947 , 1465. 

4. Coebel, U.S. 2,436,181, C. A. 42, 3425g TT948) . 

5. Kosolapoff, J. Am. Chem. Soc, _67, 1180 (1945).. 

6. Lugovkin and Arbuzov, Doklady Akad. Nauk S. S. S. R., 59, 
1301 (1948); C. A., 42, 7265g (1948). 

7. Kosolapoff, J. Am. Chem. Soc, J37. 2259 (l945) . 

8. Kosolapoff, ibid., 69, 2248 (1947) . 

9. Kosolapoff, ibid., 69, 1002 (1947). 

7 * 

k ■ '. 


10. Kosolapoff, Ibid., 68, 1103 (1946). 

11. Chevane and Rumpf, Compt . Rend. 225 , 1332 (1947)-. 

12. Arbuzov and Kamai, J. Gen. Chem.~Tu. S. S.R. ) JJ?, 2149 (1947) ; 
C. A. 42, 4523g (1948). 

13. Arbuzov, Konstant inova and Anzyfrova, Izvest. Akad. Nauk 
S.S.S.R. Otdel. Khim. Nauk, 1946 , 179; C. A., 42, 6315b (1948). 

14. Arbuzov, Arbuzov, and Lugovkin, Bull, acad, sci. U.R.S.S., 
Olasse sci. chim. , 1947 , 535; C. A., 42, 1886h (1948). 

15. Kabachnik, Rossiiskaya and Shepeleva, ibid., 1947 , 163; 
0. A., 42, ,41521 (1948). 

16. Kosolapoff, J. Am. Chem. Soc . , _66. 109 (1944). 

17. Kosolapoff, ibid., '66, 1511 (1944). 

18. Saunders, Stacey, Wild and Wilding, J. Chem. Soc, 1948, 699. 

19. Kabachnik and Rossiiskaya, Izvest. Akad. Nauk. S.S.S.R. Otdel. 
Khim. Nauk., 1946, 295; C. A. 42, 7241f (1948). ' 

20. Kabachnik and Rossiiskaya, Bull. acad. sci. U.R.S. S. , Olasse 
sci. chim., 1947 , 389; C. A. 42, 1558d (1948). 

21. Kabachnik and Rossiiskaya, ibid., 1947 , 95; C. A. _42, 5846e 
(1948) . ' 

22. Kabachnik, ibia., 1947 , 631; C. A. 42, 5845f (1948 ) . 

23. Rossiiskaya and Kabachnik, ibid., 1947, 509; C. A. _42, 2924b 

24. Arbuzov, Zoroastrova, Rizpolozhenskii, ibid., 1948 , 208, 
C. A. 42, 4932g (1948). 

25. Kamai and Belorossova, ibid., 1947 , 191; C. A. 42, 4133f (1948) 

26. Arbuzov and G-rechkin, J. Gen. Chem. (U.S.S.R.) 17, 2166 (1947); 
C. A. 42, 4522h (1948). 

27. Arbuzov and Pudovik, ibid., 17, 2158 (1947); C. A. 42, 4522a . 


Reported by Claire Blue stein 
December 10, 1948 

» * 


The punched card is a mechanical device which can be used to 
reduce the laborious repetitive work for literature searching. 
It has already achieved wide use in business and government applica- 
tions, and is finding increasing use in science. 

Punched cards are divided into two main classes according to 
the way they are sorted. 

%• Machine-- Sorted Punched Card 

This is the type used in accounting and computing operations; 
it has little use for the individual organic chemist. Where files 
can be planned of more than about 10,000 items, it should be in- 

A. Advantages 

1. cheap - about $1.10 per thousand. 
2« large punching capacity - 80 twelve-punch columns. 
3. machine operated - this card can be punched, inter- 
preted, verified, sorted, serialized 
alphabetized, duplicated, collated 
and tabulated by machine. 
H» Hand- Sorted Punched Card 

This is the type recommended for personal files and individual 
applications. The cards may be obtained from the McBee Company, 
("Key sort" ) , Athens, Ohio, and the Charles R. Hadley Company, 
("Rocket"), Los Angeles, California. 

A. Advantages 

1» large informational capacity - body of card on both 

sides can be used. 
2. simple sorting - knitting needle and gravity. 

B. Sorting Procedure 

Punched cards to be sorted manually have holes cut in 
them near the margins. When this margin is cut away from a par- 
ticular hole on a card, leaving an open slot, the card may be 
sorted by means of a knitting needle. The needle is thrust through 
the deck at the particular position and lifted; slotted cards will 
drop out. A double row of holes will give three categories. 

C. Coding Procedure 

The most difficult and most important part of preparing 
a ounched card file is the selection of a suitable code. This 
should be based on an outline covering the complete subject. For 
a most useful procedure, see Cox, Bailey and Casey (6). This 
procedure will insure coding into the file only information likely 
to be sought later. 

There are three general methods of coding information 
on punch cards. 



1. The so-called "direct" coding (better described as single- 
position punching, or unit punching) uses only one hole for a 
specific item of information. The presence of carbon and nitrogen 
in a compound can be indicated as 

This type uses the most holes, 
one pass selects the desired cards, 
exclusive , 

but is the 
The items 

easiest to 
need not be 


mut all y 

2. Numeric (and alphabetic) coding makes use of a group of 
adjacent holes, called a "field," to indicate an item. To code the 
number 952 (which might indicate a process like hydrogenation of 
C-0 double bonds, or a concept like the theory of reaction mechan- 
isms at C-X linkages) the following punching would result: 

1 <± Z I 



7 V a. / 

T £" N S 


7 4 I I 

u n ) rs 

o o 

O H 5" Z. 

The principle can be applied to alphabetic coding by number- 
ing each letter, ANM would be coded 

N-Z 7 4 2 1 



O o o'o ^AJ c o~o\J 

N-Z 7 4 2 1 



cXAAy o 

N-Z 7 4 2 1 

O o 




More complex systems have been developed which effect a saving 
in holes (4-7). 

This coding allows a great many items of information to be 
entered, but only one item per field, and a separate code index is 
required. A feature of this type is that if the field or group ui" 
fields is sorted in order, hole by hole, from right to left acrccs 
the card, with the cards that drop placed in back, the resulting 
deck will be in serial or alphabetic order. 

3. Random coding* consists of the superposition of randomly- 
chosen designations on the same set of holes, allowing a statistical 
distribution to control the number of cards which result from chance 
sorting, This can be controlled to any degree of fineness. 

Sorting is more complex, but allows the use of a file contain- 
ing unrelated topics. The best plan for "unrelated topics," for 
the individual, is to make a separate file for each. This makes 
for ease of sorting and specificity. 

* The patent status of this system is unclear at present and some 
caution should be used in its application, 

s. t:. 


"Direct" coding should be used where space is available, but 
generally both it and numeric (or alphabetic) coding will be used 
on the same card. Where random coding is used, it is best to employ 
all the holes on the card as one field. 

A point of first importance is to leave room on the final card 
lay-out for expansion. 

D. Uses 

In general, there are two types of files which are of 
interest to the individual organic chemist: a chemical compound 
file and a literature reference file. 

The compound file usually contains information on elements 
present, physical properties, and some structural indication, among 
others. Complete structural codes are in the process of development 

A file of literature references is ordinarily kept alpha- 
betically, but punched cards can be kept in, and sorted from, randor. 
order. Other items often included are date of publication, major 
and minor subjects treated, language of original, junior authors. 
etc. Each of these types of information takes the place of another 
whole file of ordinary cards. 

Correlations can be made by simultaneous or serial sorting, 
Thus, if one is interested in all organic compounds containing both 
N and As, and having a density between 1.5 and 2.0, it is relatively 
easy to arrive at these compounds using punched cards, and somewhat 
difficult other ways. More complicated multiple sorts can be made 
in subject classifications. It is here that the advantages of a 
well-made outline and coding system appear. 

- - 3 - - 
The author wishes to thank Mr. Dan Merrick, MoBee Co. . Athens, 
Ohio, Mr. C. M. Oehmke, Supervisor, U. of 111. Tabulating Office, 
Mr. J. ¥. Perry, Chairman, ACS Punched Card Committee, MIT, Cam- 
bridge, Mass., and Dr. E. J. Seiferle, General Aniline and Film, 
Easton, Pennsylvania. 


A. Bibliography 

1. Ferris, Taylor and Perry, "Bibliograohy on the Uses of 
Punched Cards." Available from Mr. Perry, Room 4-463, MIT, 
Cambridge 39, Mass. 

B. Books 

2. Baehne , G-. W. , et al., "Practical Applications of the Punched- 

Card Method in Colleges and Universities," Columbia Univ. Press- 
New York, 1935. 

3. Perry, et al., a general book on the use of punched cards 
in science, to be oublished early in 1949. 

C. Journal articles a. General 

4. Bailey, Casey and Cox, Science, 104 , 181 (1946). 

5. Casey, Bailey and Cox, J. Chem. "Education, 23, 495 (1946). 

6. Cox, Bailey and Casey, Chem. Eng , News, 23,~T623 (1945). 

7. Cox, Casey and Bailey, J. Chem. ^EducationT 24, 65 (1947). 

8. Gull, Special Libraries, 37, 223 (1946). — 

b. Scientific. Applications 

9. Aldous, Science, 106. 109 (1947). 

10. Allen, Brambe 11 aird"Mills. J. Exp. Biol.. 23, 312 (1947), 

11. Buhle, et al . , J. Chem. Education, 23, 375^:1947). 

12. Burkhard. et al.. Chem. Revs., v 41. V7 (1947). 

13. Clarke, Nature , 137, 535 (1936). 

14. Cox. Dodds, .Dixon and Matuschak, J. Dental Research, 18 , 

15.. Frear, Chem. *Eng. News, 23, 2077 (1945). 

16. Frear, Seiferle and King?- Science, 104, 177 (1946), 

Karl F. Heumann 
December 10, 1948 


\ •• .- . 



I. Nomenclature 

Carbodiimides have the general structure R-N:C:N-R'. If R = 
cyclohexyl- and R' = phenyl- , the conroound may be called carbocyclo- 
hexylphenyldlimide,or cyelohexylphenylcarbodiimide. 

II, Sterlochemlstry 

Carbodiimides of the type A and B are theoretically structur- 
ally similar to the allenes (l) . 

K ^C N 7 

R^ \^V 


If the nitrogen atom has a fixed tetrahedral structure, a pair 
of mirror images may exist, since the two substituted R groups are 
in planes different from those of the C=N linkages. No optically 
active carbodiimides of this type have been reported. 

Optically active carbodiimides have been prepared by the use of 
optically active reagents (2)» Dibornyl- and dimenthylcarbodiimides 
were prepared from the corresponding thioureas. Their optical 
rotations lie between those of the corresponding ureas and thioureas, 

III. Preparation 

Carbodiimides are best prepared by desulfurization of 

(I): SC(NHR) 3 + HgO -* HgS + H 3 + C(:NR) 2 

Numerous side reactions are also possible: II, the addition of 
water to the carbodiimide forming the urea; III, the rep ct ion of 
the thiourea and carbodiimide to form the guanadine and isothio- 
cyanatej IV, the reaction of urea with carbodiimide to form the 
isocyanate and guanidinej V, polymerization reactions. 

In the original method of Weith (o) , the aromatic thiourea was 
boiled in benzene with mercuric oxide. This treatment was too 
vigourous, resulting in low yields as a result of reactions II and 
V. Rotter (4) attempted to avoid reaction II, but the urea forma- 
tion could not be prevented. 

For the preparation of aliphatic carbodiimides, pure, freshly 
prepared dry "thioureas and freshly precipitated yellow mercuric 
oxide were shaken "in dry ether, benzene, or carbon disulfide at 
room temperature, thus repre ssing polymerization reactions (5,6,7.8)- 
In cases where urea formation was greatly favored, the desulfuriza- 
tion velocity was increased by using freshly prepared, undried 

-2- 99 

mercuric oxide; the carbodiimide wag formed before reaction II 
could occur. Di-isopropyl-. di-n-propyl- 5 propyl- cyclohexyl-, 
and propyl- isopropylcarbodiimides were prepared in two to fifteen 
minutes in over 90 per cent yield by this method. 

The applicability of the above method is limited by the 
solubility of the thiourea in the solvent at room temperature. 
A modification of the method was subsequently sought which would 
be applicable to the difficulty soluble aromatic thioureas (9). 
The de sulfur ization velocity is increased by the use of sulfur as 
a catalyst, by increasing the repctive surface of the metal oxide, 
and by using acetone as a solvent, thus preventing harmful accumu- 
lation of water on the metal oxide- sulfide surfaces. Sulfur also 
inhibits urea formation and resinif io.ation. 90 per cent yields of 
most aromatic carbodiimide s were achieved. This method is not 
applicable to aliphatic thioureas (9)e 

Carbodiimide s are also prepared by the action of phenyliso- 
cyanate on phosphinimines, and aromatic acid chlorides on cyanamino- 
ethyl alcohol. 

IV. Stability toward polymerization 

Aliphatic carbodiimide g with two primary residues are very un- 
stable; stability in the primary residue increases with the number 
of carbon atoms. The stabilizing effect of secondary groups is 
larger than a proportional increase in the primary group size (?) . 
Two secondary groups show greater stability, while tertiary 
residues are the most stable (8). 

There is a wide variation in the polymerization tendencies of 
aromatic carbodiimide s. For example, di-p-lodophenylcarbodiimide 
and carbo-p-dimethylaminophenyl-phenyl- carbodiimide polymerize 
readily; diphenyl-, p-tolyl-p-bromophenyl-, di-p-bromophenyl-, 
and di-p-tolylcarbodiimide moderately; di-p-dimethylaminophenyl- 
and di-a-pyridylcarbodiimide tend to polymerize only slightly. 

V. Reactions 

A. With G-rigard Reagents- R*MgX adds across one C=N double 
bond of R-N:C:N-R forming the addit ion product, R-N=C(R ! )-N(MgX)=NR, 
which is subsequently hydrolyzed to a substituted amidine, 
R'C(NHR)=NR (lb). 

B% With Phenols- Diphenylcarbodiimide when heated with ■ohenol 
yields the 0- ether of diphenyl- Y~ carbamide, BhN: C (OPh)NHPh. A^id 
treatment yields phenol and diphenyl urea, p-cresol, a~ and £- 
naphthol were also used (ll) , 

C. With Aromatic Amine s - Aromatic guanidine derivatives. 

D. With Qarboxylic Acids - (See Uses). 

E * With Hydrazoic A.cid - Carbodiimides add HN 3 forming 1,2,3,4- 



t '- 

. '. . 



F« With Dlazomethane - The reaction of carbodiimides with 
diazome thane yields triazoles. 

VI, Uses of Carbodiimides 

A, Characterizat ion of Carboxylic Acids 

1. As Ureides 

Depending upon the solvent, the temperature, the carbo- 
diimide, and the acid, carbodiimides react in two ways 
with carboxylic acids (12,13,14): 

a) With one rnole of acid to form the ureide (N-acyl, 
N, N f -disubstituted urea. 

b) With two moles of acid, forming the acid anhydride 
and a disubstituted urea. 

If carbodicyclohexylimide is boiled in alcohol with 
carboxylic acids, it is possible to make ureide formation 
the chief reaction. Ureides of butyric, benzoic, and 
stearic acids are readily prepared. A similar reaction 
with aromatic carbodiimides in ether or benzene solution 
at room temperature results almost exclusively in the 
formation of the anhydride. 

2. A Test for cc-, g-unsaturated Acids (12) 

Ureides with C ($NC 6 H 4 N-Me 3 -p) s and RCOCH are colored 
when R = R T CH=CH- or R'Os C-; those having an unsaturated 
link in any position other than cc,£-, are colorless. 

3. A Test for a- halo aliphatic Acids (9) 

Like the a, p-un saturated acids, the cc— halo aliphatic 
acids form colored ureides with C(: NC e H 4 N-Me 2 ) 3 . 0,3T- 
and other halogens do not produce a deepening of color. 
Bromide and iodide cause a greater effect than chloride. 

4. Detection of Free Carboxylic Acids in Anhydrides 

As little as O t l per cent free acid will form a pre- 
cipitate with carbodi cyclohexylimide (13), 

5. Preparation of Acid Free Anhydrides 
Same method as above, 

B. Industrial Applications 

1. Deacidif ication of animal and vegetable oils and fats. 

2. Textile finishing and impregnating agents. 

3. Preparation of films, fibers, and various molded 



1. Adams, R. , and Roll, L. JV, J. Am. Chem. Soc . , _54, 2494 (1932). 

2. Zetzsche, F., Ber., 7J5B, 1114 (1940) . 

3. Weith, W-, ibid., 7, 10 (1874) . 

4. Rotter, R., Monatsch., 2±) 353 (1926). 

5. Schmidt, E. , Per., 71R, 1933 (1938). 

6. Schmidt, E., ibid., 75B, 286 (1940). 

7. Schmidt, E., ibid., 74R, 1285 (1941). 

8. Schmidt, E., Ann., 560 , 222 (1938). 

9. Zetzsche, F., Ber., 75B, 50 ; 465 (1940). 

10. Busch, M. , ibid., _40, 4296-8 (1907). 

11. Busch, M. , J. prakt. chem., 2Ji> 513 (1909). 

12. Zetzsche, F., Ber., 71B, 1088, 1516, 2095 (1938). 

13. Zetzsche, F., ibid., 72B, 365, 1599, 1735, 1477 (1939). 

14. Zetzsche, F., ibid., 74B, 1022 (l94l) , 

Reported by Edward F. Elslager 
December 17, 1948 



A new aldehyde synthesis, disclosed in a patent (l) issued to 
Roelen in 1943. involves the combination of olefins with carbon 
monoxide and hydrogen in the presence of a catalyst containing 
cobalt. The aldehyde slurry is then subjected to hydroge nation re- 
sulting in reduction of the aldehydes to primary alcohols. The 
alcohols are then separated from the reaction mixture by distil- 

A wide range of Pressures and temperatures may be used in both 
stages (?)c The pressure range usually employed is 150 to 200 
atmospheres. The temperature in the first stage is usually about 
140° C, and in the second or hydrogenation stage about 180° C. 

Mechanisms have been proposed for this reaction but sufficient 
evidence has not been put forth which proves any postulated. The 
reaction may be illustrated by the following: 

H 3 

CO H 3 _j> RCH s -CH 3 -0IIO -> RCH 2 CH 3 CH 3 OH 

RCH=CH 3 -> [ RCH-CH 3 ] -*- ' U 

H ; 





CH 3 0H 

As a general rule, about 60 percent branched and 40 percent 
straight chain alcohols are obtained, 


(l) Primary Reactions* 

(a) Formation of aldehydes-* 

v RCH 3 CH 3 CH0 
RCH=CHp + CO + H s 

' RCH-CH 3 


(b) Formation of ketones- 

2RCH=CH + CO + H. 

RCH P - CH p- CO- CH P - CH P - R 






0H 3 CH 3 
(2) Secondary Reactions: 

(a) Formation of alcohols- 
Occurs possibly as a result of a Cannizzaro-type reaction. 
Formation of acids- 

This also is due apparently to the Cannizzaro-type reactioi 
Formation of esters. 

Formation of high -molecular-weight compounds- 
Formed by polymerization and condensation reactions. 
Formation of saturated hydrocarbons- 
Corresponding to starting material. 



(e) F< 



This is an important feature of the oxo process. Using pure 
1-dodecene, in the absence of hydrogen it has been shown that all 
dodecene isomers in almost eoual ratios are formed. Using cobalt 
metal in the presence of an inert gas, under the usual conditions 
of the oxo reaction, no isomerization takes place. The isomer iza- 
tion agent is apparently the dicobalt octacarbonyl, [Co (C0) 4 ] 3 , 
itself. It has been stated that the oxo reaction and isomerization 
proceed simultaneously but that the former takes place with the 
greater velocity since when terminal olefins are used, the format ioi 
of branched products is not as great as would be predicted from 
labor? tory experiments on the isomerization of olefins in the 
presence of carbon monoxide but in the absence of hydrogen. 

Below are 

listed the results (4) obtained by using various 

Base Material 





3- Me thy 1-Pente ne~ 3 

2, 4, 4-Trimethyl-Pentene^l 

Alcohols Obtained 

60^ n-Butanol 

40% 2-Methyl-Propanol-l 

bOf n-Butanol 

bO% 2- Methyl- But a no 1-1 

50% n-Pentanol 

50^ ?,-Methyl-Butanol-l 

3-Methyl-Butanol-l (Only) 

50^ n-Hexanol 

40^ 2-Methyl-Pentanol-l 

IQfi 2-Ethyl~Butanol-l 

55,^ n-Hexanol 

35% 2- Methyl-Pent a no 1-1 

1G% 2-Ethyl-Butanol-l 

50% ft-Heptanol 

30^ 2- Methyl- He xano 1-1 

20^ 2-Ethy.l-Pentanol-l 

30^ 2,4-Dimethyl-Pentanol 
&0# 5- Me thy 1-He xano 1-1 
30^ 3-Methyl-Hexanol-l 

3, 5, 5-Trimethyl-Hexanol-l 

The results may be summarized as follows: 

(l) Addition of a formyl group to a tertiary carbon atom does not 
occur at all. No quaternary carbon atoms are formed. 


(2) Addition of a formyl group adjacent to a quaternary carbon 
atom does occur. 

(3) Addition of a formyl group adjacent to a tertiary carbon atom 
is strongly hindered, but it may occur to a small extent. 

(4) Addition of a formyl group is not hindered by an isolated 
tertiary carbon atom. 

(5) Isomerization of the double bond generally accompanies the 
formylation, but does not necessarily occur. 


(a) CH 3 =C-CH 3 -CH 3 -C"=CH 

CH 3 CH 

(b) CH 3 -C=CH-CH=C-CH S 



(c) CH 3 -CH=CH-S~CH=CH-CH 3 
CH 3 - CH= CH- 0- CH= CH- CH 3 

40^ Diol 
&0% C g Monol 

100^ C 9 Monol 

Non- identifiable 

It was assumed that the low yield of the diol in (a) was 
brought about by the oxo reaction itself since the presence of con- 
jugated bonds could not be established in the original olefin or 
after heating it with a catalyst to 130°. 

It was expected that a higher yield of diol would be formed in 
(c) since the sulfur atom or the oxygen atom should prevent the 
double bond migration to form a conjugated system. The resulting 
products boiled from 100° to 400° but no fraction had an appreciable 
hydro xyl number. 





Otto Roelen, U.S. Patent 2,327.066 (1930). 

Bureau of Mines Report of Investigations (R.I. 4270). 

References for proposed mechanisms: (a) "Cyclopropanone ring 
intermediate" by Reppe, Final Fiat Report 1000 — Department of 
Commerce. (b) "Carbonium-ion mechanism" by Wender and Orchin 
R. I. 4270. (c' 

"Free-radical mechanism" by H. S. Newman. 
"'Metal complex formation" by duPont, C. A, 42 
"Cobalt hydride intermediate" by Adkins and" 
S.,70, 383, 1948. 
J...L.J.-1. iwot-Lcmcuie and A. Kwantes (Laboratorium N.V. De 
Bataafsche Petroleum Maatschappij , Amsterdam). 
Reel No. 2, 8, B,7 Report No. PB 859 — Department of Commerce. 
R. I. 4270. 

R. I. 4270. (d 

7232, 1948. (e 

Krsek, J, A. C. 

A.I.M. Keulemans 

Reported by Edward F. Riener 
December 17, 1948 




. . - • ; - • 

• ■ . , 

i f ■ 



Because of its strong acidity and convenience for cryoscooic 
measurements, concentrated sulfuric acid is an excellent solvent 
for studying the basic ionization (in the BrSnsted-Lowry sense) 
of organic substances. The magnitude of the f reezing-ooint depress- 
ion compared to the depression produced by a non-electrolyte 
(known as the van't Hoff V factor) is accepted as a relatively 
accurate measure of the number of ions produced by the solution of 
one molecule of solute. 

Hantzsch (l) discovered that many organic oxygen compounds 
ionize almost completely as monoacid bases. Most acids and esters 
ionize as follows: 

(A) RCOGR' + H 3 S0 4 = RC00R'«H + + HS0 4 ~ ( >* =2) 

Such solutions conduct electricity and yield the original compound 
unchanged upon dilution with water or alcohol. Certain strong 
acids, such as trichloroacetic acid, behave as non-electrolytes 
(•>=1J, while dichloroacetic acid ionizes only partially. The be- 
havior of aldehydes and ketones resembles that of acids and esters, 
but most alcohols have an> factor of 3 and yield alkyl sulfuric 
acids on dilution with water. 

Further investigations by Hammett (2,3,4) and later by 
Newman (5,6,7,8) disclosed an unusual type of ionization by certain 
sterically hindered acids and esters (such as mesitoic acid): 

+ + — 

(B) RSOOR' + 3 H 3 S0 4 = RCO + H 3 + R'OS0 3 H + 2 FS0 4 

The jy value is 4 when R'=H and 5 when R' is an alkyl group. Di- 
lution with water yields the corresponding acid in either case, 
while dilution with an alcohol yields the corresponding ester. 
This is particularly surprising, since sterically hindered acids 
are not esterified by the usual acid catalysis while (as seen above) 
ordinary acids are recovered unchanged from dilution of their 
sulfuric acid solutions with alcohol. Newman (5) utilized this 
reaction as a method of preparing esters of sterically hindered 
acids (yields 60-80^). 

In the series of compounds studied by Hammett, acyllzation 
(see B) occurred only when there were two ortho-methyl groups. 2- 
Methyl-6-nitrobenzoic acid ionized normally, as did 2,4,6-tri- _ 3 
bromobenzoic acid (possibly because of its strong acidity, K=4xl0 ) 
Dibromomesitoic acid exhibited partial acylization in pure sulfuric 
acid, but when a little water was present the ^ factor sank to 2. 
Hammett therefore postulated that acylization takes place in two 
non-overlapping steps: 


RCOOH + H 3 S0 4 = RC0 3 H 3 + HS0 4 

+ + + 

RC0 3 H 3 + H 3 SC 4 = RCO + H 3 + HS0 4 

j . • 

_ . - .. ■ . .-.-.•, ...; .. .- 

■ ' 




■ ' , ' ■ ' ' li . 

1 ;■•.:. f 

-■:■. " '-':'.■ 




Confirming this, Newman found that the presence of HS0 4 ion 
partially inhibits acylization even in mesitoic acid. 

Until recently, it was believed that the effect of the 
alcohol group in an ester had no effect on acylization. However, 
Kuhn and Corwin (9) have found that such esters as cellosolve 
benzoate and trichloroethyl acetate have /^ values of 4.4 and 3.5. 
Furthermore, when the sulfuric acid solutions of these esters were 
diluted with alochol, the free acid was obtained. The investiga- 
tors postulated that the benzoyl and acetyl cations are unstable 
in sulfuric acid and react as follows: 

RCO + 


As proof, they prepared acetyl and benzoyl sulfate and found that 
they react with alcohols to form free acetic and benzoic acid. 

Kuhn and Oorwin also found that anisic acid and its esters 
form acyl cations (the first known instance of acylization without 
ortho substituents) , while Corwin and Straughn (lo) have investi- 
gated selective acylization in pyrryl dicarboxylic esters. 

Newman (6,7,8) has investigated o-benzoyl benzoic acid, which 
is unusual in that the free acid and its pseudo methyl ester (i) 
undergo acylization while the normal ester does not, Newman 
postulates a cyclic acyl cation (il) because the pseudo ester is 
obtained when the sulfuric acid solution of the acid or pseudo 
ester is diluted with methanol. He further postulated that a new 








acyl cation (III) of higher energy content is obtained upon heating 
(il) . This would explain why anthraquinone is not obtained in the 
cold, since (ill) can react to form the quinone while (II) cannot. 
Newman also found that both the normal and pseudo methyl esters 
of 2-benzoyl-6-methyl-benzoic acid are acylized.. 

Smith and Smith (II) recently attempted to prepare the 
methyl ester of triphenylacet ic acid by Newman's method, but the 
product obtained was the methyl ether of triphejpylcarbinol . Pre- 
sumably the intermediate acyl cation (C s H 5 ) 3 CC0 is formed, which 
then loses carbon monoxide to form the triphenyl carbonium ion. 
However, this reaction suggested a simplified procedure for the 
preparation of such ethers from triphenyl carbinol in concentrated 
sulfuric acid (yields 86-97$). 




1. Hantzsch, Z. physik. Chem., 61, 257 (1908); 65, 41 (1909). 

2. Hammett, "Physical Organic Chemistry", McGraw-Hill Book Co. 
New York, N. Y. , 1940, pp. 45-8, 54-6, 277-85. 

3. Hammett and Deyrup, J. Am. Chem. Sec, 55, 1900 (1933). 

4. Treffers and Hammett, Ibid . , 59, 1708 (1937). 

5. Newman, ibid . , 63, 243T~Tl94l7T 

6. Newman and McCleafly, ibid., 63, 1537 (1941). 

7. Newman, ibid . , 64, 2324 (194277 

8. Newman, Kuivila, and Garrett, ibid. 704 (1945). 

9. Kuhn and Corwin, ibid . , 70, 3370 '(1948). 

10. Corwin and Straughn, ibid . , 70, 2968 (1948) . 

11. H. A, Smith and R. J. Smith, ibid . , 70, 2400 (1948). 

Reported by R. G. Bannister 
December 17, 1948 


When there is a methylene group situated between two carbonyl 
groups in a 1,3-diketone the existence of tautomeric forms is very 
probable. The nature of the structure of the tautomeric enolic 
form, if stable, depends largely upon the groups or radical 
attached directly to the carbonyls. 

In the aromatic series it has been shown that substitution 
in the ring influences the enolized form, also certain relationships 
as to o, m or p- substitution in the ring have been pointed out 
i 1,2,8). Attempts have been made to explain these substitution 
effects by the various theories of electron releasing tendency 
and electron attracting ability of the substituent groups. 


Compound Enolized Form 

1. p-bromodibenzoylme thane resonating forms 

2. p-methoxydibenzoylmethane P-CH 3 0-C S K 4 -C~CH=C!H-C6H 5 

'6 - OH 

3. m-nitrodibenzoylmethane m-N0 3 -C s H 4 -C-CH=CH-C 6 H s 

'6 OH 

4. 3-nltro-4' methoxydibenzoylmethane 3-N0 3 ~C 6 H 4 C-CH«gH-C s H 4 -0CH^ 

0* OH 

5. Benzoylmesitoylme thane C S H B C=GH-C-Mes 


6. o-methoxybenzoylmesitoylmethane 0-CH 3 0-C 6 H 4 -C=CHC-Mes 

OH b 

7. Mesitoyl-o-nitrobenzoylmethane MesC-CH=C-C s H 4 -N0 3 -0 

b HO 

8. Me sit oyl-m-nit rob enzoylme thane MesC=CH-C-C s H 4 -N0 3 -m 


9. Me sitoyl-p-nitrobenzoylme thane MesC=CH-C-C 6 H 4 -N0 3 -p 

ok '6 

The comparitive substituent effect in the direction of eno- 
lization can be seen from this table. 


NaOH Br 3 

RC=0 + gH 3 C-R« -> RCH=CHC-R f -> RCH-CH-C-R' 

H 10% Q £ r Br 

J KOH in 

r^n *„ ~ ^. \bmethanol 


6h 5 «~ 0CH 3 8 



This method is well known and the mechanism is firmly estab- 
lished (3,4). Isomeric |3- unsaturated ketones may be prepared in 
this way by the proper selection of aldehyde and ketone for con- 


Series A Series B 

R'C-CH=0-R ' R'C=CH-C-R 


i H « NQH 1 

R'9-CH=C-R R^C=CH-C-R 

N-OH 6H 5h HO- ft 

J i 

R i? — u H * CH — P- R 

!l j! Isomeric II I 

N C-R isoxazoles R'-CH N 

The enollc compounds listed in Table I gave rise to a single 
isoxazole, indicating only one form. The structure of the iso- 
xazole was determined by isoxazoline formation and oxidation ac- 
cording to the following scheme (6): 

1 H 3 NOH 

4 1 


[ R X-H CH 1JT R ] ' R & CH =£r R ' J 


RtrjH CH 



Isomeric || jj 

NH C-R isoxazoline s NH 

x o' 


rnuo « !f? iso 2 e r ic isoxazolines are then oxidized to the iscaazolej 
ihe position of the nitrogen is thus fixed and the direction of 
enolization of the (3-diketone shown. - 

thP 1°T+°^ B 7' 8 ' 9 in T * ble *» be ^use of the presence of 
the mesityl nucleus together with a nitro group substituted in 
the other ring, exhibit some interesting and varied properties. 



1. Kohler and Barnes, J. Am. Chem. Soc, _56, 311 (1934). 

2. Barnes and Delaney, Ibid . , 64 , 2256, 226, (1942). 

3. Bodforss, Ber., 49. 2795 (1916). 

4. Sorge, Ber., 35 ,~ 1068 (1922). 

5. Barnes and Brandon, J. Am. Chem. Soc, 65, 1070 (1943). 

6. Blatt, J. Am. Chem. Soc, j33, 1133 (193TJ. 

7. Barnes and Spriggs, J. Am. Chem. Soc, 67, 134, 1945. 

8. Fuson, Soper, J. Org. Chem., 9, 193, 1944. 

9. Barnesand Da Costa, J. Am. Chem. Soc, j69, 3129, 1947. 

Reported by Alfred S. Spriggs 
January 7, 1949 



In 1908, Meldrum (l) reported that he had treated a suspension 
of malonic acid in acetic anhydride-sulfur ic acid mixture with 
acetone and obtained a crystalline product melting at 97°. 

(CH 3 ) 3 C0 + CH 3 (C00H 3 Ar3 -2r sS 4 H 3 + C 6 H Q 4 

Room temp. 

This compound proved to b e a monobasic acid, the properties 
of which did not correspond to those of the previously prepared 
(2,3,4) isopropylidene malonic acid or p-methyl crotonic acid, so 
Meldrum concluded that the compound must be p, |3- dime thy 1-|3- 
propiolactone-cc-carboxylic acid (i) . 


Ri R 3 yP 

(CH 3 ) 3 C CHCOOH R 3 -C C-C00R 4 (CH 3 ) S C XH 3 


6=0 C=0 T) 

la II 

Since that time his conclusion hpd been accepted as demon- 
strated fact by numerous investigators until Davidson and Bernhard 
(5) recently proposed that the properties of the compound did not 
Justify Meldrum' s conclusion. They have pointed out the fact that 
the reactions of Meldrum 1 s acid indicate that the methylene group 
of malonic acid still exists in the molecule, and that the compound 
is actually isopropylidene malonate (il). 

In support of this conclusion, they quote the following 

A. There is no good evidence for the existence of the bond between 
the a- and p- carbon atoms; rather, all the reactions of the 
acid indicate a strong tendency to regenerate acetone and 
malonic acid (or its derivatives and decomposition products) . 
(See, for example, l f .5,6,7,8,9.) 

B. Other than the fact that titration with dilute alkali shows the 
the compound to be a monobasic acid, there is no evidence for 
the presence of a free carboxyl group. 

(1) Attempts to form the ethyl ester or nitrile corresponding 
to (i) by condensations involving ethylhydrogenmalonate and 
cyanoacetic acid have been unsuccessful (6,9). 

(2) Treatment of the silver sftlt of the acid with methyl iodid* 
gave a mixture of products which were stated to correspond with 
structures (i), (ill), and (iv); 

* «. '■ % 



(CH 3 ) a C— C-COOH 
I ! 

— c=o 


9H 3 
(CH 3 ) S C— C— -C00CH 3 

I ! 

0— -c=o 









C(CH 3 ) 


(CH 3 ) 3 C .0H S 

N CH 3 — G' 


No evidence was presented in support of structure (IV), but it was 
reported that pyrolysis of this derivative yielded acetone, carbon 
dioxide, and dimethylketene (6,13,16). On the basis of this fact, 
Davidson and Bernhard theorized that the compound represented by 
(IV) is actually isopropylidenedimethylmalonate (v). 

C. Hydrolysis of the dimethyl derivative (v) with dilute hydro- 
chloric acid gives high yields of dimethylmalonic acid, where-^ 
as structure (IV ) would be expected to give monomethylmalonic 
acid (5). 

D. Meldrum 1 s acid has several properties which are the same as 
those of methone (Va) , to which the proposed structure (II) is 

Reaction with: 

(a) NaN0 3 (ll) 

(b) Br 3 

(c) RCHO 

Meld rum 1 s acid 
8.0 x 10" 6 

gives purple product 

reacts with 2 moles 

gives precipitates in 
the cold 

Metho ne 
6.3 x 10" (10) 

gives violet salt of 
nitroso derivative. 

reacts with 2 moles (lc 

gives precipitates 
in the cold (12) 

Numerous other compounds with structures represented by (la) 
have been synthesized by the investigators whose reports have been 
quoted here, usually by Meldrum* s method or by Ott's modification. 
Experimental data on these compounds are so incomplete that it is 
not possible to reach a definite conclusion at this time, but 
their reported properties have all been the same as those of 
Meldrum' s acid. In view of this fact, it seems that these com- 
pounds should be thoroughly investigated on the basis of the 
strong possibility that they have structures of the cyclic tvpe 

t .... 



R/ C^ S R 4 




1. Meldrum, J. Chem. Soc, 93, 598 (1908), 

2. Massot, Ber. 27, 1225 (1894) . 

3. Meyenberg, ibid., 28, 785 (1895). 

4. Knoevenagel, DRP 162281. 

5. Davidson and gernhard, J. Am. Chem. Soc, 70, 3426 (1948). 

6. Ott, Ann. 401, 159 (1913). 

7. Michael and Ross, J. Am. Chem. Soc., 55, 3684 (1933). 

8. Michael and Weiner, ibid., 58, 680 (l"9~36). 

9. ibid., p. 999. 

10. Schwarzenbach and Felder, Helv. Chim. Acta 27, 1701 (1944). 

11. Clarke, "Handbook of Organic Analysis", Ed. IV, Arnold and Co. 

London, 1931, p. 215. 

12. Vorlander, Z. Anal. Chem., 22> 241 (1929). 

13. Organic Reactions III, 124. 

14. Kandiah, J. Chem. Soc, 1952 , 1215. 

15. Salkowski, J. Prakt, Chem. JL06, 259 (1923). 

16. Bockstahler, Organic Seminar, University of Illinois, 

November 21, 1947. 

Reported by Richard H. Tennyson 
January 7, 1949 

• f ' 


- . . , 

. . ■ 





I ■ -. ■ toJk 

: . - 

.,.::■. r w i 


♦ ■ ( 


. i •-''- ■'.'" H 
■ . ■■ : 




Criegee (1,2) in 1944 reported that the benzoate (II) of 
decalin hydroperoxide (i) rearranged into the isomeric benzoate 
(III) on warming and that alkalihe saponification of III yielding 
cyclodecanol-6-one (IV), thus providing a convenient route from 
decalin to the cyclodecane series. 




— > 

C 5 H 5 N 




Me OH 
— > 




CH 3 CH 3 CHCH a CH s 




Wieland and Maier (3) had previously described a similar rearrange- 
ment in the attempted preparation of benzoyl triphenylmethyl per- 
oxide. When trityl hydroperoxide (v) was treated with benzoyl 
chloride in the presence of sodium hydroxide, the only product ob- 
tained had none of the characteristics of a peroxide and subse- 
quently was assigned the structure of the benzoylated hemiacetal of 
benzophenone and phenol (vi) . 

j6 3 C00H -> fb s G 




More recent work 
mechanism of the 

by Criegee (4) has been undertaken to study the 
rearrangement of esters of decalin hydroperoxide. 

The rearrangement 
factors: l) the stren^ 
of the peroxide, and 3, 
stronger the acid used 
for rearrangement.. Of 

has been shown to be dependent on three 

;th of the acid in the ester, 2) the acidity 

the solvent used. It was found that the 
for esterif ioation, the greater the tendency 
the numerous hydroperoxides prepared, only 

'• " . : : ' . . 


. . 






.... - - . 

. - 


. . 

. ■ - 

- ' " 

3.S ' ■■. 

-.. . i., rjj 






■ ' 

■■■■■■■ ,. 

. ■ . ■■■ ; ;: " - c 




trityl and decalin hydroperoxide fail to form sodium salts with 50% 
sodium hydroxide (5), thus Indicating their weak acidity. Because 
these two hydroperoxides are the only ones for which the rearrange- 
ment hae been demonstrated, • it is assumed that the rearrangement 
kes place only if the peroxide is a very weak acid. In general, 
rearrangement is much more rapid in polar solvents. 


The experimental facts, indicate that the cause of the rearrange, 
ment may lie in a strong polarization of the 0-0 bond in the per- 
:ide ester, the decomposition of which results in the formation 

of a benzoate anion and a 
deficient oxygen atom. A 
VIII, which combines with 
ester III. 

species which contains an electronically 
Whitmore shift produces the carbon iura ion 
the benzoate anion to yield the rearranged 

/ 0-0- 8/6 




p- © 

vii + p6o\ 




In accord with all carbonium-like processes, complete dissociation 
of the benzoate anion is not assumed. However, influences which 
favor a polarization of the 0-0 bond in the direction of such a 
dissociation, i.e. a strong acid used in the esterif ication, a 
weakly acidio peroxide, and a polar solvent, facilitate the re- 

Several other reactions may be considered as involving cationic 
oxygen in an intermediate state. 


l) If the conversion of cyclic ketones to lactones with Ca 
acid involves first the addition of the reagent to the darbonyl 
double bond as postulated by Stoll (6), the formation of the inter 
mediate X would follow analogously to that from decalin peroxide 
benzoate. Rearrangement of X to the carbonium ion XI, followec by 
the loss of a proton would yield the lactone (XII). 




■00S0 3 H 


J 1 

/N° H 



L xi 






2) Tetralin hydroperoxide (XIIl) on attempted benzoylat^on 
loses a molecule of water to form cc-tetralcne (XV) . The inter- 
mediate XIV is stabilized by the loss of a proton. 










3) The derivatives of hydrogen peroxide which are the 
strongest oxidizing agents in the cold are the peracids. Presum- 
ably this is due to the strong electron attracting nature of the 
carbonyl group which tends to weaken the 0-0 bond and thus facili- 
tates the formation of + 0-H. The assumption of such an inter- 
mediate will account for the ready formation of epoxides, bulf oxide r 
and amine oxides when olefins, thioesters, and tertiary amines &i 
treated with peracids. 


$C00H ~+ j6flqi 
d) RCH=CH 2 + I OH 

b) R 3 Si + !0H -h 


c) R 3 Nf + (OH 


[RCHCH 2 0H] ~> RCH-CH. 

© -if 




R 3 S0 
~* R 3 N-0 




Criegee, Ber. , 77, 22 (19.44). 

Criegee, ibid ., 77, 722 (1944). 

Wieland and Maier, ibid . , 64 5 1205 (1931). 

Criegee, Ann., 560, 127 (1948). 

Criegee, ibid, .. 560, 135 (1948). 

Stoll and Scherer, Helv, Chim. Acta, 13, 142 (1930) 

Reported by Jean V, Crawford 
January 7, 194 9 

: - 



The Arndt-Eistert synthesis (l,2) offers a means of convert- 
ing a carboxylic acid into its nevt higher homologue or a deriva- 
tive thereof. The following reactions are involved. 

S0C1 3 CH 3 N 3 AH 

RCOOH -» RC0C1 -» RCOCHN 3 ~* RCH 3 00A 
etc. Ag 

Where A=HO, R»0, NH 3 , or RNH 

The accepted mechanism (g.,5) for the last step of the 
synthesis is as follows. 

H H 

-♦ ©o:c.' + ;n;::n: -* '6."C::g:r 

I -0: J 



it it ■< 

A H 

On the basis of this mechanism the use of higher diazohydro- 
carbons, such as diazoethane or 1-diazopropane , in place of diazo- 
methane would be expected to lead to carboxylic acids, or deriva- 
tives, bearing an alpha-alky 1 substituent. It is rather surprising 
that until very recently there had been only one recorded attempt 
to extend the generality of the reaction in this manner. In 1941 
Eistert (4) reported the successful rearrangement of the diazo- 
ketone from p-nitrobenzoyl chloride and diazoethane to the anilide 
of alpha- (£-nitrophenyl)propionic acid. 

This year Wilds and Meade r (5) have reported the successful 
use of diazoethane and diazopropane in the Arndt-Eistert synthesis 
of a variety of carboxylic acids. In connection with this' work 
a new and more dependable method for rearranging dia 7,0k e tone s was 
developed. B 

In the early stages of the work it was learned that the dia- 
ketones obtained from diazoethane would not consistently undergo 
rearrangement under the usually employed conditions, i.e., in the 
hp e p??2%°£ c ° lloidal silve r in methanol. The rearrangement could 
atr^l- tt> £? wever > b y using the already known procedure of 
S • p^, / kZ . 0ket0 , ne int0 boll ^S aniline (l)". This method 
had the disadvantage that the dif f icultly hydrolyzable anilides 

^L. L rearrsnged acids Were ob ^^, and so Wilds aund Meader 
devised a new procedure. According to "thair innovation the 
diazoketone is heated to 170-180* in a ml*tur& of gamma- coll id ine 
ana benzyl alcohol, whereupon the easily saponified l*enzyl esters 


< ■ ■■''- -■ 







. ' 

' - n : ■ ■ 

. ■ ;> : ' i 


■ ■ - 

■ — 

.-■ . • • -. - : .- •?•! si :i* ■' - • i ' '■• •• ■ • 

- - -■ , j . ■ ■ • .- ■ ■ • - •■ 

■- • • —- 

■ r ' ' ■ ■ ■. ■ ' 




.-■'"■'• " 



of the rearranged acids are formed. This method not only made 
practicable the use of diazoethane and diazopropane in the Arndt- 
Eistert synthesis but was also found to be an improvement over the 
colloidal silver-methanol procedure when applied to diazoketones 
made from diazomethane. The new method gave more consistent re- 
sul t s . 

It was found that other tertiary amines and high boiling 
alcohols could be substituted for gamma-collidine and benzyl 
alcohol without greatly lowering the yields in the rearrangement, 
and the reaction temperature proved not to be highly critical. 

In preparing diazoketones from acid chlorides and higher 
diazohydrocarbons the reaction temperature had to be controlled 
more carefully than when diazomethane was used, and even under 
optimum conditions the diazoketones often were not completely cry- 
stalline, part of the product appearing as an oil. For instance, 
in the preparation of 1-p-chlorobenzoy 1-1-diazoethane the most 
suitable temperature was found to be -20°. Changing the tempera- 
ture by 10°, up or down, caused the yield of crystalline diazoketoru 
to fall from 71 to 61$. Diazomethane, on the other hand, gave 
practically quantitative yields of crystalline diazoketones at 
temperatures ranging from 0° to room temperature. 

In the following tables are presented the results obtained 
when Wilds and Meader subjected various acids to the Arndt-Eistert 
reaction using diazoethane and diazopropane and carrying out the 
rearrangement of the diazoketones in gamma-collidine-benzyl alcohol 
mixture . 


Arndt-Eistert Synthesis Using Diazoethane 
Starting acid 

1- Naphthoic 
2- Naphthoic 
pr op ionic 

Yield of 

71$ + 7% oil 
51# + 16$ oil 



Yield of rearranged acid 
based on starting acid 





Arndt-Eistert Synthesis Using Diazopropane 

Starting acid 




Yield of 
diazo ketone 




Yield of rearranged acid 
based on starting acid 



■■ • ■ 

•■ '") 

. .. ■ <■, ■ ■ 


, ■ 
I . ■ 

''■:•■ + Mb 


: ^ • 

■: ■ ■ 


- ' ; . 

;s y-'.> 

':.. -j: ::r.u 


■ ■ <■-.!«■■ 

* . - •■ - ■ 

■_.^- ; - 




1. Arndt and' Eistert, Ber . , 68, 200 (1935). 

2. Bachmann and Struve, "Organic Reactions" , John Wiley and Sons, 

Inc., New York, N. Y., 1042, Vol. 1, pp. 38-62. 

3. Huggett, Arnold, and Taylor, J. Am. Chem. 3oc, _64, 3043 (1942), 

4. Eistert, Angew, Chem., _54, 124 (1941) . 

5. Wilds and Meader, J. Org. Chem., 13, 763 (1948). 

Reported by Paul M. Mader 
January 14, 1948 



Recently it has been suggested ( 
consideration in aromatic substitutio 
the intermediates through which the r 
commonly, the tacit assumption is mad 
the point of highest electron density 
focused upon the factors which produc 
of the energy of the transition state 
lation of several seemingly unrelated 
predominance of the para isomer in _or 
benzene series; b) the almost complet 
in such compounds as p-naphthol and i 
position to which entering groups are 
aromatic and heterocyclic compounds. 

l) that the most important 
n is the energy content of 
eaction progresses. More 
e that substitution occurs at 

and attention has been 
e such charges. Consideration 
, however, permits the corre- 

phenomena, including: a) the 
tho - para substitution in the 
e inertness of the 3-posltion 
soquinoline; and c) the 

directed in polynuclear 

It is generally accepted that aromatic substitution proceeds 

as follows: (2) , _ , «. 

'catalyst _ _ 

H : Z f~ > U + Z^ 

+ Z 








H' X Z 

The transition state theory proposes that the reaction will 
proceed through the intermediate (transition state complex) which 
is most readily formed (the one with the lowest energy content). 

Examples : (The intermediate of the favored product is under- 
lined. ) 

A. Para substitution predominates over ortho. 







Explanation: The p-quinoid complex is more readily formed by 
analogy with the fact that p-quinone has a loiter energy content 
than o-quinone. (As shown by their oxidation-reduction potentials. 

B% Position of substitution in other aromatic systems. 

1. The 3-position in p-naphthol is inert. 
H Z 

^— OH 



2. The substitution of indole. 











3. Heactivity if the the methyl groups in 1- methyl and 3- methyl 

n: e 



4. Positions of substitution in oolynuclear hydrocarbons. 
(Only the favored intermediates are shown.) 


- ■ 









Explanation: Reaction occurs at such points that benzenoid 
resonance is interrupted in the fewest number of rings. (Note the 
similarity of Fries rule.) (4) 

C. Substitution of 5- hydroxy indane. 



Explanation: XIV 
rings is stretched b 
of a single bond. (5 

is favored because the bond common to both 
the five membered ring to a length near that 

It has been suggested that free radical substitution can also 
be explained by similar considerations. It would not be surprising 
if elucidation of the free radical mechanism proved this so. (1,6) 





Waters, J. Chem . Soc . , 1948 , 727. 

Price, "Mechanisms of Reactions at Carbon - Carbon Double 

Bonds", Inter science Publishers, Inc., New York, 1946, Chapt-2, 
Hammick and Mason, J. Chem. Soc, 1946 , 640. 
Fries, Ann., 454, 121 (1927). 
Wheland, "The Theory of Resonance", John Wiley and Sons, 

1944, p. 270-2. 
Wheland, J. Am. Chem, Soc, 64, 900 (1942). 

Reported by G-eorge R. 
January 14, 1949 


• . ., r 


.. - : . -• 

CHEMISTRY 435 II Semester 1948-49 

The Naphthenic Acids 

H. E. Baumgarten, February 11 1 

Recent Syntheses of OXazoles and 2-0xazolines 

R. M. Ross, February 11 6 

The Structure of Patulin and Investigations Relating 
to its Synthesis 

Robert A. Hardy , Jr. , February 18 9 

Hexahydroxybenz ene (Benzene Hexol) 

I. Moyer Hunsberger, February 18 12 

Synthesis of t-Qarbinamines 

Karl F. Heumann, February 25 16 

Ant ihistaminic Drugs 

George I. Poos, February 25 19 

Orientation in Aliphatic Ohlorination 

John Lynda Anderson, March 4 24 


Claire Blue stein, March 4 27 

The Use of LiAlH 4 and Related Comoounds in Organic Chemistry 

H. Wayne Hill, Jr., March 11 30 

Recent Developments in the Catalytic Hydroge nation of 
Organic Compounds 

Melvin I. Kohan, March 11 33 

The Willgerodt Reaction 

P . D . Cae sar, March 18 37 

The Mechanism of the Fries Reaction 

Robert E. Ca.rnahan, March 18 41 

The Autoxidation of Tetralin 

Robert G. Bannister, March 25 44 

Synthesis of the Cyclobutane Ring 

Aaron B. Herrlck, March 25 47 

Survey of Decarboxylation Reactions 

H. A. DeWalt, Jr., April 1 51 

The Use of Sodium in the Preparation of Tertiary Alcohols 

Carl S. Hornberger, Jr., April 1 ' 55 


Some 1,4- Addition Reactions of Conjugated Systems 

Emil W. G-rieshaber, April 3~ 58 

Recent Studies on Furan 

William R. Miller, April 8 61 

Ethylene Imine Ketones 

Sidney Baldwin, April 22 64 

Factors Influencing Elimination Reactions 

George R. Coraor, April 29 68 

An Electronic Interpretation of the Decomposition Re- 
actions of Aromatic Diazo- compounds in Aqueous 

K. H. Takemura, April 29 72 

The Elbs Persulfate Oxidation of Phenols 

H. A. DeWald, May 6... 76 

Peroxide Catalyzed Reaction of Alkyl Benzenes with 
Maleic Anhydride 

Edward F . Riener, May 6 79 

Aiken e Sulfides 

E. F. Els lager, May 13 82 

The Synthesis and Structure of Sempervirine 

Charles W. Fairbanks, May 20 87 

Isomerism of Lysergic Acid 

Allen B . Simon, May 20 90 

Reported by H. E. Baumgarten February XI, 1949 

The naphthenic acids are saturated alicyclic aoids, the majority 
of which have the empirical formulas, C n H 3n _ 3 3 and C n H 3n _ 4 © 3 . They 
are probably naturally occurring constituents of all crude petroleum 
oils, occurring in amounts variously estimated as between 0.03 and 
Z>% (l). The methods used for the separation of the naphthenic acids 
from petroleum, their purification, and their uses have been dis- 
cussed in a review (l) . 

Structure Studies. 

The first acidic compounds from petroleum were reported by 
Pebal (2) in 1860 (of., however, (3)) and from that date until about 
1930 the major contributions to the knowledge of the naphthenic 
acids structures were vague and often incorrect generalizations. 

1. The Naphthene Nucleus * 

As early as 1874 Hell and Medinger (4) suggested that the 
naphthenic acids have a cyclic nucleus. Markovnikov (5) related the 
acids to the cyclic hydrocarbons, the naphthene s, and was the first 
to call them " naphthene sauren. " Actually most petroleum acids are 
mixtures of aliphatic (fatty) and alicyclic (naphthenic) acids in 
which the former usually occur to the extent of approximately 5f C) 
the exact amount depending on the source and treatment of the crude 
acids. In general crude naphthenic acids contain aliphatic acids 
up to C 10 , monocyclic naphthenic acids from C s to C 13 , bicyclic 
naphthenic acids above i3 , and poly cyclic naphthenic acids above 
about C 14 (some as high as C 39 ) (6,7,8,9,10,11). 

2. Size of the Ring . 

For many years some workers believed that there were 
either very few or no natural naphthenic acids having the six- 
membered ring (1,12), but today we know from complete structural 
studies that both cyclopentane and cyclohexane derivatives are found 
in the naturally occurring naphthenic acids. Apparently five- 
membered rings predominate. G-oheen (ll) reports that very high 
molecular weight naphthenic acids contain two to three f ive-membered 
rings per molecule. As yet no one has reported the presence of the 
six-membered ring in the polycyclic naphthenic acid molecule. 

3. Linkage of the Carboxyl . 

Primary, secondary, and tertiary acids have been isolated 
and identified from naphthenic acid fractions. Xn general primary 
and secondary acids predominate (6,7,8) with the primary probably 
the more abundant of the two (of . .however, (6) J. For new structure 
studies the methods used for determining the linkage of the carboxyl 
are of considerable importance. The method of von Braun (7,13) is 
probably the most satisfactory. 


(a) RCHaCONHR' + 3PC1 5 -> 
RCC1 2 G-C1 + HoO 

RCC1 3 G-C1 + P0C1 3 + 2PC1 3 + 3HC1 

RCC1 3 C0NHR' + HC1 

1» 2PC1 

(b) R 3 CHCONHR' . L* R 3 CC1C0NHR' 

2* H 3 

1, PC1 5 

(c) RaCHCONHR 1 ) R 3 CCONHR» 

2. H 3 

R = aliphatic or 

R r = ethyl, methyl, or 

The methods of Chichibabin (6) and Lapkin (8) are less satisfactory. 
The method of Whit more and Crooks (14) has given erroneous results 
in the naphthenic acid series (15). 

Although all of the primary acids identified to date have been 
acetic acid derivatives, von Braun (7,13) has presented evidence 
that there may be as many as three methylene grouos between the ring 
and the carboxyl in some naphthenic acids, i.e., the structure 
-CH 3 CH 3 CH 3 C0 3 H. ' 

C 9 H 17 COOH 

1. P + Br. 

2. EtOH 



C 7 H 13 CH=CHC00H 

H 3 S0 4 

+ H 3 

CC 6 H 11 GH=CHCH 3 COOH]^ 

C 6 H X1 CH-CH 3 -CH ? -C=0 


H0 3 C-CH 3 CH 3 -C0 3 H 

(yield of lactone: 
20^ overall) 

4* Isomers. 

most of 
of such 


A tremendous number of structural isomers are possible for 
the naphthenic acids and the indications are that a large 
isomers do exist. Both optical (16) and cis-trans (17) 
occur in the natural naphthenic acids. 

The Individual Acids . 

At the present time about a dozen naphthenic acids have 
been Isolated from petroleum and identified more or less satisfact- 
orily. Some of these have been isolated and identified through 
rather conventional analytical procedures; others have been identifiec 
through degradative studies. All of the acids whose structures are 
reasonably well known are listed in Table II. Some of the more in- 
teresting studies are listed here. 


\ ".. 

' »s' , ■' <-- 



a . Hrans-''2,2,6-Trimethylcyclohexanecarboxylic Acids . 

^ i n o^Y* 3 ' H ^ re cz.V, Wash, and Lochte (15) isolated a solid 

acid im.p, 83 C.) from California petroleum. Application of the 
von Braun method showed it to be secondary. The acid was degraded 
as follows: 

HN 3 HN0 3 

C 9 H 17 C00H -> C 9 H 17 NH 3 -> 



■CH S 3 


•CH 3 



C=CK 3 
CH 3 

CH 3 jCH-CH. 

The ketone was identified through comparison with an authentic 
sample. bince a Dem'yanov rearrangement was possible in the nitrous 
acid treatment, all of the amines which could rearrange (by Whitmore' 
theory (18j) to give the olefin indicated were formulated.' Then all 
?i ^f fcids corresponding to these amines were synthesized. Final 
identification was through comparison of the natural and 'the 
synthetic acids. 


5-Dimethylcyclopentaneacetin A c id. 


„_, .. . _ Ne i> Crouch, Rannefeld, and Lochte (17) isolated 

acid through fractional distillation of the methyl ester and 
fractional^ neutralization of the acid. The von Braun method indi- 

methn/fn^ lm f r ry a 2i d *,. The acld Was ^graded by the Barb ier-Wie land 
method following the directions of Skrauo and Schwamberger (19): 

C7Hi 3 CH 3 C00H 

2$MgBr OH 

7 H 18 CH a 0OOMe -* C^ 3 CH 3 C (C 6 H 5 ) 

'6^5 I 3 

Cr0 3 
C 7 H 13 CH=C(C S H B ) 2 -* C 7 H 13 C00H 


^!« a °^ d ° f ° n f less carbon a tom than the original was identified 

neacelic'aMd^ 011 ?$t *? a f hentic ***#*< 2,3-Dimethylcyclooenta- 

° ^ S s ^ nthesized and was shown to be identical with the 

*-'"'- -^o J- na J. a c id . 

c# ^, 5, 4-Trlmethylc.y clopentaneacetic Acid . 

acids from varlnn^cnn^ (7 'u 0) isolat;ed a lo fraction of naphtheni 
touehd7J?^«? sources. He was unable to purify it satisfactory 
tnrough distillation of the acids or their methyl esters Purifica- 

crystalliz'ftion^of 1 ^' 10 "- ' the ™ Xm ^ Schmidt reaction an™* 
S^Sd^SAS 8 .?^^*? ° Xalate WaS m ° re Batisfacto^, 

C 9 H 17 COOH 



^ sHl 5 CHgNH; 

3 Aldehydes 
— » Ketones 
Condensation products 

C 8 H 15 CH 3 (CH 3 ) 3 0H 

H s 2 


C 8 Kx 4 3 + C 8 H 14 

One- third of the final product was the ketone, C e H 14 0, which was 
purified through distillation and crystallization of the semicarba- • 
zone. The ketone reacted with two moles of o-nitrobenzaldehyde to > 
give a di-p_-nitrobenzal derivative, indicating the structure' 
-CH 3 -CO-CH 2 -. This bit of evidence eliminated all but eleven of the 
eight carbon ketones, von Braun was able to synthesize or obtain 
data on only ten of these ketones, von Braun' s ketone and its de- 
rivatives did not correspond to any of these ten ketones, so by 
elimination von Braun concluded that his ketone must be the ketone 
he could not synthesize, 3,3, 4-trimethylcyclopentanone. From the 
purified ketone he regenerated the original acid (through a 
Reformatsky reaction, followed by dehydration, then reduction). For 
ketone properties see Table I. 

In 1942 Buchman and Sargent (2l) were able to synthesize 3,3 4- 
trimethylcyclopentanone by two independent routes (the product of 
the second route was identified in terms of the first). Comoa risen 
of their synthetic product with the degradation product of von Braun 
indicated to them that the two ketones could not be identical. Tv>u . 
they claimed von Braun' s structure to be in error. For comparison : : 
of ketones see Table I. 

In 1948 Mukherji (22) synthesized 3,3, 4-trimethylcyclopentanone 
D.y a third route and obtained a ketone having very nearly the same 
properties as the von Braun ketone. The ketone was converted to the 
napnthenic acid by approximately the same route as that used bv 
von Braun; the properties of the synthetic acid were very nearly 
identical with those reported by von Braun for the natural napnthenic 
acid. See Table I. 

The von Braun ketone has been obtained as an impurity in the ''■ 
degradation products from a nine carbon naphthenic acid isolated 
: r r *°J! ^ Petroleum (23). A ket one that appears to be identical ' 
with the y on Braun ke tone was isolated from wood extracts by Prine;- 
sheim (24). See Table I. 6 


von Braun 
Buch man- 
Mukhe r j i 
Aruba acids 




m.p. Semi- m.p.Di-p- 

carbazone nitrobenzal b.p. Oxime 


(12 mm.) 
at 25°. 

213. 5- 




a204, 7-05.1 



116- 20 (14mm.) 
(m.p. = 99. 8- 

115 (12mm.) 

1.4390 0.895 
1.4386* 0.392 : 




• ' • 

. . 






• . - : 

. ■ ■:;::; _ • ■ ' 


. ■ 

■ - . 




Naphthenio acid identified: 











I (28) 


Cyclopentanecarboxylic acid 

Cyclohexanecarboxylic acid (hexahydrobenzoic) 

2-Methylcyclopentanecarboxylic acid 

3-Methylcyclopentanecarboxylic acid 

p-He xahydrotoluic acid 

"Syclopentaneacetic acid 

3-Methylcyclopentaneacetic acid 

2, 3-Dimethylcy elope ntaneacetic acid 

jcis-2, 2, 6-Trimethylcyclohexanecarboxylic acid 

trans- 2, 2, 6-Trimethylcyclohexanecarboxylic acid 

dl- Camphonanic acid (l, 2, 2-trimethylcyclopentanecarboxylic 

3,3,4-Trimethylcyclopentaneacetic acid (?) 

♦Abstract of (27) called acid m-hexahydrotoluic acid, but properties 
listed were those of o-hexahydrotoluic acid; hence, the listing here, 


1. Littman and Klotz, Chem. Rev., 30, 97 (1942). 

2. Pebal, Ann. Chem. Pha rm. , 140 , 19~7l860) ; Chem. Zentr. , 1860 832. 

3. Zelinsky, Ber., _57B, 42 T1924) . 

4. Hell and Medinger, Ber. 7, 1216 (1874); 10, 451 (1877). 

5. Markovnikov, Ber., _16, 1873 (1883); J. prakt. Chem., _45 , 574 (18 92) , 

6. Chichibabin, Compt . rend. acad. sci. U.R.S.S. , 1930A, 382; 
Chemie and Industrie, _27, Special No. 3, 306 (I93^y7 

7. von Braun, Z. angex^, Chem., 44, 661 (1931) ; von Braun, Keller, 
Weissbach, Ann., _470, 100-179 (l93l) . 

8. Lapkin, Bull, inst . recherces biol. Perm., 8, 51 (1931). 

9. Balada and Wegiel, Chem. Obzor., 11, 187 (1936); C.A., 31, 2399 
(1936) . 

10. Harkness and Bruun, Ind. Eng. Chem,, 32, 499 (1940). 

11. G-oheen, Ind. Eng. Chem., J32, 503 (l 94*57 . 

12. cf, reference (3); Aschan, Ber., J>4, 2617 (1891); Ann., J324, 1 
(1902); Kharichkov, Chem., Ztg., 36, 1378 (1912). 

13. von Braun, Allgem. 01. u. Fettztg., 25, Minerole, 1; Chem. Zentr, 
19251 , 2876. 

14. Whit more and Crooks, J. Am. Chem. Soc, ,60, 2078 (1938). 

15. Shive, Horeczy, Wash, Lochte, J. Am. Chem. Soc, 64, 385 (1942). 

16. Bushong and Humphrey, Orig. Com. 8th Intern. Congr. Appl. Chem. 
6, 57 (1911). 

17. Ney, Crouch, Rannefeld, Lochte, J. Am. Chem. Soc, 65, 770 (1943) 
Quebedeaux, Wash, Ney, Crouch, Lochte, Ibid . , 767. 

18. Whitmore, J. Am. Chem. Soc, _54, 3274 (1932). 

19. Skraup and Schwamberger, Ann,, 462, 141 (1928). 
20,. von Braun, Mannes, Reuter, Ber., 66, 1499 (1933). 

21. Buchman and Sargent, J. Org, Chem., 7, 148 (1942); Sargent, Ibid. 
154. *~ 

22. Mukherji, Science and Culture, 13. 296 (1948). 

23. Baumgarten and Richter, Reg ional^Mee ting of A.C.S., 
Texas, December 1947. 

24. Pringsheim and Schrciber, Ce 1 luloschemie, 8, 45 (1927). 
§§* T^ 1 ^ 301 ^ pimitrie. Isacescu. Volrap, BerT, 71, 2056 (l938h 

26. Schutze, Shive, Lochte, Ind. £ng. Chem., Anal—£d. , 12, 262 (1940 

27. Pekov, J. Russ. Phys. Chem. Soc, 46, 178 (19: 

28. Hancock, Lochte, J. Am. Chem. Soc.T~51, 2448 




Reported by R. M. Ross 

February 11, 1949 

Oxazoles and 2-oxazoline s are represented by the following 
structural formulas: 





-fi ^ c " 


an oxazole 






S \ /3 

a S-oxazoline 

Interest shown in this class of compounds stems from their 
pharmacological action, their close relationships to naturally occur- 
ring products, and their unusual chemical properties. Although two ' 
review articles are available which discuss these het erocycle s, the 
chemistry of oxazoles and 2-oxazolines is relatively incomplete. 

The remainder of this seminar will be limited to a discussion 
of some newer preparations of various oxazoles and 2-oxazolines. 
No attempt will be made to cover the entire synthetic field; for 
such information those interested are referred to Wiley's publi- 
cations (1,2) and the Ph. D. theses of Leffler (3) and Sparks. (4) 

From q-Amino Acids 

Oxazole s 

Starting with certain a-amino acids, Wiley (5,6) has modified 
Wrede's and Feurriegel's (7) early work to the point wherein quite 
respectable over-all yields of substituted oxazoles may be obtained. 

C S H S 9HC0 3 H 


Ac 3 


C 6 H 5 9HC0CH 3 H 3 S0 4 
NHCOCH 3 -» 

-H 3 

C s H 5 gKC0 2 H 

CeHc C- 





Ac 3 


C 6 H 5 pHC0CH 3 


The final step in the process, i.e., dehydration of the N-acyl 
ketone, is a classical oxazole synthesis to be credited to Robinson 
{&) . Using Wiley's procedure it is oossible to prepare 2,5-dimethyl 
derivatives with varying substitutents on carbon atom four. The 
method is straightforward and easily carried out. Thus far, Wiley's 
procedure and that of Wrede and Feurriegel have been applied success- 
fully to the following a-amino acids: glycine, alanine, valine, 
leucine, phenylalanine, tyrosine and glutamic acid; the use of 
asparagine, tryptophan and formyl glycine has been unsuccessful. 


The Cornforth Synthesis (9) 

Recently, Cornforth and Cornforth reported an excellent syn- 
thesis of oxazoles which not only offers good yields, but which 
shows promise of being applicable to a wide variety of substituted 
oxazolest Starting with ethyl iminoacetate and ethyl glycinate 
hydrochloride, the following process leads to the formation of either 
2-methyl-4-carbethoxyoxazole or 2-methyloxazole , 





CH 3 NH 3 C0 ; 



Et 3 

CH 3 C=NCH 3 C0 3 Et 

KOEt, HC0 3 Et 

CH 3 (? 





CH 3 C = N 
i ! 

OEt /,CC0 3 St 



C0 3 Et 



■ N 

1. KOH 



CCH 3 2. -CO; 








Until the Cornforth synthesis was reported, no preparation of 
oxazole itself had been effected. A minor variant of the procedure 
shown x\ras employed by the Cornforths to yield the parent member of 
the series, oxazole. The method was extended to the preparation 
of 2-benzyl-4-carbethoxyoxazole (lo) last year with good results. 
Because of this, the Cornforth synthesis would seem to be applicable 
to the synthesis of 2-phenyloxazole, which was obtained for the first 
time in 1942 by Cass, (ll) in quite poor yields. 

It should be pointed out that the intermediates in the Cornforth 
synthesis are attacked readily by moisture, air, etc. Therefore, no 
undue delay should be allowed in carrying out the preparation. 

From N-Acyl-fl-Amino Alcohols 

Cyclization of N-acyl-p-amino alcohols using sulfuric acid (12) 
or thionyl chloride (13) has resulted in the formation of a number 
of 2-oxazolines in very good yields. 





NHCH 3 C< 

N N-CH 3 

In 1948 an application of the thionyl chloride cyclization proved 
most fruitful in obtaining 2, 5-dimethyl-4-carbethoxy-2-oxazoline , 
the key intermediate in a novel synthesis of DL- threonine . (14) 



^0 gH 3 J) CH 3 30C1 3 /0-CHCH; 

CH 3 C X C=0 -* CH 3 C HOCH -> CH 3 C | 

NrTTTATT^/-v TT.U \T,TTTATTn^ T^+- V* 1 


DL- threonine 


Ring closure "by the sulfuric acid method is limited to amides in 
which the hydroxy 1 group is on a secondary or tertiary carbon atom. 
Amides containing hydroxyl groups on primary, secondary or tertiary 
carbon atoms, however, have been cyclized using thionyl chloride. (3) 
Some 2-oxazolines prepared by these routes are: 2,5-diphenyl-, 2- 
phenyl-5- carbomethoxymethyl- , 2-phenyl-4-carbomethoxymethyl-, and 

From Imlno Esters 

Bockmuhl and Knoll (15) reported successful condensations of 
imino ester hydrochlorides, derived from fatty acids, with a-amino- 
0-hydroxy compounds to produce substituted 2-oxazollnes. A similar 
type of condensation has been applied recently to the preparation 
of 2-benzyl-4-carbethoxy-2-oxazoiine (10) with good results, 

» <?0 3 Et 

C0 3 Et HN „ 25 OH. N 

CCH 3 C S H 5 -► j || + NH 4 C1 

IC1 + EtO^ 

CH 3 0H 

CHNK 3 HC1 + EtO^ CH 3 CCH 3 C S H 5 

\0 X 


Among the 2-oxazolines reported by the Bockmuhl and Knoll 
process are 2~pentadecyl-5-diethylaminomethyl-2-oxazoline and 2- 


1. Wiley, Chem. Rev. 37, 401 (1945). 

2. wHey and Bennett, Chem. Rev., (in press). 

3. Leffler, Ph. D. Thesis, University of Illinois (1936). 

4. Sparks, Ph. D. Thesis, University of Illinois (1936). 

5. Wiley, J. Org, Chem. 12, 43 (1947). 

6. Wiley and Borum, J. Am. Chem. Soc. , ^70, 2005 (1948 ). 

7. Wrede and Feurrlegcl, Z. Physiol. Chem.. 218, 129 (1933). 

8. Robinson, J. Chem. Soc. , 95, 2167 (1909) . 

9. Cornforth and Cornforth, J. Chem. Soc, 96 (1947). 

10. Ross, Ph. D. Thesis, University of Wisconsin (1948). 

11. Cass, J. Am. Chem. Soc, 64, 785 (1942) 

12. Takeda, J. Pharm. Soc, Japan, _426, 691 (1917). 

13. Bergmann and Brand, Ber. , 56B, 1280 (1923). 

14. Pfister, Robinson, Shablca and Tishler, J. Am. Chem. Soc, 70, 
2297 (1948). 

15. Bockmuhl and Knoll, U, S. Patent 1,958,529 (1929). 


Reported by Robert A. Hardy, Jr. February 18, 1949 

Patulln is a bactericidal compound obtained from a variety of 
mould organisms, and has been variously named according to the 
source from which it is isolated. Patulin from Penicillium 
patulum Banier (l), claviformin from Penicillium claviforme (2), 
clavacin or clavatin from Aspergillus clavatus (No. 129) (3), and 
expansine from Penicillium expansum Westl. (4) are the same com- 
pound as is conclusively shown by comparison of the Physical and 
chemical properties of these substances (4,5,6,7). Structural 
investigations (1,4,8) have shown that this compound is probably 

anhydro-3-hydroxymethylenetetrahydro-Y'-pyrone-2-carboxylic acid' (i) 
and/ or its keto-enol isomer (II) . 

!H 3 C=CH -+ CH^ "C=CH 

L jy> *- i 

Ha CH-CO CH 3 

? > 


Patulin, C 7 H 6 4 , is an optically inactive, neutral compound 
which is soluble in water and most organic solvents. The Presence 
of one carbonyl group is shown by the formation of a mono-phenyl- 
hydrazone and a mono- oxi me (8). Patulin forms an easily hydrolyzed 
mono-acetate (l) (and other esters (4)); the mono-acetate when 
treated with a HC1 solution of phenylhydrazine gives the same phenyl- 
hydrazone as that formed from patulin itself. A Zerewltinoff deter- 
mination shows the presence of one active hydrogen per molecule. 
This evidence would indicate a keto-enol grouping. Deeolorization 
of cold alkaline permanganate (l), bromine titration (4), and 
perbenzoic acid oxidation (4) show the presence of at least one 
double bond; one mole of bromine adds very rapidly followed by 
gradual utilisation of 1-2 additional moles which 1 may involve 
cleavage of the molecule. A freshly prepared aqueous solution of 
patulin does not give a coloration with FeCl 3 , or a Schiff test, 
but reduces cold ammoniacal silver nitrate and Fehling's solution 
when warmed (1). After standing, the aqueous solution becomes acid 
and now gives a positive Schiff test and a typical enol reaction 
with FeCl 3 (4). No methoxyl groups could be found by Zeisel deter- 
mination (4) and only traces of C-methyl were found by the Kuhn- 
Koth method (1,4). An attempted periodic acid oxidation showed 
that patulin does not contain two adjacent carbon atoms bound to 

The behavior of patulin im alkaline solution (1,4) suggests a 
lactone as the ring is slowly opened forming an acid, and two 
moles of alkalai are consumed. Also, a lactone has been isolated 

- .. . 

. I ■ 


■ ' 





:-> > ■ ■ : ■-.'[ 


■ :. . I ■•' ■■■. ■ '■ ■ ■■■ ■ "■' ■ ■■■ 

s ':''■■ . • ■''•' 

• i ■ : ■ ; . • 

! ' 


" I • •■ 

....... :■ ■ w ; 

..." . - 

. . . ' - 


'"■'.' j- .. 

\ - I A ! J . . . 3 • 



. "■ j * k ' ; v, x. ■ ' . ' * -' ' 

■ \ .. : '-- .-■•■ ' - - ■'" - ■' • ••■"' 

,-■■'. .. . j. . . • •..,.- 

» - ... - . ' ■ ! 


, •• • . - ■ ' 

> ■ "- •". 

.... ' ■ 



as a degradation product of patulin; hydrogenation followed "by 
treatment with HBr and a second hydrogenation (to remove bromine) 
has yielded p-n-propyl butyrolactone (III) (8). The reduction of 
cold ammoniacal silver nitrate and a positive color test with 
sodium nitroprusside (Legal* s test) (4) indicate thpt patulin con- 
tains an unsaturated lactone grouping, probably a Ap>" -unsaturated-* 
T- lactone. Piecing this information together patulin must contain 
the grouping shown by IV, which is the lactone of a p-aldehyde 

^h 8v ^\ ;{ x 

CH 3 OH— CHg, -C- C=CH X H 3 -C- x C=CHOH -> -C^ x C-CHO 

I I s° \ I > ~> ■ | I f- | | 

CH 3 CH 3 -C0 -C- -0-C0 x -C- -C-COOH -C- --C-COO' 

1 l • i I i 


Opening the ring to form an acid which colors FeCl 3 solution and 
also gives a positive Schiff test would be represented by conversion 
to structures V and VI. The structure shown in IV would' also ex- 
plain the formation of formic acid on ozonolysis (4,8), the for- 
mation of a dimethone derivative (4), and the slow uptake of a 
second molecule of hydroxy lamine during titration (4). 

A study of the products of acid hydrolysis will settle the re- 
maining structure of the patulin molecule, including the position 
of the free carbDnyl group, Raistrick and co-workers (l) have iso- 
lated one mole of formic acid and a 10% yield of inactive tetrahydro- 
ZT-pyrone-2-carboxylic acid (VIl) . This establishes the ^T-pyrone 
ring and also the location of the free carbonyl group, and leads to 
structure I for the patulin molecule. 

CH 3 X GH 3 CH 3 "CH 3 CHf V CH-CH 




Other degradation products which have been isolated are f- 
keto-C-iodo-n-hexanoic acid (VIII) by treatment with concentrated 
HI (1), T-keto-$-methyl-n-hexanoic acid (IX) (8), the lactone of 
p-me thy 1- "^hydroxy- n-hexanoic acid (l) ^nd p-methylcaoroic acid (l). 
After ozonolysis (4,8) the products isolated include one mole of 
U0 3 , one mole of formic acid, glycolic aldehyde, glyoxal, and 
oxalic acid. AH of these degradation products are consistent with 
structures I and II for the patulin molecule. 






Attempts at synthesis have not yielded a compound identical 
with the natural product, but have given an ■isomer which i s an 4h,- 
^-unsaturated lactone while patulin contains the A |3, JT- unsaturated 
lactone ring. This synthesis has been carried out by two different 
groups of investigators (9,10) working independently. The general 
method involves the Claisen condensation of the appropriate methyl 
ketone to yield a 2, 4-diketo ester, followed by treatment of the 
sodium enolate with formaldehyde. This gives the corresponding 
a-keto-p-acyl-butyrolactone which is cyclized to the dihydropyrone. 
The following reactions illustrate this synthesis: 

8 J 

GK^ ^CH 3 NaOCH 3 CH^ " X CH g NaOH 
COOR -» j | -► 


— ^n 

GH 3 GH 3 


CH C-CH 2 , 

CH 2 C-CO^ 


The product of this synthesis is not identical with patulin. It 
has a lower melting point, very little bacteriostatic action com- 
pared to patulin, and differs from patulin in its chemical be- 
havior (10) . Attempts to form a mono-acetate under the same 
conditions which patulin forms a mono-acetate left X unchanged. 


1. Raistrick, et. al., Lancet 245 , 625 (1913). 

2. Chain Florey, and Jennings, British J. Exptl. Path. 23, 202 
(1942). " 

3. Waksman, Horning, and Spencer, Science 96, 202 (1942), and 

J. Bact. 45, 233 (1943). 

4. Nauta, et. al . , Rec. trav. chim. 65, 865 (1946). 

5. Bergl, et. al., Nature 152, 750 (1943). 

6. Hooper, Anderson, Skell, and Carter, Science J39, 16 (1944). 

7. Chain, et. al., Lancet 246 , 112 (1944). 

8. Bergl, Morrison, Moss and Rinderknecht, J. Chem. Soc. , (1944) 

9. Puetzer, Nield, and Barry, J. Am. Chem. Soc, 67, 832 (1945). 
10. Fttldi, Fodor, and Demjen, J. Chem. Soc, (1948T~1295. 



•• ■ i ■ ■ i ' 

R e ported by I. Moyer Hunsberger February 18, 1949 

I. Syntheses of Hexahydroxybenzene . 

A. In 1862 Lerch (l) unfittingly prepared hexahydroxybenzene (i) 
by the reactions outlined below. Because of the peculiar nature of 
this synthesis (2,3,4) the structures of II and I were not elucidated 
until 1885 (5). At present this method possesses only historical 










K0-,/ W-OK 





B. The following synthesis (6,7,8) has definite preparative 
value despite the highly reactive intermediates. Very recently, 
the overall yield has been reported (9) as only 20^, but earlier 
claims were considerably higher. 



— > 




HN0 3 

H 3 S0 4 

N0 2 - 





SnCl 3 ,HCl NH 3 - 
88-96^ H0- 

Nitranilic Acid (ill) 

HN0 3 ( =1.4) 
§0%, crude 


I <- 

SnCl 3 ,HCl 




Triquinoyl (V) 

C, Tetrahydroxy-p-benzoquinone (vi) , prepared either by 
oxidative self-condensation of glyoxal (8-11) or by controlled 
oxidation of meso-inositol (VII) (12), can be satisfactorily reduced 
to I using either tin (8) or stannous chloride (9), but 45^ 
hydriodic acid apparently is most convenient and Rives a 70% 
yield (12). The nitric acid oxidation of VII gives a variety of 
products (13-15) unless a mixture of hydrochloric and hydriodic 
acids is added to stop the oxidation at VI (12). The preparation 



Of I from VII appears to be the moet desirable of all the available 
methods. OH 


CHO atm.O. 


' II I 

CHO alkali carbonateno 



3 J 







OH 1,3/2,4,5,6 

II. Properties of He xahydroxy benzene. 

A. Miscellaneous . I crystallizes from water (stannous chloride) 
on addition of hydrochloric acid. Pure I is an infusible grayish 
solid only slightly soluble in organic solvents. It instantly 
reduces cold silver nitrate and gives a transient violet color 
with ferric chloride (6) . Either I or II readily forms a hexa- 
acetate. On distillation with zinc, I yields benzene and diphenyl 

B. Oxidation of Hexahydroxybenzene . The most convincing evidenc 
for the trihydroquinone nature of I is afforded by its stepwise 
oxidation to VI, VIII, and V, procedures being available for iso- 
lating each of these in a pure state (1,6,15,16). Recently the 
oxidation-reduction potentials and ionization constants for this 
series have been determined (17). I, VI, VIII, and V all revert 




HO-^^s-OH [0] HO-f/X-OH [0] 'H0~/\ = 

II l| 

OH <- HO-i* Ji-OH <- HO- 





[H] V 

Rhodizonic Acid (Vlll) 



under alkaline conditions to croconic acid (IX), which in turn 
is easily oxidized to leuconic acid (x) . This formation of a five- 
from a six-membered ring (18-20) presumably proceeds via a benzil- 
benzilic acid type transformation (21) followed by decarboxylation. 
The structure of X follows from its hydrol ysis to glyoxal and 
mesoxalic acid (22) . 









0=C C=0 HC1 

I |/C0 3 K 

o=c c 


0= V 



N C X N 0H 









ft 5 , 

o=c — c=o 






ene. Wieland and Wishart's 
23) could not be repeated by 
recently (9) I was hydro- 
150°) to a complex mixture 
re isolated by tedious 
nd scyllitol were obtained in 
c process was responsible 

C. Reduction of Hexahydroxybenz 
catalytic hydrogenation of I to VII ( 
later workers (9,24), However, very 
genated (Raney nickel; 100 atm.; 125- 
from which five isomeric cyclitols we 
fractionation. Meso- inositol (VII) a 
_ca. equal amounts. That the catalyti 

for the isomerization is indicated* by the fact that meso-inositol 
(VII) remained unchanged under the conditions used to hydrogenate I, 

III. Syntheses of Hexarnethoxybenzene , Hexamethoxybenzene (XI ) 
became available in 1941 by two different routes (25,26) from 2,6- 
dimethoxyquinone (XIl), which in turn is prepared in excellent 
yield by nitric acid oxidation of pyrogallol trimethyl ether. Very 
recently XI has been produced in high yield by methylating I with 
excess dipzomethane (9). 

MeO-/\- OM e 1. -OCl! vA-OMe 1. Red.Acetylf X 

(Quant -) I 2. Hydrolysis 

~* .\ J -OMe — > 

2. NaOHe MeON^ 

3.Me 3 S0 4 , 



-OMe CH,N 






Na 3 S 3 4 (95^) 

Me 3 S0 4 +NaOH(84^ crude) 




Ac oi yy 



A1C1 3 


-OH I. Dakin(62#) 

v "J-C0CH 3 2.Me 3 30 4 +NaOH 

T (93#, crude) 

OMe ' 






. l.AcCl,AlCl 3 (25# 
2. Dakin(48^ ; crude 
3.K0H,Me 3 30 4 (25^, 
OMe crude) 


IV. Miscellaneous . 

A. Some twenty aliphatic and aromatic esters of I have been 
prepared. Interesting correlations exist be-tween the structure and 
melting points of these esters (8,12). 

B. I, VI, and V increase the electrical conductivity of boric 
acid T27). 

C. It seems reasonable that the I-VI-VIII-V equilibrium may 
be involved in the oxidation- reduction processes of living cells 
(17), for VII is widely distributed in nature and is an accessory 
growth factor for many organisms (28). Furthermore certain 
bacteria can convert VII to calcium rhodizonate (29), and the rat 
is able to convert VII to glucose (24), 


1. Lerch, Ann., 124 , 20 (1862). 

2. Liebig, Ann,, 11, 182 (1834). 

3. Brodie, Ann., 113 , 358 (i860). 

4. G-melin, Ann. Phys., 4, 31 (1800). 

5. Nietzki and Benckiser, Ber., JL8, 1833 (1885). 

6. Nietzki and Benckiser, Ber.. 18, 499 (1885). 

7. Henle, Ann., 350 , 330 (1906), 

8. Backer ard van der Baan, R e c trav, chim., 56, 1161 (1937). 

9. Anderson and Wallis, j. Am. Chem. Soc,, 70, 2931 (1948). 

10. Homolka, Ber., _54, 1393 (1921). 

11. Nomolka, D.R.P. 368, 741 (1°22); Chem. Zentrell., 1923 , II, 911. 

12. Neifert and Bartow, J, Am. Chem. Soc . , 6>5, 1770 (1943) . 

13. Maquenne, Compt . rend,, 104 , 298 (1886). 

14. Contardi, Gazz. chim. ital . , 51, J f 107 (1921). 

15. G-elormini and Artz, J. Am. Chem. Soc, _52, 2483 (1930). 

16. Preisler and Berger, J. Am. Chem. Soc, J34, 67 (1942). 

17. Preisler, Berger, and Hill, J. Am. Chem. Soc, 69, 326 (1947). 

18. Nietzki and Benckiser, Ber., 19, 293 (1886). 

19. Nietzki and Benckiser, Ber., 19, 772 (1836). 

20. Nietzki, Ber., 20, 1617 (188777 

21. Karrer, "Organic Chemistry," Elsevier Publishing Co. (1946) , p. 641 

22. Homolka, Ber., _55, 1310 (1922). 

23. Wieland and Wishart, Ber., 47, 2082 (1914). 

24. Stetten and Stctten, J. Biol. Chem., 164, 85 (1946), 

25. Robinson and Vasey, J. Chem. Soc, 66"cTTl941). 

26. Baker, J. Chem. Soc, 662 (1941). 

27. Boeseken and Meuwissen, Rec. trav. chim., 45, 496 (1926). 

28. . Eastcott, J. Phys. Chem.., _32, 1094 (1928). 

29. Kluyver, Hof, and Boezaardt, Enzymologia, 7, 257 (1939). 


Reported by Karl F. Heumann February 25 1949 

Carbinamines are related to methylamine in the same manner 
that carbinols are related to methyl alcohol. 

The earliest preparation of a Jt-carbinamine (specifically, 
t-butylamine) is that of Linneman and Brauner (l) who used the 
following procedure (yield about 45^>: 

1. 2AgNC0 + (CH 3 ) 3 CHCH 8 I -» Agl + (CK 3 ) 3 0=GH 2 + OC-N-Ag 


2. 0=n-M-Ag 

HN-6=0 + (CH 3 ) 3 CHCH 3 I-»AgI + 0=<J-N-C(CH s ). 


3> 0=9-N-0(CH 3 ). + 4K0H->2K 3 C0 3 + NH 3 + (CH 3 ) 3 CNH 3 


Coleman, et al, (2) reported the preparation from t -butyl- 
magnesium chloride and chloramine (NHgCl), but the instability of 
the latter made the reaction undesirable: 

RMgX + HH 3 C1 -► RNH 3 + MgXCl (60fc) 

A high yield (85fo) characterized the preparation of Klages. 
et al. (3): & ' 

/ \ HC1 H-9-Ni 

(CH a ) 3 C=N-N=C(CH 3 ) 3 + 0H 3 MgBr-» -> t-BuNHNH 8 *HCl ~* * t-BuNH 3 

ice 17qo 

A general method was reoorted by Mentzer, et al. (4): 

R p NaHF R 

ArOH 3 01 + Na<J-C-0 -» ArCH a fi-g-/jJ - a ArCH 3 C-CONH 2 

R'O ft«'o Jt s 


-* ArCH 3 (?-N=C=0 -* ArCH 3 (J-NH 3 -HCl 

R» R . 

(R and R 1 may be alkyl, aryl or aralkyl) 

«Ptinn G ^ e ' AU !? and ^ e8 } le (5 ) Prepared carbinamines by the re- 
action of an active nitrile and a G-rignardr 

EtOOH 3 CN + 2CH 3 =CHCH 3 MgBr -> (CH 3 =CHCH 3 ) 3 C-NH 3 

CH 3 OEt 



An interesting reaction was used by Karabinos and Seriian (6) 
to prepare t-BuNH 3 in 70^ yield: ; 

H 3 ,Ni 
(CH 3 ) 3 C ! - ■ CH a -> (CK 3 ) 3 CNH 3 (no isobutylamine) 

X N' 130 , 700# 


Campbell, Sommers and Campbell (7) repeated the hvdroee nation at 
low pressure (60# a 60°, Raney Ni, in dioxan) n ^ aro e e natlon a ^ 

Cairns T] ?8)^ 2 "" dimethylethyleneimine W&S made by the mGthod of 

/ s ^^ 4 Na 0T J 

(CH 3 ) 3 C-CH 3 0H -+ (CH 3 ) 3 9-CH 3 030 3 H -> (CH 3 ) 3 C-CJ 

3/ 3^ ~ ^n 8 


N T ato S m C °bS? U i? Ill n^ Slnal J y r epar6d t0 test the symmetry of the 
u atom, but it was not resolved; see also Adams and Cairns (9)). 

of three^tenffq^S Tf"f ° f **?&***■* of t-butylamine consisted 
01 -cnree steps I Smith and Emerson (lo)): 

H 3 S0 4 
1. jt-BuOH + urea -+ t-BuNKCNH a 



2. t-BuNH,C"NH a + phthalic anh. 




N-C(CH 3 ) 



^V \ 



N-C(CH 3 ) 3 + NH,NH a - H,0 


t-BuNH 3 *HCl (89$) 


0- m et hyl & n x^LSe ne o S n ( t 1 ^ EJJSSS %S ?$?* * """ ° f 

SRMgCKor Br) + MeONH. 



methods? Dre°mra?ion n o'f t^fn?" h * S b86n turn0d to a S enera l 

ing as occurring: ™ Smides ' and Postulated the follow- 

(CH 3 ) 3 C=CH 3 + H 3 30 4 

H 3 
CH 3 C=N-C(CH 3 ), A 

0S0 3 H 

(CH 3 ) 3 C-030 3 H + CH-ON 

[CH 3 C=N-C(CH 3 ) 3 ] 

CH 3 JJ-NH-C(CH 3 ) 




This has been tried on a number of compounds and appears to be 
general; it is recommended as a source of solid derivatives for the< 
identification of both nitriles and olefins. When HCN is present 
in the nitrile used, N-alkyl formamides are formed. 

Ritter and Kalish (13) employed as starting material a tertiary 
alcohol or alkene in glacial acetic acid solution to which an 
equivalent of NaCN has been added; the reaction occurs spontaneous- 
ly when sulfuric acid is added and simple dilution generates the 

„ m H -n HCN OS0 3 H H 2 

R 3 COH ] M 14 J * R3C-OSO3H -* R 3 C-N=CH 4 

r 3 c=chr(h) j 3 

aq. alk. 
[R 3 C-N=CH] -► R 3 CNHCHO -> R 3 C-NH 3 


The N-alkylformamides were hydrolyzed with aqueous alkali to t- 
carbinamines. Other amides are more difficult to hydrolyze, ~ 

The following compounds were reported in reference (13) (with 
approximate yields)? 

(CH 3 ) 3 C-NK 3 .HC1 /50H 3 -0-NH 8 CH 3 -b CH 8 -5-NH 3 />CH 3 -CH-NH 

(i) (bo%) 3 3 0Hs * CHs 

(II) {55%) (Hi) (70%) (IV) (30^) 

Compound (IV) is amphetamine, included because of the interest 
in beta-phenylethylamines as medicinals. It was not obtained from 
a formamide but by hydrolysis (with FC1 for 11 hours) of its acetyl 
derivative formed from allylbenzene and acetonitrile . 


1. Linneman and Brauner, Ann., 193, 77 (1878) 

2. Coleman et al. J Am. Chem. Soc. 51 567 (1929). See also, 
ibid. 50, 1193 (1928); 55, 3669 (1933); 58, 27 (1936); 
Sheverdma and Kocheshkov, J. Gen. Ghem. TU3SR), 8, 1825 (1938). 

3. Klages, et al., Ann., _547, 24 (1941), 

4. Mentzer, et al,, Bull. soc. chim., 9, 813 (194*). 

5. Henze, Allen and Leslie, J. Am. Chem. Soc, 65, 87 (1943). 

6. Karabmos and Serijan, ibid., _67, 1856 (194577 

7. Campbell, Sommers and Campbell, ibid., 68, 140 (1946). 

8. Cairns, ibid., 63, 871 (1941). — J ' 

9. Adams and Cairns, ibid., 61, 2464 (1939). 

10. Smith and Emerson, ibid., 67, 1862 (1945). 

11. Brown and Jones, J. Chem. Soc., 1946, 781. 

i, *}tter and Minieri, J. Am. Chem. - ^., 70, 4045 (1948). 
13. Ritter and Kalish, ibid., 70, 4048 (19487. 


Reported by George I. Poos February 25, 1949 

It was reported (l) in 1937 from the Pasteur Institute that 
certain organic compounds exert a specific antagonism to the power- 
ful physiological action of histamine. Of more than thirty aryl 
ethers and amines investigated, 2-thymoxyethyldie thylamine (F929) (i) 
and NjN-diethyl-N'-ethyl-^-phenylethylenediamine (F1571) (II) proved 
to be the most active although both were found to be too toxic for 
human use. 

CH 3 m a 

r /^-OCH s CH 3 N^ y l C 3 H 5 / a H 8 

X C 2 H 5 S/ ^N-N-CHpCHoN 

3 n 5 ^/ \j>,-iM-un 3 un 2 

N CoH 




The first compound with antihistamine activity to receive 
extensive clinical trial was N-benzyl-N* ,N T -dimethyl-N-phenylethyl- 
enediamine (Antergan) (Dimetina) (R f P. 2339) (ill) (2) . The following 
scheme has been patented for its preparation (3): 

,CH 3 
H C 3 H 5 MgBr Mg3r C1CH 3 CH 3 N { ^ 

j6ch 3 n/> -> j6CH 3 N-j6 -> \5H a 

a — v v r - 3 P* 

// NN-N— CH 3 CH 3 N 

N CH, 

Several types of more active histamine antagonists appeared 
soon after the success with Antergan had been announced. Rieveschl 
and Huber investigated (4) a number of benzhydryl alkamine ethers, 
prepared from diphenylmethyl bromide and appropriate amino alcohols 
(5). The 2-(N,N-dimethylamino)ethyl benzhydryl ether (Benadryl) (iv) 

j6 3 CHBr + HO(0H 3 ) n N]!l 



proved the most effective of twenty-one compounds tested (6) and has 
received widespread clinical use. The marked side reactions induced 
by Benadryl, especially the saporlfic action, make its use less desir- 
able than some of the newer less toxic drugs. 



Another important agent is N-benzyl-N 1 ,N' -dime thy 1-N-fe-pyridyl}- 
ethylenediamine (Pyribenzamine) (v) , the most active of twenty 
similar compounds prepared (7). The method of preparation (used 
in general for compounds of this type) involves successive alkyla- 
tions of the appropriate heterocyclic amine with alkyl and aralkyl 
halides in the presence of sodamide or llthamide (8). 




X or LiNH s ,A 
R f - heterocyclic 
R" - alkyl or aralkyl 
X - halogen 


R"X B." / 

-> R»-N-CH 2 CH 3 N V 
NaNH 3 r- 
etc. / R T - 2-pyridyl 
V A R"~ benzyl 

N,N- dimethyl 

Pyribenzamine has been used extensively as a histamine 
antagonist and is probably less toxic than Benadryl (9), 


Many analogs and substituted derivatives of Pyribenzamine equal 
excell the potency of the parent comoound. N-Benzyl-N 1 ,N'- 
di$ethyl~N-(2-pyrimidyl)-ethylenediamine (Hetramine) (Vl) (10); N, N- 
dimethyl-N'-P-methoxy benzol- N- (2-pyrimidyl)-ethylenediamine (Neo- 
hetramine).(Thonzyl amine) (VII) (ll); and N,N-dimethyl-N'-p_-methoxy- 
benzyl-N-(2-pyrldvl)~ethylenediamine (Neoantergan) (Pyranisamine) 
(R. P. 3786) (VIII) (12) increase in activity in the order given (8b). 




.CH 3 

ch 3 ch 3 n' 

X CH, 

V 4 

^ CH 3 


0CH 3 

J-N-GH a CH a m 




Recently it has been shown that certain Pyribenzamine thenyl 
analogs compare very favorably with Pyribenzamine. N,N~Dimethyl~ 
N , -(2-pyridyl)-N'~(2~thenyl)-ethylenediamine (Histadyl) (Thenylene) 
(Antihistamine 01013) (IX) (13,14) and the corresponding 2-halogen- 
2-thenyl compounds (Chlorothen and Bromothen) (14) have been prepared 
and tested (15). N, tt-Dlmethyl-N»~phenyl-N , - (2-thenyl)-ethylene- 
diamine (Diatrin) (W-50) (x) (16) and N,N - dime thyl-N'-f urf uryl-N'- 
(2-pyridyl)-ethylenediamine (Xl) (17) are reported to be very 

• •' - 




."; ' ' 


— 7— 

effective and of low toxicity. Other thiophene analogs of anti- 
histamines thus far prepared and tested have less activity than 
the corresponding phenyl compounds (18), 

^-N-CH 3 CH 3 N 


^2 /CH 3 



/ CH : 




Replacement of the dime thylaminoethyl grouping by the 2-methyl- 
imidazoline group leads to another series of active drugs. 2-lN- 
Benzyl-N-phenylaminomethyl)-imldazoline (Antistine) (Phenazoline) 
(XII) (19) ia well established while 2- (aryloxymethyl)-imidazolines 
(20) such as 2- (benzhydryloxymethyl)- imidazoline (XIIl) show pro- 

It has been reported (2l) that certain phenothiazines such 
as N-(dimethylaminoisopropyl)phenothiazine (R. P. 5277) (XIV) have 
a very high order of antihistamine activity subsequent tests 

(2b) have shown the phenothiazines to be too toxic for human use. 




yH 3 ^N-CH 3 
X N NH-CH 3 






*N— CH 3 S 




£H 3/ CH 3 




2-Dime thylaminoethyl ethers of 2 
methanols have recently been reported 
gonists (22). The most active of the 
me thy lam inoe thoxy ) -a- me thy lbenz y l] -p y 
(XV) . Among other compounds reported 
histaminic activity may be included ? 
tetrahydro-1-pyridindene ( Thephorin) 
pheny 1-1- (2-dimethylaminoe thy l) -piper 
(XVIII) (25). 

',- substituted pyridine 

to be active histamine anta- 
se compounds is 2-[ct-(2-di- 
ridine (Decapryn) (Doxylamine) 

to have specific anti- 
!-methyl-9-phenyl-2. 3,4,9- 
(Phenindamine) (XVI) (23); 4- 
azine (XVIl) (24) and 1- phenyl- 
(Trimeton) (Prophenpyridamine) 




^-c-och 3 ch 3 n' 

X N X CH 3 V CH 3 




/CH 3 

N' X N-CH 3 CH 3 N 

N / X CH 


CHCH 3 CH 3 N 

CH 3 

The results of a recent clinical comparison of seven important 
histamine antagonists show the following order of decreasing anti- 
histamine activity (26): Neoantergan > Histadyl> Antistine> 
Pyribenzamine>> Benadryl/' Neohetramine > Thephorin. 


1. Bovet and Staub, Compt. rend. soc. biol. , 124 , 547 (1937); 
C. A., 31. 3988 (1937). Staub, Ann. inst .""Pasteur., 63, 400, 
485 (1939); C. A., _34, 5163,2069 (1940 ) . 

2. a. Halpern, Arch, intern, pharmacodynamic, _68, 339 (1942); 
C. A., 38, 5957 (1944). b. Winter, J. Pharmacol., 90, 224 
(1947). — 

3. Dan. 63,614; C. A., 40, 4394 (1943). 

4. Rieveschl and Huber, Abstracts 109th ACS Meeting, 50K (1946). 

5. Rieveschl, L r . S. 2,421,714; C. A., 41, 5550 (1947). 

6. Loew et al, J. Pharmacol., 83, 120 (1945). 

7. Huttrer ejt al, J. Am, Chem. Soc, 68, 1999 (1946). Mayer et al. 
Science, JL02, 93 (1945). — ~~ 

•8. Djerassi et al, U„ S. 2,406,594; C. A., 41, 489 (1947). 
9. Feinberg, J. Am. Med. Assoc, 132, 702 (1946). 



10. Feinstone et al, Proc Soc. Exptl. Biol. Med., 63, 158 (1946). 

11. Reinhard and Scudi, Ibid , 66, 512 (1947). Dreyer and Harwood, 
^"bid , 66, 515 (1947). Bernstein and Feinberg, J. Allergy, 19, 
393 (1948). ' — ' 

12. Bovet .et al, Compt. rend. soc. biol., JL38, 99 (1944); C. A. 
39, 3070 (1945) . Friedlaender et al, J. Lab. Clin. Med., 32, 
47 (1947); C. A., 41, 7499 (19477. ' — " 

13. Weston, J. Am. Chem. Soc, .69, 980 (1947). 

14. Clapp et al, ibid , 69, 1549 (1947). 

15. Feinberg. Quart. Bull. Northwestern Univ. Med. School, 22, 
27 (1948); 0. A., 42, 3080 (1948). Roth et al, Arch, intern, 
pharmacodynamic, 7FJ 362 (1948); C. A. 427 "6454 (1948). 

16. Leonard and Ulrlch, J. Am. Chem. Soc, .70, 2064 (1948). Ercoli 
£t al, J. Pharmacol. Exptl. Therap., 93, 210 (1948); C. A., 
42, 7434 (1948). — /,*.*., 

17. Vaughan and Anderson, J. Am. Chem. Soc, _7Q> 2607 (1948). 

18. Kyrides _et al, Ibid , 69, 2239 (1947). (Compare: Leonard and 
Solmssen, ibid. , 70, 2064 (1948). 

19. Meier ejtal, Schweiz, med. Wochschr., .76, 294 (1946). Brit. 
589,504; C. A,, 42, 616 (1948). Miescher and Klarer, U. S. 

. 2,449,241; C. A., 43, 692 (1949). 

20. Djerassi and Scholz, J. Am, Chem. Soc, 69, 1688 (1947). 
Djerassi and Scholz, J. Org. Chem., 13, 830 (1948). Cavallinl 
and Mazzucchi, Farm. sci. e. tec, 2, 273 (1947); C. A., 42, 
1664 (1948). Dhalbom and Sjogren Scand. Chem. Acta. 1, 7777 
(1947). -' ' 

21. Halpern and Pucrot, Compt. rend. soc. biol., 140, 361 (1946): 
0. A., 41, 4234 (1947). Halpern, Bull, soc dhim. biol., 

29, 309^1947). ' 

22. Tilford £t al, J. Am. Chem. Soc, 70, 4001 (1948). Brown et 

•riLo\ in 5 er ? and Ber nstein, J. Lab. Clin. Med., 33, 325, 319 
TI948); C. A., 42, 3846, 6005 (1948). — 

23. Lehmann, J. Pharmacol. Exptl. Therap., 92, 249 (1948): 
C. A., 42, 3849 (1948). — ' 

24. Cerkovnikov et al, Arhiv Kem., 18, 12, 37, 87 (1946); C. A., 
42, 1938, 1942, 3394 (1948). * ' 

25. Schering Corp. 

26. Swartz and Wolf, J. Allergy, 20, 32 (1949). 

■■ ■' .-- •. 


Reported by John Lynde Anderson March 4, 1949 

Although orientation in aromatic substitution reactions has 
been rather thoroughly investigated, the directive effects in ali- 
phatic substitutions have received little attention until recent 
years. Both ionic and free radical substitution reactions are 
known in the aliphatic series. The free radical aliphatic chlori- 
nation reaction has been studied in the most detail, and this 
seminar will be limited to a discussion of this type of reaction. 

Chlorinations of unsubstituted aliphatic hydrocarbons : The vaoor 
phase chlorination at 300° and the liquid phase chlorination at 
25 of unsubstituted paraffins have been investigated (l). It has 
been shown that hydrogen atoms are replaced in the order primary / 
secondary < tertiary; the relative rates are 1.00 to 3.25 to 4.43. 
As the temperature of both types of reaction is increased, the 
ratio of rates approaches unity, the limiting ratio in liquid Phase 
reactions being reached at much lower temperatures. 

Chlorinations of substituted aliphatic hydrocarbons : The directive 
effects of various substituents in aliphatic hydrocarbons have been 
indicated by a number of recent investigations ♦ Thus the liquid 
phase chlorination of 1-chloro-, 1, 1-dichloro-, and 1,1,1-tri- 
chlorobutane using sulfuryl chloride as the chlorinating agent 
(2,3,4) shows that the effect of each chlorine substituent is to 
direct further chlorination to more remote positions in the mole- 
cule (see Table I) . Ash and Brown (5) believe this effect is due 
to deactivation of the adjacent carbonhydrogen bonds rather than 
to activation of the more remote bonds. From an inspection of 
these data, it is apparent that two or more chlorine atoms are 
sufficient to deactivate the beta position* 

The directive influence of fluorine atoms and the orienting 
effect of the trichlorosilyl grouo are also shown in Table I. The 
products are in accord with Prediction, for fluorine is more 
electronegative than chlorine and carbon is more electronegative 
than silicon. 

The chlorination of acids and acid chlorides is markedly af- 
fected by the reaction conditions employed. Thus chlorination of 
butyryl chloride in the presence of iodine or phosphorus leads 
only to alpha substituted derivatives (via the ionic reaction); 
however when very pure reagents and equipment are used, the 
peroxide-catalyzed chlorination of n-butyryl chloride leads to only 
three percent alpha substitution (4). 

The directive effects of the acetoxy (4), the trichloro- 
acetoxy (6), the methyl (l), and the phenyl groups (4) are 
indicated in Table I. Obviously the effect of the methyl and 
phenyl groups is to activate the neighboring position in contrast 
to the deactivating effects of the other groups. 


— o_ 

TABUS I (5) 

Relative percentage substitutions in the chlorination of 1- sub- 
stituted propane s. 

C - C - C - X 








CH 3 




CH 2 C1 




00CCH 3 




SiCl 3 























CC1 3 



CF 3 . 

Discussion: Ash and Brown have discussed the theoretical aepeots 
of aliphatic free radical chlorinations in their most recent paper 
(5). Agreeing with Tischenko (?) , they consider the best explan- 
ation of these directive influences to be due to the inductive 
effect of the group X. Thus the separation of a hydrogen atom 
from the 2- carbon atom in 1-chlorobutane is oredicted to be more 
difficult than in butane itself. This prediction is in accord 
with the observed fact. That the effect is additive as more 
chlorines are introduced into the 1-carbon of normal butane is in- 
dicated by the results for the chlorination of the three chloro- 
butanes. In Table I the groups X are listed in the order of de- 
creasing activating effect, that is increasing negative inductive 

In vapor phase chlorinations above a critical temoerature, 
little or no 1,2-dichloroalkanes are isolated. For example, when 
1-chlorobutane is chlorinated at temperatures in excess of 312°, 
the 1,2-dichlorobutane is absent or present in only very small 
amounts (8). Ash and Brown postulate that this apparently anoma- 
lous result, the "vicinal effect", is due to the instability of 
the free radical (I) which eliminates a chlorine atom to form the 
olefin (II). 

CH 3 CH 3 CHCH 3 C1 ",- i CH 3 CH 2 CH=CF 3 + CI- 

(I) (II) 

Analogously, ethyl chloride in the absence of free halogen is 
quite stable up to 415°, but in the presence of chlorine, it de- 
composes to ethylene and hydrogen chloride at temoerature 3 as low 
as 280 (9). This "vicinal effect" has been observed only in 
vapor phase reactions. 



The free radical stability factor in the chlorination reaction 
is well indicated by a comparison of the attack of & chlorine atom 
and a methyl free radical on isobutyric acid. In the first case, 
15 percent of alpha- and 85 percent of beta-chloroisobutyr ic acid 
is formed (10), while the reaction of isobutyric acid in the 
presence of acetyl oeroxide leads exclusively to tetramethyl- 
succinic acid (ll). Ash and Drown believe that the initial 
attack is at the beta position predominantly (a) which is in 
accord with the data in Table I. They also postulate that the 
reaction with chlorine is very fast (c). However, in the case of 
the methyl free radical initiation, the beta free radical, pre- 
dominantly formed at first, reacts with unchanged isobutyric acid 
to form the alpha free radical which is considerably more stable 
due to resonance (b), The slow coupling reaction leads only to the 
formation of tetramethylsuccinic acid (d) . 

(a) CH- 

(CH 3 ) 3 CHC00H + CH 3 '(C1-) -> CHCOOH(P.) + CH 4 (H0l) 

v <CH a " 

(CH 3 ) 2 CCOOH(T*) 

CH 3x 

(b) P« + (CH 3 ) 3 CHC00H ^r±: ^CHCOOH + T- 


(a) P- + 01 a -> PCI + 01- 
T* + Cl 3 -> TCI + CI- 


(d) T- + T- — > TT slow 


1, Dunstan, "The Science of Petroleum," 1st Ed., pp. 2787-2794 
(by H. B. Hass) , Oxford University Press, London, 1938. 

2. Brown and Ash, Absts. of the 109th Meeting of the Am. Chem . Soc 
Atlantic City, N.J. , April 1946, p. 21M. 

3. Brown and Ash, Absts. of the 112th Meeting of the Am. Chem. 
Soc, New York City, September 1947, p. 12L. 

4, Brown and Ash, Absts. of the 114th Meeting of the Am. Chem. 
Soc, St. Louis, Mo., September 1948, p. 3N. 

5. Ash and Brown, Record of Chemical Progress, 9, 81 (1948). 

(Contains an excellent bibliography). 

6, Gaylord and Waddle, J. Am. Chem." Soc, J53, 3358, (1941). 
7- Tischenko, J. Gen, Chem, (U. S.S.R.), 7, 897 (1937); C. A., 

31, 5755 (1937). 

8, Vaughan and Rust, J. Org. Chem., 5, 449 (1940). 

9. Rust and Vaughan, ibid., 6, 479 (1941) . 

10. Kharasch and Brown, J, Am. Chem. 3oc, .62, 925 (1940) . 

11. Kharasch and Gladstone, ibid., 65, 15 (1943). 



Reported by Claire Bluest e in 

March 4, 1949 

The hydroxypyrazines are of interest because of their relation 
to physiologically active compounds. Early investigators were 
limited in working with the isolated pyrazine nucleus because of 
low yields in the methods of preparation and the resistance of the 
ring to the usual aromatic substitutions- A thorough review of 
pyrazine chemistry up to 1946 has been made by Krems and Spoerri 
(1). Since that time there have been further extensions. 

Syntheses of the Ring 

1. The method developed by Tota and Elderfield (2) appeared to be 
a very general one for hydroxypyrazines. However, more recent work 
(3,4) has shown that the method is applicable mainly to the o re- 
paration of 5,6-disubstituted or 3, 5, 6-trisubstituted-2-hydroxy- 
pyrazines. This method is outlined in equation I.. 


NH a 'HCl 












I I 

x NH y 

R/ ^V-R" 


The condensation between the a-aminoket one and the bromoacyl bro- 
mide is best carried out by using N-methylmorpholine in anhydrous 
chloroform (4). 

When R=H, it is necessary' to protect the aldehyde group before 
the final condensation with ammonia. The only feasible way of doing 
this is to prepare the thioacetal and later to cleave it in the 
usual manner with HgCl 3 and CdC0 3 (3). These added steps, however, 
reduce the overall yield considerably. 

2. It 

ketone in this 
II (4). 

is impossible to make a 3, 5-di substituted- 2- hydroxy pyrazine 
H) by the above method. The intermediate bromoacylamido- 

case condenses with ammonia only as shown in equation 





3H 8 NHCOi 





R // "yNHCOCHgR" 



This reaction works well for , 

none thy 1 ketones and yields 3,6- 
The corresponding hydroxypyrazines 



can be obtained by treatment with nitrous acid (5) or nitrosyl 
sulfuric acid (6) . 

3. Diketopiperazines, which are amino acid anhydrides, are tauto- 
meric with dihydrodihydroxypyrazines. They are most conveniently 
prepared by heating the amino acid with ethylene glycol (7). There 
is no direct method of oxidation to the hydroxypyrazines, but 
Baxter and Spring (8) have achieved conversion to the mono- or di- 
chloropyrazines by use of phosphorus oxychloride (equation III). 









P0C1 3 R/^ \-Cl 





The monochloropyrazines can be converted to hydroxypyrazines 
by hydrolysis methods, and the dichloropyrazines to hydroxy chloro- 
pyrazines and also to dialkoxypyrazines, but it hss been impossible 
to obtain a dihydroxypyrazine because the ring cleaves first (9). 

4. Another method, the use of which has 
G-astaldi (10) . The starting material is 
The. steps in the improved synthesis (ll) 

been extended, is that of 
an oximinomethyl ketone, 
are outlined in equation IV, 

NaH30 3 ^SO a Na 


x NHS0 3 Na 





A ^CN 

HC1 N NH. 


" NT 

x NHS0 3 Na 



v N A N 

R{/ \CN " R^ ^-OH 
I 15^K0H 

^ yR H0 s C 


X N X 

be decarboxylated to the 

If desired, the final product can easily 
3, 6-di substituted- 2- hydroxypyrazine. 

5. A new general synthesis of hydroxypyrazines involves the con- 
densation of 1,2-dicarbonyl compounds with a-amino acid amides 
(12;. This is the simplest and most direct method for obtaining 
compounds with a variety of substituents on the ring, and the 
yields in most cases are high, 75-95#. The reaction, as shown in 
equation V, is best carried out at -10° to -20°C in water or 
methanol solution with one equivalent of NaOH present. Unsymmetri- 
cal dicarbonyl compounds yield only one product. 



R— C=0 




H a N-CHR 


Reactions of the Ring 



Due to the deactivation of the pyrazine ring towards electro- 
pnllic substitution, there are few methods for introducing sub- 

I - 



stituents directly into the ring (l). Hydroxypyrazines are ob- 
tained chiefly through synthesis or by replacement of the amino 
group as mentioned previously. A chloropyrazine, if available, can 
be hydrolyzed conveniently with KOH (8). Recently a fairly direct 
method for introducing chlorine easily into the pyrazine nucleus 
has been deVised (13). The pyrazine is treated with hydrogen 
peroxide, which gives the mono- and di~N~oxides. These can be 
separated by means of their solubility, and further treatment 
of each with phosphorus oxychloride yields the mono- or dichloro- 
pyrazine, respectively (equation VI), 

R H 9 

/ <N-R 

3 U 3 





If there is already a CI on the pyrazine ring, the peroxide 
oxidation will yield only one rnono-N-oxide, that in which the 
oxidized N is not adjacent to the CI (9). 

Treatment of the N-oxide shown in equation VII with phosphorus 
oxychloride yields a chloromethyl pyrazine in addition to the ex- 
pected chloropyrazine (9). 

1 s 

tf N N 

VCH 3 Cl/ ^CH* / ^>CH a Cl 




X N X 

3 V 




40, 279 (1947). 
Chem. 7, 313 (1942) . 
J. Chem. Soc, 1947 , 370. 
ibid., 1948 , 1855. 

1. Krerne and Spoerri, Chem. Rev 

2. Tota and Elderfield, J. «rg. 
5. Baxter, Newbold, and Spring, 

4. Newbold, Spring, and Sweeny, 

5. Newbold and Spring, ibid., 1947 , 3737 

6. Erickson and Spoerri, J. Am. Chem . Soc, 

7. Sannie, Bull. soc. Ohlm. 9, 487 (1942). 

8. Baxter and Spring, J* Chem. Soc, 1947 , 1179. 

9. Baxter, Newbold, and Spring, ibid. 7^948, 1859. 

10. G-astaldi, ftazz. chim, ltal. J51 I, 233 (193l). 

11. Sharp and Spring, J. Chem, Soc, 1948 , 1862. 

12. Jones, J. Am. Chem. Soc, 71, 78 (1949). 

13. Newbold and Spring, J. Chem. Soc, 1847 , 1183. 

68, 400 (1946). 

* :;. 



■' ■ '■ ■■ " ' ■■»! — — i— -I, i ^ii i"^ ■ ii i - 1 ■■■■■■■ m —i-m - — ^i m ■ ■ ii " " ^ f -■ ■ '■ p * '<• '*■ ■*■ ■■ ■■ '■■* ■*■ ■■■-■■ « ^.1 — 

Reported by H. Wayne Hill, Jr. March 11, 1949 

Since the discovery of LiAlH 4 , much attention has been given 
to its use both as a reducing agent and in quantitative organic 

Lithium aluminum hydride is readily prepared by the action 
of A1C1 3 on LiH under anhydrous conditions in ether solution (2). 

4LiH + AlClg -> LiAlH 4 + 3LiCl 

Both LiH and LiAlH 4 are now sold by Metal Hydrides, Inc. of 
Beverly, Massachusetts. 

Reductions with LiAlH 4 Lithium aluminum hydride is a powerful re- 
ducing agent for organic comoounds, its chief advantage being that 
side reactions are at a minimum. Most of the reductions to be 
mentioned in this seminar are conveniently carried out at room 
temperature, the general technique being quite similar to that used 
in G-rignard syntheses. 

The equations for the reaction of carboxylic acids will serve 
to indicate the stoichiometry of the reduction (10) . 

4RC00H + 3LiAlH 4 -► LiAl(0C^ 3 R) 4 + 2LiA10 3 + 4H 3 

LiAl(0CH 2 R) 4 + 4H 3 -* RCH 3 OH + LiOH + Al(OH) 3 

This represents an excellent means of reducing an acid directly to 
the corresponding alcohol. 

In order to effect the reduction of alkyl halides, it is 
necessary to use somewhat more vigorous conditions. The reduction 
is conveniently carried out in boiling tetrahydrofuran (b.p. 65o) 
as solvent using LiH with a small amount of LiAlH 4 as the hydrogen 
carrier (5) . 

LiAlH 4 
RX + LiH -> RH + LiX 

Alicyclic and aryl halides are very unreactive . 

Although carbon-carbon double bonds in general pre not re- 
duced by the reagent, it has been noted that ethylenic bonds of the 
type C S H 5 CH=CHX where X is N0 3 , COOH, CHO, COR, etc. are reduced 
in the normal reduction procedure (addition of a solution of the 
compound to the hydride solution) . However if the order of 
addition is reversed, it is possible to reduce X without affecting 
the carbon-carbon double bond (4). Reduction of double bonds pro- 
ceeds slowly. 

The following table illustrates the types of compounds which 
may be reduced and their reduction oroducts. In many cases the 
yields are almost quantitative. 


Type Compound 




Ref e rence 


.LiAl(OCH 3 R) 4 


R 3 CO 

LiAl(OCHR 3 ) 4 




LiAl(OR') 2 (OGH 3 R) 3 




L1A1C1 3 (0CH 3 R) 3 



(RCO) 3 

LiAl(OCK 3 R) 4 




LiAl(OGH 3 R) 4 








Lr-gCH 3 J 3 

1~o6h 3 J 3 







LiAl(NCH 3 R) 3 



ArN0 3 



RN0 3 

LiAl (NR) 3 

RNH 3 







LiAl (ArCH 3 NAr) 4 

ArCH 3 NHAr 



LiAl[OCH(CH 3 )R] 4 




RCH 3 NH 3 

11, 14 

.Reductions with NaBH^ Since sodium borohydride is DotentiPlly a 
cheaper material than LiAlH 4 , it has been of interest to investigate 
its action as a reducing agent in organic chemistry (l). it has 

NaBH 4 can be used in aqueous or ethanolic solution, 
great advantage in convenience as compared with 
also be possible to reduce ether- insoluble comoounde 
by this method, LiAlH 4 is of little use in the re- 

thus offering a 
LiAlH 4 . It may 
such as sugars, 
duction of such 


Sodium borohyd 
ductions. Thus in 
duced to alcohols, 
and acid anhydrides 
the alkaline condit 
able groups. Acid 
while acids, acid a 
affected under the 
in such cases. 

superior to LiAlH 4 in selective re- 
solution, aldehydes and ketones are re- 

ride is 

whereas acids, acid chlorides, esters'," "nitriies 
are not reduced. However in aqueous solution, 
ions may effect hydrolysis of easily hydrolyz- 
chlorides may be reduced in inert solvents, 
nhydrides, esters and nitro compounds are not 
same conditions. LiAlH 4 affords no selectivity 


Actlve Hydrogen Det erminations with LiAlH A (3,7,8,15) Many com- 
pounds containing active hydrogen atoms decomDose ether solutions 
of LiAlH 4 to liberate hydrogen. Thus by measuring the hydrogen 
evolved, it is possible to calculate the number of active hydrogens 
in a molecule . to 

4R0H + LiAlH 4 -» 4H 3 + LiAl(OR) 4 

u ^ 1 _ In ^ t ? e ma J° rit y of cases, results similar to those obtained 
by the Grignard method were observed. One notable difference was 
with keto-enol tautomers. These compounds react with the Grignard 
f?A?S nt 4., as thou S h * he y exist in the anol form only, whereas with 
L1AIH4 they behave as though they are only partially enolized. 

With compounds containing more than one active hydrogen such 
as primary amines and unsubstituted amides, a rather long~rea ction 
time (an hour or more) was required for the complete liberation of 
hyarogen. However in most of these cases, the first hydrogen was 
liberated rapidly (5-10 minutes) . 

Determinatio n of Reducible Groups with LiAlH. (3.15) In connection 
T ^J h the determination of active hydrogens, LiAlH 4 "may be used to 
determine reducible groups, such a s: carbonyl, ester, carboxylic 
acid, nitrile and amide. The procedure consists in treating a 
weighed amount of the compound with a known amount of the hydride 
and measuring the hydrogen gas evolved to get the number of' active 
hydrogens. The reaction mixture is then allowed to stand for some 
time (during which the reduction occurs) Hnd is treated with alcohol 
to decompose the excess hydride. Again the evolved hydrogen is 
measured. Thus the amount of hydride used in effecting the re- 
duction is obtained by difference, no hydrogen being liberated 
in the reduction process. 


1. Chaiken and Brown, J. Am. Chem. Soc, 71, 122 (1949). 

2. Finholt, Bond and Schlesinger, ibid., W. 1199 (1947). 

3. Hochstein, ibid., 71, 305 (1949). 

4. Hochstein and Brown, ibid., 70, 3484 (1948). 

5. Johnson, Blizzard and Carhart, ibid., 70, 3664 (1948). 

6. Karrer Portmann and Suter, Helv. Chim7"Acta., 31, 1617 (1948). 
'• Krynitsky, Johnson and Carhart, Anal. Chem., 207~3ll (1948). 

o f T ry ? ltsk ^ Johnson and Carhart, J. Am. Chem. "So"c . , 70, 486 (1948 

9. Nystrom and Brown, ibid., 69, 1197 (1947). — 

10. Nystrom and Brown, ibid., _69, 2548 (1947). 

11 * Nystrom and Brown, ibid., _70, 3738 (1948). 

12. Nystrom, Yanko and Brown, ibid., 70, 441 (1948). 

13. Seven, Organic Seminar Abstracts, November 7, 1947. 

14. Uffer and Schlittler, Helv. Chim. Acta., 31, 1397 (1948). 

15. Zaugg and Horran, Anal. Chem., 20, 1026 (1948). 


Reported by Melvin I. Kohan March 11, 1949 

Catalytic hydrogenation is largely an emoirical art so that 
success in any given case' is assured only by experiment. Not only 
the extent but also the direction of a reduction depends important- 
ly on many factors: the kind, method of preparation and amount of 
catalyst; the compound to be reduced (hydrogen acceptor); temper- 
ature, pressure, solvent, promoters and poisons. Any process" de- 
pending upon the balance of so many variables is obviously comolex 
but^an attempt can be made to understand many Dhenomena on the 
basis of the Langmuir concept of a unimolecular film adsorbed on 
the catalyst and proximate centers of activity (l) . This seminar 
reviews the recent work and the explanations of the results ob- 

I. Hydrogenation of the Benzenoid Ring (2). 

1. Polyalkyphenols (3). Whitaker studied the effect of 
steric hindrance on the reduction of phenols and observed the 




Raney Ni 

n-Ri (R-Ni) 

/-R 3 140-240° 
V 1500- 
R 3 2500 psi 

;-Bu-l/\.-R x t-Bu-KXl-R,. t-Bu-^v/Nl-R, 
I or or 


CH 3 


t-Bu H 

H~or t-Bu 
H or t-Bu 
CH 3 or t-Bu 


III only 
II or III 
I or II 

R 3 

The following guide for the hydrogenation of ohenols over R-Ni (in 
the absence of promoters) is therefore proposed: 





one ortho position unsubstituted 
one ortho position occupied by 
CH 3 and one by t-Bu 
both ortho positions occupied by 

cycl one xanol derivative only 
cy clone xanol or cyclo- 

hexanone derivative 
cyclohexanone or cyclo- 

hexene-one derivative 

-m„+?; g-Naphthol. Using R-Ni Stork (4) found that in basic 

t , °lr^1 U ° iS - the tetralol > but in neutral or acid media 
the unsubstituted ring is attacked preferentially. The effect of 

hPPn ^^r 6 ? 8 ^ *?? ease of red ^tion of phenolic compounds had 
been established earlier by Ungnade and co-workers (5). Since 
copper-chromium oxide, which attacks carbon-to-oxygen in preference 
to carbon-to-carbon double bonds, can be used to hydrogena?e 
?S nllZ co ^ Dound s but not their ether derivatives (6) and since 
n? \ltl ?S? fu° U ? iS kn ° Wn t0 react more readily in the presence 
tt r Il'tA'v, f , ef f ect , can be interpreted as facilitating 
tautomerism to the ketonic structure. However, the methyl ether of 




p~naphthol exhibits a similar behavior over R-Ni so Stork proposes 
that the effect is one of increased adsorption of a positive cente 
on the catalyst through coordination with the base. On this basis 
also, he explains the fact (8) that by the addition of a little 
NaOH the R-Ni hydrogenation of benzylcyanide gives almost exclusive- 
ly the primary amine. 

It has also been shown that this behavior of p-naphthol is 
essentially independent of the type R-Ni (9) used and that triethyl- 
amine can serve as the base instead of NaOH (10) . 

3. Pyrogallol (ll) . This compound (l, 2, 3-trih.ydroxy benzene) has 
been hydrogenated as the monosodium salt over R-Ni to the ene-diol, 
dihydropyrogallol (2,3-dihydroxy-2-cyclohexene-l-one) . The stopping 
of the reduction at this stage is attributed to the resonance 
stabilization of the anion of the 1,3-diketone system* 

4. Hydro xyphenyl Aliphatic Acids. The reduction of this type 
compound has been complicated since hydrogenolysis of the hydroxy 1 
group occurs with noble metal catalysts and decarboxylation with 
R-Ni (12). This problem has been resolved by use of R-Ni with the 
ester to which 0.3 mole per cent of the sodium salt has been added 
(13). (.Of. base effect, above.) 

5« Use of Adams Catalyst at High Pressure (14). The hydrogenation 
of an aromatic ring using the Adams Pt catalyst at low pressures 
can be accomplished if glacial acetic acid (15) or alcohol con- 
taining HC1 or HBr (16) is the solvent employed. The high pressure 
reaction also depends oritically on the solvent and, in general, 
proceeds more readily although an anomalous failure was observed in 
the case of aniline. 

6, Polymethylbenzenes and Polymethylbenzoic Acids. Smith and co- 
workers have undertaken a study of the hydrogenation of the benze- 
noid nucleus using the Adams datalyst in HOAc at low pressures. 
The behavior of phenyl substituted aliphatic acids and alkylated 
benzenes (l?) indicated the importance of symmetry, e.g. p-cymene 
reduces faster than i-propyl benzene. Examination of almost all 
Of the polymethylbenaenesand polymethylbenzoic acids (18) confirmed 
this fact: as the number of groups increases, the rate of hydro- 
genation (19) decreases; for a given number of groups, ag the 
symmetry of the molecule increases the rate increases. 

II. Hydrogenation of Esters to Alcohols (20, 21, 22, 23). The use 
of a Xtl or higher ratio of catalyst to hydrogen acceptor lowers 
the temperature required for the hydrogenation of esters and, in 
so doing, gives high yields of alcohols where previously only the 
corresponding saturated compounds have been obtained. Thus, 
benzoates give benzyl alcohols; malonates, 1,3-glycols; p-keto and 
-hydroxy esters, 1,3-glycols; a-keto and -hydroxy esters, 1,2- 
glycols; a- and (3-amlno esters, 1,2- and 1, 3~amino->alcohols. High 
pressures (5000 psi) and temperatures of 125-150°C with CuCrO or 
25-75 with W-6 R-Ni are used. Copper- chromium oxide is preferred 
in general and in particular with the |3- substituted esters unless 
temperatures below 100 are necessary. If R-Ni is used the amino- 
esters reduce faster than the keto or hydroxy esters unless tri- 
ethylamine is added (Cf. base effect, above). This development is 



considered to be the most important advance in recent years in 
catalytic hydrogenation. 

III. Reversibility of Catalytic Hydroge nations. The following 
equilibrium over copper- chromium oxide at 240-60° C and 200-300 
atmospheres of hydrogen pressure has been definitely established 

RC00CH 3 R f + 2H 3 =f=^ RCH 3 0H + R»CH 3 0H 

It is evident that compounds such as RC00CH 3 R, R» C00CR 3 R» and 
R C00CH 3 R are possible in such a reaction. The isolation of such 
compounds has never been accomplished since the concentration of 
ester is only about 1%, Subsequent experiments (25) used for the 
sake of simplicity compounds which can give only one ester and one 
alcohol: CH 3 (CH 3 )eC00(CH 3 ) 7 CH 3 , (CH 3 CH 3 ) 3 CHC00CH 9 CH(CH 3 CH 3 ) s and 
(CH 3 J 3 CC00CH 3 C(CH 3 ) 3 . The position of the equilibrium was shown 
to depend profoundly on the hydrogen pressure; at 10 atmospheres 
the cone, of ester is 80$. This would indicate the reaction has 
potentiality for the preparation of hindered esters from the 
corresponding alcohols. A similar reaction attempted over R-Ni 
yielded only a hydrocarbon with one less carbon (25, 26). 



Similarly, over copper- chormium oxide at 150-70 C (25, 27): 

50$ at 30 atm. 


VV *=H or CH 3 \^\^ 

90$ at 400 atm. 


2, 3-D lme thy 1 indole gives a lower cone, of the indoline at compar- 
able pressures. However, the reduction of 3,3-dimethylindole pro- 
ceeds irreversibly at 35 atmospheres of hydrogen pressure. 

IV. Preparation of Raney Nickel Catalyst. A number of catalysts 
(9) have been prepared from a 50$ Ni-Al alloy, but there are basic- 
ally three kinds: a slightly alkaline catalyst (commercial, ¥-3,- 
4,5; a highly alkaline catalyst (¥-7), and one with a large amount 
of adsorbed hydrogen (¥-6). The commercial type serves for most 
purposes. ¥-6 is the most active and is sometimes the only 
effective catalyst; it has been used especially in the low temper- 
ature hydrogenation of esters (above) and in the low pressure 
technique ordinarily employed with the noble metal catalysts 
(9d). ¥-7 is especially useful, even at low oressures for 
aldehydes, ketones and nitriles (9d) . 

The affect of the composition of the alloy has been subjected 
to further study, and it is claimed that an alloy containing 
only 20$ Ni gives not only a superior but also a non-pyrophoric 
catalyst (28). 

V. Selective Hydrogenation of cc, |3-Un saturated Ketones (28,29). 
l he low pressure hydrogenation of a, p-unsaturated ketones over 
R-Ni is known to give a saturated alcohol, but a recent oaoer 
states that if CHC1 3 or HC1 is present the saturated ketone is ob- 
tained. For example, the reduction of benzalacetone to benzyl- 
acetone and dibenzalcyclohexanone to dibenzylcyclohexanone are 
cited. It is also claimed that comolex systems such as difurfural- 
of e CFCl °ldded selectivel ' y hydrogenated by controlling the amount 


; ■ 

• I ' f. 

" -i •. ' 


.• . . 

. . • 



G-eneral References 

Adkins, "Reactions of Hydrogen, etc.". Univ. of Wisconsin Press, 
1937. ' 

G-ilman, "Organic Chemistry", 2nd Ed., 1943, Vol. l,pp. 779-834. 

"Newer Methods of Preparative Organic Chemistry", Interscience, 
1948, pp. 61-130. 

Billman and Binder, Univ. of Illinois Seminar, II Semester, 1939-40, 


1. Adkins, Ind. Eng. Chem. , 32, 1189 (1940). 

2. Rothstein, Univ. of Illinois Seminar, Jan. 16, 1948. 

3. Whitaker, J. Am. Chem. Soc . , 69, 2414 (1947). 

4. Stork, ibid., 576 (1947). 

5. Ungnade and co-workers, ibid., 66, 118, 1218 (1944). 

6. Musser and Adkins, ibid., 60, 664 (1938): Wilds and McCormack, 
ibid., 70, 4128 (1948). "**" 

7. Delepine and Horeau, Compt. Rend., 201, 1301 (1935). 

8* Fluchaire and Chambret, Bull. soc. chem., 5, 11, 22 (1944). 

9. a. W-l: Covert and Adkins, J. Am. Chem. Soc, 54, 4116 (1932). 

b. W-2: Mozingo, Org. Syn., 21, 15 (1941). 

c. W-3,4: Pavlic and Adkins, J. Am. Chem. Soc, 68, 1471 (1946' 

d. W-5,6,7: Adkins and Billica, ibid., 70, 695 (1948). 

e. Commercial Raney Ni, G-ilman Paint and Varnish Co., 

Chattanooga, Tenn. 

f. Commercial Ni on kieselguhr, Universal Oil Products Co. 
10. Adkins and Kresk, J. Am. Chem. Soc, 70> 412 (1948). 

11,. Pecherer, Jampolsky and Wuest, ibid., 70, 2587 (1948). See 

also Org. Syn., 27, 21 (1947). 

12. Levin and Pendergrass, ibid., 69, 2436 (1947). 

13* Ungnade and Morris s, ibid., 70 > 1898 (1948). 

14. Baker and Schuetz, ibid., 69, 1250 (1947). 

15. Adams and Marshall, ibid., _50, 1970 (1928). 

16. Brown, Duran* and Marvel, ibid., _58, 1594 (1936). 

17. Smith and co-workers, ibid., 67, 9.72, 276 and 279 (1945). 

18. Smith and Stanfield, ibid., 71, 81 (1949). 

19. Fuzek and Smith, ibid., 70, 3743 (1948). 

20. Ovakimian, Kuna and Levene, J. Biol. Chem., .135, 91 (1940) . 

21. Adkins and Pavlic, J. Am. Chem. Soc, 69, 3039 (1947). 

22. Mozingo and Folkers, ibid., _70_j 227, 229 (1948). 

23. Adkins and Billica, ibid., _70, 3118, 3121 (1948 >. 

24. Burks and Adkins, biid., 62, 3300 (1940). 

25. Adkins and Burks, ibid., 7£, 4174 (1948). 

26. Wojcik and Adkins, ibid., _55, 1293 (1933). 

27. Adkins and Coonradt, ibid., 63, 1563 (1941) . 

28. Cornubert and Phelisse, Compt. Rend., 227, 1131 (1948). 

'- * 


Reported by P. D. Caesar March 18, 1949 

Reviews of the Willgeredt reaction and its various modifi- 
cations oomplete through 1946 can be found in "Organic Reactions," 
vol* III, and in a past seminar (l). 

Mechanism Studies 

Two mechanisms have been proposed, either of which can be 
used to explain most of the experimental facts. 

A. The mechanism proposed by Carmack and coworkers (2) 
follows the route: 

^ — > — * > 

Ketone «_ hydroxy lam ine <_ unsaturated amine <_ acetylene ^ un- 
saturated amine 
When the labile amine group reaches the end of the chain, an 
irreversible oxidation reaction takes place between the sulfur and 
the nitrogen compound to produce a thioamide. Subsequent hydrolysis 
gives amide and acid* 

Objections to this theory include the need of a separate 
mechanism to get the labile group past a branched chain, the un- 
substantiated unsaturated amine intermediate, and the unexplained 
final oxidation step* 

B. King and coworkers (3) suggested the route; 

Ketone ^t thioketone T^- thiol 7^- olefin ~£L thiol, . . . 
Again when the labile group, this time a mercaptan group, reaches 
the end of the chain an irreversible reaction with sulfur takes 
place to produce a thioamide. 

This mechanism offers an adequate explanation for the migra- 
tion of the SH-group past a side chain 5 

5=C-C <r- RC-C-C <- RC-C*( 



and accounts for the poor yields obtained, since t-thiols, when 
used as starting materials in the Willgerodt reaction gave very 
poor yields of the amides. Moreover, with the aid of some' recent 
investigations all intermediates proposed in this mechanism have 
been shown to be feasible starting materials, and the steps in the 
final oxidation reaction have been thoroughly analyzed* 

Thioketones as Intermediates 

Confirmation of the probable participation of thioketones 
in the Willgerodt reaction was obtained (4) when trithioacetophenonc 
was found to give better yields of the thioamide than did aceto- 
phenone # 






+ HIT 7 S} 

5, 150° <^\cH 2 b-N x ^0 




Moreover, a, p-unsaturated acids lose the original carboxyl 
carbon with conversion of the a-carbon to a carboxyl (5). Using 
cinnamic acid as a typical example, King's mechanism could be 
applied in this manner. 

C 6 H B CH=CHb0H <r 

-C0 3 S 

— » C 6 H 5 CCH; 



C B n B CHCtl— C 

N 0H 


?H ,9 ox. S Q 

C 6 H B CHCH 2 COH -> C 6 H B CCH 2 C0H 

C R Hc,GH 9 C00H 


These experiments coupled with the negative results obtained 
with certain secondary alcohols give substance to the ketone- 
thioketone-thiol-olef in route. However, easily dehydrated alcohols 
can be used as starting materials, so that analogous ketones may by- 
pass the thioketone step. 

Oxidation of Primary Thiols (6) 

The oxidation reaction of a primary amine to a carboxyl ic acid 
derivative is unique. The steps will be illustrated, where possible 
by compounds actually employed. 

1. C S H B CH 2 CH 2 SH + S morpholine(R g NH) x C S H B CH 2 CH 2 SSCH 2 CH 3 C 6 H B 



/NRg H3O 

2. (I) + Sdm.eq.) 2R 3 NH C 6 H 5 CH 3 0-H -^ C 6 H 5 CH 3 CH0 

-» N NR 3 R 3 NH 


In this step phenyl acetaldehyde was isolated upon hydrolysis 
of the product when the sulfur used was insufficient to carry the 
reaction to completion. The best of several methods of accounting 
for this seemed to be the assumption that the amine and sulfur 
react to form a thiohydroxylamine (III), 


a. R 2 NH + S -> R 3 NH ^- R 3 NSH (ill) 

which then reacts with disulfide. 

2R 2 NH ,KHa 

b. (I) + 2(111) -* 6 H B CH 3 CHS3CHCH 3 6 H B ~* 2C S H B CH 2 C-H (II 

NR a NRp N NRo 

3. (II) + S-» G 6 H B CH 2 b~NR 2 98^ 




In this step it is assumed that the amino groups tend to 
activate the hydrogen at the central bond so that it can be 
attracted away from the molecule, leaving it anionic 

,NR 3 cs yNR 3 

a. (II) + R 3 NH p C 6 H 5 CH 2 C^9 

+ R-NHsT 

0«H R 0H 3 &-3B 

N NR 3 

'6 n B' 

(7) ' f 

C 6 H 5 CH 3 (S-NR 3 
+ 2R 3 NH 

It should be noted that this course of oxidation can not be 
applied to tertiary thiols (8) ^ast steo 1. or to secondary thiols 
past step 2. Therefore, the reaction must proceed to the primary 
thiol before reaching the irreversible oxidation stage. 

. . 14 

Tracer studies (9) using in the carbonyl group have shown 
that in addition to the major yield of unrearranged amide, a sub- 
stantial quantity of acid is formed in which the C 14 has migrated 
to the carboxyl group. This work in its complete form is now in 
press and t^e findings do not differ from those specified in the 
previous Communication to the Editor. 

H"£o CuCr 4 

e H 5 0*HpCONH ? -» -> 

C 6 H 5 C*CH 3 

5 + amine 


14% C 6 H 5 CH 3 C*00H 

003^(2^ init.act.) 
COg^ (75^ init.act.) 

Miscellaneous Contributions 

1. The conversion of a-tetralone (10) to 4- (S-naphthyl)' 
morpholine represents the first use of a cyclic ketone in the 
Willgerodt reaction. 


s + A 




. ^ — IT x 



2. Oximes and phenyl hydrazones (ll) have been converted to 
the expected acids, usually in very low yields. 

(J * 






3. A substance (13) idientified as dithiooxalodimorpholide 
was isolated from a number of runs using olefins as starting 
materials, particularly those in which the reaction was accompanied 
by substantial cleavage at the double bond. 



4. Several heterocyclic ketones have been employed (13,14). 
a-Thienyl methyl ketone gave tarry products only, but the pyridyl 
and quinolyl methyl ketones reacted normally. 

H ?* — Nr /\ A <^ • t3 


1. Leubner, Organic Seminar, p 36 (1946). 

2. Carmack et al, J. Am* Chem. Soc, 68, 2025, 2029, 2033 (1946). 

3. King et al, ibid., 68, 525, 632, 1369, 2335 (l946). 

4. Campaigne and Rutan, ibid., 69, 1211 (1947). 

5. Davis and Carmack, J, Org. Chem., JL2, 76 (1947). 

6. McMillan and King, J. Am. Chem. Soc, 70, 4143 (1948). 

7. Scott and Watt, J. Org. Chem., 2, 148 Tl937). 

8. Small, Bailey and Cavallito, J. Am. Chem. Soc, 69, 1710 (1947) 

9. Dauben, Reid, Yankwich and Calvin, ibid., 68, 2117 (1946). 

10. Horton and Van Den Berghe, ibid., _70, 2425~Tl948) . 

11. Stanek, Coll. Czechoslov. Chem. Commun., JL2, 671 (1947). 

12. McMillan and King, J. Am. Chem. Soc, 69, 1207 (1947). 

13. Schwenk and Papa, J. Org. Chem., 11, 798 (1946). 

14. Malan and Dean, J, Am. Chem. Soc, J39, 1797 (1947). 

15. Ott, Mattane, and Coleman, ibid., 68, 2633 (1946). 



Reported by Robert E. Carnahan 

March 18, 1949 

The Fries reaction consists in the conversion of a phenyl 
ester into an ortho - or para- hydro xypheny Ike tone under the action 
of an acid catalyst such as aluminum chloride. In most cases by 
the proper choice of the reaction conditions, primarily one isomer 
or the other may be obtained (l)» 


Previous to 1940, three mechanisms for the Fries reaction had 
been presented (2). Each of these had evidence which supported 
it but did not exclude the others. 

I. Cleavage of the ester to yield an acid chloride which could 
then acylate the aromatic nucleus. 

QC0CH 3 QA1C1 3 

+ A1C1. 


This was supported by the fact that acetyl chloride could be 
isolated when m-cresylacetate was submitted to the Fries reaction 
in the presence of _o-chlorobenzoyl chloride. 

II. Acylation of one molecule of ester by another. 








COCH3 + 

'/ ^ 




Cross-products are obtained when p-cresylbenzoate and 2- 
chloro-4-methylphenylacetate are mixed and submitted to the Fries 
reaction. This was objected to on the basis that an acyl inter- 
change could precede the actual rearrangement of the ester. 

III. True intramolecular rearrangement. 

The absence of the meta- isomer from the oroducts of the Fries 
reaction compared to its production in the corresponding Friedel- 
Crafts synthesis was supplied for evidence of this view. 








Actually the two intermolecular mechanisms above are one and 
the same (3). Each of these mechanisms is basically a Friedel- 
Crafts type of reaction. The crucial part of such a mechanism in- 
volves the attack of an oxo-carbonium ion on the aromatic nucleus. 
The most logical route for the production of this ion is as follows. 



6 OCR 

+ AlCl 

C 6 H 5 OCR 
ftlCl a 


C1 S A10C ( 





The possibility for an acyl interchange to precede an intra- 
molecular rearrangement in a case where mixed products are obtained 
is excluded, This would involve the attack of the oxo-carbonium 
ion, RCO®, on a molecule of ester to form the ion C e H5o(C0R)C0R' 

which would then dissociate into a molecule of 
oxo-carbonium ion, R'CO^ 

e n 5' 
ester and 

the other 

CftR-cOCOR* + RCO 


C 6 H 5 0(C0R')C0R 

C s H 5 0C0R 



Since these reactions are run in the presence of an excess of 
aluminum chloride, the ester would be completely tied up as the 
aluminum ohlorlde complex and thus would be unavailable for such 
an attack. 

Kinetic studies have shown that the reaction is not second 
order (3). This excludes mechanism II above. Step 2 is probably 
the rate determining step in the formation of the oxo-carbonium ion. 

Dilution studies have shown that the reaction at best only 
simulates an intramolecular mechanism. Experiments run at high 
dilution in the presence of a competing nucleus have shown that a 
significant amount of product is obtained by the chanoe attack of 
the oxo-carbonium ion on the nucleus from which it seceded. 

Ortho- , Para - Orient at ion 

An advantage of the Fries reaction is that it is possible to 
predict the orientation of the acyl group. The structure of the 
group and the reaction temperature are the determining factors (4). 



Aoyl G-roup 

aroma tic- aliphatic 
(e.g. phenylacetyl) 

(e.g. acetyl) 

(e.g. benzoyl) 

.Isomer Produced 


ortho- or para - de- 
pending upon the 


In the case of the aliphatic esters, a high reaction temper- 
ature (about 160°) in the absence of a solvent leads to crtho - 
orientation. A low reaction temperature (60° or less) in the 
presence of a solvent such as nitrobenzene leads to para-orienta- 
tion. It is interesting to note that with aromatic-aliphatic and 
strictly aromatic esters, orientation is temperature independent. 
A correlation between the enolization tendency of the ester and 
the ortho -orientat ion of its acyl group has been made (4). 
It has been found that those esters which have a strong tendency 
to enolize give rise to ortho - orient at ion while those which cannot 
enolize give only the para - isomer regardless of the reaction 
conditions. On the basis of this, an intramolecular mechanism 
for the ortho - shift, involving enolization as the first step, was 


1. A. H. Blatt, Organic Reactions, Vol. I. John Wiley and 
Sons, New York, N. Y., 1943, p. 342. 

2. A. H. Blatt, Chem. R e v., 27, 429 (1940) . 

3. R. Baltzly and A. P. Phillips, J. Am. Chem. Soc, , _?2; 4191 

4. Or, I. aershzon, J. Gen. Chem., 13, 82 (1943); C. A. 38, 
1220 (1944). 


Reported by Robert G-. Bannister March 25, 1949 

Because of its activated methylene groups, tetralin undergoes 
autoxidation with measurable speed under mild conditions; moreover, 
its primary oxidation oroduct, tetralin hydroperoxide, is a re- 
latively stable compound which can be isolated in crystalline form 
at least 98/ pure. The reaction has therefore been studied in 
detail in an effort to shed light on hydrocarbon oxidation in 
general. Previous work has been concerned chiefly with kinetic 
studies based on oxygen uptake, but recently the identification of 
numerous by-products has made possible a more thorough understand- 
ing of the reaction. 

When a current of air is passed into tetralin at 76 for 50-80 
hours, a viscous reddish-orange oil is produced. The oxidation has 
been shown (1,2,3) to take place in four distinct stages in which 
the reaction is at first barely perceptible, then accelerates 
rapidly, reaches a steady state, and finally tapers off after about 
30/ of the tetralin has been oxidized. It is impracticable to 
oxidize the last 30-40/ of the tetralin. 

The mechanism of the primary oxidation process involves the 
chain formation of peroxides: this reaction accounts for at least 
95/ of the oxygen uptake (2) . 

(a) > OH' + 3 ~t >CH00- 

(b) >CH 3 +>CH00* -> >CH00H + ^>CH- 

Tetralin hydroperoxide is itself an autoxidation catalyst, so that 
the initiation of the above reactions can be attributed to the 
minute quantities of the peroxide always found in tetralin as well 
as adventitious catalysis by active spots on the surface of the 
vessel, etc. After 80 hours reaction time the concentration of 
tetralin hydroperoxide is 25$; at earlier stages it is even higher 
(35-40/) . 

After the reaction mixture has been heated to destroy the 
peroxides (130-150°) and the unreacted tetralin has been re- 
moved, the chief reaction products, oc-tetralone and cc-tetralol, may 
be distilled off. Several workers (3,4,5) have recently shown 
that cc-tetralol, which cannot be separated from the ketone by 
distillation alone, constitutes 20-30/ of this fraction. 
Robertson and waters (3) were also able to isolate the following 
by-products in small amounts from the residue: (i) -V-o- 
hydroxyphenylbutyraldehyde, (il) (9-o-carboxyphenylpropionic acid, 
and (III) Y-0"--hydroxyphenylbutyric acid together with polymeric 
oroducts and saponlfiable substances, probably esters of 
tetralol with acids (il) and (.III). The tetralin fraction was also 
found to contain a small percentage of 1, ki-dihydronaphthalene . 
All of these products were also obtained by thermal decomposition 
of tetralin hydroperoxide. 



Robertson and Waters account for the production of tetralone 
and tetralol by reactions of the type: 

(c) CioHnOOH -> C 10 H 10 + H 3 (see below) 

(d) C 10 H ll OO* + .OH -» C 10 H X1 OH + S 

(e) doHnO* + RH -> C 10 H X1 OH + R< 

The fact that equation (d) plays some part in the formation of a- 
tetralol is shown by the observed evolution of oxygen; however, 
most of the tetralol is probably derived from Ci H lx O« radicals 
which have oxidized (dehydrogenated) other organic matter accord- 
ing to equation (e) . 

The aldehyde (i) is regarded as a product of the direct decom- 
position of the peroxide, analogous to the known decomposition of 

(f) H 3 + ^V\ 



Cfi-» — CH- 



trlphenylme thane hydroperoxide to benzophenone and phenol. Most of 
the peroxide, however, decomposes to form tetralone, as shown also 
in equation (c), indicating "the tendency to break the C-H bond 
rather than the C-aryl link* 

The investigators have found that the phenolic aldehyde (I) 

and acid (ill) are inhibitors of the autoxidation when added at any 

stage of the reaction and therefore attribute to their formation 

the fact that the autoxidation does not proceed to completion. 

The dicarboxylic acid (il) probably results from the further 
autoxidation of a-tetralone in the p position, followed by fission 
of the a, 0- carbon- carbon bond. Pure a-tetralone undergoes slow 
autoxidation at 100 to form the 1,2-diketone and unidentified 
acids, while the analogous autoxidation of cyclohexanone gives 
both the diketone and adipic acid (as well as its hemi-aldehyde) 
in good yield-. 

The investigators regard the phenolic acid (ill) as the 
product bf a reaction between a-tetralone and tetralin hydrooeroxide 
since they were able to prepare the same acid (in the form of the 
lactone) by oxidation of tetralone with Caro r s acid (which is, 
of course, Itself a hydroperoxide), They have postulated the 
following general mechanism for peroxide oxidation of- ketones: 


i. ' ■■■ 




... •■ 

■■• i-: O .1 " 



^C=0 -> 

R »t/ 

(R=C lo H 11 ,S0 3 H, etc.) 


R f » x OOR 

R' OH 

Jr'v x o (or - ) 

R T ,OH 

X - 

R 1 COOR K + H 


The postulated mechanism seems to be applicable to many types 
of peroxide reactions: Caro ! s acid oxidation of ketones, peroxide 
ester rearrangements as reported in a recent seminar (6), per- 
benzoic acid oxidation of ketones (7): C 6 H 5 COAr — » C s H 5 OOCAr, and 
probably also the Dakin reaction; 
OH * ~ OH 

// N - ocho /y N - OH 



H a 0. 


9 X-nrrwn 


The formate intermediate in the above reaction has been isolated 
by carrying out the reaction with peracetic acid in glacial acetic 
acid (8; . 

It is interesting to note the similarity of the above mechanism 
to those of the Beckmann and pinacol rearrangements; 

R 7 0H 





— 0. 


H + ^C=N + ( " H) 




L ( 0H ) 

^H J 


.C C^ 

R / j + X R 
OH /0H\ 

















rt R 





Waters, Ann. Reoorts 42, 130 (1945) , 

Robertson and Waters, Trans. Far. Soc, 42, 201 (1946). 

Robertson and Waters, J. Chem. Soc. , 1948, 1574-90. 

Sully, Trans. Far. Soc. _42, 261 (1946) . 

Mentzer and Billet, Bull. Soc. Chim. JL5, 835 (1948). 

Jean Crawford, Organic Seminar, Jan. 7. 1949. 

Friess, J. Am. Chem. Soc. 21> 14 (1949;. 

Robert Bauman, Organic Seminar, Nov, 1, 1944. 

Reported by Aaron B. Herrick March 25, 1949 


Introduction : The synthesis of cyclobutane and its derivatives is 
difficult because of the strain involved in forming four-mernbered 
rings. Few synthetic methods leading to cyclobutane s were known 

until recently, and most of these result in poor yields. Xn re- 
cent months cyclobutane has been prepared for the first time in 
good yield; a few general method of preparing alkyl cyclobutane s 
has appeared, and fluorinated cyclobutane derivatives have been ob- 
tained in large numbers. 

Historical : Perkin in 1883 prepared the first cyclobutane deriva- 
tive, the mono-carboxylic acid, from trimethylene bromide and 
malonic ester in the presence of sodium ethoxide, followed by 
saponification and decarboxylation (2). The two dicarboxylic acids 
were also prepared by Perkin in a similar manner. A method using 
sodium cyanide as the condensing agent affords the 1, 2-dicarboxylio 
acid in better yield (3) . In a similar fashion trimethylene 
bromide and benzyl cyanide condense in the pre se nee of sodamide to 
yield 1-cyano-l-r-phenyl- cyclobutane. 

The only other general source of cyclobutane derivatives is 
a number of dimerization reactions. Cinnamic acid dimerizes to a 
mixture of truxinic and truxillic acids; divinyl acetylene di 
to a cylcobutane derivative (i), and octafluorocyclobutane 
tained upon heating tetrafluoroethylene . (ll) (4). 

is ob- 


= CH-C=C-CH— OH. 

p — \j:\— \j— \j— oil'—" Ch - 

! I 

OH 3 = OH-OSC-GH— 0H a 


CF 3 

CF 3 

~ CF 8 

~ CF 3 

Preparation of cyclobutane : Willstadter obtained a small amount 
of cyclobutane hy a series of reactions in 1907. However this 
compound was not prepared in appreciable yield until Cason (5) 
developed the following synthesis: 

,C0 3 H 



Br 2 
in CC1 4 


Bu 3 




Cason also obtained cyclobutane in 7^ yield from tetramethylene 
bromide by a Wurtz reaction employing sodium in boiling xylene. 
Although this aooears to be a poor yield, the best previously ob- 
tained was approximately 1/. 

Preparation of tetrafluorocyclobutanes : A large number of tetra- 
fluorocyclobutane derivatives have been reported recently (6), They 
are produced by the reaction of tetrafluoro ethylene with a variety 
of unsaturated compounds. The products are one to one adduots, and 



they are obtained by heating the reagents at 100-150 in a high 
pressure bomb. o 

100-150 CF 3 ~CH 3 

CF 3 = CF 2 + CH 3 = CH - X -* I | 

bomb CF 3 -CH-X 

Table of Representative Tetraf luorocyolobutanes 

Starting material 

CH 3 = CH S 

C 6^5 OH = CH 3 

0H 3 

CH 3 

CH 3 

CH 3 

CH 3 

CH 3 

CH 3 





CH-CH 3 0H 



C = CH 3 


-C 6 H S 




-C" 3 0H 

-C0CH 3 

-CH = OH a 

= CH 3 

Percent Yield 


These "cycloalkylations" proceed in higher yield at lower tempera- 
tures than the dimerization reaction mentioned previously. The 
relative ease of reaction is: dienes > CH 3 = CHR^) RCH = CHR. 

Preparation of 'fflkyl cyclobutanes : Boord (7) and coworkers have 
recently outlined a novel synthesis of alkyl cyclobutanes from 
neopentyl type tribromides using a zinc dehalogenation in molten 
acetamide (the Hass-MoBee procedure). The yield of alkyl idene 
cyclobutane is usually 40-50^ of theory. Some of the other 
products of the reaction, differing only in the position of the 
double bond, can be hydrogenated to the same alkylcyclobutane , 
The synthesis of ethyl cyclobutane illustrates this method. 

CH 3 0H 
CH 3 0H 

CH ,CH= 

CH 3 Br 
PBr 3 CH,C^ 3 CCH 3 Br Zn, CH 3 C0NH 3 


!H 3 Br 

Na 2 C0,. Nal 




CH 3 =CH 3 



> (189?) 

10,% ) 

CH 3 CH 3 -C=CH 3 



CH 3 CH 3 

X 0, 
CH 3 S 

^CH 3 
V CH 3 


Starting with the appropriate tribromide, the corresponding methyl 
and isopropyl cyclobutane s were also obtained by this method. 
Methylene cyclobutane and methyl cyclobutane were obtained in 
better yield from the dehalogenation of the tetrabromide from pen- 
taerithritol using zinc in ethanol (G-ustavson procedure) (8). How- 
ever, the Boord synthesis appears to be the only general route to 
monoalkyl and monoalkylidene cyclobutane s. 

A mechanism suggested by Poord to account for the products 
obtained (above) involves the preliminary formation of a bromo- 
ethylcyclopropane (a) which is further converted by zinc bromide t< 
the carbonium ion (B) , which then undergoes a rearrangement 
(Whitmore shift) to the cyclobutyl carbonium ion (C). By a 
"hydride shift" (6) rearranges to (D), another carbonium ion. 
Familiar operations on these four intermediates lead to the 
products obtained. 

CH 3 Br 0H 3 Br r H 3 Br -+; 

C 2 H 5 -9-CH s Br Zn C 3 H 5 -C-CH 3 Br -ZnBr 3 C 3 H 5 -C-CH 3 ZnBr 3 C 3 H 5 0H 3 
CH 3 Br -> CH 3 ZnBr -> CH 3 -* " ,x CPf 

(A) CH 2 CH 3 

V r*s (B) 

Whitmore CH 3 -C-C-CH 3 "Hydride" CH 3 -C-CH 3 -CH 3 

-» H CH 3 -CH 3 -> H CH 3 -CH 3 
shift shift 

(0) (D) 

CH 3 -Q-C-CH 3 -H CH 3 CH = C-CH 3 

H 0H a -CH a -> \ \ 


CH 9 -CH 

2 Vli 3 


^) Br ZnBr 

CH 3 -CH 3 -p-CH 3 Br CH 3 CH 3 -C-CH 3 Zn CH 3 CH a -q-CH 3 HQ 

0H 8 >CHa -> CH 3 ^CH 3 -> CH 3 -CH 3 -► 


C 3 H 5 CHqH 3 + ZnQBr 
CH 3 CH 3 

6 @ 

CH 3 -g~9H-CH 3 -H CH 3 = CH-CH-CH 3 

H CH 3 -CH 3 -► CH 3 ->CH 3 

(D) G> 


C 3 H 5 /CHs electron- C 3 H 5 ^CH 3 Br Zn HQ C 3 H 5 .CH- 

>C/" pair' shift ^ Ch -*_>-» ^C / 

CH 3 -CH 3 -> GH 3 -CHf CH 3 CH 


C 3 H 5 ^ /CH 3 Br C 3 H 5 .CH 3 

^-C Zn HQ ^G^ 

CH 3 CH 3 -> -> CH 3 — CH 3 




1. G-ilman, "Organic Chemistry", pp. 65-116 (by Reynold C. Fuson)-, 
John Wiley and Sons, New York, 1943. 

2. Perkin, Ber. , JL6, 1793 (1883). 

3. Fuson and Kao, J. Am. Chem. Soc, 51, 1536 (1929). 

4. Henne and Ruh, ibid., 69, 279 (1947). 

(Footnote 1 contains r eferences to the patent literature). 

5. Cason and Way, J. Org. Chem. 14, 31 (1949) . 

6. Coffman, Barrick, Cramer and Raascb, J. Am. Chem. Soc. 71, 
490, (1949). 

7. Derfer, Greenlee, and Boord, ibid., 71, 175 (1949). 

8. Shand, Schomaker, and Fischer, ibid., J36, 636 (1944). 


Reported by H. A. DeWalt, Jr. April 1, 1949 

The well known use of decarboxylation reactions in organic 
synthesis has instigated numerous studies of this reaction. The 
influences of substituent s, pH, solvents, biochemical enzymes, 
and optically active catalyst for asymmetric decarboxylations have 
been studied (1,3,3). Unfortunately, all of these investigations 
with the irdifferent points of view did little to correlate this 
general reaction. From the results of these investigations and the 
modern concept of organic reactions Sohenkel et al (4,5,6.7) have 
treated decarboxylation reactions as outlined in this abstract. 

In general, the decarboxylation of an acid follows the 

R: J -0: | H -* R:H + C=0, The bond between the R 
group and the carboxyl group is cleaved with retention of the 
electron pair by the R group. This is followed by the lost of 
carbon dioxide with simultaneous electroohilic attack of the 
carbanion by the proton. The following two eleotrophilic 
mechanisms have been formulated and verified by experiment. 


Sfcl H + R-COOH -» RH + C0 3 H 

The rate determing step is the displacement of the carboxyl group 
by the hydrogen ion. Anthracene-9-carboxylic acid decarboxylates 
according to this mechanism, and the orediction that the rate of de- 
carboxylation increases with increasing acidity of the solvent has 
been experimentally verified. 

Se^ RCOOH -> R:^ + COOH * -> C0 S + H® 

R: w + H -> RH 

The rate determing step is the dissociation of the bond between the 
R and the carboxyl groups. The strong organic acids - such as 
trichloroacetic, tribromoacet ic, nitroacetic, and 2,4,6 trinitro- 
benzoic acids — were found to decarboxylate by this mechanism. 

*■£• The Catalyzed Reaction . — The sensitivity of certain decarboxy- 
lation reactions to catalysts is well known. Schenkel (6) explains 
the action of the catalyst with the assumption of a donor-acceptor 
reaction between the carboxylic acid and "the catalyst exclusive of 
the dielectric properties of the solvent. 

Further correlation is possible by considering the separate 
influence of the R and the carboxyl groups upon the decarboxylation 

A. Reactions with the COOH, CO 

1. Where the acceptor molecule is an acid molecule, the 
following reactions are possible. 



I 01 ~> A^ 


r> -c-oh 

■C-OH + A— I © 

(^ -0-0 + AOH^ 

Thru polarization of the carbonyl group, the CO grouo becomes 
positive or electron deficient and impedes decarboxylation. 
Studies on the energy of activation for decarboxylation of tri- 
chloroacetic and trinitrobenzoic acids in water dioxane solutions 
show a decrease energy requirement with increasing diox»ne content. 
Hydrogen bonding between the oxygen atom of the carbonyl grouo and 
the hydrogen atoms of the water molecules causes the CO group" 
to become positive. A higher energy of activation is then reauired 
to push the electrons towards the CO to complete the electron 
octet and permit decarboxylation. 

2. When the donor molecule is a basic molecule, the CO 
group possesses two acidic atoms, the hydrogen atom and the carbon 
atom according to the Lewis concept. Either of these acidic atoms 
can be neutralized singly or at the same time by basic molecules. 

^ -C-0 + HD a 
-C-OH + D 

<? V © 



-C-OH £■ 

1 01© 

+ 21) _♦ -C-o® + HD ® f 


In each of the above equations the CO becomes negative ard will 

*„ 4.? 0Ut ^ a i ? the x " 7e11 known decarboxylation in alkaline medium. 
Equation IB explains the catalytic effect of tertiarv amines on 
decarboxylation of beta ketoacids. The following experimental 
facts are offered as evidence for this mechanism: 

(1) Since the anion is stable towards decarboxylation, 
the increased decarboxylation rate of the free acid 
in the presence of an amine can not be due to route 

(2) The catalytic effect of the tertiarv amine in acid 
solution is retarded due to the lose of its co- 
ordinating electron in salt formation. 


Route V Is observed by the kinetic investigations on the rate 
of decarboxylation of trichloroacetic acid in aniline - benzene or 
toluene solution. The greatest yield of product was obtained when 
two moles of aniline were present for every mole of acid. 

B. Reaction with the rest of R- 

The general treatment of catalytic addition to the R group 
of carboxylic acids must be postponed until this part of the 
problem has been more throughly investigated. However, the follow- 
ing cases represent two interesting examples. 

Alpha ketoaol&e * — These acids can be catalytically decarboxyl- 
ated by the SeI mechanism. The required primary amine converts the 
ketoacid into an iminoacid which on account of the strong basicity 
of the nitrogen atom exists in the immonium carboxylate form (i). 

-C-C0 2 H 4-» -C-C0 3 «-> -0-0=0 

(I) (II) 

The resonance of (i) to (II) permits the formation of a strongly 
positive or electron deficient alpha carbon atom which attracts 
the electron pair of the bond connecting the negative CO group 
to itself. Decarboxylation then stabilizes the molecule. 

Beta-ketoacids . — Since these free acids are readily decarboxyl- 
ated, Schenkel (6) proposes the following mechanism. 


- A -c=o 

-c-c- r -o=o 


-C=C< + CO; 

This decarboxylation proceeds through intramolecular neutralization 
of the polarized beta carbonyl group by a proton from the CO 
group 'Hie beta carbon atom becomes positive and induces the 
dissociation of bond between the alpha carbon atom and the 
negatively charged CO group. Double bond formation between the • 
alpha and beta carbon atoms neutralizes the latter' s electron 
deficiency. This mechanism predicts the following: &l) Beta- 
keto-acids should decarboxylate according to Sgl. (s) Anions 
can not be decarboxylate. (3) [5,5] bicyclo beta-keto-acids should 
be stable to non catalytic earboxy la t ion since double bond 
formation between the alpha and beta carbon atoms is prohibited 
by Bredt's rule. 

Experimentally, beta-keto-acids have been found to decarboxy- 
late according to Se 1 and their anions are not decarboxylated. 
7,7 dimethyl-bicyclo-[l,2, 2]-?-heptanone-l-carboxylic acid has 
been found stable towards non- catalytic decarboxylation. 

V * 


This same decarboxylation mechanism has been applied to 
acetoned- oar boxy lie, dihydroxymaleic, and dibromomalonic acids 
where the carboxyl group beta to the carbonyl is the one that de- 
carboxylates. However, in these acids the mechanism fails to 
explain why the free acids decarboxylate slower than the singly 
charged anion. 

III. Miscellaneous Reactions . — Hammick and coworkers (9,10) in 
their study of decarboxylation repctions boiled aloha picolinic, 
quinoline-2-carboxylic, and isoquinoline-1-carboxylic acids in 
benzaldehyde and isolated secondary alcohols instead of the 
expected products. When acetophenone or benzophenone was used 
as solvent, tertiary alcohols were isolated. The products of the 
decarboxylation reaction were similer to those obtained if the 
previously mentioned carbonyl compounds were treated with alpha 
pyridzyl, 2-quinolyl, 1-iatiquinolyl magnesium bromides. These 
workers offered the following mechanism. 

R 3 C00H -> R 3 ^ + C0 3 + H® 

R ® + Ri-C-R 3 + H® -> 

Further studies to prove the generality of this reaction indicated 
that only those heterocyclic acids containing the structure -SNfl' 
would produce carbinols when decarboxylated in the presence of 
carbonyl compounds. The -^=Q / structure is similar to the 
cyanide ion [N^C'J whe^e one'of the nitrogen to carbon bonds is 
replaced by a'ring, [N=C,T 5 '. It, therefore, follows that the 
formation of carbinols by decarboxylation of these three alpha 
imino acids is similar to the analogous cyanohydrin formation 
with hydrogen cyanide and carbonyl compounds. 

Of all the various carboxylic acids isomers of the pyridine, 
quinoline and isoquinoline series, only those molecules containing 
the alpha imino acid group decarboxylate most readily. This 
case of decarboxylation can be explained as due to the resonance 
of the imino group which is described in the alpha keto acid section 
of this abstract. 


1. Cazeneuve, Bull. Soc. Chim. France [3] 2, 550 (1892). 

2. Ibid [3] 15, 72 (1896). 

3. Hemmelmeyr, F,, Monatsch 34, 365 (1913). 

4. Schenkel, H. and Klein, A., Hel . Chim. Acta. 28,1211 (1945). 

5. Schenkel, H., Hel. Chim. Acta. 29, 436 (1946). 

6. Schenkel, H. and Schenkel-Ruden, M., Hel. Chim. Acta. 31, 

514 (1948). — 

7. Ibid, Hel. Chim. Acta. 31, 924 (1948). 

8. Conway, E. J. and MacDonald, E., Nature 156, 752 (1945). 

9. Dyson, D. and Hammick, D. Lt., J. Chejn. Soc, 1937 , 1724. 
10. Ashworth, M. r. p., Daffern, and Hammick, I>. Lt., Ibid 

1939 , 809. 

'.:■!'*.■ ■ ■"'-•" v 



Reported by Carl S. Hornberger, Jr. April 1, 1949 

The preparation of tertiary alcohols by the reaction of 
G-rignard reagents upon carbonyl compounds is limited by steric 
factors to components which are not highly hindered. Often with 
branched aliphatic reactants, normal addition is not found and the 
reaction products are those obtained by dehydrohalogena tion, 
enollzation, coupling, and reduction (1,2,3). These difficulties 
have been overcome in part by the use of organolithium compounds 
which have a greater tendency to undergo normal addition to the 
carbonyl group (4). However, these seem to be limited to compounds 
which are no more highly branched than in the case of diisopropyl 
ketone and isopropyl lithium which react to form triisopropyl 
carbinol (5). 

An early work by Morton and Stevens (6) indicated that ketones 
and organic halides could be condensed in the presence of sodium 
to give carbinols analogous to those obtained through the use of 
G-rignard reagents. 

Na . * 

+ CH 3 CH 2 COCH 3 CH 3 -> <(/ ^>COH(C 3 H s ) 3 (25#) 

In 1945, the first successful synthesis of tri-t-butyl carbinol 
was achieved by an extension of this reaction (7). 

(CH 3 ) 3 CCOC(CH 3 ) 3 + (CK 3 ) 3 CC1 Na [CCH 3 ) 3 C] 3 COH (8.5#) 

The synthesis of carbinols with a great degree of branching was 
thus facilitated by the utilization of increasingly specific 
reagents. (Na>Li>Mg) 

Recently, the use of sodium has been investigated more 
thoroughly so that now the reaction seems applicable to a wide 
variety of highly branched alcohols (8). 

Preparation from ketones When a solution of ketone and organic 
halide is added to sodium sand dispersed in solvent, a reaction takes 
place which yields after hydrolysis an alcohol. 

(1) (CH 3 ) 3 CHCOCH(CH 3 ) 3 + CH 3 CH 3 CH 3 C1 Na [ (CH 3 ) 3 CH] 2 COH (CH 2 CH 2 CH F 

(2) (CH 3 ) 3 CHCOCH(CH 3 ) 3 + (CH 3 ) 3 CHCH 2 C1 N -^ [ (CH 3 ) 3 CH] 3 COH[CH 3 CH (CH 3 ) 3 : 





(3) (CH 3 ) 3 CCOCH(CH 3 ) 3 + (CH 3 ) 3 CHC1 -> [ (CH 3 ) ,C] COH[CH(CH 3 ) J 3 


(4) (CH 3 ) 3 CCOCH(CH 3 ) 3 + (CH 3 ) 3 CC1 Na [ (CH 3 ) 3 C] 3 COHCH(CH 3 ) s (16$) 

The diisopropyl ethyl carbinol corresponding to the first 
example has "been prepared using the ethyl G-rignard reagent in a 
somewhat greater yield (9), This seems to indicate that on the 
basis of yield, there is little choice between the two methods 
when using a straight chain halide. The isopropyl G-rignard reagent 
will not add to diisopropyl ketone but gives 68$ enolization and 
21$ reduction of the ketone (10) « Isopropyl lithium will add to 
this ketone to give the carbinol in 19$ yield. 

With the methyl ketone, pinacolone, the reaction was one of 
self condensation. 


(CH 3 ) 3 CCOCH 3 + GH 3 CH 3 CH 3 CH 3 C1 -► (CH 3 ) 3 CC(CH 3 ) = CHC0C(CH 3 ) s 

Preparation from esters The reaction with esters leads to 
the formation of several products* 

Na (G-rignard 7$) 

( 6) (CH 3 ) 3 CHCOOCH 3 + (CH 3 ) 3 CC1 -* [ (CH 3 ) C] 3 C0H[CH(CH 3 ) 3 ] (19$) 


(•0H 3 ) 3 CGXH(CH 3 ) 3 (50$) 

(CH 3 ) s CGH0HCH(CH 3 ) 3 (7$) 

(?) CH 3 CH 2 CH 3 C00CH 3 + (CH 3 ) 3 CCl-> (CH 3 ) 3 GC0CH 3 CH 3 CH 3 (6$) 


(CHg) 3 CCHOHCH 3 CH 8 CH 3 (5$) 

(8) (CH 3 ) 3 CC00CH 3 + (CH 3 ) 3 CC1 -> [ (0H 3 ) 3 C] 3 G0H 

L(CH 3 ) 3 C] 3 C0 ^ (71$) 

[(CH 3 ) 3 C] 3 CH0H ) 

Oxidation of the crude carbinol- ketone mixture from reaction 
number eight hps been found more convenient for the preparation 
of di-t-butyl ketone than the previous methylation of pen tame thyl 
acetone or t-butyl methyl ketone (7). 

The reaction fails when applied to other acid derivatives 
like amides or anhydrides or when tried on aldehydes* 



Mechanism Since it has been established that both carbonyl 
compounds and esters form a disodium derivative, a metathetical 
reaction may take place (11). 

R01 + Na-C-ONa -> NaCl + (R) 3 CONa 

A more likely mechanism seems to be based on the reaction of a ketyl 
free radical to form the sodium alkyl as follows! 

RCOR + Na* -» (R) s CONa 

(R) 9 CONa + RC1 -» R« + NaO-C-Cl 

NaO-C-Cl -> NaCl + (r) 2 CO 

R- + Ka. — > RNa 

RNa + (R) 3 CO -> (R) 3 CONa 

When this reaction is used in the laboratory, it seems to be 
easier to carry out than the corresponding G-rignard reaction. It 
is a one step reaction which is suitable to rather large scale 

(65 mole) and is one from which a major portion of the unreacted 

starting material may be recovered. 


1. Conant and Blatt, J. Am. Chem. Soc. , 51, 1227 (1929). 

2. Whitmore et al. ibid . , 54, 1239-1251 Tl942) ♦ 

3. Sampson, Org. Seminar, U. of 111., Dec. 9, 1942. 

4. Ziegler, Angew, Chem., 49, 455 (1936) . 

5. Bartlett, Swain, and Woodward, J. Am. Chem. Soc, 63, 3229 (l34l). 

6. Morton and Stevens, ibid ., 53, 2244 (1931). 

7. Bartlett and Schneider, ibid ., 67, 141 (1945). 

8. Cadwallader, Fookson, Mears, and Howard, J. Research Nat. Bur. 

Standards, 41, 111 (1948). 

9. Howard, Mears, Fookson, Pomerantz, and Brooks, ibid . , 38, 365 

(1947) . 

10. Young and Roberts, J. Am. Chem. Soc, _66, 1444 (1944). 

11, Kharash et al. J. Org. Chem., 5, 362 (1940) . 



Reported by Emil W, Grieshaber April 8, 1949 

The ability of a conjugated system containing a hetero atom 
at a terminal position to undergo 1,4-addition has been related to 
similar abilities of longer conjugated systems to add 1,6 and 
possibly 1,8. As a result, the term "Conjugate Addition" has been 
proposed as a general name to include all such addition reactions. 
This seminar is limited to examination of the mechanism, to a brief 
consideration of some normal 1,4 addition reactions and to © review of 
an abnormal addition of this type. 

A simple conjugated system may be described by the following 
resonance structures: 

R R R 

RCH=CH-'C-0 f-> RCH=CH-C=0 *-* RCH-CH=C-0 (l) 


The addition of a negative fragment to positions 2 or 4 in these 
systems may proceed Kith either polarized structure I or III with 
the attachment of a positive fragment to position 1 followed by re- 
arrangement of the enol so formed. Support for the polarized 
structures as indicated is lent by the fact that the negative ion 
invariably attaches itself to a carbon at position 2 or 4, usually 
the latter. 

R R -> R 

RCH=CH-C=0 + HX -> RCH-CH=C-OH «- RCH-CH 2 -C=0 (2) 

V < 4 3 3 1 


In many instances the product of a 1,2 addition is unstable 
and the reaction is reversible so it is not isolated. On the other 
hand, 1,4 additions are thought not to be readily reversible. That 
the addition is actually 1,4 and not 3,4 as an examination of the 
product would lead one to believe is established by the addition of 
a G-rignard reagent to a conjugated system. 

C s H B CH=CH-C=0 5 + C e H 5 MgBr -> -> (C 6 H 5 ) 3 CH-CH 3 -C=0 5 (3) 

Had the addition gone 3,4 the carbonyl group would have been exposed 
to further attack by excess G-rignard Reagent to yield a carbinol. 
Further, enols have been isolated in the form of peroxides (6). 

Reagents which add to the conjugate system in this manner in- 
clude water, hydrogen halides, sulfhydryl compounds, ammonia and 
amines, Grignard reagents, hydrogen cyanide, sodium bisulfite and 
active methylene compounds. The latter group is usually classified 
as a Michael condensation and is not considered here. The general 
reaction is given in equation (2). Special examples include: 

i a). Addition of ammonia to phorone to form triacetoneamine 



(CH 3 ) 3 C=CH-C-CH=C(CH 3 ) 3 + NH q -► (CH 3 ) 3 -C-CF 3 -b-CH=C= (CH 3 ) 

NH 3 

H,A,H 2 

(CH 3 ) 



=(CH 3 ) 

b) . Addition of cinnamalhydrazone to itself to form 5-phenyl 

C 6 H 5 CH=CH-CH=N-NH ! 

C 6 H 5 -CH— CH 3 


c) . Addition of 2-mercaptoethanol to acrylonitrile 
HO-CH 3 -CH 8 -SH + CH 3 =CH-C=N -> 


H0-CH 3 -CH 3 -3-CH 3 -CH 3 -C^N (6) 
d). Addition of hydrogen cyanide to ethylbenzalmalonate (7) 


C 6 H 5 CH=C(C0 3 C 3 H 5 ) 3 + 

C 6 H 5 GH-CH(C0 3 C 3 H 5 ) 3 

The above reactions involve systems which are terminated by a keto 
carbonyl group, a nitrile group, an imino linkage and an ester 
carbonyl group. 

Recently a 1,4 addition reaction accompanied by replacement 
has been reported by Richtzenhain (9, 10, 11). 2, 5-Dimethoxybenzo- 
nitrile IV (4,8) was found to react with ethylmagnesium bromide to 
give 2-ethyl-3-methoxybenzonitrile V. This amounts to replacement 
of the 2-methoxyl group by the alkyl group of the Grignard reagent. 
A mechanism which allox^s for 1,4 addition with subsequent elimin- 
ation of methyl alcohol or its equivalent is necessary to rational- 
ize the course of this reaction (5,9). 




-och 3 


C 3 H 5 MgBr 






H a 





~C 3 H 5 



+ CHoOH 



The related 2, 3-dimethoxy~5-methylbenzonitrile underwent a similar 
replacement when treated with ethylmagnesium bromide. It was found 
that methylmagnesium bromide gave only the normal 1,£ addition pro- 
duct which could be hydrolyzed to the 2,3-dimethoxyacetophenone . 
Other ^ alky 1 G-rignard reagents found to add as does the ethyl reagent 
are listed with yields of methoxyl group replacement oroduct as 

U V 


Grignard reagent Percent yield 






Although phenylmagnesium 
add 1,2 to IV, (2) Richtzenhai 
of 1,2 and 1,4 addition. Appa 
not anticipating a replacement 
condense with IV (l) . In gene 
with lower yields of rep lace me 
Richtzenhain therefore employe 
aromatized the addition produc 


bromide had been reported earlier to 
n obtained approximately equal amounts 
rently the earlier investigators were 
Benzylmagnesium chloride does not 
ral, aromatic G-rignard reagents add 
nt product than do alkyl reagents. 
d tetrahydro G-rignard reagents and 
t to obtain better yields (lo). 

The function of the adjacent methoxyl groups is not yet known. 
2-Methoxy-l-naphtho nit rile VI would seem to offer enhanced possi- 
bilities of 1,4 addition since it is known that the naphthalene 
derivatives possess a greater double bond character between the 1- 
and 2-positions. The behavior of 2-methoxybenzonitrile VII should 
test the necessity of the methoxyl group in the 3-position. If this 
group is exerting an important influence on the cyano group, 2,5- 
dimethoxybenzonitrile VIII should undergo replacement. A doubled 
opportunity for replacement is available in 2,6-dimethoxybenzonitrile 

IX, which might exhibit steric inhibition of the 'competing 1,2 
addition to the cyano group. None of these compounds could be shown 
to give the desired replacement; rather, good yields of the normal 
products were isolated (3). 

C=N C=N 




-0CH 3 

k V\^ J 


CH 3 

















D.C. Heath 

^y* 6 - 

Baker and Eastwood, J. Chem. Soc, 

Baker, VI. and Smith, A. R. , Ibid., x 

Chadwick, Ph. D. Thesis, U. of 111 .7 

Fieser and Fieser, Organic Chemistry , 

Boston, 1944, p. 697. 

Gaertner, Ph. D. Thesis, U. of 111.', 

Gilman, Organic Chemistry , 

Vol.1, p. 672. 

Gilman, Organic Sftntne s6 s , John Wiley and Sons, 

Vol. I, p. 451. 

Noelting, Ann. chim. phys. 

Richtzenhain, H. Ber. , 77B ', 1-6 

Richtzenhain, H. and Miedreich, A., Chem. Ber., 81, 92-7 (1948). 


John Wiley and Sons, 

and Co . , 

New York, 
New York, 



502 (1910). 

Richtaenhain, H. and Nippus, P., Ber., 77B, 566-72 (1944). 


Reported by William R. Miller 

April 8, 1949 

Structural considerations and the common reactions of furan 
were discussed in the last seminar on this subject (l). The con- 
clusion reached in that discussion was that furan possesses a 
degree of unsaturation less than that of a 1,3-diene but greater 
than that of benzene. This seminar will discuss the more recent 
work on furan which, in general, appears to bear out this con- 


The methods of synthesis of substituted furans have been well 
discussed in a recent article by Wright and G-ilman (2) . A new 
general synthesis was reported last year (3); An alpha , beta -un- 
saturated ketone :1s treated with sulfuric acid and acetic anhydride 
to form a del_fca- suit one. This compound is then pyrolyzed to give 
the furan. The synthesis of 2, 4-dimethylfuran illustrated the 

(CH,) 2 C=CHCOCH. 



CH 3 C- 













It is necessary that the ketone be branched in the position beta 
to the carbonyl. This synthesis may be adapted to give 2,4-, 
2,3,4-, 2,3,5- and 2, 3, 4, 5- substituted furans. 

Nuclear Oxidation 

The reaction of furan which has been most studied recently is 
that with alcohols and halogens to give 2, 5-disubstituted-2, 5-di- 
hydrofurans (4). This reaction is of special interest in that the 
products are the cyclic acetals of dialdehydes or diketones and 
that hydrolysis will produce these compounds (4a): 


X, RO 





H a 






This reaction may be carried out using a wide variety of sub- 
stituted furans. Acetic acid may be used in place of the alcohol 
to give the corresponding acetoxy derivatives (4b). 


The acylation of furan has been extensively studied in the 
past two years. Acetic anhydride reacts with furan to produce the 
2-acetofuran under the catalytic influence of boron trifluoride 
etherate (5a) and methyl alcoholate (5b), phosphoric acid (5c), zinc 
chloride, acid clays (5d) and hydriodic acid (5e). The longer the 
acid chain the better were the yields. Propionic and n-butyric 
anhydrides have also been used (5a). 2,5-Diphenylfuran may be 


acylated in the 3-position by means of acetic anhydride and stannic 
chloride (6). 

Sulf onation 

Russian workers have made an extensive study of the sulfonation 
of furan (7). Furan is best sulfonated by pyridine- sulfur trioxide 
in a sealed tube at 100° for eight to ten hours. Furan gives the 
2-furansulf onic acid. Sylvan (2- me thylfuran) gives the 5- and 3,5- 
disulfonic acids. 2, 5-Dimethylfuran gives the 3-sulfonlc acid (7b). 


Furan can be chlorinated to give 2-chlorofuran provided that 
the temperature is maintained at 50° and the HC1 formed is immed- 
iately removed from the reaction mixture (8a) . Low temperature 
chlorination (at -40° to -20°) will give a mixture of mono-, di-, 
tri- and tetraehlorofurans as well as 2, 2, 3,4, 5, 5-hexachloro-2, 5- 
dihydrofuran but the 2-chlorofuran may be separated by distillation 

The bromination of beta- (2- fury 1) -acrylic acid and its esters 
may be so regulated as to give a variety of products of both 
addition and substitution ($). 

Other Reactions 

Both sylvan and furfuryl alcohol will undergo the Mannich 
reaction to give the 5-aminomethyl derivative (10). 

Methyl fur oat e can be oh loro me thy la ted to give the methyl 5- 
chlorome thy If urate. The alpha -chloroethyl derivatives can also be 
readily prepared (ll). The chloromethylation of 2, 5-diphenylf uran 
gives only the 3, 4-di- (chloromethyl)-2, 5-diphenylf uran (12). 

Furan is metalated in the 2-position by n-butyl lithium (13a) . 
With sodium and amyl chloride, followed by carbonation and reaction 
with diazomethane, 27$ of methyl furoate and 19^ of dimethyl 2,5- 
furandicarboxylate are obtained (13b) . 

Furan reacts with diazonium salts in the presence of base or 
with N-nitrosoacetanilides to give 2-arylfurans (14). With ]9- 
nitrobenzenediazonium chloride in alcohol solution, however, 2,5- 
dimethylf uran is cleaved and the final product is 1-p-nltrophenyl- 
3-acetyl-5-methylpyrazole (15). 

Sylvan condenses with alpha , Jbeta- unsaturated ketones and 
aldehydes to give beta - (5-methyl-2-furyl) carbonyl compounds (16). 

Diene Reactions 

Furan derivatives have been condensed with diethyl acetylene- 
dicarboxylate to give an intermediate which, on partial hydrogena- 
tion loses ethylene to form the 3, 4-di carboxy- 2- substituted 
furan (l7)j 




HC-O-CR -* + C 3 H 4 

R CCOOEt EtOCOC==CCOOEt 180° 11 'J R 

acid adds 

Furan will condense with ethylene, in the presence of a little 
hydroquinone, to form 3,6-epoxycyclohexene (19). 


1. McBride, Organic Seminar Abstracts, 22 Sept. 43. 

2. Wright and G-ilman, Ind. Eng. Chem. 40, 1517 (1948). 

3. Morel and Verkade, Rec trav. chim. 67, 539 (1948). 

4. a. Fakstrop, Raleigh and Schniepp, Abstracts 114th ACS Meeting, 

14L (1946). ' ' 

b. Clauson-Kaas, Acta Chem. Scand. 1. 379 (1947) [C. A. 42. 
5447c (1948)3. ~ "~" ' 

c. Clauson-Kaas and Fakstrop, ibid., 415 [.C. A. 42, 5901g 
(1948)]. Clauson-Kaas and Limborg, ibid., 619 [0. A. 42, 
50902a (1948)]. Jones, Brit. Pat. 595,041, 25 Nov. 47, 

C. A. 42, 2992b (1948). 

5. a. Heid and Levine, J. Org. Chem. 13, 409 (1948). 

b. Hartough and Kosak, J. Am. Chem. Soc. , 70, 867 (1948). 

c. Ibid., 69, 3093, 3098 (1947). 

d. Ibid., 1012, 1014. 

e. Ibid., _68, 2639 (1946). 

6. Lutz and Rowlett, ibid., 70, 1359 (1948). 

7. a. Terent*ev and Kazitsyna, Zhur. Obschchei Khim. (j. Gen. 

Chem.) 18, 723 (1948) [C. A. 43, 214d (1949)]. 

b. Kazitsyna, Vestnik Moskov Univ. 1947, No. 3, 109 [C. A. 48. 
3751i (1948)]. -*-•' 

c. Terent'ev and Kazitsyna, Compt . rend. acad. sci. URSS 55, 
625 (1947) 51, 603 (1946) [C. A. 42, 556c (1948), 41,~~ 
2033a (1947 )T7 — ' — ' 

8. a. Cass and Copelin, US Pat 2,443,493, 15 June 48, C A. 42, 

?340e (1948) . — 

b. US Pat 2,430,667, 11 Nov. 47, C. A. 42, 2284c (1948). 

9. Dann, Chem. Ber. J30, 435 (1947). 

10. Holdren, J. Am. Chem. Soc. 69, 464 (1947): Holdren and Hixon, 
ibid., 68, 1198 (1946). ~~ 

11. Bremner and Jones, US Pat 2,450,108, 28 Sept. 48, C. A. 43, 
1065h (1949). — ' 

12. Lutz and Bailey, J. Am. Chem. Soc. 68, 2002 (1946). 

13. a. Benkeser and Currie, ibid., _70, 1780 (1948): 
b. Morton and Patterson, ibid., 65, 1346 (1943). 

14. Johnson, J. Chem. Soc. 1946 , 895. — 

15. Eastman and Detert, J. lrriT~Chem. Soc. 70, 962 (1948). 

16. Alder and Schmidt, Ber. 76B, 183 (194377 

17. Hofmann, US Pat 2,382,4IB7-i4 Aug. 45, C. A. 40, 365 a (1946): 

to £' ^' 9 hem ' Soc - 67 ^ 421 (1945). v ~~ 

18. Woodward and Baer,-Ibid. , 70, 1161 (1948). ~ 

19. KE n £ erg ' U l £ a J 2,405.2677 6 Aug. 46. C. A. 40 A 6500 (1946): 
Nudenberg,and Butz/J. Am. Chem. loo. 66, 307 71^44 ). " 



Reported by Sidney Baldwin April 22, 1949 

I * Synthesis ; (1,3,4) 

Cromwell and coworkers have prepared ethylene imine ketones 
by the reaction of the corresponding a, |3-dibrorno ketones with the 
respective amines thus: 

?r §r <40° 3 3 y ®> Q 

RCH-CH-CR' + 3R"NH 3 -> R-CH-CH-CR' + 2R H MH 3 Br 


The following ethylene imine ketones have been prepared in about 
25% yields by this method: l-benzyl-2-phenyl-3-benzoylethyl- 
enimine (i), l-cyclohexyl-^2-phenyl-3-benzoy 1 e thylenimine (il), 
l-methyl-2-phenyl-3-benzoy le thylenimine (III), l-benzyl-2- (m- 
nitrophenyl)-3-benzoylethylenimine (IV ), l-benzyl-2-phenyl-3- 
(p-toluyl)-ethylenimine (v) , and l-benzyl-2- (p-tolyl)-3-benzoyl- 
ethylenimine (VI) . In addition, (l) and (II) were prepared from 
the corresponding amine and oc-bromobenzalacetophenone ; while (I) 
was also produced from benzylamine and a, p-dichlorobenzy laceto- 
phenone , 

II. Isomerism and Ultra-Violet Absorption Spectra : (6,2) 

The yield of ethylene imine ketones Is low, and 
diphenyle thylenimine is known to exist in cis and trans forms (9), 
This indicated that ethylene imine ketones may also exhibit cis - 
trans isomerism. The low yields may have resulted from the 
failure to isolate the more soluble isomer, or from side product 
formation, such as piperazines and cc-imino ketones. When benzyl- 
amine was allowed to react with a-p-dibromobenzyl-p-methvlaceto- 
phenone in dry benzene at 20°, the isomeric products (VA) {29%) 
and (VB) (3?^) were isolated. 

Hs. .C0-C s H 5 -CH 3 -p C 6 H«= C0-C 6 H 5 -CH 3 -p 

,c— cr -* ^c — cr 

,«\ f\ *- * "\ /\ 

OsH s V *'H sunlight H ^"* N ^H 

(VA) CH 3 C 6 H 5 R -T. CH 3 C 6 H 5 (VB) 


trans m.p. 120-121 cis m.p. 70-74 


The low melting isomer (VB) was partially decomposed and rearranged 
to the higher melting form (VA) when its saturated petroleum ether 
solutions were exposed to sunlight at room temperature. Scale 
models of the isomers demonstrate that race mate (VB) would be a 
more highly strained structure than racemate (VA) . Thus the more 
labile form (VB) might be expected to rearrange to the less strain- 
ed form (VA) . Only one form of the imine (I), however, could be 
Isolated. (m.p. 108°, presumably trans) . 

In contrast to the unsaturated amino ketones, these 
ethylene imine ketones, which do not have conjugated unsaturation, 
show only a maximum similar to that of the parent unsaturated 



ketones. The maximum absorption band of the isomer (VB) was 
30-80 A nearer the red (visible), with a 2000-3000 greater 
extinction coefficient value, as compared with (VA) . This is in 
agreement with the view that the more strained structure (VB) 
should absorb light of longer wave-lengths. 

III. Reaction with Hydrogen Bromide: (1,3) 

a-Bromo-p-benzylaminobenzylacetophenone hydrobromide (VII) 
was produced from the reaction of the imine (I) with aqueous 
hydrogen bromide. On the other hand, if the imine (D is treated 
with dry hydrogen bromide in dry benzene solution, the cleavage 
product is the isomeric fi-bromo ketone hydrobromide (VIII) . An 
authentic sample of (VII) was also prepared from (IX) as shown 

benzene Br 

C 6 H 5 -CH— - CB>C0-C G H 5 -> C s H 5 -CH-CH-CO-C 6 H 5 
\ / dry HBr / 

N Jj' eBrH 2 NCH 3 C 6 H 5 (VIIl) 

CH 3 C e H s vHBr <© 

H 3 ^ C 6 H B -CH-CH-C0-C 6 H s 

I | dry HBr C 6 H 5 -CH-CH-CO-C 6 H 5 

C 6 H 5 CH 3 NH 3 Br0 < ] { 

(VII) ether CsH B CH 3 NH Br 


The a-bromo ketone (VIl) released iodine (3) from acidified 
potassium iodide solutions at room temperature in thirty minutes, 
whereas the p-bromo-ketone (VIIl) gave no reaction under identical 

IV. Reaction with Hydrogen Chloride : (1,3,5,6,7) 

Excess amounts of dry or aqueous hydrogen chloride react 
with the imine (i) (presumably trans ) and with the trans imine 
(VA) to yield the p-chloro-cc-aminoke tone hydrochlorides. With 
minimum amounts (i.e. the proper number of equivalent amounts) of 
dry hydrogen chloride, the trans imine (i) gave 73.5^ of the a- 
chloro ketone and 26.5^ of the p-chloro isomer, while the trans 
imine (VA) gave 78^ of the cc-chloro ketone and 22$ of the p-chloro 
ketone. The cis form (VB) reacts very rapidly with excess or 
minimum amounts of dry hydrogen chloride to yield nearly equal 
amounts of the a-chloro and p-chloroketones in both cases. 

A mechanism consistent with the above results is 
proposed as follows: (?) 



(E) + C 6 H 5 -CH CH-Ar <; 

R-NH C1C1 


or HC1 

C s H B -CH-(pH-CO-Ar 


Route [1] 

(e) (a) r 

/<9 ga A r HCl C 6 H 5 -CH CH-CO-Ar 

C 6 H 5 -CH->- OH-)- 6=0 -> \.' /> 

1 .'NIH \ 

L [2]' HH 

(E) i 


Route [2] (S) 


C s H B -CH-gH-CO-Ar HCl or C S H B ~CH„ CH-CO-Ar 

CI NH 3 C1 i CI NH 

+ R (S) R 


A hydrochloride of the type (s) has been isolated (l). An S n l 
mechanism would cleave the ring mainly according to scheme [l], if 
the cc-carbon atom has the higher electron density. Course [2] 
should be followed if an S n 2 mechanism came into play, especially 
if the f3-carbon atom is more electrophilic than the a. Excess 
chloride ion concentration should favor the S n 2 mechanism and 
course [2]. 

Under the influence of excess hydrogen chloride, the salt 
might also undergo ring cleavage via a carbonyl directed 1,4 
attack as outlined below; 



<H > 0^ 




X 1/ 


Cl^HrN^CHpCsHq - 

H-p— C= 

Ar &H a 

CH 3 C 

2 H 



p a 
Ar-CH- CH- 


^H 3 
GH 3 CsH 5 

Scale models demonstrate that trans structures could form a 
transition complex (T) with ease, but that the cis form should not 
because of the repulsion of the chlorine in hydrogen chloride by 
the aryl group on the p-carbon atom. 

V. Reactions with Phenylhydrazlne: (8) 

Trans ethylene 
glacial acetic acid to 
acetylphenylhydrazone s 
sulfuric acid produced 
Pyrazoles correspondin 
prepared in this way. 
yields of the intermed 
stable in neutral or b 
acid solution to produ 

imine ketones react with phenylhydrazlne in 

produce the corresponding pyrazoles and N- 
, Hydrolysis of the latter with boiling 2®^ 

the respective pyrazoles in 90-100^ yield, 
g to (I), (V), and (VI) (all trans ) were 

Cis ethylene imine ketones give excellent 
iate 4-amino-pyrazolines. The latter are 
asic solution but lose benzylamine in strongly 
ce the pyrazoles. 




N-CH 3 C S H 5 

trans imine 



1/ \* 

C 6 H 5 ~N C-R* 

C 6 H 5 NHNH : 



= CH 


C S H 5 NHNH 2 
c is - imine — > 

C«H K -N' C-R" ! 

C S H 5 -N / X p-R" 

. * 7 


\^ CH 2 C 6 H 5 
C 6 H 5 -N' N N-R" 

U U 

R!_C C-H 

r 1 — ^ 

1 h :n H 


etc. C S H 5 -N 


3 u S n 5 




'6 n 5' 

\ I 



H N C-H 

R 1 Hf:H j 
CH 2 C 6 H S 


R* HN: 
C S H 5 CH 3 


C 6 H 5 -N / ^C-R" 





group on 0-4 
and H on C-5 
are on same side 
of ring. 

Groups on 0-4 
and C-5 are on 
opposite sides of 

(XI) + H 3 S0. 


VI. Reactions with G-rignard Reagents : (4) 

G-rignard reagents add to the carbonyl group of ethylenimine 
ketones to give a neitf class of compounds, enimine carbinols, in ex- 
cellent yields. The addition reactions took place rapidly, in- 
dicating that the imine ring offers very little hindrance to the 
carbonyl group. Furthermore, the G-rignard reagent does not open 
the imine ring, nor does it cleave the aliphatic chain. This 
shows that the imine ring is stable and is further chemical 
evidence for the ethylene imine ketone structure. 



Cromwell, Babson, and Harris, J. Am. Chem* Soc 
Cromwell and Johnson, ibid., J55, 316 (1943). 
Cromwell and Caughlan, ibid., 67, 2235 (1945). 
Cromwell, ibid., _69, 258 (1947T7 

70, 1320 (1948). 
, 71, 708 (1949). 

71, 711 (1949). 
, 71, 716 (1949). 

Weissberger and Bach, Ber., j64, 1095 (1931). 

Cromwell and Wankel, ibid 
Cromwell and Hoeksema, ibid 
Cromwell and Wankel, ibid., 
Cromwell and Hoeksema, ibid 

65, 312 (1943), 


Reported by George R. Coraor April 29, 1949 

Most elimination reactions proceed "by one of the following mechan- 



«■* 6 - c • 





fast - = C- + H 

E ft Base 

+ _ 5^ c - x 

+ x 


(Occurs in neutral or 
weakly basic media.) 

H-Base + - 0*=. C- (Occurs in 

I I highly basic 
media. ) 

Elimination and substitution proceed under the same conditions, 
hence are always in competition. A knowledge of factors which 
favor the reaction desired is of practical as well as theoretical 
importance . 

A. Concentration of base : 

The rate of E x , like S ni , is unaffected by changes in the con- 
centration of base. The proportion of olefin to substitution 
product is thus unaffected, (l) 

The rate of E 2 reactions is increased greatly with increasing 
concentration of base. The olefin proportion is unchanged because 
the rate of Sn g reactions increases in a parallel manner. 

In mild alkali, both Ei and E 2 processes proceed simultaneous- 
ly. This situation caused many disagreements in the older liter- 
ature over the ease of elimination in secondary and tertiary alkyl 
halides. The diagram below explains how differing results are 






t-alkyl halides 

sec-alky 1 halides 


P. Polarity of Solvent: 

Solvent effects arise from the difference between the 
solvation energy of the transition state end the initial state. 



Solvation results from the attraction of the dipole charges of a 
polar solvent for partial or unit charges on the solute. If the 
magnitude of charge on the solute is decreased or spread over a 
larger volume, solvation is decreased. To decrease solvation, 
energy must be supplied because the net effect is that the forces 
holding solvent to solute have been overcome. To decrease solvation 
of a more polar solvent (one held more firmly because of its 
greater dipole charges) more energy is required. Therefore, if the 
charge on a reactant is decreased or more widely dispersed in the 
transition state than in the initial state, the reaction will be 
hindered by a polar solvent.- Conversely, if the charge is in- 
creased or concentrated, a polar solvent will facilitate the re- 
action. The olefin proportion is affected by virtue of the fact 
that the dipole charges are spread over a 5 carbon system in the 
transition state of elimination, but only over a 3 carbon system 
in the transition state of substitution. Hence, if the charge is 
dispersed in progressing to the transition state, it is more widely 
dispersed in elimination than in substitution. The table below 
summarizes the predicted solvent effects on two common types of 
elimination reactions. Using the considerations stated above, 
similar predictions can be made of other types of elimination. 

Rxn Disposition of charges Effect of Effect of more solvent on ~ 

initial transition activation Reaction Olefin 
state state on charges rate proportion 

Sn 3 £=* + RX yJ._. k ..X^ ^ f small small 
E2 " rq> " Y%".H. . Or. C. .X" (dispersed 1 decrease decrease 
as (OH + RC1) ; 

Sn 3 ^+ R#> Yt.R..X^~ £ Z) /"large 

E2 " r-, " ©YT.H. .C.-.C..3T /reduced 1 decrease ? 
as (OhP + RNMe,) 3 l 

The reaction rates are affected as predicted above. A decrease in 
olefin proportion iwas also observed in the cases so predicted. In 
the case designated as questionable, no trend could be discerned 

C. Temperature : 

The proportion of olefin increases with temperature for both 
first and second order reactions. No adequate explanation has as 
yet been offered (4). 

D. Const itutional Factors: 

There are two empirical rules of elimination: 

l) Hofmann rule: Elimination in quaternary amine salts will 
yield the olefin with the least alkyl substitution on the (3 carbon, 

CH 3 CH(NMe 3 )CH 3 CH 3 E 3 CH 3 -CH : CH-CH 3 + CH 3 -CH 3 -CH: CH 3 

-+ 26fo 74/c 

f v 


2) Saytzeff rule: Alkyl halides dehydrohalogena te to yield 
the olefin with the most alkyl substitution on the p carbon. 

CH 3 CHBrGH 3 CH 3 E t or E 3 CH 3 -CH: CH-CH 3 + CH 3 -CH 3 -CH: CH 2 

-> 82?o ISfo 

A careful study (5) has revealed that the rules are more 
general than stated above. Their applicability is as follows: 

1) Only bimolecular eliminations of onium ions follow the 
Hofmann rule. 

2) Unimolecular onium ion eliminations and eliminations of 
all neutral molecules follow the Saytzeff rule. 

Interpretation of Hofrrtann type elimination: (6-) 

The proton is removed by collision with a basic ion. 

R ~CH ~ CH 3 - c NMe 3 
^ ir* © 

The positive nitrogen induces a partial positive charge on the 
carbon atoms of the chain, thus facilitating the hydrogen's re- 
moval. If, however, the grouo R is electron releasing and 
neutralizes the partial positive charge on the (3 carbon atom, the 
hydrogen is less easily removed. Consequently, the hydrogen 
attached to the |3 carbon atom with the least alkyl substitution 
is removed. 

Rationalization of the Saytzeff type elimination: (6) 

The weakening of the carbon- hydrogen bond in the Hofmann 
type elimination is the result of induction. Apparently, in the 
Saytzeff type elimination hyperconjugation is more important 
than induction. Olefin (a) below is more stabilized by hypercon- 
jugation than olefin (b). Quantum mechanics suggests that the 
transition state leading to (a) is also more stabilized than that 
leading to (b) . Consequently, the olefin formed will be the 
one with the greatest number of allyl hydrogen atoms. Here, a 
alkyl substitution as well as substitution must be taken into 
consideration, for all allyl hydrogen atoms can participate in 

f A 


H. . .Base 

(f. CH 3 -CH 3 -CH:C(CH 3 ) 3 

CH 3 -CF 3 -CH-C-C(CH 3 ) 3 — >• predominates 

* ^ (8 hypercon jugation 

Base7\ Br^ forms.) 

CH 3 CH 3 CHCH(CHg) 3 

n no \, H...Base 

aase^ CH 3 -CH:CH-CH(CH 3 ) 3 

CH 3 -CH^CH-CH(CH 3 ) 3 ~* (4 hyperconjugation 

• j. o i m s • / 



1. Cooper, Hughes, Ingold, and MacNulty, J. Chem. Soc . , 1948 , 

2. Hughes, Ingold, and MacNulty, J. Chem, Soc., 1940, 899-=912. 

3. Cooper, Dhar, Hughes, Ingold and MacNulty, J. Chem. Soc, 
1948, 2043-49. 

4. Cooper, Hughes, Ingold, Maw, and MacNulty, J. Chem. Soc, 
1948, 2049-54. 

5. Dhar, Hughes, Ingold, and Masterman, J. Chem. Soc, 1948 , 

Dhar, Hughes and Ingold, ibid., 1948 , 2058-65; 2065-72 
Hughes, Ingold, and Maw, ibid., 1948 , 2072-77 
Hughes, Ingold, Maw and Woolf, ibid., 1948 , 2077-83 
Hughes, Ingold and Woolf, ibid., 1948, 2084-90 
Hughes, Ingold and Mandour, ibid, 1948 , 2090-93 

6. Dhar, H U ghes, Ingold, Mandour, Maw and Woolf, ibid., 1948 

< • 



Reported by K. H. Takemura April 29, 1949 

Decomposition reactions of aromatic diazo- compounds in 
aqueous solutions have been explained largely by assuming that the 
diazo-compound decomposes into molecular nitrogen and free 
radicals. Recently Hodgson (2) has presented an electronic theory 
to explain some of these reactions without the use of this free 
radical hypothesis. 

Decomposition in the Presence of Mild Reducing Agents . 

1. Prior to the work mentioned above, Hodgson and co-workers 
(3) had potulated essentially the following series of reactions to 
explain the decomposition of a number of aryl diazo- compounds in 
sulfuric acid solution in the presence of cuprous hydroxide as a 
reduxing agent; 

I. Ar;N:::N: + e — » Ar:N::N»-* Ar« + :N:::N: 

II. Ar* + Ar* — > ArrAr 

• • • • .... 

III. Ar* + Ar:N::N* — > Ar:N::N:Ar 

IV. Ar- + H* -* Ar:H 

The course taken was found to depend upon the posit ivity of 
the carbon atom to which the diazo-group was attached. If this 
positivity was great, the odd electron of the free radical was 
so restrained that it could combine only with nascent hydrogen; 
e.g., diazotized 2-nitro-l-naphthylamine gave only the nitronaph- 
thalene on decomposition. With slightly less restraint, biaryl- 
formation occurred; and when the restraint was small or even re- 
versed, azo-formation occurred. Any one or all of these reactions 
took place to a greater or less extent depending upon the nature 
of the aryl diazo-compound and the conditions. 

2. Saunders and Waters (6) arrived at another series of re- 
actions from their work with diazo-compounds in aqueous solutions 
near the neutral point in the presence of ammoniacal cuprous oxide. 
Their mechanism differed primarily in the initial step in which a 
homolytic cleavage of the diazo-hydroxide to give free radicals 
with the liberation of nitrogen was Postulated (VI) . 

V. Ar:N:::N: ^=^ Ar:N::N:X ^ > Ar:N::N:OH 

X~ (X- any anion) 


VI. Ar:N::N:0H — > Ar* + ;N:::N: + • OH 

VII. Ar. + Ar:N:::N: + e — > Ar:N::N:Ar 



Saunders and Waters regarded the mechanism postulated by 
Hodgson et al as Improbable since they considered the driving force 
q.f the reaction to be the liberation of nitrogen gas through the 
homolysis of the covalent diazo-compound with the simultaneous 
formation of two free radicals. Hodgson (2) later pointed out that 
his mechanisms dealt with ionic diazonium compounds in acid solu- 
tion and not with covalent compounds in neutral solution. He then 
set forth an electronic theory to explain a number of decomposition 
reactions without the use of the homolytic cleavage exemplified by 
the free radical hypothesis. 

Electronic Theory versus The Free Radical Hypothesis . 

1 » Hrtho, Para-Substitution . According to the free radical 
hypothesis, the invariable p- and/or o-sub stitut ion in reactions of 
the G-omberg type has been explained by assuming that the free aryl' 
radical which is formed (VI above) is amphoteric in type; that is, 
the free radical may function either as a cationoid or as a 
anionoid reagent as the occasion demands (5) . Thus the reaction of 
diazotized aniline with nitrobenzene has been found to give a 33$ 
yield of 4-nitrobiphenyl; and the diazotized n- n it roan i line with 
nitrobenzene gave a 69^ yield of 4,4*-dinitrobiphenyl (l). 

In as much as nitrogen, :N:::N:, is evolved in the reaction, 
Hodgson considers it more reasonable to have the nitrogen in the 
triple-bonded state as is the case in the diazonium ion, with only 
one bond to break, rather than the double-bonded state with two 
bonds to break in the case of the covalent compound. 



Ar:N OH 






According to Hodgson's theory, the diazonium ion attacks the 
anionoid reactant (usually benzene; at an anionoid carbon causing 
its hydrogen to become cationoid and to attract the anion of the 
diazonium salt (usually hydroxide or acetate). Eauation IX. 


£ * — c ^ 



The hydrogen is then assumed to split off as a proton, with electron 
release, to form HOH, with the liberation of nitrogen and 
formation of the biaryl. 

2. Non-Formation of S.vm-Biaryls . This mechanism does not 
involve any free radicals and hence the formation of sym-biaryls 



does not occur. This point is another difficulty found in the 
free radical hypothesis, according to which the assumed free radicals 
react preferentially with relatively unreactive molecules, rather 
than with themselves to form sym-biaryl compounds. Hence, bi- 
phenyl is formed in the G-omberg reaction from diazotized aniline 
only in the presence of benzene. 

3. o,p-Activity of Njtrohydrocarbons . The anomalous behavior 
of nitrohydrocarbons at the o- and p-positions has been one of the 
main arguments in favor of the free radical hypothesis. 

















] N a + 

(similar forms 
for ortho) 


The diazonium 
bond as in IX 

j+ H £-• 


ion and 
to form 

the hydroxide ion then attacks the 
the biaryl compound. 



5 + 

— H 

Further Applications of the Electronic Theory . 

1. Azo-Formation . The formation of azo- compounds can be ex- 
plained "hy the attack of the diazo-compound in the form VIII (b). 
Whether an azo-compound x^rill form depends upon the anionoid 
character of the second reactant. IT this is sufficiently great, 
the diazonium ion can attack the anionoid carbon as shown in XI: 


:N::N: <- C<^ %£~R 

OH -> H 

Thus , the re 
and partial 

is complete azo-formation with 
azo-formation with biphenyl. 

phenol and {3-naphthol, 

The main driving force in the usual azo-coupling reactions is 
the formation of water or weakly- ionized acetic acid. Where the 
anionoid character of the second reactant is small, e.g. benzene, 



the driving force must be augmented by the liberation of nitrogen 
and biaryl formation occurs. Usually both reactions occur to vary- 
ing degrees depending upon this anionoid character of the second 

2. Decomposition of p-Nitrodi azonium Sulfate in Neut ral Solu- 
tion. (4). An aqueous sulfuric ac: 

nitroaniline, neutralized with cal 
deoompose yielded a solid product 
dinitrodiazoaminobenzene (ca. 40^) 
azophenol (ca. 10^); and a small a 
amine (f ) . The remainder of the s 
tar. A small amount of p-nltrophe 
residual liquid of the decompositi 
these products can be explained by 

id solution of diazotized p- 
cium carbonate, and alloiired to 
from which was isolated (d) p,p f - 
; (e) 4-nitro-2-p-nitrobenzene- 
mount of p,p'-dinitrodiphenyl- 
olid, ca. b0% f was an inseperable 
nol (g) was isolated from the 
on mixture. The formation of 
the equations below: 

3 N<^~ ~^N OH =^ 

3 N<^ ^>N=N-0H^0 3 N<^~ ~^> I ) IH 
(b) (c) 

(c) + H a 0-> ? N^ 



C^/f ~^C -> N 3 OH 

(a) I fa 

i-0— H b+ 


H s -> 2 N<X N ^N=NNH<X X>N0 3 
+ HONO (d) 



0») + (g) -* <? ^>N=N^V ^>N0 ; 

3 N 





+ HoO 




Reduction of the reaction mixture gave no benzidine which 
indicates that no sym-biaryl, p,p'~dinitrobiphenyl, was formed. 






Bachmann and Hoffman, "Organic Reactions", Vol. II, John Wiley 
and Sons, (1944), pp. 254; 856. ' ' 

Hodgson, J. Chem. Soc, 348, (1948). 

Hodgson, Leigh, and Turner, ibid., 744, (1942). 

Hodgson and Norris. ibid.. 87, (1949). 

Grieve and Hey, ibid., 1797, U934). 

Saunders and Waters, ibid., 1154, (1946). 


Reported by H. A. DeWald May 6, 1949 

Phenols may be oxidized by potassium persulfate in alkaline 
solution to give hydroquinone or catechol derivatives. If the p_- 
position to the phenolic group is free, then hydroquinone deriva- 
tives are produced; if the para position is occupied a derivative 
of catechol is formed although usually in much smaller yield. 

The reaction was first studied by Elbs (l) in the favorable 
case of o-nitrophenol (i) which gave nitrohydroquinone (III) in 
30-40$ yield, about half of the starting material being recovered. 
Early German work showed that the reaction proceeds via the 
intermediate formation of a hydroxyphenylpotassium sulfate (II) 
which is subsequently hydro lyzed in acid solution to the hydro- 
quinone . 


V \-N0. 



— > 





«<^\-N0 : 




A recent study of the oxidation of a number of simple phenols 
has established certain optimum reaction conditions (2) and may 
indicate certain structural requirements of the phenols in order to 
obtain respectable yields. In the para-oxidations, the yields 
are increased by (a) the presence of an electron attracting group 
and (b) by increasing ring substitution-particularly if the sub- 
stitution exerts an effect to make the position para to the 
hydroxy 1 group relatively rich in electron density. The ortho 
oxidation of p-substituted phenols gives catechol derivatives 
in very poor yield and tarry matter is simultaneously produced. 

The oxidation is effected quite simply by the slow addition of 
a saturated solution of potassium persulfate to a stirred solution 
of the phenolic compound dissolved in excess 10^ sodium hydroxide, 
kept at 20° or lower. The reaction mixture is allowed to stand 
overnight, acidifed to Congo red, and extracted with ether to re- 
move unreached starting material. The aqueous layer is then 
treated with excess HC1, heated for a short time and the dihydric 
phenol extracted with ether. 

Although the oxidation product is generally quite pure, a 
side reaction has been observed in several instances. (3) When 
2-hydroxy-5-methoxyacetophenone (iv) is oxidized with alkaline 
persulfate, the principal product is a biphenyl derivative (V) with 
only a small yield of the expected catechol (vi). 




CH 3 

CH, \) 



Synthetic Applications 



The reaction is sometimes a convenient method for the 
introduction of a para hydroxy 1 group into a phenolic compound 
having an oxidizable side chain. G-entisaldehyde (2, 5-dihydroxy- 
benzaldehyde) has been prepared by persulfate oxidation of 
salicyladehyde or m-hydroxybenzaldehyde (4). 

The phenyl potassium sulfate derivative is stable under 
alkaline conditions and may be alkylated, then hydrolyzed in acid 
solution to give a hydroquinone monoalkyl ether of known orienta- 
tion (2). 

OH fH 

*rVCH a f<^\-CH 3 (CH s ) 2 S0 4 


S0 3 OK 

Of particular interest is the application of the reaction by 
Seshadri and coworkers to the synthesis of many flavone and 
favanol derivatives (5). 


5 II 






By alkaline persulfate oxidation of various phenolic derivatives 
and the use of known flavone condensation reactions, 5,8-; 5, 7-j 
5,6,?-; 5,7,8-; and 5,6,7,8-hydroxy flavones and flavanols have 
been synthesized in a fairly direct manner. The general procedure 
may be illustrated by the synthesis of a 5,6,7,8 tetrahydroxy- 
f lavone. 





(CH 3 ) 2 S0 4 

K 3 C0 3 

K3S 2 S 
CHaO^^-OH 1. -"►• 

— \j— uH 3 2. — b. 

H 3° 


CH 3 0,<Z \,-OH 



vera trie 

anhydride 0H 3 OwV 

Na-veratrate KO 



OCR 3 
<^>-0 C H 3 




/^V \ — <^ ^V-oM 





(CH 3 ) 2 S0 4 
* K 2 C0 3 

1. K 2 S 2 s CK 3 


2. H 3 0' CH3O 

40^ yield OH 


H 3 


Bibliograph y 

1. Elbs, J. prakt. Chem., 48, 179 (1893). 
S. Baker, J. Chem. Soc, 1948 , 2303. 

3. Baker, J, Chem. Soc, 1939 , 1922. 

4. Neubauer and Flatow, Z. physiol. Chem., _52, 380 (1907), 
Hodgson and Beard, J. Chem. Soc, 1927 , 2339. 

5. Seshadri et al, Proc. Indian Acad. Sci., 25A , 262 (1946); 
27 A , 220 C1948); 28A, 1 (1948). 




Reported by Edward F. Riener 

May 6, 1949 

The reaction 

+ CH-C 
I V) 



■ H CH 3 -C 

was discovered by Dr. Joseph Binapfl (1,2,3,4,5) of Germany before 
1935. The conditions used were temperatures of the order of 300 
and pressures of about 400 psi. Reaction times were short, of the 
order of 5-30 minutes under the cited conditions. 

It was claimed that any a, (3- unsaturated aliphatic acid was 
operable, such as itaconic or citraconic anhydrides, maleic acid, 
fumaric acid, mesaconic acid or citric acid (which dehydrates on 
heating). Binapfl described the use of iodine, sulfur, or copper 
bronze as catalysts, but his evidence does not show that the 
catalysts were at all necessary. 

An aliphatic side chain with an alpha hydrogen was necessary 
for reaction under these conditions; neither benzene nor naphthalene 
reacted. It was found that the reactivity of the side group was in 
the following increasing order: 

-CH 3 -CH 2 - -CH 

When the aromatic nucleus contained more than one side chain, as 
in diethyl benzene or triisopropyl benzene, only one group reacted. 

Clar (6) reported that when fluorene 
gently heated for forty hours at 210°C, 
per cent yield of 
male ic 

and maleic anhydride were 
there was obtained a 28 
9-f luorenylsuccinic anhydride. Acenaphthene and 
anhydride gave 23 per cent yield of acenaphthenylsuccinic 


Beavers (7) recognized the 
investigated it further. He 
ually a free radical reaction 
by peroxide catalysts. Using 
covered that the reaction cou 
conditions. The addition of 
to a hydrocarbon solution of 

rise to good yields of 

a. a 

possibilities of this reaction and 
postulated that the reaction was act- 
and, therefore, should be catalyzed 
various peroxide catalysts, he dis- 
ld be run under comparatively simple 
a hydrocarbon solution of the catalyst 
maleic anhydride at about 110 C gives 
dialkyl benzylsuccinic anhydrides. 

Below is a list of hydrocarbons which react with maleic anhy- 
dride and respective yields: 








Tetramethylbenzene (mixed isomers) 


m- Xylene 



Tri isopropylbenzene 


23. b% 
25 . 9% 
15. 4# 

Under the same conditions, the expected product was not ob- 
tained in significant amounts from any of the following donors: 

Toluene (8) 
Ethyl acetoactate 
Diethyl malonate 
2r Me thy lthiophe ne 
o_- Nit ro toluene 

With isopropylbenzene as the donor in all capes, the following 
potential acceptors did not give the expected products: 

Chloromaleic anhydride 

Mesityl oxide 

Dibutyl maleate 

n-Butyl crotonate 

Crotonic acid 

Qui none 

n-Butyl vinyl ether 


Fur an 

The mechanism of the reaction is a free radical chain reaction 
since only 0.04 mole of peroxide per mole of maleic anhydride is 
sufficient to catalyze it. This mechanism is illustrated as 



\co| 3 o. 



— » (*• + CO 2 + other products 

■C-H -> / X\,-C« + $ H 

R» R' 

1 // 



+ l\ )P 

oh- a 

xN o 





S-C— CH-C* 
R' 1 > 



St | 


This mechanism formulates a cycle by which one free radical 
from the peroxide may cause the formation of many molecules of the 
product. The initiating step, to give (A) , has been demonstrated 
by Kharasch. (9) 

Notable is the fact that when isopropylbenzene was treated 
with an unsaturated carbonyl compound under the conditions stipu* 
lated before, symmetrical diphenyltetramethylethane was obtained 
in low yield. 

Marvel _et al, (lO) found that when dimethyl maleate was 
treated with peroxide in the presence of dioxan, dimethyl 
dioxanylsuccinate was obtained as a by-product. The mechanism of 
the formation of this compound is the same as that between alkyl 
benzenes and maleic anhydride. 







G-erman Patent 
G-erman Patent 
G-erman Patent 
I.G-. Reports, 
Production of 
Microfilm PB 


No. 550,706. 
No. 607,380. 
No. 623,338. 

Microfilm Reel-2(tJerdingen) : 
Triisopropylbenz ene " . 
17657 (0-60): Frame Nos. 1691-1698. 

"The Commercial 

Chem. Abstracts, 41, 6553 (1947). 

Work done by Dr. Ellington Beavers at Rohm and Haas Laboratories 

Philadelphia, Penn. Patent applied for. 

Journal of the American Oil Chemists Society, .45, 251, (1948) 

It has been found that using a large amount of benzoyl peroxide 

(5 g. beroxide to 10 g. maleic anhydride) a z&f- crude yield of 

benzylsuccinic acid is obtained. 

Kharasch, McBay and Urry, J. Org. Chem., 10, 401, 1945. 

Marvel et al, J. Am. Chem. Soc . , _69, 52, 1947. 





Reported by Edward F. Els lager 

May 15, 1949 


Compounds containing the ethylene sulfide ring (thiirane), by 
analogy to those composed of the ethylene oxide ring (oxirane), 
would be expected to be very reactive and should undergo numerous 
transformations involving the degradation of this ring. 


A* Addition of sulfur to olef inic bonds 

Several investigators (1,2) have reoorted the formation of 
alkene sulfides by the addition of sulfur across olef inic double 
bonds. Complex mixtures were formed, and the yields of olefinic 
sulfides were very low. This procedure needs further investigation, 

B, From 1, 2-dithiocyano- and 1-chloro- 2- thiocyanoe thanes 

The action of sodium sulfide on l-chloro-2-thiocyano- and 1,2- 
dithiocyanoe thanes in aqueous solution yields ethylene sulfides 
in low yields (3,4,5), This method was one of the earlier methods 
and is not extensively used at the present time. 

■On + Na 3 S H 3 X C- -CfC + NaSCN + HC1 


C. From analogous epoxides (6,7,8,9,10,11) 

Ethylene oxide and its derivatives, when treated with aqueous 
potassium thiocyanate or thiourea, are converted to the correspond- 
ing ethylene sulfide derivatives. The yields are very good, and 
this is by far the baafc method for the preparation of alkyl olefin 





CH 3 -CH CH 2 

N S ' 

+ KOCN 81% 

It was found that the action of two moles of metallic xanthates 
ethylene oxide yielded 97 per cent of ethylene trithiocarbonate 
by means of the intermediate (il) (10), 









'*. -. 


>? ■;■;.; .-V 
T ; M f ■ ■ 



D* Dehydrohalogena tion of an g-chloro thiol 


the presence of an alkaline reagent, o-chlorothiols are 
dehydrohalogenated to ethylene sulfides (12) . According to the 
patent, 50 to 90^ yields of ethylene sulfide are obtained, and this 
procedure may be commercially important. 

NaHC0 3 
— ► 


CH ; 



E. From G-rignard reagents 

Tetraarylethylene sulfides are produced by the action of 
a Grignard reagent on a diarylthioketone (13,14). 

2 Ar 3 0S + 2 Ar'MgX 

Ar 2 C CAr 3 + Ar'Ar 1 + MgX s + MgS 

This reaction may be regarded as the bimolecular reduction of 
thiocarbonyl compounds by G-rignard reagents, and the reaction is 
very general* This reduction is also effected with magnesious 
iodide (15)* 

F. From diary Id iazome thanes 

Another synthesis of tetraarylethylene sulfides involves 
the reaction between a diaryl diazomethane and a diaryl thioketone. 
It i s believed that an unstable 4,4,5,5-tetraaryl-4,5-dihydro- 
1,2,3-thiadiazole is first formed; it subsequently loses nitrogen 
to form the olefin sulfide (16,17,18,19). In contrast to the above 
method, this scheme is used when one wishes to make Ar and Ar 1 

Ar 2 CN a + Ar'-CS 






• N 



Ar a C/ 


3 l 

G-. From substituted oxadiazoles 

2,2, 5,5-Tetraaryl-2,5-dihydro-l,3,4-oxadiazoles when 
treated with hydrogen sulfide in ethanolic solution yield the 
corresponding thiadiazoles, which decompose into nitrogen and 
the ethylene sulfide derivative (20). 

Ar 3 C0 + NH 2 0H-*Ar 3 C=N-0H K 4 Fe(CN) 6 N-C-Ar 3 H 3 S 

-* I > -* 

NaOH N-C-Ar 2 EtOH 

N-C-Ar 2 

II > 

N-C-Ar 3 

Ar 3 C 

l> + N 3 
Ar 3 C x 

This synthesis is especially useful in those instances where the 
starting diarylthioketone is not readily available. 

i i 

i ■ r. 



III. REACTIONS - Reactions C, D, I, J and K appear to be the most 


A. CH S CH 2 . + HN0 3 -» H0 3 S-CH 3 COOH + H0 3 SCH 3 CH 3 SCH 3 COOB (Bl) 

B. « HC1 

-* amorphous polymer (21) 

C. M 3 Pts HC1 

-> HSCH 3 CH 3 C1 + HSCH s OH a SCH s CH s Cl (21 ) 

D. » NaHS0 3 + 

-» HSCH 3 CH 3 S0 3 Na H acid (22) 

Bz. 50 — » 

E. " RC0C1 

-► 01CH 3 CH 3 SC0R (23,24) 


F. " + H 3 S -> SHCK 3 CH 3 SH + S(CH 2 CH 2 SH) s (ll) 

49 fo 16$ 

G. " + CsHnSNa-* O s H xl SCH 3 CH s SH + CgR - ! 1 SCH 2 CHgSCH^CH 2 SH 

75^ 2S% 

H. " I s 

-> ICH 2 CH 3 S«-SCH 2 CH 3 I (25) 

weak alk. 


I. " *hNH 3 PhNHCH 2 CH 3 SH CH 3 CH 2 PhN(CH 3 CH 2 SH) 2 

100 -> 


J. " R 3 NH R 3 NCH e OH 3 SH CH 2 CH 2 R 3 NCH 3 CH 3 SCH 3 CH 3 SH 

-> vs 7 

100° -> 

<?Hg Xfl. CHg CHg 

K. CH 3 -C CH 2 RSH CH 3 -C-CH 3 SH + CH 3 -C-CH 2 SR 

N S' -* SR SH 


The reaction products are capable of reacting further as follows: 

R 3 C-CH 3 SH + R' 3 C- — 0H 3 -+ R 3 C-CH 3 SC(R , 3 ) CH 3 SH (27,8) 
SR x S y SR 


'■■> f 

.1 : .. 






L. With alcohols - The same authors (8) reported the condensation 
of primary alcohols with alkene sulfides in the presence of boron 
fluoride catalysts; the products were £-alkoxy mercaptans. 

M. Vfith benzene and AlCl a - Propylene sulfide reacts with benzene 
in the presence of A1C1 3 to yield a polymer, which, when heated, 
yields 1, 2-diphenyl propane (24). 

N. With acetic acid - Acetic acid gives only a lb% yield of 
primary addition product when reacted with propylene sulfide. 
This product is a mixture of the isomers: 

CH 3 -gHCH 3 0C0CK 3 and HSCH 3 gHOCOCH 3 (24) 

SH CH 3 

0. Synthesis of 2- iminoth iophane s -* Ethyl cyanoacetate and olefin 
sulfides were found to react in the presence of sodium ethoxide to 
give 2-iminothiophanes (9). 

^ SH 

CH 3 C=N 

\CH 3 CH 3 C0 3 Et 

CH 3 C=N 


0H 3 — -CH-C0 3 Et 

chC x c=nh ch 3 c-nh 3 

II I II 2Z% 

CK 3 CH-C0 2 Et CH 3 C-C0 3 Et 

P. Polymerization - Polymeriaat ion was one of the first reactions 
€ the olefin sulfides to be noted, and although the polymeric 
sulfides have been described as amorphous solids, little work has 
been done to determine their structure. When treated with mineral 
acids or concentrated alkali, the polymerization proceeds with the 
liberation of much heat. Ethylene oxide polymerizes spontaneously 
even at 0°; by the addition of small amounts of aliphatic mercaptan, 
the polymerisation is inhibited. Cyclohexene sulfide can be stored 
in the refrigerator for several days without polymerizing, and iso-' 
butylene sulfide has been stored at room temperature for several 
months without any appreciable change (8). 

Tetraarylethylene sulfides when heated decompose to the corres- 
ponding olefin and sulfur, or ring close with the loss of KC1 
forming the corresponding benzothiophene derivative. 


The products resulting from the action of amines on olefin 
sulfides are useful in the preparation of dyes, vulcunization 
accelerators, and textile assistants* Ethylene sulfides are found 
to react with wool fiber forming a polymer which greatly decreases 
the shrinkage characteristics of the wool. The condensation of 
ethylene sulfide with cyanamide in water produces a substance 
which has good insect icidal properties. 


[•■ :-■ . 

f .. 

, • i: '• 4 1 < 

,*• i . ■ ■ 1 -. 



i. ■: 

•; • 




1. We st lake, H. E. Jr., J. Am. Chem . Soc, J38, 748 (1946). 

2. Jones, S. 0., Ibid, 60, 2452 (1938). 

3. Calingaert, G-., Bull. soc. chim. de Belg. 31, 109 (1922). 

4. Mousseron, M., Compt. rend. 215, 201 (1942*77 

5. Youtz and Perkins, J. Am. Chem." Soc. 51, 3508 (1929). 

6. Fr. Pat. 797,621; Chem. Abstr. 30, 7122 (1936). 

7. Barr, T., and Speakman, J. B., J. Soc. Dyers and Colorists 60, 
238 (1944). — ' 

8. Snyder, H. R., Stewart, J. M. , and Ziegler, J. B., J. Am. Chem. 
Soc, 267g (1947). 

9. Snyder, H. R. , and Alexander, W. , Ibid, 70, 217 (1948). 

10. Culvenor, C. C. J., J. Chem. Soc, 1050Tl946)< 

11. Meade, E. M. , Ibid. 1894 (l948>. 

12. Chem. Abstr. 34, 2395, 2863, 6302 (1940) , 

13. Schonberg, A., Ber. 58B, 1793 (1925). 

14. Schdriberg, A., Ann. 454, 37 (1927). 

15. Schonberg, A., Ber. 60B, 2351 (1927). 

16. Standinger and Siegwant, Helv . Chim. acta. 3, 833 (1920) . 

17. Standinger and Siegwant, Ibid, 3, 840 (1920"7. 

18. Schfinberg, A., Ann. 483, 176 (1930). 

19. Schttnberg, A., Ber. £4, 1390 (1931). 

20. Schonberg, A., J. Chem. Soc, 1074 (1939). 

21. Delepine, M., Bull. soc. chim. 33, 703 (1923). 

22. Chem. Abstr. 35. 5909 (1941 ). 

23. Ibid, 35, 463~Ti94l) . 

24. Stewart, J. M. , A.C.S. Meeting, San Francisco, April 1, 1949, 

P. 69L. ' * 

25. Alexander W. . Thesis, Doctor of Philosophy, University of 
Illinois, (1946). 

26. Reppe, W. , and Nicolai, p., Chem. Abstr. 30, 6008 (1936). 

27. Ibid. 35, 5909 (1941), — 

28. Ziegler, J. B. Thesis, Doctor of Philosophy, University of 
Illinois (1946) . 

29. Delepine, M., Compt. rend. 171, 36 (1920) . 

30. Delepine, M., Ibid. 172, 158~Tl92l). 


f * 

* . 




Reported by Charles V, Fairbanks 

May 20, 1949 

The American "Yellow Jasmine", gelsemium sempervirens , has 
afforded the following crystalline alkaloids (2) : G-elsemine 

4 N 2 ) 

(^3oHs2°4N 3 ) , sempervirine (C lg Hi 6 N 3 ), and gelsemicine (C 20 H 2S u 4 r 
as well as other amorphous constitutents of unknown composition. 

By treating gelsemium root in a modified Sayre and Watson pro- 
cedure sempervirine and gelsemine were obtained in a ratio of 19 to 

Sempervirine absorbs three molecules of hydrogen over pallad- 
ium and five molecules of hydrogen over Adams catalyst. The former 
product could not be crystallized nor could any crystalline deriva- 
tives be isolated as the material rapidly resinified. The latter 
product could be crystallized but contained oxygen and rapidly 
absorbed more oxygen from the air. Sempervirine affords a quater- 
nary mono-methiodide uoon treatment with methyl iodide. No defin- 
able product could be obtained on degradation with ao x ueous alkali. 
No useful results were obtained by oxidation with permanganate, 
nitric acid or hydrogen peroxide - osmium tetroxide; by heating 
with palladium in air or oxygen, or by potash fusion (2). 

The free alkaloid crystallizes from chloroform in reddish 
brown needles, is slightly soluble in alcohol and water and is al- 
most insoluble in ether, benzene and petroleum ether. The hydro- 
chloride is readily soluble in water and alcohol and is precipitat- 
ed by nitric, tannic and picric acids. Yellow precipitates are 
obtained by treatment with potassium chromate, platinic chloride, 
sodium chloride and sodium nitrite (5). 

Sempervirine has an active hydrogen as shown by means of a 
Zerwitinoff determination. The N-methyl determination was 
negative. The ultraviolet absorption spectra of sempervirine shows 
a strong series of absorption bands; these bands are almost identi- 
cal in alkaline and neutral alcoholic solutions (4) . 

Sempervirine is isomeric with yobyrine (i). In attempting to 
relate the two it was found that upon heating sempervirine with 
selenium it was changed to yobyrine, as determined by mixed melt- 
ing points and ultraviolet absorption spectra. Sempervirine, when 
heated with Raney nickel in xylene solution, gave poor yields of 
tetrahydroisoyobyrine (II) as determined by mixed melting points 
and ultraviolet absorption spectra (4) . 








These experiments appeared to have clarified the ring 
structure of sempervirine; however, they still left the position 
of the double bonds undetermined. Upon considering the ultraviolet 
absorption spectra, which requires an extended chromophoric system 
conjugated with the aromatic system, Prelog proposed the structure 
(ill) as a possible formula for semperYirine. 

All N-unsubstituted indole derivatives are characterized by 
an intense sharp band at 2.9. Sempervirine shows no such band. 
When sempervirine methochloride is treated with selenium a new 
base, N-methylyobyrine, is obtained. Its ultraviolet spectrum 
is nearly identical with that of yobyrine and its infrared spectrum 
possesses no NH band. The base was identified by direct comparison 
with a synthetic sample. 

These considerations led Woodward (6) to propose a new 
structure (IV) for sempervirine. This structure implies an 
important contribution of the fully aromatic ionic structure (v). 
This view explains the color of the alkaloid and its high basic- 
ity (pK 10.6). 







M © 





The formation from sempervirine of a mole of methane in the 
Zerewitinoff determination can be attributed to the presence in 
(IVO V) of a virtual (substituted) Y-pioolinium system. 

Final proof of the structure of sempervirine was obtained 
through the synthesis of sempervirine mothosalts by an unambiguou 

In a model experiment, the lithium derivative of a-picoline 
was condensed with isopropoxymethylene cy clohexanone (Yl) , salts 
of the dehydroquinolizinium cation (VII) being readily obtained 
from the acid-treated reaction-mixture. 






\=CH-0CH(CH 9 ) 3 H 







In a similar reaction, the lithium derivative of N-methyl- 
harman (VIIl) led to the smooth synthesis of salts of the methyl- 
sempervirinium cation (IX). 




=CHOCH(GH 3 ) 3 + 





N v © 





Synthetic samples of sempervirine methopicrate and semper- 
virine methochloride showed no depression in melting point on 
admixing with the corresponding salts prepared from the natural 
sempervirine. Further corroboration for formula (IV) was obtained 
through the reproduction of the characteristic ultraviolet ab- 
sorption spectra. 


1. Auwers, Ber. 71, 2082 (1938), 

2. Forsyth, liarrian and Stevens, J. Chem. Soc, 1945 , 579. 

3. Johnson and Posvic, J. Am. Chem. Soc, 69, 1361 (1947) . 

4. Prelog, Helv, Chim. Acta., 31, 588 (19487. 

5. Sayre and Watson, J, Amer. Pharm. Assoc, 8, 708 (1919) 

6. Woodward and Witkop, J. Am. Chem. Soc, 71, 379 (1949). 

7. Woodward and Witkop, J. Am. Chem. Soc, .71, 379, (1949). 





Reported by Allen B. Simon May 20, 1949 

Importance of Lysergic acid . 

Lysergic acid is the most important fission product of the 
ergot alkaloids, and the only product common to all ergot alkaloids 
upon alkaline hydrolysis (2). The great difference in physiolog- 
ical activity between the almost inactive dextrorotatory and the 
active levo rotatory series of alkaloids clearly depends only on 
the lysergic acid moiety (5) and seems to be determined by a steric 
shift on an asymmetric center, as will be shown in this seminar. 

Ergot, the source of the alklaoids, is a fungus which grows on 
the rye plant. Eating of the infested plant causes a severe form 
of gangrene and, in pregnant women, abortion (3). When pharmaco- 
logically administered, ergot induces a prolonged, rhythmic con- 
traction of the puerperal uterus. 

Lysergic acid is also of chemical interest since the ergoline 
ring system, the basic tetracyclic ring structure of lysergic 
acid, represents the only known example of an indole derivative 
condensed in the 3,4 position to other nuclei (7) Diagram I, 

Previous Work . 

The basic ring structure of lysergic acid and of its isomer, 
isolysergic acid, was confirmed by the synthesis of dihydro-d,l- 
lysergic acid (8;. This, however, still left unanswered the 
question of the position of a non-aromatic double bond present 
outside of the indole nucleus. Ultraviolet absorption studies 
indicated that in both lysergic acid and isolysergic acid this 
double bond is conjugated with one of the double bonds in the 
indole nucleus. Jacobs (5) assumed that the isomerism between 
lysergic acid and isolysergic acid is brought about by a shift in 
the position of this double bond. Positions 4-5 and 5-10 would 
place the double bond equidistant from the NCH 3 group (position 6). 
These positions were excluded when the difference in basicity of 
the tertiary amine groups in the two compounds were ascribed to 
difference in distance from the non-aromatic double bond to the 
NCH 3 . The basic group in lysergic acid is weaker than that in 
isolysergic acid (2); by analogy with the findings of an earlier 
study on dissociation constants (4), it was concluded that the 
double bond in isolysergic acid is 9-10, the farther position, 
while the double bond in lysergic acid is 5-10, the nearer position 
(2). When further study showed that vinyl tertiary amines are 
more basic than unsaturated tertiary amines not in the vinyl 
position (l), it was proposed that the positions of the double 
bonds are reversed. 

J' I. • • 

r, ■ i • ■> .. 

> . . I 't 

ii .'»' 


Diagram I 




P0 3 H 
CH— CH. 

Jo * e>-CH, 
^ B C, 

H *C0 3 H 

HsCgt-^io 'M 

lCH : 


5> 10 

C0 3 H 

0_CH V CH 3 

Lysergic Acid 

Isolysergic Acid 

Lysergic Acid (A ^ >10 
Isolysergic Acid ( S± ) 


Isomerism Explained as Stereoisomerism (6) 
Formation of the Lactam 

Stoll, Hofmann, and Troxler began their investigation of the 
position of the non-aromatic double bond by the removal of the C8 
asymmetry. This was unexpectedly accomplished when treatment with 
acetic anhydride yielded the lactam. Both lysergic and iso- 
lysergic acids yielded the identical lactam which was opitcally 

From this experiment the following conclusions were drax^n. 
The double bond in lysergic acid and isolysergic acid is 9-10. 
The large displacement of the ultra-violet absorption spectrum of 
the lactam towards the region of the long wave lengths indicates 
that the new double bond in 7-8 is conjugated with those already 
present. That could only be possible if the non-aromatic double 
bond is 9-10. 

Lysergic acid and 
arrangement about C8 

isolysergic acid differ only by the steric 
since removal of C8 asymmetry produces the 

identical compound from both acids. Therefore, they are diastereo- 
isomers and not structural isomers as hypothesized by Jacobs. 

Diagram 2 

"C0 3 H 

*CH CH 3 . 

/a i 6 N-CH. 

(AcO) 3 
— > 

7,5 H a [, 


Hofmann Degradation 

The removal of asymmetry from 0^ was attempted by the Hofmann 
degradation method, using stable derivatives of the lysergic acids 
as starting reagents. The products from both reagents proved to 
be identical, optically active, and with C5 still asymmetric* 
The ring had broken between positions 6-7. 

fcj 'W 


The degradation was continued to give a product which was no 
longer optically active, confirming the assumption that C5 and C8 
are the only asymmetric atoms in lysergic and isolysergic acids. 

Number of Isomers 

If the double bond is fixed in the 9-10 position in both acids 
then upon racemization two racemates should be formed. Two race- 
mat I'a of the lysergic acids and two of the hydrazides are known. 
If the formulas of Jacobs are correct, then one racemate of lysergic 
acid and two racemates of isolysergic acid are possible. Stoll, 
xiofmann, and Troxler have made repeated attempts to discover a 
third racemate but have always been unsuccessful. 

The saturation of the double bond, 9-10, of lysergic acid with 
hydrogen causes the formation of a new center of asymmetry at C5 
with the possibility of two stereoisomers. Up until now, however, 
only one isomer has been found. With isolysergio acid the satura- 
tion of the double bond again causes the formation of a new center 
of asymmetry at C5 with the possibility of two isomers. Here both 
isomers have been found. Under definite conditions one of the 
isomeric dihydroisolysergic acids can be irreversibly converted 
into the dihydro lysergic acid. Therefore, the two dihydroacids 
differ only in the steric arrangement about 08. The steric arrange^- 
ment about the newly formed center of asymmetry, CIO, is identical. 

Importance of 9-10 Unsaturation for Isomerism 

Lysergic acid and isolysergic acid are easily converted one into 
the other. If the carboxyl group is replaced, however, the inter- 
conversion can no longer be brought about. Ester derivatives do 
isomerize and the alkyl portion of the ester influences the speed 
of isomerization. If the 9-10 double bond is saturated, isomerism 
can no longer occur except that the one isomer of dihydroisolysergic 
acid can irreversibly change to dihydrolysergic acid. 

Although most of these observations had previously been used as 
evidence for the hypothesis that the isomerism is due to a shifting 
of the double bon& (5), Stoll considers them readily explainable 
on the basis of nls- theory. The presence of the 9-10 double bond 
enhances the enolization of the carbonyl portion of the carboxyl 
group; it permits the formation of a completely conjugated do* le 
bond system from the enol double bond to the double bond system of 
indole. Since the enol form of C8 is symmetrical, it permits the 
formation of equal amounts of the enantiomorphs upon tautomerization 
back to the keto form. Therefore, it could be expected that sat- 
uration would hinder isomerization. 


1. Adams and Mahan, J. Am. Chem. Soc, 64, 2588 (1942). 

2. Craig and others, J. Biol. Chem. 125"7"289 (1938). 

3. Coodman and G-ilman, The Pharmacological Basis of Therapeutics , 
The Macmillan Co., New York. 1941, , . 

4. Hixon and Johns, J, Am. Chem. Soc, 49, 1786 (1927). 

5. Jacobs and Craig, J. Biol. Chem, 115, 227 (1936). 

6. Stoll, Hofmann, and Troxler, HelvT~Chim. Acta 32, 506,(1949). 

7. Uhle, J. Am. Chem. Soc. 71, 761 (1949). „ — 

8. Uhle and Jacobs, J. Org. Them. ■ 10, 76 (1945).