(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Organic seminar abstracts"

L I B R.AR.Y 

OF THE 

UN IVERSITY 

or ILLI NOIS 

54T 
»938/39 



fk 



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

University of Illinois Library 



L161— II41 



Digitized by the Internet Archive 

in 2012 with funding from 

University of Illinois Urbana-Champaign 



http://archive.org/details/organicsemi38392univ 



}'^3?f^i SEMINAR TOPICS 

^ Chemistry 135b II Semester 1938-39 



The Periodic Arrangement of the Amino Acids in a Protein 
Molecule Madison Hunt, February 8 

Enzymatic Synthesis of Peptide Bonds 

R. Mozingo, February 8 

Enolization and Acidity T. A. G-eissman and E. F. Rogers 

February 15 

The SyntJriesis of Chroraane Derivatives 

H. M. Teeter, February 22 

Transformations of the Steroid Crroup 

E. H. Riddle, February 22 

Nitrones B. R. Baker, March 1 

The Allyl Rearrangement A. H. Land, March 1 

Toad Poisons E. C. Horning, March 8 



S 



V 



The Formation of Sulfur-containing Rings 

S. C. Kelton, March 8 



\ Rearrangements of Aryl Salicylates and Compounds of Similar 

\ Constitution S. L, Scott, March 15 

\ 

"^ The Structure of Rottlerin R. 0. Sauer, March 15 

^ Phenolic Resins E. H. Dobratz, March 22 

The "Oxidizing Action" of Alkalies 
V J. F. Kaplan, March 22 

4 

\ Vitamin Bq E. Welch, March 29 

The Relationship between Fluorescence and Chemical Constitution 

M. H. Gold, March 29 

IS 

Organic Compounds Containing N 

L. C. Behr, April 5 

Natural Azulenes J. J. Denton, April 5 

Steric Hindrance in Substituted Benzaldehydes 

R. V. Lindsey, Jr., April 12 

Relation of Basicity and Solubility to the Toxicity of Amines 

J. H. McCracken, April 12 



The Use of Metallic Complexes in the Determination of 

Configuration ' W. H. Sharkey, April 19 

Organic Compounds Containing Selenium and Tellurium 

D. E. Burney, April 19 

The Chemist rj' and Structure of Lignin 

L. A. Patterson, April 26 

Aliphatic Diazo Compounds and Their Reactions with Carbonyl 
Derivatives W. H, Rieger, April 36 

Some Recent Advances in Chemiluminescence 

R. B. Moffett, May 3 

Action of Organic Acid Chlorides upon Aliphatic Ethylenic 
Hydrocarbons in the Presence of Stannic Chloride 

H. E. Conde, May 3 

Peganine •> Vasicine J, Harkema, May 10 

The Tautomerism of p-Hydroxyazo Compounds 

S. J. Circle, May 17 

/5^-Keto Bases J. H. Burckhalter, May 17 

Isatogens and iso -Isatogens F, C. Dietz, May 24 

Mechanism of Ketone Formation from Carboxylic Acids 

J. F. Mcpherson, May 24 

Molecular Dissymmetry due to Restricted Rotation in the 
Benzene Series: An Optically Active Ethylenic 
Derivative F, Richter (not reported) 

The Formation of Quaternary Ammonium Salts from Dihalogeno- 
paraffins, etc., in Aqueous Acetone Solution 

R. W, Kell (not reported) 

Y -Substitution in the Resorcinol Nucleus 

J. W. Shackleton (not reported) 

Raman Spectra in Organic Chemistry 

A. W. Anderson (not reported) 



THE PERIODIC ARRANGEMENT OF THE AMINO 
ACIDS IN A PROTEIN MOLECULE 

Bergmann and Niemann -- The Rockefeller Institute 
for Medical Research, New York 

The work of Fischer and Hoffmeister has estahlished the peptide 
hond as the basic linkage between the various amino acids in the 
protein molecule. No conclusions could be drawn, however, regarding 
the arrangement of these various amino acid residues in the protein 
molecule. 

Bergmann and his associates havf; undertaken a careful study of 
the amino acid content of various proteins with rather striking 
results. Gelatin was the first protein analyzed by Bergmann. The 
results of the analysis for glycine, proline, and hydroxyproline are 
shown in the table below. 

Gram Molecules Ratio 
Found Calcd. 

0.34 0.36 6.0 

.17 .18 3.0 

.11 .12 2.0 

The data indicates that in a gelatin molecule one-third of the 
amino acids are glycine, one-sixth are proline, and one-ninth are 
hydroxyproline. These amino acids can be arranged in periodic order 
so that every third amino acid is glycine (G) , every sixth is proline, 
every ninth is hydroxyproline; the letter P represents either proline 
or hydroxyproline in the structures postulated below. 

1. -G-P-X-G-X-X-G-P-X-G-X-X- 

2, -G-X-P-G-X-X-G-X-P-G-X-X- 

Grassmann and Riederle have isolated lysylprolylglycine from the 
hydrolysis of gelatin. This would tend to support the occurence of a 
prolylglycine (P-G) unit in gelatin. This unit is found in structure 
2 but not in 2» 

The data on analysis of egg albumin is shown below. 



Amino acid 




Per cent 


Molecular 
Weight 


Glycine 




25.5 


75 


1,-Proline 




19.7 


115 


_l-Hydroxyprol 


ine 


14.4 


131 



Amino acid 


Per cent 


Molecular 


Gram Mo 


lecules 


Ratio 


Fr 


equency 






Weight 


Found 


Calcd. 




of 


Occur- 
rence 


Glutamic acid 


14 


147 


0.095 


0.101 


36 




8 


Asparti,c acid 


6.1 


133 


.045 


.045 


16 




18 


Methionine 


5.2 


149 


.034 


,033 


12 




24 


Lysine 


5.0 


146 


.034 


.033 


12 




24 


Arginine 


5.6 


174 


.032 


.033 


12 




24 


Tyrosine 


4.2 


181 


.023 


.022 


8 




36 


Histidine 


1.5 


155 


.009 


.011 


4 




72 


Cysteine 


1,3 


121 


.010 


.011 


4 




72 


















o,e: 






biiu-y^i. 



1^1 






II iYt^C/'fr 









■..■■15 or;^-ip.l . • b^,■^^ns^'rtB ocf .■:.;•;> :.;;J:o.n oain^> o?;?)rh- ' .^^itii-tqicXa-r: " • 
.:.ii3-q HI iy..x/-.; v:i9T:? ,(0; !J:t.::,v.:Tj ^j /;i,j^ oviirut biiri^ V-r^.y., ;t.. 



- y ~X ^ 41 - 'i - T ^ p. ^ y • V • ■ ,.■,-■; ,. ' '. 









T? y ,;•; 









i xTi.^■ 






nii :'^;xv .f!:.«f 



'r''': ."^s-.-'^-'T-. 



f.X: 



is 






i^'::A. ■tart'': 


T-;i:v;c^iaM 


) bLiiso'^ 


td^xow 


• ;;0 ». 


"^i»i; 


iioT' 


§^x 


'-^^:. 


^'ii^I 


^'.^a.. 


a^'i; 


'"•■.';'r'. 


^-Tr. 




':gr. 


^^O^i, 


«3i: 



TO*i &•i■0^:^ O.t^/IX.; 



■>i 


h 


ir 


1^ oiai^j;/;"" 




.-V 


•-'^' 




0..a 






Oiiiav. . 


^,.^r 






3Tfir; :'--'.; 


a:.^ 






orui; 


v't,-!'. 






onibiiihlii 



i.v.' 



-2- 

Good analytical methods for valine, leucine, and isoleucine are 
not available; however, good methods for glycine, alanine, proline, 
hydroxyproline, tryptophane, tyrosine, cystine, methionine, histidine, 
aspartic acid, glutamic acid and arginine are available. 

The average molecular weight of the amino acids formed on 
hydrolysis of egg albumin is 142; therefore the average weight of an 
amino acid residue when combined in the protein molecule is 18 less 
than this, or 124. Thus 100 grams of egg albumin should yield on 
hydrolysis 0.806 gram molecules of the hypothetical average amino 
acid. V/ith this value the frequencies of occurrence in the last 
column were calculated. Since there are four histidine and four 
cystine molecules per molecule of protein it follows that the minimum 
number of amino acid residues in a protein molecule is 288. This 
value multiplied by the average molecular weight of the residue gives 
a value of 55,700 for the total molecular weight. This value is in 
good agreement with that determined by other methods. 

A similar calculation from the analysis of blood hemoglobin gave 
a minimum of 576 amino acid units and a molecular weight of 69,300. 
This value is also in good agreement with the value obtained by other 
methods. 

From a consideration of the analytical data on a variety of 
proteins Bergmann concluded that the total number of amino acids in a 
protein molecule could be represented by the arithmetic expression 
2^x3™ where m and n are small whole numbers. Likewise the number of 
units of an individual amino acid and its frequency of accurrence in 
the chain could be represented by the same expression where m and n 
are either zero or small whole numbers. 

Silk fibroin composed largely of glycine and alanine yielded 
very interesting results when subjected to analysis. 



Amino acid 



Glycine 

Alanine 

Tyrosine 

Arginine 

Lysine 

Histidine 



er cent 


Molecular 




Weight 


43.8 


75 


26.4 


89 


13.2 


181 


0.95 


174 


0,25 


146 


0.07 


155 



Gram Molecules Ratio Frequency 



Found Calcd. 
0,584 0.584 1296 



.296 
.072 
.005 
.001 
.0004 



.073 
.005 
.001 
.0004 



648 
162 

12 

4 

1 



of Occur- 
rence 
2 
4 

16 

216 

648 

259 2 



The average molecular weight of the amino acids is 102 which 
gives 84 for the average weight of the amino acid residue in the 
protein molecule. Therefore on hydrolysis 100 grams of protein 
should yield 1.190 gram molecules of the hypothetical average amino 
acid. Thus the molecular weight is 2592 x 84, or 217,500. 

Partial hydrolysis of silk fibroin has yielded glycylalanine 
(G-a), glycyl tyrosine (G-T), alanylglycine (a-G) and alanylglycyl- 
tyrosine (a-G-T). It seems unlikely that there are two glycine 
molecules or two alanine molecules linked together since no alanyl- 
alanine or glycylglycine was isolated. Therefore, each glycine unit 
must be separated from every other glycine unit by one other amino 
acid. Likewise each alanine unit must be separated from each other 



,::*:r/:rolrftGffr ^-rcxc^-aY' ,arfiaoi\;^ «^j:-vsrfvjv- =i^O^ , .?uii:o•V;|Y:^;;:l^v•| 



1 L'O 






vh- f^i.if->i,^?-: ciiJ Vc cl-;-f\>lo\v t.--.i:j;f. ionr ;7:5;rf -^-rr ^'r(.t vc! b^irqi^rXum -uUr^ 












Y+=>ni;::v b rfo s:tai> 1^,,^ 



iCLTs -Miit 'Jto ;toi Ji^^-i-ibir-'tOi/ ^^ iso^ct 



;i.£ e^r■:)f• onifri.6 1;- •to^^J;;/;- I.l'c; :;d' erfcf ^j:^:i;t BfjbiLrXo.iOC; aiii'm'^-'^'^ eni^ito^q 



.STDOmi'ff 3 Lon^v il.f^f'V^ '.'C '.J- 



; ^r .L •... 



■ai'ij;-!^; LriF ^niovl:^ '^io v 



•:'■! Tj;. .: 



t-, ,;, ■ 



•;^i. 



w :ilC1-il 






■J.. llrv 


I'i 


'^ --• r ■''"■■' 


;jf:XJ^ 


'(J 


"" cfifei 


ir^^;. 


f, 








Qb 


■- '■ '•-■ t 




rax 






:[Vr 


iOC, 




dri 


^QOu„ 




rir] f 



8 




€1 


1- 






S 




^X 


rJO 




■•^> 








vo 




•.J 



■} ■■ 4'f .:! 



I' II t \r'r ■ 



.::y.-: 


















^SiXyCO^Ivr:; ^U^^f^l:!; 



oqv^ ri'fj 1:0 SviXu'i&Xoin ^^y^fi^ OQi*l hX;>i^ i.^ u;<*/^ji 



■i^^iltii^lBlvsrv-jC;^ ii 






.r'^f » fSi"; 



J 4... I W / '.I i V 



'JO •"^Isv'ioifcvrf i;?4:j-t^!2'^' 



iS::'^^^/::,^- i&.a£- ^,5-Ai a^iio^lBX^cfi^Xir j-.'^'-^'* t^aixi^rxt Ix^-^^-i 



in:Dvf-s| :iri- Oa;^ ^i-i'i;^ 



•*■ -»■ 



^■•y 



X^'^y^iUiSJ si2;>e;v il 



4V-u)~a) ;v;^ 






"oss KOTl- X''^*.p'ii!'r?3e 



:::iO it-;;;:'^ vlnu ifr^fSiiXf; .(•';.;rfi7 -Sft i-v«r;^;f/li.y 



JJ..:" 



-3- 

alanine unit "by at least three other amino acids. Similar deductions 
may "be drawn regarding the other amino acids present. Hence a partial 
structure for silk fibroin may "be written as follows, where G = glycine, 
A = alanine, T = tyrosine, Ar= arginine, and X = other amino acids « 

G-A-G-T-G-A-G-Ar-G-A-Cr-X-G-A-G-X- 
( G-A-G-T-G-A-G-X-G-A-G-X-G-A-G-X^ 2 

G-A-G-T-G-A-G-X-G-A-G-X-G-A-G-Ar- 
( G-A-G-.T-G-A-G-X-G-A-G-X-G-A-G-X-4i 3 



Bihliography ; 

Bergmann, J. Biol, Chem. , 110 ^ 471 (l935). 
Grassmann and Riederle, Biochem. Z., 284, 177 (1936). 
Bergmann and Niemann, J. Biol. Chem., 118 ^ 301 (1937). 
Bergmann and Niemann, J. Biol. Chem., 122 . 577 (1938). 



Reported by Madison Hunt 
February 8, 1939 



:^iS-jijr:.!b 'r:;_;ii:ixL 



-- ..v. ;v-:x,Tii5 TonJo '-v-cri:' j-c.:. . : /• x^i" -^ 



.i;.' 






-'rA<-0'-A':!--V-0-,\-0-X-0-A-r-.?-0-A-i 



. .- ^ 












ENZYMATIC SYNTHESIS OE PEPTIDE BONDS 

Bergmann -- Rockefeller Institute for Medical 
Research, New York 

The discovery of quantitative rules which govern the "biological 
synthesis of a protein make the process of synthesis of an individual 
protein molecule appear to he a process involving a high specificity 
on the part of the enzyme. Not only must the enzyme be capable of 
synthesizing a peptide bond, but it must also be capable of selecting 
precisely the structural unit to be used in the process to give the 
special pattern of the individual protein. The experiments reported 
here are devoted to a study of this specificity in enzymes which 
synthesize peptide bonds. 

Previous investigators have supposed that under certain conditions 
hydrolytic enzymes may also function as- synthetic enzymes under 
different conditions of temperature, pH, oxygen tension, etc. These 
experiments have all been carried out on complex mixtures of protein 
digestion products. In order to simplify the experimental conditions, 
the present work has been carried out on very simple substrates. The 
enzymes which attack proteins act on these substances provided they 
meet the specificity requirement of the enzyme. 

Using the intracellular proteinase, papain, Bergmann and his 
coworkers were able to show that three different types of reaction 
occur: 

1. Hydrolytic effect. Benzoylglycine amide is hydrolyzed to 
give hippuric acid and ammonia. 

CeHsCONHCHgCONHs ^ C5H5CONHCH2COOH + NH3 

2. Synthetic effect. Benzoylglycine and aniline yield 
benzoylglycine anilide. 

C6H5CONHCH2COOH + CgHgNHs > C6H5CONHCH2CONHC6H5 

3. Replacement, (a) When benzoylglycine amide is treated with 
aniline, benzoylglycine anilide is produced. 

C6H5CONHCH2CONH2 + CgHsNHg > CeHsCONHCHsCONHCeHs + NH3 

The rate of this reaction is greater than for reaction 1 and can not 
be a combination of reaction 1 and 2. 

(b) Benzoylleucine and glycine anilide react in 
the presence of papain to give benzoylleucine anilide and glycine. 

CeHgCONHCHCOOH + NHsCHgCONHCgHg «. CeHsCONHCHCONHCgHs + NHgCHsCOOH 

C4H9 C4H9 

All of these reactions occur under identical conditions of pH, etc. 

Just which one of these reactions will occur at a given time 
under a given set of conditions is determined by two factors: the 
specificity of the enzyme and the substrate present. This may be 



'i-^.u 



I -..■.• ;, t- 



= ''^; -V- 






( ■ .' J* '.4 i,'i; 






^■)r;Mr-' 



^•ti:-:- Jit 



^^i^^ :-f -, 



■■■.i cv 



:.»'£> 1 -X • 






*-— . 



-2- 

illustrated "by the following examples (with papain, except St). 

1. Only the natural form of the amino acid or dipeptide takes 
part in any of the enzymatic reactions. 

2. The arrangement of amino acid residues in the substrate 
determines the subsequent reactions. 

a. Bz-L-G-G > Bz-L-G + G while 

Bz-L-L-G ? Bz-L + L-G 

b. (with the enzyme, chromotrypsin) 

Bz-T-amide > no hydrolysis 

Bz-T-G -^ no hydrolysis 

Bz-T-G-amide ^ Bz-T + G-amide (quick) 

Bz~T-G-G-amide ^ slow hydrolysis 

c. Bz-L + L-anilide > Bz-L-L-anilide 

Bz-L + G-anilide > Bz-L-anilide + G 

d. Bz-L-L-G > Bz-L + L-G 

Bz-L + L-anilide > Bz-L-L-anilide 

(Bz = benzoyl; L = leucine; G = glycine^ T = tyrosine; and A = alanine, 
or their residues.) 

Because of this very complex specificity of proteinases and their 
ability to act on a variety of substrates a mechanism for protein 
formation may be advanced. Since on any substrate a given proteinase > 
in general, produces only one reaction, the process may be considered 
to be of the following type (where R is rudimentary fibroin molecule 
at any given stage in synthesis): 

R + G > R-G 

R-G + A ^ R-G-A 

R-G-A + G i> R-G-^A-G 

R-G-A-G + T > R-G-A-G-T 

R-G-A-G-T + G >, R-G-A-G-T-G 

R_G-A-G-T-G + A V R-G-A-G-T-G-A 

R-G-A-G-T-G-A + G -> R-G-A-G-T-G-A-G , etc = 

Bergmann's interest in this problem has been to obtain information 
about the physiological and pathological processes that are dependent 
upon the formation, the presence, or the transformation of proteins. 

Bibliography ; 

Bergmann, J. Biol. Chem. , n9, 707 (l937); ^24, 1^ 7, 321 (l938), 
Bergmann, Chem. Reviews, 22, 423 (l938). 

Reported by R. Mozingo 
February 8, 1939 



A . 






(i 



r>vT,?- < ; 



^. ....... 



>VO.;.c 4r-' •-' 



■■•I ^ . 0.' 



'.*»■•«. ^t" 



if-. } '■ :y.. 















ENOLIZATION AND ACIDITY 

Arndt -- University of Istanbul 

Compounds containing two or three of the following substituents 
on a -CHg or -CH group generally possess the property of forming salts 
with alkalies: 

COOR, COR, CHO, SO2R, SO2OR, SO2KR2, CN, ITO2. 

This property will be called the "empirical acidity" and means simply 
that in such compounds there is a relatively loosely bound hydrogen 
atom which can be removed by alkali with the formation of a salt anion. 

The older researches of Claisen, K. Meyer and others were chiefly 
on di- and tricarbonyl systems, and led to the conclusion that the 
acidity of these substances was a property of the enol hydroxyl group 
since the enol forms were isolable as such and proved to be more acidic 
than the keto forms. This conclusion, coupled with Thiele's suggestion 
that the empirical acidity is a consequence of enolization and enoli- 
zation a consequence of the formation of a conjugated system, led to 
the formulation of the salts of disulfones and sulfonamides as 

p 

R-S=CH-S02R and R-S=N-R 

6h oh oh 

^C«OR • 
in formal agreement with R-(^=C-(j;=0 and CH 

OH R ^COOR 

The subject of enolization and acidification has been reexamined 
by Arndt in the light of modern electronic concepts and with the 
diagnostic aid of diazomethane as a reagent for the location of acidic 
hydrogen atoms. The standard FeCls and bromine techniques are used 
also and their applications and limitations are discussed. 

The first distinction that must be made is between what is called 
"acidification effect" and "enolizing tendency". That the two can be 
distinct is seen by the acid character of compounds which, on the basis 
of the octet theory, cannot enolize, such as disulfones. Such com- 
pounds show no colors with FeCls , do not absorb bromine but dissolve 
in aqueous alkali and show C-methylation with diazomethane. 

The "enotropic" effect is the tendency for the change 

CH-C=0 ^ C=C-OH, 

The work done in transferring the proton from C to is the "preto- 
tropic expenditure of work" , and the tendency towards enolization is 
proportional to the enotropic effect and inversely proportional to the 
prototropic expenditure of work. The possibility of conjugation 
increases the enotropic effect, but the acidifying effect is entirely 
different from the enotropic effect. 

In non-enolized combinations (e.g.. disulfones) the "empirical 
acidity" is identical with the i^CH acidity; in completely enolized 



V'-T^-' 



'A ^iA #elfA2ici5'/f£ 



• ; 'I'-? V ill ! 









'^iir 'v^ i^r^i 



■I '■ '■ 






1 •■■( or . 



>v -f f \ ; .' /■ .-y h»S''-?»' "i^'^A't'-'S^ 



: ; .-t .;{ c; i- i-.! 



■i > J"-, rr. '"t .j 



»VKt .T ;il •;• i> 
r* I,'" 

5.1 .( >■» i-lfli 



•■.t^o f'f:^ i'-t^; -;i£< rt;. 
;■ ^. ■ ^^ -■. ^ ' ■ ■'■■ ■■" .-■"^■" ::^>:'^,'•J ;;v.U;:.. ;-^i<-^i-{.;f I} x^irfjhic^- 



'■-i-i^n-'- 



Si- 



-"':■:■ • • s Jii'^^^ff; if-- 



iJ'r^: 



: ■:!■' a :j t ■ 



■•fo:^o 



^•^yhidsfj^ 



' i^ '? 4 V; '^ '"^ *^- •'■> ^' «•' i'-i ;^ ■ 
■ c^. "1 j. '^3 S5 itii^ ■<;?'?'■•.'-' 



'-:s:usyOt-; 



?'1 ,- f. »3 r> rf 






tl^*c;j 






-2- 

systems ( e.g. , |5-ketoaldehydes) the "empirical acidity" is identical 
with the -OH acidity. 

Upon these general principles has heen "based the study of a 
great variety of compounds, representatives of which are given in 
the table. 

As a result of comparisons "between the large num'ber of suhstances 
containing various com'binations of the groups mentioned, Arndt has 
arrived at the following series of relative effectiveness; 

Aciilifying effect: 

NO2 > SOgOR > SO2R > CN > COOR > CHO > C=0 

R 
Electromeric effect: 

CHO > 0^=0 > CN > COOR > NO2 (groups containing -SOg- have 
R none) 

Certain of the facts adduced in these studies are not incontroverti'ble 
For example, the "indirect methylation" of nor-enolized compounds 
to enol ethers needs a more thorough explanation than has "been given, 
for upon this hypothesis rests a large portion of Arndt' s conclusions. 



Ei"bliographv ; 

Arndt and Martius, Ann., 499, 228 (l932). 

Arndt and Scholz, ibid., 510 . 62 (l934). 

Arndt and Rose, J. Chem. Soc, 1955 ^ 1. 

Arndt, Amende and Eistert, Monatsh. , 59.. 202 (l932). 

Arndt, Scholz and Frohel , Ann. 521, 95 (1935). 

Arndt and Loewe, Ber. 21, 1627 (1938) . 



Reported "by T. A. Geissman 

E. F, Rogers 
Eenruary 15, 1939 















'I. < r 



i'll-vsr^ i^ ''; 



'if; .t- 



i" <r '-^r'-"! r' ■<T,:..v?. ;^ 



s5?iftr'l •••jv.^!;:- ^^i^^^at^ra 



' ('^:^c^ci 









•^•^■^:' 



•^ V":* *7. i .* L.Ti - /'X V 



., V. H 



1 



in 



f fi .k V-'f P > f '^ '■' '•'' • 



Compound 



Br: 



-3- 

TABLE 

FeClg Type of methyla- 
tion with CHaHp. 



Remarks 



II 

III 

IV 
V 

VI 

VII 

VIII 

IX 

X 

XI. 



CHg 

^SOgOR 
CHs 

^SOgOR 

S02-CH3^ 
CHg SO3 

Tosyl-CHsCOCHg 
Tosyl-CHsCHO 



^'''^ COOR 



SO2CH2COOCH3 
Tosyl^ 
Tosyl' 
jZfOSOs, 



,/ 



CH-COOCH3 



00302'^ 



CH-COOCH3 



jZfEtNSOg^ 
jZfEtNSOa 
jZfEtNSOg 
^EtUSOs^ 



CH-CHO + 



CH-COCH3 - 



k>: 



CH; 



SO' 



f^^^^^^\ 



XII 


1 CH-COOR 


JII 


Tosyl-CHgCN 


XIV 


CH2(GN)s 


XV 


Tosyl-CH 



+ 



+ 



no reaction 



Shows S020R> SO2R 
in "acidifying 
effect" 



"Indirect" methylation 





no reaction 



Shows CHO > COCH3 in 
"enotropic effect" 
(cf. X) 















(cf. XI) 
Shows effect of " con- 
jugation partner" 



N 



'CN 



Comp. with TosylCH(COOR 
shows CN > COOR in 
" electromeric effect" 



LHAS: 



VV - V 



Cf^J;^ 



n -, rri 



qY" c:£":y 



iji^^b:! ;i4'. {:w r:nxv " 



in 



.biiuOiTfTioO 



:?.XJ'O.C:jT OI 






' '.:•■ .'■" :ir- 



c'n':^ 






nv>I::^f^lTiU'!.{^\ ^'r!-i;:--aX)X?I " 



.>i'nO':/r.'^:^~:\f<^,v-^ vr 







:::iUai..;;-j:^5a.rj: 



:);•• 



o 






>. 



, ^ ,' <i .. 



'> ^ r: --.. 



'no- i^ 



V. 



<2'.;ciicvr-C:; 






Jl .% 



^iOOO-i-IQ 



,/■.:> V"" ^'^ 



-r^X 



^ 



i^OsHO-X^-oI 



r r T ^• 









"f; 



S". \lXj ! i.>'l:.i 



H')-X^;:^.c? 7^ 



-4- 



TABLE (cont.) 
Compound 



Br. 



FeCl, 



Methylation 



Remarks 



XVI 



CH-CN 



.COOMe 



XVII Tosyl-CH 



•CN 



XVIII 



CH-COOCH3 



XIX Tosyl-CH 



,C0CH3 



"C¥ 
XX RCH2NO2 
XXI R-CH=NOOH 
XXII NOsCHgCOGEt 
XXIII TosylCHsNOs 
XXIV jZfCOCHaNOg 

-COOEt 



XXV NO2-CH 



y 



N 



N 



C,N 



IT 



Ester C=0 not as 
efficient conjuga- 
tion partner as CN 



o,c 


CO > CN 

in electrcmeric 
effect 


nitronic ester 




nitronic 




nitronic 




nitronic 





p "I 

^ nitronic; ~ enol ether 



nitronic 



COOEt 



,i^l:}ir^pi:lit^,l |iOo' 



T- N. ■ 






, '. A *J w *.l^.^ V ' 



^ Ih 






Vx^A Vi^K ^i 



A-n. 



i-KDOv--HD I 



,..^CCO,. 



..,,.4?' 



y~r:ss--r^V V 



:he synthesis of chromane derivatives 



John, Gunther, and Schmeil -- G'6ttingen 



Among the various heterocyclic ring systems which have "been 
investigated, one of great interest and importance is that of 
chromane. The chromane ring system and its numbering is shown 
in formula I. 




Not only have chromane and its derivatives heen studied in 
great detail, but the chromane ring system has been found widely 
distributed in nature. Some of the natural products containing the 
chromane ring are -tocopherol, anthocyanidines, flavones, and 
chatechines. Many chromane derivatives possess physiological 
properties. 

The first preparation of chromane was that of von Braun and 
Steindorff in 1905. They diazotized ^-chloro-^-propylaniline to 
introduce an hydroxyl and then brought about ring closure with 
alkali • 




CH3CH2CH2CI 



/X^CHsCHsCHsCl 



/V-^CH^ 



X/'^OH 



-» 



0" 



CH3 
.CH2 



This synthesis is not very useful because of the difficulties in 
obtaining the starting material and because of poor yields. In 
1919-20 the synthesis of chromane was reinvestigated by RindfusE 
and coworkers. They succeeded in obtaining two convenient methods 
of preparation. 



OCH2CH2CH2OH 



oun 

(Br) 



Pp.0 



3^5 



V 



-» 






'CH2 



Zinc chloride can also be used as a condensing agent. By the use of 
substituted phenols, chromane derivatives substituted iw the benzene 
ring are easily obtained, Chromanes substituted in the oxygen ring 
have been obtained by Claisen by the action of acid condensing agents 
on mixtures of phenols and butadienes? for example, phenol and iso- 
prene give 2, 2- dimethyl chromane. 



/VoH 



Hp.C^ 



6h CHo 



A 



-*f- 



g-cHg 

GEo 



\/^CH2 



Oxidized derivatives of chromane, namely substituted chromenones 
have been obtained by two general methods. In the first method 
0,-acetoxyphenyl ethers are treated with esters followed by saponifi- 
cation and ring closure. 



V, • .^-H't :. 



•rtiiJio-.: 
• .urn «: 



. •,■ ., !• 



! ■■■■.!- 



:3,5>^f!K. ;; 



.. . ..'f ;,■;^JJ;j^,.■■v 






"n;-i J'^iif^^l- 



-2- 



/VoCHj 



+ ROOCR' —^ — > 




coned. 



HR H2SO4, 



■^ 



'CR' 
II 
.CR 



n the second method /3-keto esters condensed with phenol. The 
eaction follov/s two courses depending upon the condensing agent 



I 
r 
used. 



/VOH 



HO-C-CH3 

-♦• C-CH3 
COOR 




C-CH3 



C-CH3 

,6=0 



Chromenone 3-carboxylic acid is ohtainable from hydroxy-acetophenone 
and oxalic ester. Upon decarhoxylation chromenone itself is 
obtained. 



0" 



-t- 



COOC2H5 
600C2H5 



/\.0H 



Ha 



COCH3 



COOOOC2H5 
CH2 



HI 



■> 




C-COOH 
it 

CH 



^b 



heat 



/V° 



\ 







CH 
•I 

CH 



Demonstration "by Karrer that oc-tocopherol (Vitamin E) has the 
structure II, has spurred chemists on to synthesize other chromane 
derivatives containing the same suhstituents on the benzene ring. 



CH 



HO^^CHs 



CH3 

iT "?^2 r ^^3 1 ! 1 

,^--^^0-C-CH3^CH2CH2CHCH2i3H CHg'X^' 
CH3 ^X^H3 - -' 



CH- 



( 



Vq^C-R 

CHc ^CHg 



R - H,CH3 ,Ci 2H25 



II 



3 
III 



The compounds synthesized have the general structure III, The 
methods of synthesis used in obtaining these compounds may be class- 
ified under three heads: 1. Simonis synthesis; 2, Friedel-Craf ts 
condensatioH; 3. Grignard reactions. These syntheses are demonstrate 
by the following examples. 



^^>y\ 



\ 



.-fl . »«. . .5>^ , 



^ i 



HO 
CH3 



CH3 
/\ C2H5OCO 
+ 0^2 

g=o 
CH3 



OH 



-3- 
CH3 



PpO 



2^5 



HO 



^ 



CH3 
IV 



C1-C=0 

CH AICI3 . HO 

IV + H — 0£l > 

C-CH3 ^ ^ CH3 
CH3 



OH, 



CH. 



cq 







CH 

H - 

C-CH3 



Hs,Pt 




CH. 



HO 



CHq- 



CH3 

V 



HO 



CHff-- 



CH3 



CH. 

^H 



S^c/ 



2 
CHCH. 



CH2 



CH3 



0^ "^CHg 



CHc 



VI 



Bo 
CH3 



CH< 



(X 



CH. 



CHgMgl 



N' 



CH; 



/ 



.c=o 



CH3 

' ,CH< 



■ .CHs 



CH. 



VI 



C12H25 



VII 



CH< 



A = 5:4 mixture of CHgMgl and CigHssMgl 



Compounds V, VI, and VII were similar to -tocopherol in that they 
are easily oxidized to quinones with simultaneous cleavage of the 
pyran ring and formation of secondary or tertiary alcohols. The 
ultraviolet absorption spectra of these chroraanes are practically 
identical with that of -tocopherol. Although certain ethers of duro- 
hydroquinone and of trimethylhydroquinone show a definite Vitamin E 
activity, nevertheless compounds V and VI were inactive in doses up 
to 50 mg. Tests on compound VII have not "been completed. 

BihliograiJhv ; 

von Braun and Steindorff, Ber.,38, 855 (l905). 

Rindfusz, J. Am, Chem. Soc. H, 665 (l919). 

Rindfusz, Ginnings, and Harnack, ibid., 42, 157 (l920). 

Claisen, Ber. , 54, 200 (l92l). D.R.P. 374142 (l923). 

Bloch and Kostanecki, Ber., ^,v.J999 [l890). 

Petschek and Simonis, Ber., 4^, 2015 (1913). 

Peters and Simonis, ihid. , 4J,, 837 (l908). 

John, Gunther and Schmeil, Ber., 7_1, 2637 (l938). 

Smith and Denyes, J. Am. Chem. Soc, 5^, 304 (l936). 



Reported by H. M. Teeter 
February 22, 1939 



r-''y 



ft' 



ii 



I 

t 

/ 



/\ 



/\ 



f . 1-. 

'S 1 



/\ 









••> 






.A V4>« !« ■.'..'-- 



U'X^i!-; 



TRANSFORMATIONS OF THE STEROID GROUP 



Butenandt -- Kaiser-Wilhelm Institute fiir Biochemie, Berlin-Dahlem 
Mamoli -- Kaiser-Wilhelm Institute fiir Biochemie, Berlin-Dahlem 
Ruzicka -- Eidg. Techn. Hochschule, Zlirich 
Reichstein -- Eidg. Techn. Hochschule, Zlirich 

The name " stercJid" was suggested in 1936 
by Callow as a convenient designation for the 
cyclopentanoperhydrophenanthrene group: 

This group includes the following subdivisions , important 
examples "being shown: 





Cu CHCH2CH2CH2CHCH3 



Ergosterol ; 
II CH 



QH3 OHg /CH3 

nu chch=ch(!:ech 

3' ^CHg 




Progesterone : 
V 




B, Sex Hormones (cont.) 



Androsterone; 



VI 



Testosterone 



VII 



CH.O 




C. Adrenal Cortex Substances 



Corticosterone : 
VIII 



CH2OH 



CHcC=0 




CH2OH 
Desoxvcorti coster one ; qh (i=0 



IX 




Other members of this group, which will not be treated in this report, 
are the bile acids, toad poisons, cardiac glycosides, and the digitalis 
saponins. 

The structures of the sterols and bile acids were determined by 
the usual methods of dehydrogenation and degradation, most of this 
work occurring in the period 1920-1930. From 1930 to the present, 
attention has been directed chiefly toward the sex hormones. After pre 
liminary investigations on the small amounts of material available, 
structures were advanced, and then definitely established by transfor- 
mations from the known sterols. Transformations from one group to 



m^m ni^ms:^' -r- CfO.8^ioiVA^>r0H£.A.'u 






; ( IX "T?*.!-. J ■> i.ns^ri r> 3 .: 






UUifi5^S ^9iyii> 






'W ^ X -. 



V- ""J i"< - 






I. ;j 






■ii\p,3x^ . ^i'iol-^ fr^ji^fftjjs ^^iu^^ni|d7 vil<^ aefci/iv'ii? 



;nvv 



.. "I \\i ■■■ i- ■ ■ ■■ 

. i .f aa 



,X^fJi ,?i. 






f j 



XV 





^^^:?-rs:r^{0^::V3i 


:iiv 



r^^--^^l^i4L^4y5ils:y^ ' 



»'-, 






.♦ ■'■■ 



ixiv 



1 I 






4V- 



rU.y.'SJtj.i:. 



• *S!.';t" ^J i. .• Xl. ^^_ 












; .1 "X : 



V.' - ■. : S »/ '-. i . 






V - 



3ijr: 









i/._4;ia;. f ■ ^ 



Vi 






f 1 



^'^d^ 



• k;-i 



v 















■i3;ijQ 



«- " ■ " ' " X i::,,-: . ^Oti:Sr,o-.:(.U toil.,'- x^dj ni •^>7i^^i.,.iao 

■^ '^-- ... -Tsrii ^rjsifoj xi'^-^^^'^ fcsjoa^iip a--' ^ ^' " '. ^^ 



OT -If/V'Xli 



,-: ;.:7'Y--TP 



-2- 



anether are thus important for proof of the interrelationship of the 
groups, and in some cases (particularly for progesterone, V) as prepa- 
rative methods for less available substances, 

I. Stereochemical Aspects of the Sterols (F-155; G-1251,1377) : 

(References to Fieser's "Natural Products Related to Phenanthrene" 
and to -Strain's chapter in Gilman will he designated as P and G, fol- 
lowed hy the page number.) 

All natural and related products belong to two epimeric ring 
systems, differing in the configuration of the C5-H: 



CH.R 



1. Cholestane ; 

C5-H is trans to Ciq-CEqI 



(in these formulas, dotted lines indicate bonds going into the plane oj 
the paper; solid lines indicate bonds coming out of the plane of the 
paper, referred always to the fused carbons C5 and Cjo* See G-1252 foi 
illustration of space models.) 





2. Coprostane ; 

C5-H is cis to C10-CH3: 



CH.R 





Alio - refers to C5-H when it is trans to C10-CH3 (i.e. cholestane 
might be called allo -coprostane. ) 

Hvdroxvls ; (e.g. in cholesterol, I, or androsterone, VI) 

Epi - refers to C3-OH when it is trans to the Cio-CH3. 

Configurations are deduced by several methods, the most important 
being the rule of v. Auwers-Skita (G-1257): neutral media favor forma- 
tion of trans forms; acid media favor cis (see G-1258 for a chart 
showing application of this rule). 

Marker uses oc and f3 to designate the configuration of hydroxyls: 

oc = trans 

fS = cis 

II . Transformations of cholesterol (l) to progesterone (V)^ 
androsterone (VI) ^ and testosterone (VII) ; 

The older methods, where they have been replaced by new and bette 
ones, will not be discussed. In cases where it is difficult, on the 
basis of the literature, to judge the relative values of the methods, 
each one is described. A general outline of the steps in the various 



referred to nearest angular methyl group 






j|'' ^ 
.-^ ^ 


















i^V) 



t, , .;. .<■ ~. \ ,- ,^ r 

.'1 ->.b,C'iS' ■;>'-•■ -^ 



H»?l^ 



^4;^^56^^^V 



,-...../ 



ksM''Ci^> ^^ s^ 



i 
t 



'■ a>-'^-: 



I. A - J •;:' i ! O ' 



c t 



Ori.C'iJ E. 






:r p: -; 



t . on/ -V •: 0';r 00-01 i i: nr/J"i.:HO :>0 TPi ; ■ 



ilX^Jii^Ji* 



ii^Vi 



^^4i^' 









* /-V ^ ■!*• r- -- • f- 



^roii;';: eifid- 'r 






. ♦^■~ » "^ t ** 



n: 



'.r -i- 



^d:?- iv^AjS^i^M^ Ov ^ i>«:.e ■» 5^5^.^ -v 



if<JO^-i] 






.i 



? n f ; ji g i^ e- ;o-' 



iKte^.-.".\>t^gi*. :'i iSVr 












.V.:;;;u i.>;^.;j<iV' ;,i;iU>.-ii 



^v -Hyj^f-: 


;';'i «/>{> 


-'T of.' 


wi; ^.J 




.%'X:;n-:;-V 



-3- 



conversions is given first, the letters referring to the descriptions 
of the methods below. 

98^17 ^0 

i-SJ > /^ "^^^-^^^ dehydroandrosterone 

cholesterol (l) 




(d) 




testosterone (VI I) 
(c,g) 





androsterone ( VI) 
^ic) 





^r^ 



^■^-androstenedione 
(XIV) 

COCHg 



(e) 



COCH3 



^i*-cholestenone (XV) 



•progesterone (V) A -pregnenolone (XIIl) 



HO^^sX^ 



a. Dehydroandrosterone (F-232; G-1375): This important intermedi- 
ate is now an industrial product, although the overall yield from 
cholesterol is only 2.8^ (F-234; G-1381): 



CqHiv 




CfiHi? 



-» 



cholesterol (l) 



AcO 




dehydroandrosterone 
(XII) 

This is the standard method for protecting an ene-ol when oxidizing 
off the side chain. 



b. Tavatsherna (7) has obtained progesterone (v) in yields of 
10-15^ by oxidation of cholestenone with permanganate: 




-> 




COCH3 



<!i^- cholestenone (XV) 



progesterone (v) 






^ 



(nx) 



^. 




;^--' 'X 



(r; 












r-f 



uU 






( I ) Xot;3d riOioio 



(^) 



■•^. 



---^- 

k^*- 






r i 



(IIV) o:.^,-r-- 



-I r - • . 



\ 



* 

I.! 









I 



V- 



•■ ■■••k" 



-V 






£\ 

















J I 



(VX; n?no-r:A^f%.5|j^.f., ^j^yTVj^ 



-.t;^ jiT'Tstrti 






■"'"-K'N 






.>'>.. 



r 



■-^;^>-^.^-' 









v,^ 






^•^ 



r T 1 

T 1 "i 



{ 1 I 
\ - / 



iS'ioiorio 



.nij^no rihir- 3:f^ "tic 



rri^ ,",!•>. 



* i 









I 



'1 \ y . r> 3 yi ' 



>-' 



IV-, \ 



Y 



^^.....- 



V-"' ""■n''-' 



\- 



.'>^-'^ 



-4- 



c. Mamoli' s method (3): Mamoli has been ahle to carry out 
enzymatic hiological hydrogenations and dehydrogenations of steroids 
with the bacteria in fermenting yeast: 



yeast in 




sugar soln. 




-^ 




HO' "^ "^ ^ 0' ^ "*^ 0' 

dehydroandrosterone (XII) A*-androstenedione (XIV) testosterone (VII 

Migration of the double bond in the first step from Cs-e to C4-5 is 
characteristic of the oxidations of A^-ene-3-ol compounds. In the 
same way, pregnenolone (XIII) is converted to progesterone (V). 

Mamoli has recently been able to convert dehydroandrosterone (XII 
directly into testosterone (VIl) by treatment with a bacterial mixture 
cultured in yeast water. After shaking for 48 hours under oxygen, 
the reaction was interrupted, freed of bacteria by filtration, and the 
filtrate added to completely fermented baker's yeast in sugar solution 
An 81^ yield of testosterone (VII) was obtained. 

The latest application of this method is in converting 
A*-androstenedione (XIV) directly into androsterone (Vl): 






-> 



A*-androstenedione (XIV) 



dp 



H 



HO'" 




androsterone (Vl) 



d. Q-p-penauer' s method (4): The specific reducing action for 0=0 
of aluminum isopropylate in isopropyl alcohol has been reversed by 
Oppenauer, and good yields of oxidation products are obtained from 
secondary alcohols by using aluminum tert -butylate in acetone; all 
the applications are of the general types 

Al(0CMe3)3 




MegCO 



"> 




Examples from the transformations outlined on p. 3 of this abstract: 

dehydroandrosterone (XII) ^ ilf -androstenedione (XIV) Q^% 

A^-pregnenolone (XIII) ^ progesterone (V) 75^ 

cholesterol (l) ^ A^-cholestenone (XV) 94^ 



Westphal (4b) has shown that aluminum isopropylate can be substituted 
for the .tert-butylate in this reaction. 









• f <i;.';?-^'vL ^i:i •.;• " ■ 5 -^ulT^ ' 






^-^:--- 



_,^--^.---s. 



.-'-•^\X 



^^ 








-f w-xr + 



-' -p-^^T' Jmcr^^ •^^;:'v: +*?£nrlrT ..k^^ r-" •■-- 









;;^ijj<A 






^i^^,- 



*< 






I,,,'; ,;:;>::- "A 



^•r-v 






,•''•-■•5. 



r ^ 



:~---\ 



{ V'-iy.-;, "^TtXit/iTij^fJ^^^^KSf^jv... ^,s^ 



;iii-^^j^. 






-t: :y:'::\-r((v':qt}r r :^f:ilmi^ 









^^ .■"":•.;■ =^niyA • 



^t:^Tae} ^ :?:/v^ ^.^.---S^-^. 



^:«S^X-.>^ 



.-J ; vsi: 



A-: 






-;?^^;iriivd;Q^«-^v^. 












-5- 



e . Butenandt's method ( 5 ) : 


HO CN 
AcOH 



AcO 




KGN 



dehydroandrosterone (XII) 
( acetate) 



<i^^-3-hydroxy-etiocholenic acid (XVI) 
(see p. 6 of this abstract) 



HO CN ON 

I 1 pyridine J„ 1 

l.Hyd. 
2.Redn. 

COOK 



CHgMgBr ^y 



CHg 

C=0 



Raney 

Ni 



H, 



COCH3 



NaOH 



«/ 



^ COCH3 




A --pregnenolone 
(XIII) 

f» Dirscherl (6) has been 'able to oxidize the side chain of 
A*-cholestenone (XV) directly with CrOs , without protecting the 04.5 
double bond. Progesterone (V) is obtained in small yield (side chain 
oxidized at a), most of the product being A'^-androstenedione (XIV) : 




CrO, 



40-50' 








■o^ 



^^-cholestenone (XV) 



progesterone (v) A'^-androstenedione (XIV 



g» Serini (8) has converted A*-androstenedione (XIV) into testo- 
sterone (VI I) by purely chemical methods (cf. Mamoli's method above): 




HC(0C2H5)3 



orthoformic ester 




1. Na + 
PrOH 





A*-androstenedione (XIV) 

III. Ruzicka' s method (2b): 



C2H5O 
enol ether 



HCl 




testosterone (VIl) 



Ruzicka has prepared compounds of the progesterone (v) type by 
adding acetylene to dehydroandrosterone (XIl), followed by acetic 
acid (2a) : 

HO C = CH 




HC = CH in 



liquid NH9 



■> 




dehydroandrosterone (XII) 



(Xlla) 



^^Tilt^HC 



n' 






-'i nib live; 






i J 






''"^ 






i>/-t- 



•( IIX^ ^n'''i:>j c;03:hC'^'r.; -^•■ 






.^'f! 



■•Ai 






i 









f V T }■ X ■ 



/x 






vjp. -j^ «;• 



^^-^s 






(•••'•to 



( i 



'A 



■N. 



1 i C' TT Tj o> 4i ; 






a 









-<; 






^^.!. 









• ' 'T -^^ .' f" t"',"* '.T -i .^ •.'- ' •■ ' '- "'-.:•■',' 



Ill 



i I I i> 'J ■'-' '"I ."^ ^'"r r- '! •'. ""x ''*i'"- •'V - . '^ V*- w s p *.' : r. ^ ■4- - -# . r" . , ^w , 



-i J.. , ... 

I 






ai 



.Vl- -«- 



... i t'-. 



-l. >..i. 



^^ -i-''*. 



J, 



(*^ 






'-.•X/'v 



-6- 



Before adding acetic acid to this hydroxyethynyl compound (Xlla) , the 
C3-OH is oxidized by Oppenauer's method: 




C=CH 

Al(0CMe3)3, 






C^CH 



90^ HCOOH 



CiECH 



(Xlla) 




OCH. 



HgO 
BPg-ether 
catalyst 



NaOH 



(2c) 



AcOH 
AC2O 

OAc 
AcO 6=CE2 



-> 





60^ 

l7-hydroxyprogesterone (XVIl) 

IV. Adrenal Cortex Hormones : 

Corticosterone (VIII) has "been shown hy Reichstein (lOa) to he 
related to the sex hormones hy its conversion to all^-pregnane (XIX) 
the alio isomer (Cf^~H trans to C10-CH3) of the parent hydrocarbon of 
the pregesterone group: 




COCH2OH 

1 complete redn. 



CH2OH 

6hoh 



(200 mg.) 
corticosterone (VIIl) 

CH2CH3 




HIO, 



■^ 



(170 mg.) 



CHO 








COCH3 



CHgMgBr 

CH3 
CHOH 



Zn-Hg 



(11 mg.) 



allo-pregnane (XIX) 







,CD 




CrO. 



::i 



H 



(70 mg.) 
(XVIII) 



(150 mg,) 



The relation of the adrenal substances to the progesterone group 
is further shown by the recent isolation (lOc) of desoxycorticosterone 
( 21-hydroxy-progesteroney IX) from adrenal cortex, and the discovery 
that it has 5-7 times greater cortin activity than corticosterone. 
Reichstein had already synthesized this compound from an etiocholenic 
acid, using a method of Arndt and Bradley (for references see 10a): 



COOH 



COCl 




CHal^ 



AcO 

A^-5-hydroxyetiocholenic acid (XVI) 

(acetate; see p. 5 of this abstract) 




COCHNg 






>f' 



K — 



riOOOH e^O*^ 












-^ 



^^V 



HO-dA /O^n 






! I ! 



/ 



oH^:-^: ..-Oil 



TJ 1^, f^, *\ ,-'-. 












<i\X/' 






'.'•J 






(^.s 



^/.\rf ...rrs'f..^:. •.. '^. i^Hn^oiD o:r ^^^vjl H^;0^ T^rrrosi oii^''-dJ 



'"-OTj ^jno'l^:?!-!:);^^- 






■■f-" •■- 






( . ^OT 00:" ^ 



V.A.^!v 



[lllV] jriO'X:r.i-::;J;^'So; 



2 ,1 



.•,i-:.'^00 









^'i^jrlVX; ^ 



4., .-s 

,1 




:"3 1'i .13- n 5 D T(V - 0. i i! i?. 






■U v:^ 



XA^L^u lUJ-rjjro 'xor.ci^ts a^iiaj- ^^-d. ~^r{. ;i;i j^j:' 






i' 



X. 



K"-' 



r 



■^-^'^--.. 






rIODD 






' -' U.-_i^ 



S^-.Sii2iiL 



--'*^*a^-^v 



■^>' ^^- • 






-7- 




C0CHN2 



_Ji AcOH^ 



COCHgOAc 




OCHgOAc 



1. NaOH 

2. Zn 

desoxycorticosterone ( IX) 
( 21-hydroxyprogesterone) 

V. Preparation of Estrone from Ergosterol (G-1361)j 

In 1936, Marker and his associates reported the important con- 
version of ergosterol (ll) to estrone (IV). When ergosterol is ex- 
posed to visihle light in the ahsence of O2 , it suffers dehydrogena- 
tion and yields a substance of unknown structure called ergopinacol , 
which by pyrolysis loses the C10-GH3 as methane, giving neoergosterol 
(XX) in Z0% yield: 



CH399^17 



T CqH- 




^ ergopinacol 



heat 





C9H-L7 



CrOg oxid. 



acetate 




estrone ( IV) 

Windaus (9a) later at 
neoergosterol (XXI) with s 
compound XXII, but he obta 
questioned the formation 
Marker (9b) on the reducti 
tiate his earlier work, I 
Bamberger and Kitschelt (9 
/3-naphthol by reduction 



XXII 
HO ^" ^ - ^ 

(also some non- 
phenolic products) 




neoergosterol 
XX) 



^9^'l7 



dehydro neoergosterol 
(XXI) 



tempted to repeat this reduction of dehydro- 
odium and amyl alcohol to give the phenolic 
ined no alkali-soluble products, and he 
f estrone by Marker. The recent article by 
on of equilenin (XXIIl) seems to substan- 
n support of this. Marker cites the work of 
d) who obtained an Q% yield of ar- tetrahydro- 
f /3-naphthol with sodium and amyl alcohol. 




equilenin (XXIII) 





Pvuzicka (9c) has also recently studied the reduction of equilenin 



1 j 



OiH 






,,'■'.)■> '-,111 . 



S,L 



/ 






Y-""1''"> 



-..'<>^ 






;i5 



^"^H"'"l 



I 






' 1 . , I 1 - ;• '•• -•■>- .. ^^ I -;.'-' •' . -^ l y •'■'rv 



S 



HQO 












;-■■: \ 






■<r — 



^;^^:f 



.oo.i^^'i'fo^-i-x.. 






1 

--A1 






^^^--^-^ 



,.-"-'. --i^-^ 






N^-' N 



L ( 



i 






•v^ 



b txo r--:'^ 



r 



. 






••:iOi^. v;flO<=! Jo : . 



i.-^ir)^ -.1 






w .'^^viqirrfi Tv,i3 Ti^;^J?JU \a<:f 



: T\V O "; rV- 









r 



•'^. 



■">:r' 



> 



^■0., 



■OiStA 



J' .-J 



i 



"■ •■ --k 



-^-■. 



V'' ■^-x'* 



if 



( -r 



hi.-:i.\.iiyr:r:. 



3i p;j;i-.c? \:;X j-TG^'ijT; Oa.U; :,*;![ ( OV) ) .^;S?;. 



-8- 

with sodium and amyl alcohol and 20% of his reaction mixture con- 
tained alkali-soluble (phenolic) products, which he did not 
characterize. The non-phenolic products, amounting to 80^, were 
the same as those obtained by Marker. 

Bibliography ; 

1. Ruzicka, Furter and Goldberg, Helv. Chim. Acta, 21, 408 (1938). 

2. a. Hennion, Hinton and Nieuwland, J. Am. Chem, Soc, 55., 2858 

(1933). 

b. Ruzicka and Hofmann, Helv. Chim. Acta, 20, 1280 (1937); 

ibid. 5 ^, 88 (l938); Ruzicka, Hofmann and Meldahl, 
ibid., 11, 371 (l938); Ruzicka and Meldahl, ibid., 21, 
1760 (1938) . 

c. Inhoffen, Logemann, Hohlweg and Serini, Ber., 21> 1024 (l938). 

3. a. Mamoli and Vercellone, Ber., 22, i'^O (l937). 

b. Mamoli and Vercellone, ibid., 70, 2079 (l937). 

c. Ercoli and Mamoli, ibid., 7J^, 156 (l938). 

d. Ercoli, ibid., n, 650 (l938)? ibid., 72, 190 (l939). 

e. Schramm and Mamoli, ibid., 21, 1322 2698 (l938). 

f. Mamoli, ibid., 21, 2278, 2701 (1938). 

4. a. Oppenauer, Rec. trav. chim., 56, 137 (1937). 
b. Westphal and Hellmann, Ber., 20, 2136 (1937). 

5. Butenandt and Schmidt- Thome, Naturwiss,, 16, 253 (1938); Ber., 21, 

1487 (1938); ibid., 22, 182 (1939TT 
Kuwada and Mujasaka, cf. Pharm, Soc. Japan, 5.8, 540 (1938); 
C. A., 32, 7474 (l938). 

6. Dirscherl and Hanusch, Z. physiol. Chem., 252, 49 (l938). 

7. Tavatsherna, Arch. sci. biol. (USSR), 40, 141 (1936); C. A., 31, 

6670 (1937). 

8. Serini and K'6ster, Ber., 21, l'^66 (1938). 

9. a. Windaus and Deppe , Ber,, 2Q, "^^ (l937). 

b. Marker, J. Am. Chem. Soc, 60, 1897 (1938). 

c. Ruzicka, Miiller andM'6rgeli, Helv. Chim. Acta, 21, 1394 (l938). 

d. Bamberger and Kitschelt, Ber., 23, 885 (l890). 

10. a. Steiger and Reichstein, Helv. Chim, Acta, 20. 1164 (1937). 

b. Steiger and Reichstein, ibid., 21, 161 (1938). 

c. Sutter and Reichstein, ibid., 21, 11^7 (1938). 

11. Miescher and W. Fischer, Helv. Chim. Acta, 22, 158 (1939). 
Miescher and Kdgi , ibid., 22, 184 (1939). 



Reported by E, H, Riddle 
February 22, 1939 



..p- 






■V^ " .W 



^' 









;c'^: ^ 03-'^ r 






ff?^:'''' ) 



^ ,.hicfl ::v^?:t'.::) 0(^'d ,il ^.i^xcfl ,i 



, r»"f '; 









'. ' 






: <'?:«;?} ^?^r^ *Vl ,.iO^": ^rrni^nr-. -.i;.; r.:.rlcfj::?^:l^ .^ 



\!,;.! 



( : '-■►.' 



• O. 









ii ?* n <^ 



«^L 1,;^'^'^ .ciXiiD ,Tlv;'.' jifiiiJ-gnoio;! dhj^ lo^^iio:]^ .; 



^./ffiA .m-hf:;' ,vi:?'i \To.doai? .W l}a/j ^oriOR-:ji:^ . ^ 



elb^i^I .H .:? \t<:i ^^.-■ 



NITROHES 



Krohnke -- Univ. of Berlin 



Aromatic nitroso compounds can react with active methylene 
roups in two different ways to form (l) an azomethine (anil), or 
8) a nitrone. 

CN 
CgHgCHaCN + CgHsNO > 



CeHs-^H-N-CeHs 

6h 



CeHs-CH-F-CeHs 
OH 

CN 
^ C6H5-C=N-C6H5 + H2O 



CeHs-iH-N-CgHs 

6h 



2 CgHsNO 



CN 

C6H5-6=5-C6H5 + C6H5-N=N-C6H5 



It is readily seen that an excess of benzyl cyanide will favor 
reaction 1, whereas an excess of nitrosobenzene will favor the forma- 
tion of a nitrone 2. In many cases both reactions occur, especially 
if equal quantities of reactants are used. 

If the C-atom of the active methylene contains a halogen, then 
the reaction proceeds to a nitrone in the presence of alkali with 
elimination of HX, 

3. ^-NOgCeH^CHaCl + CeHgNO > ji-NOgCeH^CHCl-NOH-RJ ^^^^ > 

? 

:e-lT02C6H4CH=N-C6H5 

No excess of nitroso compound is needed as in the second case. 

Kr'ihnke has found that the reaction proceeds more smoothly if 
the halogen compound is first converted to a cyclammonium salt with 
pyridine. 



4. RCHgX + C5H5N > [R-CH2-NC5H5] X' 



R'NO 



,NC5H5l + 



RCH 
^N-R' 



CI" 







NaOH 



R-CH=J-R' < '^^^ [rCHOH-NOH-R'] 



Instead of pyridine other teriary bases as quinoline, isoquino- 
line or triethylamine can be used. For the nitrosobenzene can be 
substituted ^-nitrosodimethylaniline . The latter is preferable as 
dehydration takes place more smoothly. 

The aromatic nitrones are well-crystallized stable compound.^, 
while the aliphatic nitrones are not stable and polymerize. The 
nitrones are more basic than oximes and form stable hydrochlorides. 



, . > , ■. 'V 



T f 



-.jr -r 



-P,- 



They are readily hydrolyzed to the corresponding aliiehydes or ketones 
with cold concentrated hydrochloric acid or warm dilute acid. This 
property offers an excellent method for converting active methylene 
groups into carhonyls. 

5. CeHsGOCHaBr ^sM-^ [CeHsCOCHgNCsHs] V" -i^i^^l^^^likl^ 

C6H5C0CH=¥-C6H4N(CH3)3 -^ CeHgCOCHO ■+ { CHg ) gNCeH^mOH 

acid 

In this case the yield of pyridinium "bromide is quantitative. The 
formation of nitrone goes in 97^ yield and the hydrolysis to phenyl- 
glyoxal in 90>b yield. 

The nitrones are also split by such mild reagents as phenyl- 
hydrazine, aniline or hydroxylamine . With one raol of phenylhydrasine 
per mol of "benzoyl N-phenylnitrone a mixture of the a: and y^-phenyl- 
hydrazones of phenylglyoxal is obtained. With two moles of phenyl- 
hydrazine the osazone is obtained in 76^ yield. 

With hydroxylamine isonitrosoacetophenone is obtained in 75^o 
yield. With two moles of hydroxylamine a 60)o yield of .gjrti- phenyl - 
glyoxime is obtained. 

With aniline phenylglyoxal dianil hydrate is formed by warming 
for a few minutes. 

With warm 2N sodium hydroxide the phenylglyoxal obtained from 
benzoyl N-phenylnitrone is converted directly to mandelio acid in 
60% yield. 

6. C6H5C0CH=fi-C6H5 _l5i5JL,|CgH5COCHO] + C6H5-W=S-C6H5 

ICeHsCOCHO] ^^^ ^ C6H5CHOHCOOH 

Nitrones can be prepared in other ways. 

A, Staudinger found that diazo compounds react with nitroso 
compounds to give nitrones. 

/V 

R'NO + R2CN2 ^R'-N N ^ R-N=CR2 + Ng 

^CRs-N'' 

The nitrones obtained were not as easily isolated pure as by 
KrAhnke's method. 

B. Aldoximes or ketoximes can be alkylated, but not arylated, 
to give a mixture of the 0-ether and the N-ether (nitrone) of the 
oxime. 

_ R2C=N0Me 
R2C=N0H + Me2S0.i + NaOH 

R2C=N-Me 






-3- ■ ^ " 

C. The condensation of a ketone with a su"bstituted hydroxylamine . 

R2C=0 + R'NHOH > R2C=N-R' 

6 

Reactions of nitrones ; 

1. Hydrolysis to carlDonyl compounds. 

2. Addition reactions. 

According to Smith, all additions take place 1-3 to the 
system C=N— »0. 

a. .V/ith the Grignard reagent 

C6H5CH=N-CH2C6H5 + CeHgMgBr > (CeHg ) sCH-N-CHgCeHs ^^^ > 



OMgX 

(C6H5)2CH-N-CH2C6H5 (N-henzyl-N-benzohydrilhydroxyl- 
6h amine) 

b. With alcoholic potassium cyanide 

? KCN ?^ ?^ -KOH 
N02C6H4CH=N-C6H5 — '^^^ ^ NOaCeH^GH-lT-CeHg > 

I 

N02C6H5-C=NC6H5 ^^^^^ > NOaCeKi-C-NCeHs 

eu C2K5OH OC2H5 . 

II III 

I was not isolated, while II can "be isolated when equivalent amounts 
of nitrone and KCN are used. Otherwise the reaction proceeds directlj 
to III in excellent yields. 

c. With phenyl isocyanate 

CeKsCH N-CeHs CeHsCH— TJ-CgHs 

II + II ^ j I 

CgHsCHg-N-^O C=0 C6H5CH2-N C=0 

Reactions a and h could be explained "by 1-2 addition, but c could 
hardly be explained in this manner without some form of rearrangement 
of the addition compound. 

d. Nitrones, in contrast to the 0-ethers of oximes, will 
undergo the Beckmann rearrangement in a trans shift in the same 
manner as oximes. 

C6H5-CH C6H5-C=0 

H > 1 

O^r-N-CHg H-N-CKg 

Cis - Trans Isomerism: The nitrones should exhibit geometrical iso- 
merism, but only in one case has this been proven with certainty. 



CH 
Ji 





CI 0<r-N-CH3 CI CH3-lT->0 



-4- 



Compound 



Yield of Yield of 

Pyridinium Nitrone 

Salt 



Hydrolysis 
Product 



Yield 



X-/ y-CHsBr 



95-99^ 



X = Br,Cl,lT02,CH3 
in 0,-, m- or ^- posn. 



52-99^ X-/ VcHO 







X-/ Vc-CHsBr 

where X = Br^I^CljCgHs 



Not given 92-99^ X-/ VcOCHO-HaO 



1^-90% 



X-COCHsBr 



Not given 



where X = thiophenyl, 
/5-naphthyl, tert -butvl 



60-99^ XCOCHO-HgO 



Bibliography ; 

Kr*6hnke, Ber., 69, 2006 (l936); ibid., 2X, 2583 (l938). 

Smith, Chem. Rev., 2Z^ 224 (l938). 

Staudinger and Miescher, Helv. Chim. Acta, 2, 554 (l919). 



>50^ 



Reported "by B. R, Baker 



March 1, 1939 



"A 



THE ALLYL REARRANGEMENT 



Mumin -- University of Kiel 



Phenyl allyl ethers quite readily undergo rearrangement to the 
corresponding ^-allyl phenols in a manner analogous to the rearrange- 
ment of enol ethers to -C-CIiR compounds. 

5 

Claisen and Tietze have shown that in some cases at least, the 
carbon which is "bonded to the ring after the rearrangement is not 
that originally "bonded to the oxygen in the ether. 



0-CH2CH=CH2 





CH2CH=CH2 



Certain imino allyl esters undergo a similar conversion to 
N-allyl amides, in the course of which the allyl group exhibits the 
same shift. 




^ 



^N-CeHs 
""O-CHgCH^CHa 




C-\ 

5 CH2CH=CH2 



A free radical or an ionic mechanism is untenable in this case 
since the rearrangement of a mixture of two dissimilarly substituted 
imino allyl esters yields no product in which an intermolecular 
interchange of groups has taken place. The formation of an inter- 
mediate ring structure by means of single electron bonds is more 
plausible: 



CeHg-C^N-CeHg 

■ CH2 

I )l 

H2C — CH 



CeHs-C-W-CeHs 
-» b CHa 

i il 

H2C — CH2 



CeHs-C-N-CeHs 

() I 

-^ GHg 

I 

HgC^CH 



Similarly, for the phenyl allyl ether: 



H 



HC 



H 



H 



CH2 



CH2 

I 

CH 



HC 

I 

HC 



Ky.K 



'c- 



CH. 



\^H--CS 



■CH 



H 

HC^ C^ 



CH; 



HCv .Cv ,CH 



/^VOH 

l<v/-CH2CH=CH2 



The dotted lines represent single electron bonds, the dashes covalent 
bonds. 



The fact that a shift has taken place is established by use of a 
substituted allyl group. The reaction follows the same course vdth 
either oc- or ^-substituted allyl groups. 



./•^'Tsitru' 



' • ■ ■ 



■."X 









•'^ 



■? • • ■ N 



r ft-' 



- ,/j 



-2- 



In the case of phenyl allyl ethers substituted in "both of the 
ortho positions the allyl group migrates to the para position. In 
this case an intermediate ring structure is impossible so the mecha- 
nism must involve a splitting into ions or free radicals, and on the 
basis of other experimental findings, the latter seems more likely. 
In some examples of this type the migration takes place v/ithout 
involving the oc- ^ shift. In others, however, the shift does 
occur. This may be explained on the basis of mesomeric polarization 
of the migrating free allyl radical. There is evidence to indicate 
a free radical mechanism may also be involved in the ortho rearrange- 
ment in some cases. 

The relationships of some of the derivatives of the OC- and (3- 
ethyl ethers of methyl _o-creosotinate are shown in the chart. It 
is of interest to note that catalytic hydrogenation of these two 
ethers results in the cleavage of the ether linkage rather than in 
the addition of hydrogen to the double bond. One possible inference 
which may be drawn from this fact is that the addition of the two 
atoms of hydrogen takes place successively rather than simultaneously, 
The mechanism of the hydrogenolysis is pictured as follows: 



C6H5-0-gH-CH=CH2 
C2H5 



H 



^ OgHs-O- 
H 
CgHsOH 







OH 
CH3 >-v^C02CH. 



The present investigation leaves some doubt as to whether IV contains 
the ethyl allyl or the ethyl propenyl group, 

C2H5 
O-CHgCH^CHCgHs 0-CH-CH=CH3 

0-CH2CH=CHC2H5 
CHa^^-^COgH 

OH "^ OH 

CHg^^^COgCHg CHg^^-^COsH 

V — ^^ M * 

C5H11 C5H11 

R = rearrangement; S = saponification. 

Bibliography ; 

Claisen, Ber., 45, 3157 (l912). 

Claisen and Tietze, Ber., 5^, 2514 (l986). 

Mumm and Moller, Ber., _70, 2214 (1937). 

Mumm, Hornhardt, and Diederichsen, Ber,, 2S, 100 (1939). 

Reported by A. H, Land 
March 1, 1939 



H- 



OH 



^5^9 



•^' ■ • ■ s • 




■ <.•■.- •■■- 


. - - •'■■ ^^ 


..i'-/Tii/i. 


. "^ -| . 


-^M-fifp 




.■.■^rvrT'- 





-■" C: 






TOAD POISONS 

Munio Kotake -- Osaka 

The toad poisons are a group of compounds present in the skin 
glands of the toad. They have "been used for centuries as ingredients 
of drugs by the Chinese (as Ch'an Su; in Japan as Sense) and in their 
pharmacological action and chemical structures are closely related to 
the cardiac poisons obtained from plants. In small amounts a typical 
emetic action is observed, while larger quantities are fatal. The 
principal toxin of the toad Bufo "bufo hufo will induce systolic cessa- 
tion of the heartbeat in the cat in doses of 0.3 mg. per kg. 

The crude poisons may be extracted from dried toad skins by 
alcohol, or obtained by expression from the parotid glands of live 
toads. The toad suffers no injury by this method. Cholesterol, 

-sitosterol (a plant sterol), adrenaline, a series of tryptamine 
derivatives, and the poisons proper are present in the crude material. 
Older methods for the separation of the poisons involved a series of 
treatments with different common solvents, but later methods have em- 
ployed these in conjunction with chromatographic absorption. The 
tryptamine derivatives have been investigated chiefly by V/ieland with 
considerable success. The cardiotonic constituents , which are sterol 
derivatives, have been studied extensively, but there is some confu- 
sion as to the relative identities of a number of compounds isolated 
by different authors. The best known components, bufotoxin and its 
derivative bufotalin, were studied by Wieland, and the structure is 
now known (as of 1936) with the exception of a double bond position. 
The structure of gamabuf ogenin has now been determined by Kotake. 

Gamabufogenin was first isolated in 1928 by Kotake from 5000 toad 
skins (B« vulgaris formosus) in 35 g. yield and given the name gama- 
bufotalin. The elementary formula C27H53O3 was determined and a 
number of derivatives prepared. Among these were the diacetyl and 
anhydro compounds, and the diformyl ester. Wieland in 1930 isolated 
gamabuf otoxin and gamabufogenin from Japanese toad skins and studied 
them. Gamabufotoxin was found to be a conjugate of the genin with 
suberylarginine, and the formulas C2QHgo*-*l0^4 ^^^ ^27^35^5 found. 
The nuclear ring system was taken as the sterol structure, since ex- 
periments on other toad poisons indicated this. Chrysene and Diels' 
hydrocarbon have been isolated in dehydrogenation studies. X-ray 
data also supports this model. A lactone ring was found, as were two 
hydroxyls which could be acetylated to give a compound very similar tc 
that obtained by Kotake under the same conditions. The toxin on mild 
hydrolysis yielded an anhydrogamabuf ogenin (a) which was not the same 
as that obtained (B) by the action of strong HCl on gamabufogenin, but 
could be transformed to the latter by HCl. Hydrolysis of the toxin 
under stronger conditions gave the compound B directly: 

C34H3505-0C0Ci3H2503lTi ^ - 'l^^^^^ > Ca^HagO^ 

toxin neaT/ anhydrogenin A 



hea?r--^aN HCl 



cone. HCl 



V 



nun cone. HCl -~.^,^ nun 

^^ai-nst^s — — — >^^^"^ ^^^■"aaOi 

genin anhydrogenin B 



\-:-PV 



*^i'.'- \- 



X-. 


















'(id vjUr: ,^i'v' 






If -1- M * .. V' t \ . 












an^^i.-^r'v 



V- ',. .• '■ 









•.I-,:-. 









■ ■':■ y-^ t/^ti v>7^ ^ii. 



f c. -•* i '1 



.T.'Ky: f.<f. 



:> ; J V. 



U ! 



i^'ji^'-i;-^- 



- ^^ .ftf■^^^■v 



-2- 



The reasons for this "behavior are not known. Suheric acid and 
arginine were the other products of the hydrolysis. On hydrogenation 
both the anhydrogenin and the genin itself ahsorhed two moles of 
hydrogen, although the former should contain three double bonds, 

Kotake accepted Wieland' s analyses and molecular formulas, but 
retained the name gamabuf otalin, and proceeded to work out more 
reactions. It was found that treatment of the genin (l) with b% KOH 
in CH3OH, followed by acid, produced an alkali insoluble product (ll) 
not containing an aldehyde group, but when this product was treated 
with dilute KOH, again followed by acid, aldehyde tests were obtained 
(ill). Alkaline saponification of (ll) or (ill) produced an acid (v). 
These reactions are typical of the side chain lactone shown when an 
hydroxyl group is present on C14, and have been studied extensively 
in other series. This same side chain is also present in bufotalin, 
and is apparently closely connected with the physiological activity 
of the material. 



CH 



CH=CH-CO 
I t 
C=CH-0 




OH 



CH=CH-COOH 



r C=CH 




CH=CH-C00CH3 
C=CHOH 




CH 




CH^CH-COOCHo 
I ^ 

CH-CHO 

III 




OH 



CH^CH-COOCHo 
I ^ 

C==CH 

II 



CH 



CH=CH-C00CH3 
C=CHOH 

IV 



n 



OH 



The ozonization products were also investigated. Diacetylgama- 
bufogenin yielded formaldehyde, glyoxylic acid, and a new acid called 
by Kotake diacetyletiogamabuf otalinic acid (VIl). This seems to con- 
firm the side chain structure, and the work of Wieland does not con- 
flict with this. 



VI 



CH 



CH=CH-CO 
C=CH-0 




OH 



■> 



HCHO 

+ 
CHO 
COOH 



CI 



COOH 
I 

c=o 




■> 



OH 



CH 



COOH 



OH 



VII 



Treatment of the genin with cone. HCl yields the anhydrogenin 
(VIII). This may be hydrogenated to two products, IX, and a compound 
X which is taken to be a dihydrocholanic acid, although there seems to 
be no particular evidence for this. The formation of the anhydrogenin 
agrees with a Ci4,-0H. The two methyl groups are placed in the usual 
sterol positions, but it remains to place the two secondary hydroxyl 
groups. One is placed at C3 by analogy with other sterols and since 
the Cii-OH acetylates with some slight difficulty while the C7-OH 



^i>::. ■}'■, 



.-r.^ ... -' 



-rito 



■'■ •«■>'. 



(f:i: 



'"-i^ I 



^01) 



t.,. 



4- 



; r • 

i' '-J 



I :; 



J 

-1. 
r 



::'Vh. 



U.4., 



:f;j;^.:d.;:>'r 



- -■;■!■■ • - / . 
- ■ '.: ••- * 



■i\n<;TlT 



I J_ - ..r'rl'W 



lio 



-3- 



acetylates easily, the remaining hydroxyl is placed in the 7 position, 
the final formulation "being XI, The Haramarsten reaction apparently 
precludes the possibility of a C13-OH. The position of conjugation 
of the suberyl-arginine remainder in the toxin is not knovm, hut it 
very probably involves a tertiary hydroxyl group. 

A number of other toad poisons have been isolated, but with the 
exception of bufotalin and cinobufagin (XII) their structure and 
identity is still fairly uncertain. 



VIII 




CH=CH-CO 
1 I 

C=CH-0 



rirj CHCHo-0 

,5^ 



CH2CH3COOH 



IX 



(.jj CH-CH3 



X 




CH=CH-CO 
I I 
C=CH-0 



CH=CH-CO 
I I 
CHgCOO^y C=CH-0 




CH=CH-CO 
CHgCOOpu C=^CH-0 




XI 



XII 



Bibliography ; 

Kotake et al , Ann., 46^, 1 (1928); ibid., 165, 11 (l928); Sci. Papers 
Inst. Phys, Chem. Research (TokyoT7l, 99, 108, 223 ( 1928) 5 
ibid., 24 39 (l934); ibid., ^2, 1, 79 (1937); ibid., 34, 
824 (1938). 

Wieland et al . , Ber., 46 3315 (1913); ibid., 55, 1789 (1922); ibid., 
64, 2099 (1931): Sitzungsber. d. Bayr. Ak, Wiss., 329 i 1920 
Ann,, 481, 215 (l930); ibid., 493, 272 (l932); ibid., 513, 
1 (193TT7 ibid., 507, 22 ( 1935T7~ibid, , 5^, 203 (1936T: 
ibid.^ 528, 234 (l937). 

Tschesche et al., Ber., 68, 1998 (1935); ibid., 6^, 2361 (1936), 



Reported by E. C, Horning 
March 8. 1939 






■"■ " ;,- t' •' 



'> -:^ 'U . 



, -T !'/•'=■"''■■■"■ ■ • "•T?-.'»-S' v,^<( -M, i;->: •.. ■ 



\'*l-^'^ 



\b"^:it T.F •:":-''.'.■,'? '^ .s. it. 



C?>jO 






i 



<.- 






■<-«■ 



.«: . • ~. i*> 



^U" 



-■r-..^r^-.l-- 



,-^v^ 



;■ T 1 .^. 

t ■ 1 7 



!<^-. 



?HE FORMATION OP SULFUR- CONTAINING RINGS 



G.M.Bennett -- Sheffield, England 
Mliller and Schlitz -- Vienna 



Investigations into the nature of cyclic sulfides were initiated 
"by Mansfield in 1886. He obtained (5) a trimer of thiof ormaldehyde 
and also ethylene and propylene polysulfide polymers believed to be 
cyclic and to have compositions corresponding to (C2H^)3S3 and 
(CaHgJsSs. Von Braun's studies { \) of polymethylene monosulfides 
placed especial emphasis upon the nature of ring closures involving 
the sulfur atom. P. C. Ra,y and his students in Calcutta have made 
contributions to the knowledge of polymethylene polysulfide rings (7). 
Since 1927 Bennett has studied extensively the penthiane (pentameth- 
ylene sulfide) (2a) series. He has also measured (2d) the rate of 
formation of various sulfide rings and has begun recently the study 
of large ring monosulfides (2e). His work has been supplemented 
consi-derably by that of Mliller and Schtitz (6) in Vienna. Emmet Reid 
investigated the cyclic and chain polymers derived from ethylene mer- 
captan and a number of polymethylene halides (8). 

In general the ease of formation of sulfur-containing rings 
varies with the size of the ring in the same manner as does the ease 
of formation of alicyclic compounds. As would be predicted, the 
cyclizations are accompanied in almost every case by linear polymeri- 
zations. The stereochemistry of sulfur rings appears to be normal. 
Bennett has found, for example, that 4-aryl penthianol sulfoxides 
exist in cis and trans forms (2a). 

Monosulfide rings have been made by the following reactions: 

I. (CH2)n + (NaHS + NaOEt) (dry) > (CH2)nS (3) 



■X 

+ KaS or NagS 

n = 3,4,5,6,12,13,14 



(CH2)j,S 



(6) 



X = CI, Br, I 



The structure of the monosulfide rings has been demonstrated 
by the following cleavage (6): 

S (CH2)nS ^^ ^^3^> [(CH2)nSCH3]l + l(CH2)nI + [(CH3)3s]l 

l(CH2)nI + 2 CeHsONa > CelhOiCEs) Y^OCelh 



II. 



(CH2)^Br 
( CHa )^^Br 



n = 1,2, or 3 
m = 0, or 1 




Na2S 




(CHa) 
( CH2 ) 



rN, 



m 



5, 6, or 7 membered rings 
were formed 



Two factors should influence the yield of a ring closure of 
this type ( 4b) : 

a. The size of the ring. 

b. The position of the sulfur atom° i.e. attached directly 
or indirectly to the aromatic ring. 



■fUTTioo-fT' ■■ "0 :lcsi?a:^c 






;.rrJ. 



/ : ;ii. ■ I .; 



•jHOtt: 



'. J.'.' 



:iui 






'T? 



ivt 






-2- 



Von Braun found factor Id to be the more important (41d). When 
m = 1, the yields approximated the theoretical while when m = they 
were ahout 40^. 



)H 



III. S-(CH2CH2C00Et)2 



NaOEt 



3H?. 



^ 



C-COOEt 



or NaNHg CHs .CH2 



1 week 







CH2 

I 

CH2 



"CH-COOEt 

I 

,CH8 



5^ KOH 







CHg 

I 

CH2 



,CH8 



The penthianone and the /3-keto ester may serve as starting 
materials for many syntheses involving the penthiane nucleus (2a). 
The phenyl and ^-bromophenylhydrazones of penthianone rearrange in 
hot acetic acid to the penthiene-indoles. 



.CH 



K 



L 



"CHj 



2 /C^ 

CH2 N-M 




HOAc 




IV. [CeHsCHgO-lCHs) J 



2^ 



fuming 
Sr 



CH2GH2 

I >-(CH2)4-0H 

CHgCK^ 

heat 

CH2CH2^ 



Br 



CH2CH2 



y 



This reaction (2b) illustrates the extraordinary ease of 
formation of five membered rings. V/ith C6H5CH20-( CH2 )n aS, where 
n = 3 or 5, fuming HBr leads to the straight chain dibromoalkyl 
sulfide. 



V. Cl(CH2)n-S-R 



heat 



dil. soln. 



[(CH2)n-S-R]c: 



heat 



(CH2)^S 



-I- 



RCl 



n 



4,5,6,14,16 



R - CH3 ,C2H5 jCeHg 



Bennett has shown ( 2d) that the rate of ring closure is decrease 
by a factor of approximately 75 as the size of the ring is increased 
from five to six to seven members. A change of R from CH3 to C2H5 
diminishes the rate by one-third. When n = 14,16, boiling aceto- 
phenone was used as the solvent. The addition of Nal was found to 
be advantageous. The alkyl iodide is formed slowly; the ring is 
closed; and the CH3I is liberated. Bennett has suggested (2e) a 
mechanism to account for the various cyclic and linear products ob- 
tained in reactions of type V: 



■•'■iy i .. 






1 <- 



.''^■,>^^'v; 



.^{i^^ 



:^ ;•:•.'•• ;).: +5d 









CHg 


-S-PX :^; ^ 




1 




-CH3 X 

k 

+ CH3X 



CH3-S-PX 

PSCHg 



-5- 
X V ^ XP-S-PSCH3 



+ CH^X -S^^Mz^etc. 



W . 






X ^ ^ S^ S + CH3X 

p^ 



P = (CHs)^; X = CI, Br, I 



VI 



N2CI 



-N, 



(4a) 




»^ 



CH2 



,/ 



H2 
H2 



VII. S(CH2CH2C1) + H2KR 



n = C6H5,CH3,G2H5, etc. 



80-85^ 

CH2CH2. 
CH2CH2 



¥-R 



Reference 8a gives a comprehensive account of the reactions of 
''mustard gas" . 

Di- and polysulfide rings have been prepared "by the following 
reactions: 

.CH2-S- CH-S- CHg^ 

VIII. C6Hi(CH2SH)2 + ArCHO -h- — > CgH^ J^gH^, 

CH2~S-CH— S- CH2 
2." sS~ J and £- Ar 

Cis - trans isomerism, contrary to prediction, could not be detected 
in the products of this reaction (l). 

R^ R 
g J(f- g 

IX. (CH2)5(SH)2 + acetone or ^ ( CHg ) 5"^ ^(CH2)5 

diethyl ketone ' ^S-^C^-S'^ 

r" R 



,/' 



X. HS-C2H^-S-S-C2H4-SH + CgHgCKCls 
(7) 



/S-C2H4-S 
S - C 2 H.j. - S 



Ce-HsCH 



/ 



S-C2H4^ 



^S-C2Hi' 



/ 



I 



' •' •'••■ ar 



t. 



I 



<1, 



<::.- 



i J 



fiV 



-4- 



XI . R( SC2H4 l^-S-CaH^-S-CsH^-Br 



/CH2CH2. 

5^ ^-1 C2H4S) j^R 
CH2CH2 



CH2CH2 
S ^S 

CH2CH2 



R'Br 



Only ethylene sulfide polymers containing halogen decompose in this 
way (2c). Those containing no halogen will give dithiane when heated 
with HBr. They may have as end groups CH2OH or CH=CH2. 



XII. BrC2H4Br + KSH ,„... » .^ 



•^ s; 



^GH2CH2^ 



CH2CH2 

XIII. NaSCHsCHaSNa + (CH2)nBr ^ ( CH2 ) 2Sa( CHg )^ 

n = 1,2,3,4,5,6 and/or j( CHg )2S2( CH2 ) J 2 

The 8 and 9 membered rings (n = 4,5) are not formed. The dimer is 
obtained in these cases. Both dimer and monomer are found when 
n = 3 (8h) , 

BihliograDhv ; 

1. Autenrieth et al., Ber., 11, 4249 (1908); ibid., 12, 4346, 4357 

(1909). 

2. a. Bennett et al . , J. Chem. Soc, 1927 ^ 194; ibid., 1929 , 2829, 

2832. 

b. Bennett and Hock, ibid., 1927, 477. 

c. Bell, Bennett and Hock, ibid., 1927 ^ 1803. 

d. Bennett, Heathcoat and Mosses, ibid., 1929, 2567; Bennett and 

Tumor, ibid., 1938 , 813« . 

e. Bennett and Gudgeon, ibid., 1938 . 1891. 
Bost and Conn, Oil and Gas J., ^, No. 3, 17 (1933); C.A. 27, 

5323 (1933); Chem. Zentr,, 33II . 2673 (1933). 

a. Von Braun, Ber., 13, 3220 ( 191071 

b. Von Braun, ibid., 58, 2165 (1925). 
Mansfield, ibid., 696"Tl886). 

6. Miiller and Schtitz, ibid., 21, 692 (1938). 

7. Ray et"al., J. Chem. Soc, 125, 1141 (1924). 

8. a. Helferich and Raid, J. Am. Chem. Soc, 12, 1208 (l920). 
b. Tucker and Reid, ibid., dd, 775 (1933). 



Reported by S. C. Kelton 
March 8, 1939 



i ^ — 



'X^i- --a-^-'v-^'-s--' ■ 



1 =» 



'^:- T Ai !\,j i^. 









REARRANGEMENTS OF ARYL SALICYLATES AND 
COMPOUNDS OF SIMILAR CONSTITUTION 

Smiles -- King's College, London 



The accidental discovery "by Warren and Smiles in 1930 that 
l-aulfhydryl-2' -hydroxy-2,1' -dinaphthyl ether (l) was rearranged by 
heat or alkali into 2,2' -dihydroxy-1 ,1 ' -dinaphthyl sulfide (ll) led 
Smiles to investigate the possibility of rearrangements in compounds 



9x 

^-^^OH HS 




9x 

\-^0H 



OH HO 




II 



of similar constitution. Smiles concluded that the above conversion 
was the result of an intramolecular change, and after numerous studies 
has found that many compounds of the general type III can undergo such 
rearrangements to give IV. All of these conversions were effected by 



■■' a:x) 



^ a:-o ■' 



heat and alkali, or, in some cases, by heat alone 



The results of these studies are summarized in the following 



X 


mav t 


)e: 








S02 


,S0,£ 


, or 









S02 


,S0, 


but ] 


act S 




S02 


,C02, 


but 


not 


SO 


or 


S02 


,(so? 


),o, 


but 


not 


S 




























0,S02 
























table 

If YH is ; 

1. NHAc 

2. OH (alkyl) 

3. OH (aryl) 

4. NH2 (aryl) 

5. SH 

6. CONH2 

7. CONHAc 

8. S02NH2,S02NHR 

The chief conditions controlling these rearrangements are as 
follows^ 

1. The positive character of the carbon atom c (see figures III 
and IV), An increase in the positivity of c should favor the conver- 
sion, not only by lessening the stability of the linkage with the 
positive group X, but also by increasing the demand for the electron 
supply offered by Y. 

2. The character of the medium in which the rearrangement is 
effected, as expressed by the tendency to remove a proton from YH. 

3. The character of the YH group, as shown by the instability 
the electron system of Y or its capacity to act as a donor to meet 
demands of the positive carbon atom c. 



of 
the 



Although all of the above rearrangements have involved mainly 



^"^"■s^ir. 



. :'Y7ii ;;o ■■'■■^■: ma^kjii^:. 



irrV-.:. 



'.■■ ■} 



-2- 

aromatic nuclei, with the exception of type 2 of the ahove table, it 
should be mentioned that at least one other example has been found in 
the aliphatic series. Thus ^-nitrobenzenesulfonylacetanilide V 
yielded the sulfinic acid VI. 

CH2SO2C6H4NO2 CH2SO3H 

V I > I VI 

CONHCgHg CO-N-C6H4NO2 

Probably the first known conversion of this general type was 
the observation by Blittcher in 1883 that the reduction of 0-benzoyl- 
^-nitrophenol yielded not the corresponding amino compound, but 
N-benzoyl-^-arainophenol . More recently it has been found that 0-acyl- 
salicylamides are transformed by heating above their melting points 
into the corresponding N-acyl isomers. It is interesting to note 

aCOJfflg ^..^^CONHCOR 

OCOR k^OH 

that this reaction is reversible, the N-acylsalicylamides being con- 
verted into the 0-acyl compounds on boiling with glacial acetic acid. 
No reversibility has been observed in the preceding rearrangements. 

In conclusion, it may be stated that these reactions probably 
have very little preparative value, although fairly good yields v/ere 
obtained in some cases. The work in this field is being continued. 

Bibliography ; 

Smiles et al . , J. Chem. Soc, 1930 ^ 956^ ibid., 1951 „ 914,2207,3265; 
ibid., 1952 . 1«40,1488,2774 5 ibid., 1933, 1490; ibid., 
1934 ,. 184,422; ibid., 1955 , 181; ibid., 1956 . 329; ibid., 
1958 . 899,1897,2052. 

Roberts and de Worms, J. Chem. Soc, 1954 „ 727; ibid., 1955 . 196,1509. 

Anschlitz, Ann., ii2, 19 (1925). 

Titherly et al . , J. Chem. Soc, 87, 1207 (1905); ibid., 89, 1518 ( 1906 

Bftttcher, Ber. , 16, 629 (l885). 

Graebe, ibid., ^, 501 (I888). 

Ransom, ibid., ^, 1060 (1898); ibid., 55, 199 (l900). 

Auwers, ibid., ^7, 5905 (l904). 



Reported by S. L. Scott 
March 15, 1959 



:.::. 'n\ 



ruf: 



■ .' '■ 



■boh 



Ti- 



THE STRUCTURE OE ROTTLERIN 

Brockmann and Maier -- G'6ttingen 
Robertson -- Liverpool 



Rottlerin, C^qH^qOo, is the active anthelmintic of the drug 
"kamala" obtained from the fruit glands of the Oriental Mallotus 
phillipinensis ; in India this reddish-hrown pigment is also used as 
a silk dye. The structure (l) has been deduced from the following 
reactions : 

A. Formation of pentaacyl or pentamethyl derivatives. 

B. Absorption of two mols of hydrogen to give tetrahydro- 
rottlerin (II), a much more stable compound. 

C. Inner condensation upon heating to isorottlerin (ill) of the 
same composition, 

D. Scission products of rottlerin, tetrahydrorottlerin and iso- 
rottlerin. 

1. Warming with weak alkali. 

a. Rottlerin gave C-methylphloroglucinol and 
rottlerone ( IV) . Dilute sodium hydroxide gave some benzaldehyde also; 
zinc dust and dilute potassium hydroxide gave C-dimethylphloroglucinoi 
and other scission products. 

b. Tetrahydrorottlerin gave G-methylphloroglucinol and 
tetrahydrorottlerone (V). 

2, Heating with concentrated alkali. 

a. Rottlerin gave phloroglucinol , acetic, benzoic, and 
cinnamic acids. 

b.^ Tetrahydrorottlerone gave only p-phenylpropionic 
acid and 5 ,7-dihydroxy-2,2-dimethylchromane . 

3. Heating rottlerin with diazoaminobenzene gave 3-acetyl- 
5-methyl-2,4,6-trihydroxyazobenzene (Vl) . 

4, Heating in vacuo: rottlerin and isorottlerin both gave 
2-methylphloroacetophenone. 

E. Miscellaneous 

1. Isorottlerin absorbs only 1 mole of hydrogen with 
palladium catalyst. 

2. Ozonization of rottlerin gave 0.9 mole of benzaldehyde; 
isorottlerin gave no benzaldehyde. 

3. Oxidation with ozcne-KMnO.^ gave 0.2 mole of acetone from 
rottlerin and 0,48 mole of acetone from isorottlerin. 

4. Oxidation of rottlerin with MlnO^ gave 1 mole of benzoic 
acid. 

5. Alkaline hydrogen peroxide gave cinnamic acid. 

6. Other oxidative degradation products are acetic acid and 
succinic acid. 

The structure assigned is 6-( 2,4 ,6-trihydroxy-3-methyl-5-acetyl- 
benzyl )-2,2-dimethyl-5 ,7-dihydroxy-8-cinnamoyl- /^3-chromene ( I) . 



CHo 
t=0 



CH— CHC gHg 

c=o 




v^'j ^g(cH3)2 



CH; 






ICH 



Rottlerin 



I) 



CH3 

t=^0 c=o 



CH2CH2C5H5 



«°fS°"«°rV'^C(CH3)a 
CHg-SAcHsAA /«^ 
OH OH ^2 

Tetrahydrorottlerin (II) 



s ■" 






■ i^q 



-VOi"T» r;'*. 



c'f 



.1: 



i-^Ji; 



?J 



« ^1 



-' -f 



■rT^ 



-?.- 



On this "basis the transformations indicated alDove may "be repre- 
sented "by the scheme shown "below. 

CH3-C=0 
CHg CgHs^CHa^ heat ^ HOv''^OH Cf,H.,-N=N-NHCpHc 



C=0 'CH 



C=0 



.0, 



Hcy^v^oH -o^.-V -C(CH3). 



CHg'X^ CH2 --%A ^^H 



vac 



OH 



■3J±Sl 



'e^^5 



OH 



heat 



2H: 




CHg-C^O 

HOy^OH ^ 



CfiH 



6 ^^5 



OH 



VI 



0=C-GH=CHC6H5 
CH 



Zn"\dil. 
dust\ NaOH 



Pd 



HOy-'^OH 

CHg^'^^V^CH. 
OH 




IV 



II 



weak alkal i ^ HO.^^-"^OH 

CHg^'^r^ 
OH 



warm 



CH2CH2C5H5 
6=0 „ 

H0^>Y^C(CH3). 



OH 



CH^ 



.(^H2 



V 



CeHsCHaCHsCOOH 



cone, alkali, heat 



^^^r'^V ^^ClCHg) 



OH 



3 ) 2 



CHc 



CH 



/ 



Boehm found that heating "bismethylphloroglucinol or similar 
compounds with sodium hydroxide, or sodium hydroxide and zinc dust, 
cleaved them at the methylene "bond. For example: 



HO^^-^OCHg CH3O, 

S^x>^--CHj 

OH OH 




OH 



NaOH 



Zn dust 



HO 






OCH^ 



* V 



CH3O 
CH 




OH 



OH 



Further, any phloroglucinol nucleus at 
methylene group yielded phloroglucinol 
analogy, then, I should give 3-methyl- 
by cleavage at oc and /3 . However, Bro 
2 hours heating with 2N sodium hydroxi 
from phloroacetophenone. This, then, 
the methyl- and dimethvlphloroglucinol 
tetrahydrbrottlerin with alkali. The 
ene group of the bisphloroglucinols is 



tached to other structures "by a 
or methylphloroglucinol . By 
and 3,5-dimethylacetophenone 
ckmann and Maier have shown thai 
de removes the acetyl radical 
explains the isolation of only 
upon heating rottlerin or 
ease of cleavage at the methyl- 
illustrated by the fact that 



: T. 



■^-w 



I 



V- 



/.\ i,.-.U:. Ki 



;'-l 'i i'- 1 "^.0'* '• > O-ritn i '■' '. •>'. .; 



-3- 

warmiMg them with diazoamino*benzene gives the trihydroxyazolDenzene in 
good yields. Hence, formation of VI under these mild conditions 
proves, beyond reasonable doubt, that the acetyl group is present in 
the original rottlerin. That it was not a larger acyl group, which 
could scarcely be decided by elementary analysis on such a large 
molecule, was indicated by the excellent agreement of the absorption 
spectra of synthetic 2,4 ,6-trihydroxy-3-methyl-5-acetylazobenzene with 
that of VI. In addition a mixed melting point showed no depression, 
while one with VI and the azo compound derived from synthetic methyl- 
phloropropiophenoiie showed a depression of 10°. 

Since ozonization of I gave essentially one mole of benzaldehyde 
while similar treatment of II gave no benzaldehyde, a benzylidene 
group was inferred. That this is part of a cinnamoyl residue v/as in- 
dicated by the fact that hot concentrated alkali upon I gives cinnamic 
acid whereas rottlerone gives /3-phenylpropionic acid. Warming 
rottlerin with dilute sodium hydroxide produces some benzaldehyde 
(Perkin); such behavior is characteristic of hydroxychalcones . 

The behavior of tetrahydrorottlerone with hot concentrated 
alkali strongly supports the structure V; the 2,2-dimethyl-j^3- chromene 
unit has been found in a number of other natural compounds, particu- 
larly those of the rotenone series. 

There are two lines of reasoning in deciding between the struc- 
tures I and VII. 

A. Brockmann and'Maier 

Isorottlerin is undoubtedly an hydrogenated flavone 
formed by the addition of a phenolic hydrogen atom to the oc-position 
in the cinnamoyl residue (see VIIl) since only 1 mole of hydrogen was 

CH3 (CHaJa 

C=0 /C-CHv CgHs^ /CHs^ 

HO>X^OH ^CH VII CH C-0 



OH \ { 



(Robertson) 



^0 




HO 0=0 VIII 



'6^^5 



easily absorbed, ozonization gave no benzaldehyde, and sublimation 
gave 3-methylphloroacetophenone, Yet, if the structure were that of 
VII there would be two isorottlerins whereas only one was isolated. 
B. Robertson 

Both rottlerone and tetrahydrorottlerone are sparingly 
soluble in dilute aqueous sodium hydroxide. Since 2,2-dimethyl-5 ,7-' 
dihydroxy-8-acetylchromane is readily soluble in 4^ sodium hydroxide, 
one might logically infer that the cinnamoyl group is then in the 
6-position. 

The argument of Brockmann and Maier is probably preferable. 



-4- 



Eibliographv ; 

Brockmann and Maier, Naturw. Umschau. Chem. Zt., 25^ 460 (l937); 

ilDid. , 26, 14 (1938) . 
McGookin, Reed and Robertson, J, Chem, Soc, 1937 ^ 748. 
McGookin, Percival and Robertson^ ibid., 1958 „ 309. 
Boehm, Ann., 302, 187 (1898); ibid., 318 ', 230 (l90l); ibid., 329 

269TT903) . 
A.G.Perkin, J. Chem. Soc, 6Z. 975 (l893); ibid., 62, 230 (l895) 

ibid., 75, 443,829 (1899). 
Dutt, ibid., 127, 2044 (l925). 
Shriner, J. Am. Chem, Soc, 5^, 2538 (1930). 
Wacek and David, Ber. , 70 j 190 (l937). 



Reported by R. 0. Sauer 
March 15, 1939 



THE "OXIDIZING ACTIOlf' OP ALICALIES 

Lock -- Vienna 

E. E. Reid -- Baltimore, Md. 

I 3 

In 1832, Wflhler and Liebig obtained benzoic acid from benzal- 
dehyde on fusion with potassium hydroxide. A few years later Dumas 
and Stas^ discovered the reaction: 



CH3CH2CH2CH2OH + NaOH ^^^^^^ > CH3CH2CK2Cf=0 + 2 H 



2 



They found that the above reaction is best for high molecular v/eight 
alcohols; e.g. they obtained palmitic acid from cetyl alcohol by 
sodium hydroxide fusion at 250 in good yields. V/6hler and liebig 
formulated their theory, whereby organic compounds under the influence 
of alkalies take over the oxygen from the water v;hich is present, and 
are oxidized v;ith the liberation of hydrogen, which comes jointly from 
the water and the carbon compound. Hence: 



C2H5OH + H2O ^^Q^ > CHjCOOH + 2 H 



2 



This old theory was accepted until 1918, when Carrol'^ introduced 
his mechanism for the conversion of alcohol to the salt of acetic acid 
in the presence of soda-lime. The reaction takes place in three 
stages; the first stage is the dissociation of alcohol to aldehyde 
and hydrogen in the presence of caustic soda as catalytic reagent: 

1st stage: C2H5OH > CH3CEO + H2 

2nd stage: CH3CHO + caustic soda ^ CH2=C=0 + H2 

3rd stage: CH2=C=0 + NaOH > CHsCOOIa 

The intermediate ketone vms not proven to be present. 

Ery^ , who studied the alkali fusion of a series of aliphatic 
compounds, concluded that the reaction involved the acidic dissocia- 
tion of the alkalies and the replacement of hydrogen or methyl radical 
by -ONa radicals. 

HONa > H- + -ONa 

In the case of the oxidation of hydroxylated toluene, benzyl 
alcohol or benzaldehydes the usual oxidizing agents fail to v/ork. By 
their use no acid is obtained while a stronger oxidation causes the 
destruction of the benzene ring. The reason for this phenomenon must 
be due to the greater resista ce of the functional group to the oxidi- 
zing agents in the presence of a riydroxyl group on the ring, and also 
to the^ lessened resistance of the liydro^q^^-carboxylic acids against 
oxidizing agents. 

Lock'° studied the fusion with alkali of these same compounds. 
He found that potassiura hydroxide reacts quite rapidly even at 105- 
110 with salicyl aldehyde converting it o^uantitatively into salicylic 
acid with the evolution of one mole of hydrogen. The m- and ^-hydroxy 
compounds on fusion decompose similarly into acid and hydrogen. Unlik. 
the ortho derivative the m- compound begins to react only at about 



:.:i:IwIa:iia ■'lo "uouok o:iio:iGixo" zv 






■■:>l1 dub t«I:I/^W . .;-jiX3XV. doc^ nl 0^'^ :r£i ;TO.U;i:n; (^j;).i;v:oT5 v,n rru/xJliOo 
. .;.:■•:.{: !?:-{t tei' ' - .--,-■...:...■. ■. j,-3-i-ro ;/;c-'--.-^:<:/ ^^^.;j;^;„;.-{:j -rj^.r;:?- ■yhBij.jic:^<yi 



■:r-->„ tfJ-^ 



40:.H,-.0 



-..ii: 



vlcvrii; rioxJ-OBOT '■>d-r tB/fc^ 5s5j;Iari:oo , cb/u;3i•|^rTr:;-■ 



i^!J 






r'^ /' 



<..; ... , 



•.1.0 i;<,o 1X1:^8 iii-iv^? ^ 

';,- ■ _..i - ■ —'Lovh '>;(;■?■ dj t.s '. 

xoiS otPJ: \;,Xt>s Cj;'rsc: r. >i«j.;'j: n("; e-'^n^'o^' 

'':':■'■ •■■:. .,- : ;/ •/ ■■■/•.ia x'r£ :' 



-2- 



o 
190 , due probably to the tendency at lo\-v temperature for , 

jsaro reaction to take place and at 190 the potassium hyd: 



Cajiniz- 
potassium hydroxide 
reacts to oxidize the alcohol. That there is not first a CEinnizzaro 
reaction with the ortho compound was indicated by the fact that 
saligenin, as well as the corresponding meta and para alcohols, and 
potassium hydroxide do not react below l6^'^ . The old rule that a 
phenolic hydroxyl group prevents the disproportionation into alcohol 
and acid, holds, therefore, only when the hydroxyl group is in the 
ortho and para position to the -CHO group. Substituted m-hydroxy- 
benzaldehydes gave similar results. 

Lock found that on the fusion of £-H0-C6H4.-CH0, 3, 4- (OH) 2CGH3CHC 
and vanillin (I), hydrogen is evolved at 110°. Isovanillin (II) does 
not evolve hydrogen below 190°. 



CHO 



CHO 



II 





OCH- 



Jacobsen reported in the literature the following fusions : 



CH3Y.--v^CH3 CH3>.^''^C00H 

\>0H \^0H 

m-Xylenol p-Homosalicylic acid 



0CH3 ^--s^COOH 
OH CHj-k^-OH 
jD-Xylenol m-Homosalicylic 

acid 



CH 



^^^C3H-7(n) ^--^COOH 
j-^X^-AdH CH3'\x^0H 



(n)C3H 



aCH3 ^- COOH 
OH (n)C3H7J\^H 



Prom these data, he gave the rule that the side-chain nearest the OH 
group is the one that undergoes oxidation in alkali fusion. However, 
Barth' showed that the methyl group is the one which is oxidized in 
thymol : 



,.0"' 



..^--v^COOH 



;Ii7"^^ 



OH 



OH 



Lock found that when he fused £-cresol with alkali in a nickel 
crucible, a rapid reaction occurred at a temperature of 300-310° and 
he obtained a 50% yield of salicylic acid in one hour and about 80% 
yield after five hours. Carrying out the same procedure in a closed 
silver crucible in an atmosphere of nitrogen, he obtained neither 
hydrogen nor salicylic acid and completely recoveied o-cresol. In a 
current of air and the same apparatus, hydrogen and salicylic acid 

Some of the salicylic acid had undergone CO2 splitting 



were obtained, 



The mechanism of the above reaction is similar to the mechanism 
of the oxidation of the substituted bcnzaldehydes and benzyl alcohols; 
that is, a hydro:-:ylation of the methyl group with a simultaneous 
dehy drogenation . 



"""] 



'V^.. 



S^ 



1 1 ' 



-v^. 






-■' ;>^ 



-^» 






-3- 

KO-CqH^-CHs + 3 ROK > K0-C5E4-C(0K)3 + 3 H2 

K0-C5H4-C(0K)3 + HOH > K0-CeH4-C00K + 2 KOH 

KO-CgH^-CHgOK + 2 HOK > KO-C6H4-C (0K)3 + 2 lig 

KO-C6H4-CHO + HOK > KO-C6K4-COOK + H2 

Alkali fusions of the sodium and potassium salts of benzoic 
acid shov; a tendency of splitting off COg as the temperature is 
raised; thus at 400 sodium benzoate v^as 100% decomposed into CC2 and 
benzene . 

Toluene and mcsitylenc were recovered unreacted even at tempera- 
tures as high as 500° . 

The fusion of benzyl alcohol gave toluene along with the 
benzoic acid. This is explained by: 

CH2OH ^^-v^COOIC 

+ KOH _-» [ T + 4H 




O 





CH2OH ^^--v^CH- 

+ 2 H -^ [ I + H2O 
nascent 

With concentrated sodium hydroxide, Maier obtained with the 
three nitrobenzaldehydes v/hcit he believed v^/as a mixture of azo- and 
nitrobenzoic acids. Lock showed that azobenzoic acid is formed from 
the o-nitro compound 

cooH coon 





and the corresponding azo:<;y compound from the meta and para . He 
proved the latter by synthesis. Similarly'" o-nitrobonzyl alcohol gave 
the azobenzoic acid while the corresponding m- and £- gave the azoxy. 
It is a two step reaction: (1) Dehydrogcnation of the -CH2OH group; 
(2) Reduction of the nitro group by the liberated hydrogen. 

In the study of the reaction of halogen substituted benzyl 
alcohols, here again he found a two stage reaction whore the first 
stage is the same as above and the second is the substitution of the 
halogen b/ the liberated hydrogen, 

Sodiuiii hydroxide and lithium hydroxide react in the same way as 
the potassium hydroxide, but higher temperatures are required. Even 
barium liydroxide dehydrated at 120° can effect oxidation. KaNHg 
could not be made to react with the compounds. 

In contrast to the aromatic methyl compounds, the aliphatic 
methyl compounds give methane, as shovm by Pry^ : 

R-CH3 + KOH — > ROK + CH4 






^^' 



'^. ^'" 



J 



-4- 

Hc shows that ethyl alcohol, acctaldchyde and acetone fused v/ith 
potassium hydroxide under mild conditions give the salt of acetic acio 
but by more drastic action the methane is obtained: 

CH3CHO + KOH --> CH3COOK + H2 

CH3COOK + KOPI — > CH4 + K2CO3 

Meth5'"l alcohol docs not C!>^vc methane, and as shown by Reid'^ 
some fozmic acid is formed in the fusion; hence 

CH3OH + 2 KOH — > 3 H2 + Na2C03 

involves the formation of the intermediate formic acid v^^hich splits 
into CO2 and hydrogen. 

Ethanol, acctaldchyde, acetone, n-propyl alcohol, isopropyl 
alcohol, and tert - butyl alcohol all give sodium carbonate, methane 
and hydrogen as products as docs glycerol, dextrose, cellulose, an.d 
other polyhydroxy compounds. Ethylene glycol, methyl alcohol, and 
formaldehyde give the carbonate ajid hydrogen. Ammonia, CIi3Mi2, 
(CH3)2M, (CH3)3N, (CH3)20, and (021-15)20 resisted the action of fused 
alkali . 

Reid'^ studied the reactions of aliphatic oilcohols on fusion 
v;ith alkali . 

(1) PrOH20H + ITaOH — > PrOOOI^Ia + 2 E2 

(2) EtOH2CIi20H — ^ Et-CH-OH20K + IlgO 

I3u 
These reactions are here associated because they so frequently go on 
in the same mixture. In reaction (2) the attack is predominantly on 
the p-carbon but may also be on the a-carbon. ITaturally the alcohol 
that is formed may also talcc part in the condensation; thus from 
butenol, a dodecanol and a hexadccanol are obtained along v;ith an 
octanol, but in smaller amounts. Reaction (1) goes rapidly at about 
320°, Several runs were made with ethanol. The conversion to the 
salt of acetic acid increased from 67/o to 3QS% v;hon the ratio of 
alcohol to caustic alkali was increased from one to three. The 
presence of water is beneficial; it suppresses reaction (2) and pre- 
vents darkening. Sodium carbonate appeared when the temperature was 
high or the time too long. Ethanol differed from the other alcohols 
in that it w s extensively dehydrated, particularly when much water 
wCvS present. 

In the case of anhydrous butanol, n-propyl- and 2-ethylhexanol 
the yield of octcjiol, etc., was Itirger when potassium hydroxide v/as 
used. With tert -butyl alcohol about one-fourth of the alkali went to 
the carbonate, no acid was isolated and the rest of the alcohol 
recovered unchanged. 

The fact that sodium acetate is used as a condensing agent in 
Perkin's Synthesis and in acctylations , suggested that the salts 
that were formed in Guerbert's c:>q:)criments''' may have been responsible 
at least in part, for the condensations. It has been found that salti 
of orgrjiic acids do effect condensations according to reaction (2). 



-5- 

Rcid roji a scries of reactions with butyl c.lcohol and the varior.s 
salts of butyric acid and found that the potass siiMi cjid sodium salts 
of fatty acids arc efficient catalysts for condensing tivo molecules 
of an alcohol into one of a hi3her molecular weight. 

Thus alkali fusion is a very useful method of oxidizing 
hydroxybenzaldelij^des, hydroxybonzyl alcohols, hydroxytoluones , and 
other substituted aromatic compounds. It is also useful for 
obtaining fatty acids from aliphatic alcohols. 

Bibliography ; 

1. Earth, Ann., __ 154, 356 (1870). 

2. Boswcll and Dickinson, J. Am, Chem. Soc, 40, 1786 (1918). 

3. Carrol, J. Phys . Chom. , 22, 128 (1918). 

4. Dumas and Stas , Ann., 35, 129 (1340). 

5. Pry, Schulze and Wcitkomp, J. J\m, Chem. Soc, 46, 2268 (1924); 
Pry cjid Schulze, ibid., 48, 958 (1926). 

6. Graebc ajid Kraft, Bcr., 39. 794 (1906). 

7. Gucrbcrt, Ann. chim. phys,, (7) 27, 67 (1902) j Compt. rend., I46, 

298,1404 (I9O8). 

8. Jacobsen, Bcr., 11, 376,570,1058,2052 (1873). 

9. Lembcrg, ibid., E2, 592 (1929). 

10. Lock, ibid., 61, 2234 (1928),- ibid., 62, 1177 (1929); ibid., 63, 

551,855 (19 30); Lock and otitz, ibid., 72, 77 (1939). 

11. Pauly, Ann., 583, 236 (1911). 

12. Re id, Worthington and Larchar, J. Am. Chem. Soc, 61, 99 (1939). 

13. V/ohler and Liebig, Ann., 3, 261 (1332). 



Reported by J. ?. Kaplan 
March 22, 1939 



"■ti- 



are 



PHENOLIC RESINS 

Httnel 

Phenolic resins, particularly phenol -fomaldeh^^de resins, „^, 
far superior to natural resins far various industrial products. Up 
until 1880, various phenol-formaldehyde products of vaiying propertie 
v;ere obtained but none of commercial value were synthesized. 

The first commercial resin was produced by Smith in 1839. 
Phenol was condensed with acetaldehyde and paraldehyde in acid solu- 
tion. The product was hard, brittle, and porous but it could not be 
moulded and so found limited use. 

In 1905, Blumer, usiii;^ tartaric acid as a catalyst, produced 
a permanently fusible, alcohol-soluble resin which was used as a 
shellac substitute. 

The name most closely associated with phenolic resins is that 
of Baekeland. The Bakelite process overcame most of the difficulties 
encounteie d in earlier processes, Baekeland classified phenolic resi 
into two groups : 

Class I : Permanently soluble and fusible resins of the type 
produced from phenol and formaldehyde in molecular proportions in aci 
medium. The product is acetone- and alcohol- soluble and has shellac- 
like properties.. It is not ''thermo-setting" . Baekeland called these 
resins "Novolaks".. They can often be converted to Class II resins by 
heating with formaldehyde. 

Class II : Insoluble, infusible resins. Class II resins cannot 
be converted to Class I resins by heating with phenol. They are pre- 
pared with larger proportions of formaldehyde and basic catalysts are 
used, 

Baekeland found that acid catalysts tend to promote formation 
of Novolaks whereas basic catalysts promote the formation of insolubl 
infusible resins almost independently of the proportions of phenol 
and formaldehyde used. 

The Bakelit e Process : The process usually is carried out in 
two or three stages the products of which are termed Bakelite A, B, 
and C which correspond to Lebach's "resol", "rcsitol", and "resit''. 

Bakelite A: Ea_uivalent quantities of phenol and 4O/0 formalin 
solution are condensed by means of a basic catalyst, usually ammonia, 
with or v/ithout heating. An aqueous layer separates and is romxovod,- 
The residue may be liquid, pasty, or an amorphous solid, depending on 
conditions and the proportion of formaldehyde used, and may be con- 
verted directly to Bakelite C by application of heat and pressure. 
The liquid is sometimes used to impregnate articles. The coating 
obtained upon Bakelizing is superior to varnish. Solid A melts upon 
heating zmd. is soluble in ordinary solvents. 

Bakelite B: Solid A is ground and mixed with an appropriate 
filler and heated. The appearance of a rubbery-like stage at about 
70*^0. indicates its conversion to Bal^ elite B. It softens on heating. 



w 






':: "Of' ;o^' 






•Y T 






brxo; 






\j:s^yyisx. 






•J-i-v^fi 



V' 



-2- 

but does not molt, becomes brittle on cooling and sv/ells in acetone. 
It can be stored as such, molded in a hot mould, removed from the 
mould and Bakelized under pressure v^rithout losinr,- its sha"DC to 
Bakelite C. 

1 Bakelite C: The finrJ. product, Bakelite C, is always formed 
by application of heat under a pressure of 50-110 po-onds to prevent 
the dissociation of fozmaldohyde v/hich caused porosity in earlier 
resins. 

Theor ies on the reactions between pheno ls and formaldehyd e : 
All theories" thus, far developed involve splitting out of water at 
some stage. Baekeland proposed the following mechanism on the basis 
of his synthesis of Bakelite from saligenin: 




+ 6 CHoO 



OH 

OCHpOH 



OH CH2-O 




CH2-O 



,-0 CH2-O CH 

da 




CH2OK 



+ 5 HpO 



CHoO 



OH CE2-O CH2-O CH2-O CH2-O CH2-O 




+ HsO 



The six-membered ring then polymerizes to the infusible resin. He 
also proposed the follov.ang mechanisms: 





+ CH2O 



+ CH2O 




CH2 ^ 
OH 



vinyl compound 
polymerizes 



■HaC jD + n2'f 



CE^^ 





OH f- 
>-v^C — 





polymerizes 



C II2 + O pH2 




C=CH. 



OH 
vinyl 
compound 
polymerize.' 



>y 



J 



<^.- 



) K- ^ i- ^ 



V-''> 



v-^- 



f"' 



:, ' 




' ' t 


* , 




: , ~ 1 


^^> ' 




''■■ J 


x- 




^ -^ - . ^^ 


-■-— ■ • H* 




%--^- ^•■•^'1 -='■ 


..:. ...^^. 




l.ri:., ^^a^--^ 



nV 



T 1 



J. 



•■V --^ 



-3- 

The isolation of 2,4'- cJid 4,4' -clihydroxyphcnylmc thanes in preparing 
the resins su^^estcd the latter mechanism. 

The follovdng mechanism was suggested by Wohl and tlylo : 




HaOH 




^ y H + H20 



-> polymer 



H CH2OH 



The vinyl theory was discarded because styrene docs not 
polymerize readily under conditions v/hich produce phcnol-iormaldchydo 
resins . 

In 1912 Raschig suggested a multi-condensation theory similar 
to the m.odern conception. Ilis m.echanism received little attention. 



^''--v^CH plOH li 



OPI 




CHp^H 



OH 



OH 



+ 




-CHoOH 



OH OH 
^'■vy^Hs >^^-''"^H2 >^^^CH2 OH 



Megson and Drummond found that 2,4' -dihydroxydiphenylm.ethane 
is consumed more rapidly than the 4, 4' compound and concluded that th' 
polymer has primarily ortho , £ara linkages. Prom £-cresol they 
isolated: 

)H OH 
/CH2* 





and proved that the phenolic hydroxy Is are not involved because it 
formed only a dibromidc. Similar intermediates were isolated from 
£-cresol. m-Cresol reacts so rapidly that intermediates are difficul" 
to isolate. 

Using alkaline catalysts they isolated mono- and polymeth^^lol- 
phcnols. Increase in tcmTocraturc and concentration dccrca.scs the 
yield of mcthylol compounds. In alkali insoluble, infusible rosins 
are obtained almost independent of the proportions of phenol and 
formaldehyde used. 

Koebner prepared chains containing from two to seven units by 
condensing p-crcsol with f omialdehyde , 



OH 



m- 




+ 2 CH2O -ili^> 



HOCK 



OH 
HOCHgY-^^CHsOH 

CH^ 




CH2OH 



OH OH 

>YH2Y-^CH2 

CH3 CH3 







jt:.: 



S'.. 



,ii 



.^ 



"^^ 



^•.X-'" 



'--Jr - 



T r 






■^^ 



'•'^. 



Vv 



7 - 



^'■■v 



-r^ 



-•'^' 



■"■v .-'■ 



.-^^ 



V 



A> 



i 



v.. J 



v 






V 



-4- 

ThcsG compounds were similcr to NovolrJk:s. Phenol and. foimal- 
dchydo could not be split from thorn by hydro].ysis. They remained 
fusible and were soluble in common solvents. The sodiujn salt of the 
scyen-membcred compound was insoluble. He could not prepare infusibl 
resins from £-crcsol in v/hich only linear (jrovrth is possible. He 
concluded that iTovolaJ^s are di-, tri- ond polynuclear linear chains 
and that resites are due to three dimensional growth of the molecule. 



Y^^CHg 




-C 



T-T 








OH 



Wovolaks which arc usually prepared from 0.5 to 1 mole of 
formaldehyde per mole of phenol have no uncombincd mothylol groups 
end hence connot further condense on heating tp foim insoluble resit 
molecules , 



Raschig' s 
mechanism. 



'multi-condensation" theory is now the accepted 



llojij other phenols and aldehydes have been investigated. 
Tri-cresol is widely used in mailing resins. The naphthols arc very 
difficult to resinify. Price is the controling factor in the use of 
aldehydes, ketones and phenols in comm.ercial resin manufacture. 

Phenolic Varnish Resins: Polymethylolphenols , or "resols"', 

when heate d with ' basi c varnish materials such as resins and drying 

oils, imjpart a superior hardness end resistivity to the final film. 
The exact nature of the reaction is not known. 

If the dimethylol compound obtained from £-cresol end. formal- 
dehyde in strongly alkaline solution is heated g'ently, only water is 
split out . Upon further hc^t;ing up to one mole of formaldehyde can 
be obtained and o. I\lovolrk: is the final product. Hdnel end Zinkc 
propose the following mechanism. 



n 



HOCHsY-^^CHsOH 



PIOCH 



R 




cH20cn2' 




CH2O 




CH2OH 



labile form 



+ (n-l)HpO 



OH OH OH 
-CH2v'-S'CH2.^>^CH2v''%-- 

O O V ^ 

R R R 



n CH2O 



final stable form 



'T'V. 



r 



^r' 



'^.- 



^ 






\ ^■ 



K.^^^' 



..•'*^! 



V 




) i 



^■Y^ 



I 






-5- 

V/hen the sniiic rosol is heated in a drying oil much loss 
f oimaldehydc is Gplit out indicating a reaction involving the mcth^lo! 
groups. The labile fozm shows whj'- low molecular weight resins have 
the same haxdening efficiency. Httnel proposes a condensation vrlVa 
methylene hydrogens or double bond liydrogcns. 

OH I OH I 

Y^CHo iOH + H^ C-H > Y^^CH2-CH 

Hilditch and Smith claim it is primcxily an addition at unsaturated 
centers because they could obtain no reaction with paraffin at 240 0. 
H6nel has since shown that resols will react even with saturated fatt^ 
acids at 250 C. 

The ''heat-hardening'' efficiency of a rcsol is dependent u^Don 
the number of methylol groups and the size of the alkyl substituonts 
in the phenol. It increases with the number of methylol groups and 
also slightly with the size of the substi'tucnt . Alkyl groups increase 
the oil compatibility whereas aryl substituents decrease the oil 
compatibility. The capacity of a resol to resinify bears no relation- 
ship to its oil compatibility. The compatibility of a resol is 
dependent upon (1) the number of reactive positions, (2) the amount 
of combined formaldehyde, (3) the size of the alkyl substituents, 
and (4) the nature of the materials with lYhich it is to be reacted. 

The following is a list of basic varnish materials arranged 
in order of decreasing compatibility. The class numbers indicate the 
number of alkyl co.rbon atoms in the meta -position or positions neces- 
sary to produce a homogeneous reaction product. 

1. Castor oil fatty acids, colophony (abietic acid) 

2. Patty acids of drying oils. 

3. Castor oil. 

4. Ester gum, cum-aron, indene resins, China wood oil, 
Synourin oil. 

5' Linseed oil, perilla oil, soya bean oil, saturated fatty 
glycerides, waxes, aliphatic unsaturated hjdxocca^'boiis . 

Thus the trimethylol compound of m-cresol is compatible with 
class 1 but the corresponding phenol compound is not and at least five 
aliphatic car dii atoms are necessary in the meta -position of a tri- 
methylolphenol to produce homogeneous reaction with class 5. In the 
type of compound shown below where the dashes indicate either 
hydrogens or metliylol groups at least 10-15 aliphatic carbon atoms 
per phenolic nucleus are necessary for complete compcitibility . 




Decreasing the amount of combined formaldehyde increases the compati- 
bility but at the expense of the heat -hardening efficiency. A recent 
process for using polymetli^/'lol compounds in varnishes involves heating 
it vath a higher alcohol. It is believed the etherification of at 
least one meth^'"lol group takes place. 



/ 



-^'K 



) 



-6- 

OH OH 

HOCHgy'^CHsOH KOCHsy'^CHsOH 




+ HoO 



CHo-0- 



Hp jOH + lip R CH2-O-R 



The final polymcthylol compound has only tv;o reaction favorable 
positions available. 

Hdncl has developed the folloiving rules for compatibility 
VI i th dry i ng oils: 

1. Any resol of the lov/ost molecular size having but two 
reaction favorable positions is com.patible providing the true resol 
stage is not exceeded. 

2. PlCSoIs having more than tv\ro reaction favorable positions 
arc generally 'thermo -setting" ojid require a certain minimum number of 
aliphatic carbon atoms for compatibility. 

a. Three such positions require at least 4-5 aliphatic 
carbon atoms per phenolic group. 

b. Pour such positions demand at least 10 aliphatic 
carbon atom.s per phenolic group providing the resol is in its lowest 
molecular* stage. 

Almost ojiy desired properties cm be imparted to a varnish if 
the proper resol cjid basic v?.rnish material are chosen. 

Bibliography ; 

Hdnel, J. Oil Colour Chcm. Assoc, 218, 247 (1938). 

'iic^QT, Bcr., 5, 1094 (1872). 

Klceberg, Ann., 263, 283 (1891); ibid., _264, 351 (1891). 

Manasse, Ber., 27, 2409 (1894). 

Lederer, J. pralrt . Chcm., 50, 223 (1894). 

Piria, Ann., 56, 39 (1845); Aim. chim. pliys . , 14 (3), 268 (1845). 

Lcbach, Angew. Chcm., 22, 1598 (1909). 

Baekeland, Ind. Eng. Chcm., 1, 149 (1909); Chom. Ztg., 33, 857 (1909) 

Traubenberg, Angew. Chem., 17, 2 (1923). 

Megson and Drui-nmond, J. Soc. Chem. Ind., 49 (2), 251 (1930). 

KodDner, Angew. Chem., 46, 251 (1933). 

Baekeland and Bender, Ind. Eng. Chem., 17, 225 (1925). 

Schciber and Baoi^thel , J. prakt . Chem., 147, 99 (1936-37). 

Hilditch and Smith, J. Soc. Chem. Ind., 5±, HI (1935). 

Granger, Ind. Eng. Chem., 2±, 446 (1932). 



Reported -by E» H. Bobratz 
March 22, 1939 






.j.urf 



• .^..;ju ,ui: 






VITMIK Bs 

' Kuhn -- Kaiser-V/ilhelrn Institrate 

The term vitamin Bg has been given by Gydrgy to that part of 
the vitamin B2-complex v/hich is responGib?^e for the cure of the 
specific dennatitis developed by young rats fed on a vitam.in B-free 
diet supplemented with purified vitamin 3, mid. lactoflavin. 

The vitamin is absorbed by fuller's earth from an acidic solu- 
tion of the alcoholic extract of autolyzed wheat gerTn. It is eluted 
with Ba(0Ii)2 and precipitated by phosphot'irngstic acid. The compound 
obtained in this m.anner is easily dializable, and heat and alkali 
stable. It has been shown, by Kuhn that the vitamin exists in yeast 
in a high molecular \7eight, not dializable, heat and light sensitive 
form. The prosthetic group can be set free from this forra by heat, 
without damage to the vitamin activity . 

The vitamin - the lo?/ molecular weight compound - is now 
called "Adeimin", a term derived from antiderraatitis . 

The structure of vitamin Bq has been shour^ by Kuhn to be: 

CH2OH 




^r^CH- 



This structure was deduced mainly from the following reactions: 

1. Precipitattion with phosphotungstic acid and no loss of 
activity by treatment with nitrous acid, indicating a tertiary amine. 

2. Formation of a monometh^/'"l ether by diazomethane and of a 
diacetyl methyl ether from the metliyl ether. 

3. Oxidation of the methyl ether with chDromic acid to form 
acetic acid. 

4. Oxidation of the meth^^l other v^rith KJ'.inO^ to form a lactone 
with loss 01 four hydrogen atoms, and more vigorously to form a tri- 
carboxylic acid from which the anhydride was produced and one molecule 
of carbon dioxide eliminated. 

5. Color reaction of the tricai'-boxylic acid of the methyl ether 
indicating an a-carbonyl group, and failure of the dicarboxylic acid 
to give this reaction. 

6. Color reaction and absorption spectra data indicating a 
p-hydroxypyridine . 

7. Partial oxidation of the metliyl ether to produce an anhy- 
dride-forming mcthyl-f3-methoxypyridinedicarboxylic acid. 

8. Synthesis of the acid in 7. 

1. Gy^rgy found that adeimin is not precipitated by salts of 
heavy metals but is precipitated by phosiohotungstic acid; it is not 
inactivated by treatment with IIiJ02 , but is inactivated by treatment 
v-ilth benzoyl chloride. Prom these observations he concluded the 
compound docs not contain a priiaary amine group, but is basic and 
probably contains a hydroxyl group. 




2. CHsOK CH2OH CH2OAC 

2J^ jl I -±.--2 — I — ). Ij j 

IT^CH3 II "^iT^CHa ^ lT^CI-13 



20^-^un Qj^^j^^ riui.2^Y^-<^-'^-3 ACoO ^-wi.'.2^>f^-^wwri3 

t- ^"^^-^^ — > no color 

Theso reactions show that adcnnin contain3 one phenolic hydroxy 1 raid 
two alcoholic hydroxj^'ls. 

Pb ( OAc ) 4 • ^ 4 • 
3 . ^^ — — -i i-=-> no oxidation 

^^ iiii:^^ > 0.06 moles IIOAc 

The failure of Pb(0Ac)4 to rcr. ct with II indicates iii at the compound 
is not an a-clycol. Criesec has shown that PbCOAc)^. reacts with a 
sugar only when -CHgOPI is joined to a carbon atom having a free 
hydroxyl group. 

0.90 moles of acetic acid was obtained by chromic acid o:Q.da- 
tion of 1 5 indicating one methyl group. 

^' ^- ^^nt ' lactone, C9H9O3K (III) 

/ ^^^'' 0^ 

II COOH j ^C=0 II ^1 IV 

HOOGy-^OCHs 0=C,,^,,.^0CH3 



Ev:di04_ 

"7(0) 



blood-red color 



■jj^COOH 
PCSO4. 




no blood- ^^^^^4 jj I V 

red color 



5. A blood-red color with .FeS04 i'^ cha?ract eristic for pyridinc- 
a-carboxylic acids. Thus the last coxbozqyl group was in the a-posi- 
tion, 

6, The absorption spectra of adcrmin in HCl and in NaOH arc very 
similar to that of p-hydroxypyridine in the same solutions. 

The Polin and Denis phenol reagent gives a deeij blue color 
vdth adermin. All the (3-liydroxyp3^ridincs tested with this reagent 
gcve this color, but no a- or of- hydroxypyri dines tested produced such 
a color. 

Prom these data it was concluded that adcrmin is a derivative 
of p -hydro xypyridine and thx'.t IV is correct for the anhydride of the 
adermin-methyl ethor-dicarboxylic acid. Synthesis of this compound 
has confirmed this conclusion. 

For the tricarbox^^lic acid there arc tv/o possible formulae: 

■ COOH COOH 

HOOCy-A^OCHs H00Cy''^0GH3 

^^ 1 J I i ^^^ 

H00C'^-;M^ ^U^GOOH 

Since adermin has one phenolic -OH, two alcoholic hydroxyls and one 
C-methyl and forms a tricarboxylic acid of the methyl ether with loss 
of six hydrogen atoms and of no carbon, it must be a m cth3''l-di( hydro xy - 
methyl ) -3 -isy^ro:>[ypyri dine . The lactone fonncd by mild oxidation with 



J>.. 



tl 
I) 



^ 



*<- 






\. 



/ 



v;- 



L i 



i:iiK0J^ 



-0- 



Idi'InO^. shows that the tv/o by droxc^^n ethyl groups arc on adjacent cojrbon 
atoms. From those considerations adcniin must have one of the 
1 1 1 o v\ring s t ru c t ur o s : 



HOHoC 




VIII 



HOI-IpC. 



:v.c 



CIIoOH 
OH 




IX 



CE3 

H0E2C.^-I^0H 



EOH2C 



'W 



7. 



CII2OH 



H0H2Cv^-^0CIl3 T3a(Mn04 ) 



COOH 




anhydride 
no color 



The failure of the acid XI to give a color v;ith PCSO4 shows that there 
is no a-hydroxymetho/'l group. Adorrnin cannot bo X. 

8. 2-Methiyl-3'-mctho:cypyridincdicarbo:-:ylic o.cid (4,5) va s synthe- 

sized riid found to bo identical with XI foiTAod '^o'j partial oxidation 
of adermin nethyl ether. 

Kuhn describes the conversion of adciiiiin r'ioth3/'l ether back to 
adorrnin by the following reactions: 



II 



I-IBr 


Br]l2Cv 


Cil2Br 

>I^K3 
I^+ Br- 


AgOAc 


66% 




II2O 



noii,c^ 



C1I2O11 

Oil 







■Foli n and Denis 
phenol reagent 



-> deep blue color 



Bi bliog raphy : 

Cricgeo, Aim., 495, 211 (1952). 
Gy^rgy, Hature,"T53, 498 (1934). 
iiirch and CryargyTT^iochcm. J., 30, 304 (19 '^6). 
Crydrgy, J. ;\ra. Cheni. Soc, 60, ^5 (1^33). 

Kulm ct al, Ber. , 71, 730, 1534 (133B); ibid., 72, 305, 309, 310, 
311 (1939)."* ■~' 



Reported by E. "/elch 
March 29, 1939 



i 



THE RELATIONSHIP BE^Tv^/ESN PLUOIfflSCElICE 
AlID CHEMICAL CONSTITUTION 

Otto Miinim -- University of Kiel 

Henri ch -- Erlanger 

Eialkovskaja -- State University, Leningrad 

The TDroblem of the rela.tionship of chemical constitution and 
fluorescence has not yet been satisfactorily ansv/eredi In fluorescenc 
the process of emission of the absorbed radiation is of very short 
duration, a time interval of only 10* to 10 seconds elapsing be- 
tween the instant of v;ithdrat'/al of the exciting radiation and prac- 
tical cessation of tJie induced emission. It is in this respect that 
it differs from phosphorescence in which the process of emission is 
extended over an easily measured period of time. 

In the case of fluorescence, the mechanism involves first the 
primary process of absorption in v/hich an atom is raised to an excited 
state. This is then followed by a redmission of the absorbed radiatioi: 
in either a single stop where the emitted quantujn has the same energy 
as that of the exciting ray (resonance), or in a stepwise fashion, 
giving off radiation of longer v;ave length and j)ossibly involving 
dissipation of some of the energy Idj inelastic collisions x^ath other 
molecules. 

It is V/aters' contention that fluorescence emission is neces- 
sarily concemod with the refomiation of stable valency bonds (or 
possible stable '"''lone pairs" of electrons) from previously activated 
molecules. In support of this ho offers that v;hcn tvro halogen atoms 
combine, the resulting molecule is so stable that energy is emitted 
as fluorescence under favorable pressuxc and temperature conditions. 
For example: 2 1 — > I2 "^ ^'^* 

It might seem reasonable to suppose that this light energy 
emission might bo associated with the chemical action of recombination 
of the neutral free atom.s to diatomic molecules. This might be ex- 
tended to other cases of fluorcGcencc which might be concerned v/ith 
the dissociation of a covalent bond into free radicals, followed by 
a recombination of these radicals once more. The emission of fluores- 
cence might, then indicate that the dissociation of a covalent bond in 
a molecule had previously occurred. 

Among some of the considerations which he offered in support 
of this suggestion are the follov/ing: 

1. Fluorescence of Na ajid Hg vapors is definitely associated 
with neutral diatomic molecules Ha2 and Hgg . These can be produced 

only b5'- association of single atoms. 

2. Fluorescent inorgfinic molecules can dissociate into free 

radicals such as NO. 

3. The inhibition of fluorescence by added substances can only 
be brought about by those compounds which can easily lose an electron 
(ex. iodide ion or Og). These same substances can also inhibit photo- 
chemica.l reactions by terminating chain processes and combine instantly 
with knovm free radicals. 

4. Only certain classes of organic molecules possessing unsatu- 
rated conjugated structures arc fluorescent and all these compounds 



:r.-)c*'!t- 



.ii'.-' r' 






*•« 



I'^f*' 



r • -■ ^ ■"» f 



O0f;;v 



: ; Si'% .V- 



'n<?v>j^;i'a:' 



-•.;l^- 



^T L'.K.:.: 



.j-.;_;ff X^. '• 



-2' 



arc capable of yielding free radicals, either by dissociation or 
addition of alkali metals . 

5. All aromatic compounds, even if not otherwise fluorescent, 
Sive characteristic "Tesla" spectra, the seat of which is the aroma- 
tic ring and v/hich it has been suggested is associated with tauto- 
merism of valency bonds betv/een two special ICekule "phases*'. Any- 
such isomerism, however depicted, must involve fission of valency- 
bonds with subsequent reunion of momentarily existing free radicals. 



M\m.m and his coworkers have spent many yeai 



\-;orking 



with com- 
pounds in the pyridine series and found that a great number of these 
compounds had "fluorescent properties. ?rom this he has formulated an 
opinion as to the chare. ct eristic structjure necessary to produce fluo- 
rescence. As a general exaniple we will consider first 1,4,6-tri- 
methyl-3-et}i3'-l-5-carbethoxypyridon (I) sjid all other products wit 
different substitucnts in positions 
substituted pyridones 



th 
3; 4 and 5. These completely 
re used in order to show that a possible enoli- 



zation of the N liydrogon to the carbonyl oxygen is not necessary in 
the mechanism. All of these pyridones show strong fluorescence even 
in the solid state, when activated by ultra-violet light. 



CpI"i502G 



CH3-^ 



Gil 




CoHr 



^^2 ^-^5 



V.' 



CH3 
12115020^^^-^^^ — C: :0: 



:K"; 



•CH- 



-OH 



II 



However, as soon as the structure cioanges from the character- 
istic pyridon grouping, the fluorescence disappears as well. This is 
a]^ o true of compounds having a structure as in II which do not show 
a trace of fluorescence. Thus he claims that only those compounds 
having the grouping -l:I(CIi3)-C0- caji be responsible for fluorescence. 

Since the possibility of enolisation has been removed by sub- 
stituting a metliyl for the IT hydrogen, that need not be considered. 
Hov/ever, the possibility of a shift to a betaine structure (III) has 
not been eliminated and according to von Auwers the spectrochemical 
data on the IM-alkyl pyridones rndicates that the struct^are is some- 
where betv/een the carbonyl and -the betaine formulae. This is explain, 
ed by a partial neutralization of the residual affinity between the 
doubly bonded oY^rQen and the nitrogen. In other vvords we have here 
a case of mesomerism and I and III are only limiting f omulae . Thus 



iiijV..© 

®CH. 




V 




for the intermediate state we can va^ite the structure as in IV (a 
free radical) which is comparable to tlic time average formula of 
benzene (V) as advocated by m.?Jiy of the modem theorists. 



di. 



Y/hen such a molecule absorbs light, the quantiLm whicli is of 
low energy raises the electron which is least firmly held in its 
nomal level, to an excited state. Thus one of the electrons shown 
between C and N or C and in IV v;ould be of this type. This action 
then disturbs the valence forces in the molecule and upsets the reso- 



X. -• ..'■ 



3ri.t -r;>- 



. v;.u.:;i:fr. 



i' .. ■^.! 



■^K : 






•i -t- 



nance of the molecule. Por example, if one of the electrons in IV 
between C and N is excited, then the free electron on the nitrogen 
will be dravi/n in closer to the triad grouping and upon emission the 
excited electron drops down to form the noirrial double bond betv/een 
K and C as in III. A similar process takes place if an electron is 
disturbed between C and in IV but in this case it drops back to 
form the nonnal double bond as in I. 

Other substances of the pyridine series which Mumm studied 
were the 1^ and a-dihydropyridines . In this case the t^dihydro com- 
poun<t; are white, fluoresce sticngly and exiiibit no basic character, 
whereas the a-dihydro compounds are mostly wealcly basic, yellow in 
color and exhibit no .fluorescence. 

If one considers the iJ-mothylatcd -f-dihydropyridines there is 
a marked similarity' in molecular str'ucture to the a-pyridoncs. Thus 
in VI we have the formula as normally written and its tautomer in VII 
Correspondingly the intermediate state can be written as analogous to 
that of the pyridoncs (VIII) end. is again a di-frec radical v/ith a 
half positive charge on the nitrogen and a half negative charge on 
the p -carbon. The same would then hold true as to light absorption 
and emission as in the case of the pyridoncs. 



VI 



•CH^ 



VII 




<$} 



•cii. 



VIII 






As support for the assumption of VIII is introduced the fact 
of the lack of basic character of the 'f-diiiydropyridines . 



although VI indicates the presence of a free electron pair 
normially easily capable of bonding with a hydrogen ion, it 
so in this case. 



Thus 5 
which is 
cannot do 



COoR 



There mcy be the objection that in this case the molecule is 
symmetrical, making carbon atoms 2 and G eqiial and that the source of 
fluorescence is due to a combined interaction of the two double bonds. 
Hov/ever, this caji be met with the fact that a substance having a struc- 
ture as in IX with only one double bond can bo prepared and that these 

substa:accs fluoresce as deeply blue as 
the if- dihydropyri dines and in which 
the basic character is also lacking. 
IX • 

Now if the a-dihydropyridinc is 
considered in the same light as the 
a-pyridones and the 'f-diliydropyri dines 
it is readily seen that here the reso- 
nance must go through a greater part of the molecule, namely up to 
carbon 5. Accordingly the negative charge is carried further so that 



RO2C. 




the mesomcric intermediate between the limiting form: 




XI 



® OH. 



XII 



X and XI is 
I 

«_c*. -C — 



XII. 



As a consequence of the greater distance over which the 






■■"I . 






J.'.'- - ^ 



X 



l: 






t .» 






_4- 

resonaiice enex^y must be distribiroed the electrons in the raosomeric 
form (X±I) are more loosely bound. Accordingly the bond between the 
nitrogen and the c-carbon is also decreased in energy and one of the 
electrons can be more easily diVe?rted to fujmish a lone electron pair 
on the nitrogen. This then accounts for the more basic charL\cter of 
the a-dih-ydro over the Tf-dihi^dropyridine . On account of this greatly 
loosened bonding of the electrons in XII there is also a lessened 
expenditure of energy necessary to raise them from their normal level. 
Thus for this pu3rr:)0se light of longer wave length is needed. This 
all correponds to the fact that the a-dihydro compounds absorb in the 
blue and are thus colored yellow, v/hercas the f-dih^dro compounds 
absorb in the ultra-violet and are v;hite. 

A further consequence of this is that upon falling back to its 
normal level, the electron in the a-diii^^dro compound emits rays of 
longer wave length which ±n general are no longer visible. Therefore 
the a-dih^drocom^DOunds do not fluoresce but when in exceptional cases 
fluorescence docs appear, the emitted light instead of being blue, as 
in those compounds mentioned earlier, is green or of longer v/ave 
length. 

It appears then that the connection between color and chemical 
constitution is alm.ost of the same nature as the f oromentioned connec- 
tion between fluorescence and constitution. The essential difference 
lies in the fact that the omission of radiation folio-wing upon the 
dropping of an excited electron consists of rcys of greater v;ave 
length which no longer belong to the visible part of the spectrum. 

Blbliograph;:^ ; 

Griffith and McEcovv-n, "Photo Processes in Gaseous and Liquid 

Systems", Longmmis Green and Co., London, 1929, pp. 244-317 • 
¥ty , "Electronic Concept of Valence and The Constitution of Benzene", 

Longmans Green and Co., London, 1921, pp. 235-56. 
Kauffmann, "Bcziehujigen Zvvischen Pb^silialischon Eigcnschaftcn und 

Chcmikcr Constitution", Pcrdinand EnlvC , Stuttgart, Gennany, 

1920, pp. 305-25. 
Muram, Ber., 72, 29 (1939). 
Waters, Nature, 128, 9O5 {1931)- 

Henri ch and Braun, J. prakt . Chem., 13_9, 338 (1934). 
Pialkovskaoa, Acta Physiochcmica, 9, 215 (1938) . 

Allen, Pranklin and McDonald, J. Pranklin Inst., 215, 705 (1933). 
Ilerjrich and Horold, Ber., 60, 2053 (1927). 
V. Auv/ers, ibid., 63, 21l6"Tl930). 



Reported by M. H. Gold 
March 29, 1939 



-Y 



V;:.^:nr:; 



J- c ■ -v' :1:'>*;. 



ORGAI^TIC COICPOUiraS CONTAIUnTG n'^ 

Schoenheimer -- Columbia University 

Since the discovery of isotopes and of improved methods for 
their concentration, attention has been turned towards their use a3 
indicators in follov/ing reactions. Radioactive isotopes have been 
used extensively in inorganic chemistry, but as yet have not proved 
to be of value in organic chemistry, since radioactive isotopes of 
organic elements of long half life have not been prepared. Stable 
isotopes have to be used; these have the disadvantage that their 
concentration is more difficult to determine. 

Schoenheimer has conceived the idea of using H'^ and D to fol- 
low the metabolism of proteins. It has been found convenient to use 
a mass spectrometer for determination of H'S. With this apparatus, 
only very small amounts of gas are necessary, and a high degree of 
accuracy can be obtained. By analysis of atmospheric nitrogen and of 
various com.pounds, the normal N'^ content was shown to be O.368 atom 
per cent. Several naturally occurring amino acids were also investi- 
gated. These had values ranging from 0.000 to 0.010 atom per cent 
excess N'^, The higher values probably result from some sort of a 
minute fractionation process. 

In carrying out a mass spectrometer analysis, the following 
technique v\/as used: 

1. Determination of total I'T, and simultaneous conversion into 
l^H^."^ by the Kjeldahl method. 

2 . Conversion of ITH4. into Ng ^^^ alkaline hypobromite : 



2 IIH^.'*' + 2 OH" + 3 NaOBr > N2 + 3 NaBr + 5 H2O 

3. Admission of gas to spectrometer. 

4. Actual measurement. 

In order that the data would be significant, Schoenheimer 
demonstrated that the C-H bond is rather stable and would not undergo 
exchange reactions, llo exchajige of 3^1 was noticed in any of the fol- 
lov/ing systems: N - NH3 or amino acids; NI13 - amino acid; amino 
acid - amino acid; hippuric acid - amino acid; urea - amino acid. 
All these were carried out at 37°. Some evidence oj -an exchange 
reaction between urea and NH^Cl was observed at 100 , probably because 
of the equilibrium MsCOMg -^rriNH^'^ + CITO" . 

The amino acids y/ere prepared according to Knoop and Oesterlin 
method and by the Gabriel synthesis. The first method involves the 
hydrogenation of an a-keto acid in the presehce of ammonia: 

RCOCOOH + IK3 + H2 -^^ RgHCOOPI + HgO 

ITH2 



Por monocarboxylic acids, two mols of ammonia were necesssiry; for di- 
carboxylic acids, three, llorleucine, phenylalanine, tyrosine, gluta- 
mic acid, aspartic acid, and alanine v/cre prepared in this way. 
Gabriel's synthesis was applied in the synthesis of dcuterolcucinc ano' 
lysine . 



aCOOH ^^'^COv 

+ M3 > [ [ it: 
COOH \xA)0 



-2- 

H + 2 HoO 



a 



CO. ^^-v^CO\ 

^mi + K >\ 1 N-K + 1/2 H2 

co^ \X^CO^ 



aCO^ /Br /COOH NEs'HCl 

N-K + RCIi > R-CH COw^^ > RCh( 

CO^ ^COOH bC 1 1 COOH 

CO^^v^ 

An. example of the use of N'^ is Schoenheimer' s v/ork on 
hippuric acid. It is a well knov;n fact that large quantities of 
benzoic acid in the organism are detoxified by glycine to give 
hippuric acid. 

CsHsCOOH + NH2CH2COOH > C6H5CONHCH2COOH 

Glycine N'^ was administered with benzoic acid, but even when there 
was an excess present, only 1/4-1/3 of the IV^ appeared in the urine 
as hippuric acid N'^. Much of the ^;lycine used in detoxification 
was therefore supplied by the tissues. 

The methods used by Schoenheimer ajjpear to constitute a 
valuable tool not only for biochemists but also for organic chemists 
in follov/ing the course of reactions. 

Bibliography : 

1. Schoenheimer et al., J. Biol. Chem., 125, 495 (1938); ibid., 127, 

285, 291, 301, 315, 319, 329, 333, 385 (1939). 

2. Urey et al , J. Am. Chem. Soc, 59, 1407 (1937); J. Chem. Physics, 

6, 296 (1938). 

3. Knoop and Oesterlin, Z. physiol. Chem., 170, 186 (1927). 

4. Gabriel and Kroseberg, Ber., 22, 426 (18^57. 



Reported by L. C. Behr 
April 5, 1939 



I^ATURAL AZULEITES 

Pfau and Plattner -- Laboratory of Pirm L. Givaudan and Co., 3. A. 

The first recorded observation of a deep-blue compound occur- 
ring in a natural oil was made in the 15th century. Since that time 
it has been shown that many ethereal oils contain these blue com- 
pounds, or the latter may be obtained from them by chem.ical transfor- 
mations. Due to their distinctive deep-blue or azure color Oessi 
named them "azulenes'' in 1863 • In all there are over 260 described 
ethereal oils which contain azulenes or azulene-like compounds, and 
which correspond to about 20/^ of the known, oils. The sources which 
have been most thoroughly investigated are camomile, elemi, eucalyptu.^ 
vetiver, guaiacol, gurjun, and milfoil oils as v\rell as brown-coal-tar 
neutral oils, certain fungi, and a synthetic tarry material called 
cuprene-tar obtained by passing acetylene over metallic copper. 

In general the azulenes are obtained from the sesquiterpene 
and ses quit cDrpene- alcohol fractions of ethereal oils by deliydrogena- 
tion with S, Se , or catalytic Ni . They are separated from other 
products by their solubility in concentrated mineral acids (50% H2SO4. 
or 80^0 H3PO4) and recovered by subseq.uent dilution of the acid solu- 
tions. Picrates, styphnates, trinitrobenzenates , trotylates, ajid 
H4.Pe(Cl0 6 addition compounds may be prepared easily, and these can be 
decomposed by (M4.)2S, NH3 , or chromatographic absorption on activatec 
alumina to give the pure azulenes. 

The name ''azulenes'' has been used for the entire class of 
compounds in the same manner that ''naphthalenes" implies- all compound; 
which have a naphthalene nucleus but various aliphatic side chains. 
In a more strict sense the name ''azulenc'' refers to the basic compoun' 
of the group with a formula of C,oHs (IX), isomeric with naphthalene. 

All azulenes have a naphthalene-like odor, and range in color 
from deep blue to violet. They react readily with Brg , ITOCl, and 
1203, but the products are too unstable for rocrystallization. 
Metallic sodium decolorizes an ether solution of an azulenc, and a 
brovjn crust is foiined on the sodium. By addition of moist ether the 
azulenc may be regenerated. The ant i- inflammatory action of the 
ethereal oil of camomile as tested by mustard oil irritation to rabbi- 
eye, and irradiation erythema in h^uman beings, rats, and pigs has bee: 
a'ttributed to one constituent, azulene . 

Various investigators have checked the molecular weight and 
degree of unsaturation of these natural azulenes and have found that 
they all possess -the formula of C,5H|q and will take up five molecule 
of 112, being transformed to colorless saturated h^/drocarbons . Each 
investigator v;ho isolated an azulene from a different ethereal oil 
gave a characteristic nome to the aoilene; for example, that from 
camomile oil was called chamazulene, that from eucalyptus oil was 
called eucazulene, ^aid that from elemi oil v/as called elemazulene. 
Such was the state of confusion of the literature when Pfau and 
Plattner took up the study. They found that treated guaiacol, after 
dehydrogenation by S or Se, gave tv;o different azulenes which they 
teimed S-guaiazulene and Se-guaiazulene . After obtaining these in th 
pure state, prepaxing derivatives, and s-tudying their absorption spec 
tra, they proceeded to compare the other knovm. azulenes v/ith these tw 



-2- 

To their surprise they found that all of the azulenes v;hich they pre- 
pared from the other natural oils by deh^^'drogenation were identical 
with S-guaiazulene with the exception of an azulene which was obtained 
from vetiver oil by dehydrogenation v/ith Se . This, then, simplified 
the study by limiting the number of natural azulenes principally to 
three: vetivazulene, S-guaiazulene, and Se-guaiazulene . 



In order to prove the structures of these azulenes oxidation 
and ozonization studies v;ere carried out. These yielded such a 
variety of compounds that Pfau and Plattner decided little progress 
could be made in that way. Similar studi'os were carried out by 
Kremcrs who suggested two possible structures both of the benzofulvenc 
type, based upon the fact that the azulenes were bicyclic, colored, 
and formed dimors by the addition of sodium. The ideas of the benzo- 
fulvenc and naphthalene nucleus were completely rejected by Pfau bo- 
cause these compounds do not form addition compounds with H4Pe(CH)Q 
while all azulenes do, and also because of some deii;}''drogenation 
studies to be discussed la. ter. In 1926 Ruzicka and Rudolph expressed 

vv.o„rv. r>^v>o«-,^n v>o. -|^];^Q structuTc of thc azulcncs by 



the general facts i<:nov/n concerning 



stating: "The color of thc azulenes is due to an cs23ccial, up-to-now 
unlaiovm type of grouping of five double bonds (without an aromatic 
ring) in a bicyclic ring of carbon atoms, v/hich is vqtj similar in 
constitution to the sesquiterpene compounds if not identical with 
them. " 



Following the sajnc procedure of Mayer and Shiffner, who found 
that certain alkj^l groups migrated from the a- to p-position on a 
na>.phthalenc ring v^hcn those a-alkyinaphthalencs were passed over 
silica gel, Pfau ajid Plattner carried out a similar reaction v;ith 
S-guaiazulcne . Good yields of naphthalene hydrocarbons were obtained 
but in m.ixturcs v/hich indicated a shift in alli;>^l groups. Hence no 
further work vms done on this reaction. 

By dehydrogenation of treated guaiacol with P and HI there was 
obtained, after separation of azulenes, a naphthalene hydrocarbon of 
the formula C|5H,q (therefore an isomer of cadaline and thc azulenes) 
Another isomeric hi^^dro carbon from thc vetivazulenc-yielding fraction 
of vetiver oil was obtained in a like manner. Thc former by picrate 
and trinitrobenzenate was shovm to be l,4-dimeth3'"l-6-isopropylnaphtha' 
lene (I) while thc latter v/as 1, 5-dimethyl-7-isopropylnaphthalcne (II 
Both had been previously synthesized and identified by Ruzicka. 






Ill 
(eudesmcne skeleton) 



Prom such results Trcibs postulated that III was the probable 
azulene skeleton, but Pfau disproved this by de'hydrogenating a satu- 
rated cudesmene-tyx)c compound to get no azulene while dccaliydro- 
guaiazulene could be tri-msformed back into an azulene. It v/as tliere- 
forc concluded by Pfau that the naphthalene skeleton was formed from 
some other ring system. Splitting the ring from a sesc^uiterpcnc com.- 
pound by mild oxidation followed by ring closure led to a cyclic 



-3- 

ketone which had one fewer carbon atoms in the ring. This ketone by ' 
dehydrogenation led to a phenol, which made the original terpene struc 
ture seem to have a seven-membered ring. 



Connecting these results then the formula of naphthalene deri- 
vatives could be explained; the rearrangement could be considered 
analogous to the retropinacolone rearrangement : 







x± 



In order to substantiate these findings several compounds of 
the bicyclo-(05 3,5) -decane system were synthesized. Ililickel and co- 
workers had previously prepared cyclop enteno-cycloheptajione (YI) in 
order to shov; that both cis and trans forms were strainless. By 
addition of a Grignard reagent and subsequent dehydration and dehydro- 
genation azulenes were obtained according to the following scheme: 






y 



•CH 



CH. 



■CH. 



x"\ 



CH2-CH2 



-H2O 



/ 

GIL 
\ 



.CH 



2V 



^m/ 






=CH 



\ 



'CH2-CH2 



CH. 



sulfur or 

catalytic 
Ni 



VI 



h^ 



2CH 



cfo 

CH ^CH 



A- 5 

CH=CK 
\ 

CH6 
// 

CH 

8 7 



IX 




C2H5 



Mixtures were obtained in the case of VIII but the true azulenes were 
separated by concentrated H3PO4.. Yields v/ere lo\7 due to the sensitiv 
ity toward acids. Compounds represented by VIII v/cre blue in color, 
had very similar absorotion spectra, and all derivatives were much 
like those of natural azulenes. These findings confirm that the 
asulene-type is of a new kind of ring system which has a five and 
seven-membered ring fused through adjacent carbon atoms. Thus 
vetivazulcne would be 2-isopropyl-4,8-dimcthylazulene (X), while 
S-guaiazulene would have the constitution of l,4-dimetli5''l-7-isopropyl 
azulenc (XI) : 



CH 
CH- 



3^ 



CH 



<^ 



X 



CH- 




XI 



n(CH3) 



3^2 



As a final discussion, of the types of bicyclic sesquiterpenes 
and their derivatives the largest nimiber belong, to the group of hydro 
genated naphthalene derivatives, bicyclo-lO, 4,4)-decanc , and may be 
traced back to three types by cyclization of three isoprcnc units 
(following the isoprcno rule) : 





XII 





XIII 



(cadinenc 



I 



u 



/ 



> 






/ 



•„sr:' .-^ 



,i-!'- ^p»v■ -^.^e* 



-4- 




'vO;) 



XIV 

( eudesmene-type ) 



The first "type (XII) , originally proposed by V/allachj has not been 
found in nature. Types XIII and XIV are well-knov/n. 

The analogous structure of bicyclo- (0,3 ,5 )-(iecane leads to 
nine possible skeletons (v/hereby it is not to be implied that a.11 of 
these' possibilities are realized in nature): 





*< 



XVI 



T 







XVII 





X 



'<>- 




XVIII 



^ 




XIX 





II. 





XXI 





XXII 





XXIII 




Type XVII is known, ivith certaintj 
to be that of vetivasulene, while 
very probably type XIX represent? 
S-guaiazulene . The further v/ork 
concerning natural azulenes and 
numerous as yet unknovm. sesqui- 
terpene derivatives, v/hich by deliydirogenation should give azulenes, 
can probably be traced to the other types. 

S ynthesis of A^^ulene : The synthesis of the T)arent hydro carbo: 
was carried out by the following method: 



OH 

CD 








li--, 



XXIV 



^^ J ' ^-octalin 
7JCV 




ITaOH 



XXVI 




XXVI- 



XXVI 



\. 



-5- 



OH 

Cn2 ^CH — CH2 

61-1 /^^2 
01x2 0112-0x12 


-HpO 


IX 


:r\ 


catalytic 


^ 



CH2 



XXIX 

This S3m.thesis also cleared up the question as to the nature of the 
blue compoimd obtained by Hentzchel on.d V/islicenus in 1893 by dry 
distilling the calcium salt of eidipic acid. Pfau repeated their vrork 
using catalytic ili, isolated and identified the blue compound as 
azulene itseif . He gives the follov/ing mechanism for the reaction: 



CK2-Cn2-C00H 



On2-CH2-OOOH 



XXZ 



HOGG 
GHpv /GOOK CHp 

/ ^\H2 \' 

CH2 ^ ^CH2 

\H2-C-CH2 — CH2 

xx:{i 



/ 

-> OK. 



HOCG 

CH2^ /OOOIT CH2 



OH 



II 

2 CHp ■ 

XXXII 



( 

/ 

■OH. 



,CH. 



-CO -^ 





II 

.0- 



•OH, 



\ 



/H2 / 
OH2 \ 

Gil/ \Ho-OH 



/ 



OH. 



XXVII 



Synthesis of Vctivazulcnc : 




HpOO 
— ^ — —5. 

HCl 




(i)03H7^;^00Et 



on2Ci 



p. 

ITa OOOEt 



hydrolysis ond 
decarboxylation 



XXXIV 



XXXVII 



EtOOC-p; 




C^-C3H7(i)^ 



Zn + HOI 



OH- 



EtOOC.^^ ^ 




m 



3^7(1) 

XXXVIII 



C3H7(i) 
XXXIX 



Sap onif i c/.it i on 
Do carbox^'"lati on 
Dciiydro g onat i on 




'pH-03H7(i) 



XXXV 



1. SOOlo 

2. AIOI3 



^cn-03n7(i) 



XXXVI 




(X) votivasulcno 



iv 



v^J 



.i. 



•. •" 



-6- 

Compound XXXVII is here formulated v;ith the arrangement of double 
"bonds which arc necessary, according to Mills and Nixon and also 
Pieser and Lothrop , for liydrindene derivatives. Addition of the 
diazoacetic ester to any of the three positions 4-5, 5-6 , 6-7 v/ould 
lead to the same end product j 4-9 and 7-8 addition would be stcrically 
difficult, while 8-9 addition seems impossible. The vetivazulenc thus 
prepared was identical in all derivatives and also in absorption spec- 
tra with that from the natural source. This, then, is the first com- 
plete synthesis of a "natural"' azulene . 

In closing it should be pointed out that different forms of 
unsymmc trie ally substituted azulcnes may exist, as follows: 

R 





The relationship here is quite similar to that of aromatic hydrocar- 
bons. An ozhaustivo discussion of this is to occur in a later i^aper 
by Plattner and Pfau. 

Recently Sklar m.adc a systematic studj^ of chromophore groups 
in relation to their absorption spectra. He worked out a method of 
calculating the absorption bands of compounds by calculating resonance 
levels. The equations used contain only one parameter which is 
entirely dependent upon the heat of hydrogenation of the compound. H* 
has successfully applied this to azulcnc and has been able, without 
the use of eiij optical data, to determine values for its absorption 
bands which agree well with experimental results. 

Eibliograpliy : 

Pfau and Plattner, Helv. Chim. Acta, 22, 202 (1939). 

Susz, Pfau and Plattner, ibid., 20, 4^5" (1937). 

Plattner and Pfau, ibid., 20, 22T"(1937). 

Pfau a^d Plattner, ibid., T|", 858 (1936). 

Ruzicka and Radolph, ibid., 9, 125 (1926). 

Ruzicka and liaagen-Smit, ibid., 14, 1104 (1931). 

Ruzicka, Pieth, Reichstein and Ehjnann, ibid., 16, 268 (1933) • 

Kremers, J. Am. Chem. Sec, 45, 717 (1923). 

Mills and Nixon, J. Chem. Soc, 2510 (1930). 

Pieser and Lothrop, J. An. Chen. Soc, 58, 2050 (1936). 

Braren and Buchner^ Ber. , 34, 982 (1901X7 

Melville, J. M. Chem. Soc., 55, 2462, 3288 (1933). 

Sklar, J. Chem. Phys . , 5, 669'Tl937). 

Mckel and Schnitzspalin, Ann., 505, 274 (1933). 

Mayer and Schiffner, Ber., 67, o7~(1934) . 



Reported by J. J. Denton 
April 5, 1939 



STERIC HINDRMCE IN SUBSTITUTED BEHZALDEHYIES 



The term, steric hindrance, is applied to the effect produced 
by groups in a molecule, v^rhich, although they do not enter into the 
reactions, exert a profound influence on the speed or course of the 
reaction taken by the functional group. The reaction which takes 
place then becomes the resultant of the reactivity of the functional 
group and the steric hindrance provided by the radical or radicals 
v;hich it holds. 

Although very little is known about the actual cause of these 
deviations from nonnal reaction course, there emerges from a large 
number of studies one verj significant if not entirely general rule. 
That is, that these effects are associated primarily with reactions 
which are fomulated as being additive rather than metathetical. 

This paper presents the results of a number of investigators 
who have studied some common reactions of variously substituted 
benzaldehyde derivatives in order to observe the influence of these 
substituents . 

Ace tal Formation ; The well knov'\ni experiments of Victor Meyer 
have sKowri that esterif ication of aromatic acids is inJriibited v^hen 
the two ortho hydrogen atoms are replaced by alkyl , Gl, Br, I, or 
WO2 groups and, conversely, that the esters 01 such di-orthosubsti- 
tuted acids are especially difficult to saponify. Since the con- 
version of aldehydes to acctals is a process entirely similar to 
esterif ication, Fischer and Giebe' studied the ease of dimethyl acetal 
formation of the aldehydes corresponding to the acids used by Meyer. 
They discovered entirely different effects. o-Nitrobenzaldehyde 
reacts easier than benzaldehyde as is the case with 2,5-dichloro- 
and 2-nitro-3,6-dichlorobenzaldehyde. In these cases the electro- 
negative substituents have exerted an accelerating effect in spite of 
their ortho position and high molecular weight. The same authors 
tried 2,4,6-trimethylbenzaldehyde . In this case the reaction was 
more difficult than v;ith benzaldehyde but the difference is not veiy 
great since, under the same conditions, benzaldehyde gave ^2% of the 
acetal while the mesityl aldehyde gave 32%. The general conclusions 
reached by these investigators was that substituents in the ortho 
position induce no great hindrance to acetal formation. 

Similar conclusions were reached by Lock^ after studying 
2,6-dichloro- and pentachlorobenzaldehydes. The 2,6-dichlorobenzal- 
dehyde reacted somev/hat slower than benaaldehyde but the yield could 
easily be increased by prolonging the reaction time. The following 
results v/ere reported: 

Compound Time in hrs. % yield of acetal 

benzaldehyde 24; 60 43; 50 

pentachlorobenzaldehyde 96 60 

2,6-dichlorobenzaldehyde 24; 96 13.6; 43 

The only example which indicates a definite hindrance is that 
of 2, 4,6-trinitrobenzaldehyde reported by Bamberger and Elger^ . 
They were able to get no acetal formation by ordinary methods but 
did obtain a slight yield after very long standing. 



'C? 






ai? 



oLv- ^i r.j 



;. e •^•. 



-2- 

Perkin Reaction : Studies on the Perkin reaction with substi- 
tuted benzaldehyde have led to some interesting results. The origi- 
nal investigations by Meyer and Beer , whose results are shown in the 
follov;ing table, indicate that, in general, mono-orthosubstituted 
benzaldehydes react more readily than benzaldehyde itself. 

Substituent in ortho position H NO2 CI I 
Yield of corresponding cinnamic acid 48 51 66 85.5 

Similar studies on variously substituted derivatives by Pieich, 
Salzman and Kav/a and by Lock^'®' indicate that no hindrance is 
observed except in the case of methyl derivatives. The combined 
results of these investigators are given in the follov/ing tables. 

Position of CI atoms 2 3 4 2,5 2,6 2,3,6 2,3,4,5,6 H 
Yield 71 63 52 78 82 66 25-30 45 

In general, all chloro derivatives gave higher yields than benzalde- 
hyde itself. The highest yields were obtained from the ortho- and 
diortho- substituted derivatives. The low yields of the pentachloro 
derivative is attributed to its great insolubility rather than to 
steric hindrance (Lock^) . 

Position of KO2 groups 2 3 4 2,4 2,6 2,4,6 K 
Yield 63 50 74 70 100 (2 hrs.) 18 

The general accelerating effect of the IJO2 group was even greater 
than that of the CI group with the exception that the £-chloro- 
reacted faster than the o-nitrobenzaldehyde. 

The trinitro derivative splits out HCOOH under the conditions 
of this reaction. This interesting behavior is typical of diortho- 
substituted benzaldehydes in the Cannizzaro reaction and will be 
discussed later. 

Position of CH3 group 2 3 4 2,6 2,4,6 
Yield all lower than none 7 -8^0 under 

benzaldehyde forced conditions 

The exceptional behavior of the metliyl derivatives has not been 
attributed to steric hindrance because the yield is lowered regard- 
less of the position of the group; that is, even meta and para 
substitution produce a retarding influence. It is interesting to 
note that mesityl aldehyde gives 2, 4,6-trimethylcinnamic ester by a 
smooth reaction in the Claisen condensation. 

General conclusions: The reaction tendency of diortho-substi- 
tuted benzaldehydes (except CH3) in the Perkin reaction increases 
with the weight of the ortho substituents contrary to the liypothesis 
of steric hindrance. Evidently stereochemical influence of the 
substituents is masked by their exalting pov/er on the reactivity of 
the CHO group . 

Cannizzaro Reaction ; Studies by Lock^ on the behavior of 
substituted benzaldehydes in the Cannizzaro reaction have shoivn that 
the reaction may take one of two possible courses depending on the 
position and nature of the substituents. In general, those derive- 



»••/ r-. 



;.i;f' 



-3- 

tives containing a single ortho substituent (OH is an exception) will 
give noimal disproportionation, while diortho-substituted derivatives, 
under the influence of alkali, lose the CHO group as foimic acid and 
fonn substituted benzene derivatives. These two types of reactions, " 
with their exceptions, will be discussed separately. 

A. Normal disproportionation: The usual Cannizzaro reaction is 
observed with all derivatives studied which have a free ortho posi- 
tion with the exception of o-hydroxy- and £-hydroxybenzaldehyde, and 
those derivatives containing tv\ro nitro groups (e.g. 2,4-dinitro- 
benzaldetxyde) . 

£-H;>''droxy- and £-hydroxybenzaldehyde are unchanged by alkali 
solution; with m-hydroxy derivatives noiroal disproportionation takes 
place as long as only one ortho substituent is present. 

Some evidence of. hindrance due to size of the group is observed 
by comparing bromo and chloro derivatives of the same parent compound. 
3,5-Dimethoxybenzaldehyde gave complete disproportionation in one- 
half hour; 2-chloro-3,5-dimethoxybenzaldehyde was complete in 3 hours 
while 2-bromo-3,5-dimethoxybenzaldehyde required 8 hours for complete 
reaction. 

B. Removal of CHO group: This type of reaction, which may be 
illustrated as follows. 



iCHO + K0:H ^> ( ) + HCOOK 





is observed with all derivatives of benzaldehyde which contain two 
halogen atoms or one halogen atom and one nitro group or two nitro 
groups in the positions ortho to the aldehyde group , and aldehydes 
with nitro groups in the 2,4-positions . Exceptional behavior is 
shown by derivatives v/ith ^-hydroxy, or o-amino substituents . 

This reaction in the aromatic series, which seems to depend 
on the position and especially on the nature of the substituents 
present in addition to the aldehyde group, is very similar to that 
observed in substituted acetaldehydes (e.g. trichloro-, tribrorao-, 
and triphenylacetaldehyde) . 

Cl3C:CH0* " V *KaH > CI3CH + HCOOK 

In this reaction, the otherv/ise very resistant C-C bond seems to be 
so v\reakened by the presence of many negative substituents that the 
cleavage can take place. That v/eight is not the only factor is em- 
phasized by^ the fact that propiolaldehyde, with very low molecular 
weight, gives the split v/ith alkali to form acetylene and potassium 
formate. 

In the aromatic scries, it is obvious that, in addition to the 
"negative effect", the substituents in the ortho positions have some 
effect, especially in view of the fact that 2,4-dihaloscn derivatives 



I 



i.f.o .!>■ 






rf- 



l^}yIO 






-4- 

givG normal disproportionation cvon though they have two halogen 
atoms present. 

Nearly all cases of 2,6-dihalogen substitution gave a practi- 
cally quantitative yield of formic acid. The amount of formic acid 
produced v;as determined by the reduction of HgClg to Hg2Cl2 (Pran.zon 
and G-reve^) . 

HCOOH + 2 HgClg — * CO2 + 2 HCl + ligsClg 

The effect of ortho and para hydro xyl groups is shovm by the 
following examples, o- Hydroxy- and p-hydroxybcnzaldehyde give no 
reaction with aqueous alkali. 2 ,4,D-Trilx7droxybenzaldehyde , when 
treated with KOH, gives phloroglucinol and carbon dioxide but no 
formic acid. Likewise, the 2,4-dihydroxybenzaldehyde gives resorcinol 
(71.6/0), hydrogen (74.3/^), carbon dioxide, and a very small amount of 
formic acid (0.9/^). 2, 6 -Dibromo vanillin gave no formic acid but the 
2,6-dibromo-3,4-dimethoxybenzaldehyde formed by methylation of the 
£-hydroxyl group induced smooth splitting to formic acid and 
3,5-dibromoveratrol. Prom this we conclude that a free hydroxyl 
group in the para position not only hinders noimal disproportionation 
to the acid and alcohol but also the elimination of the aldehyde 
group in 2,6-dihalogen substituted aldehydes. A h^^droxyl group in 
the mcta position hinders neither type of reaction. 

A similar restraining influence is shov;n by 6- substituted 
2-ajninobenzaldohydes, 2-Amino-3 5 6-dichlorobenzaldehyde, when treated 
with KOH, gave no foimic acid. However, the recovered aldehyde could 
be diazotized and converted to 2-iodo-3,6-dichlorobenzaldehyde which 
then underwent conversion to formic acid and l-iodo-2,5-dichloro- 
benzene. 

The effect of two nitre gro.ups is shown in 2, 4-dinitro- , 
2,4-dinitro-3-nicthox^r- ^ and 2 , 4-dinitro- 3-hydroxyb en^aldeh^de , all of 
which gave formic acid. 2, 4, 6-Trinitrobenzaldehyde gives foxmic acid 
and trinitrobenzene when treated with alcohol,' aniline, aimnonia, or 
dilute alkali. 

The combined effects of ortho nitre groups and para hydroxyl 
group is shown by a comparison of the 2,6-dinitro-3-methoxybenzal- 
dehyde with 2,6-dinitrovanillin. Under the influence of ^% NaOH the 
former gives 87.5/o formic acid while the latter gave only '^'^'^ 



Evidence of steric hindrance due to the size of the diortho 
substituents is indicated by the folloy\fing sequences which give the 
relative rates of reaction: 

1. For 2,6-dihalogen substituted derivatives: 

2,6-dinitro > 2,6-dichloro > 2,6-dibromo > 2,6-difluoro (gave 

slight disproportionation) 

2. For 2,6-dihalogen-3,5-dimethoxy derivatives: 

2,6-dinitro > 2-bromo-6-nitro > 2,6-dichloro > 2,6-dibromo 

This order parallels the decrease in size of the groups as determined 
by x-ray data and also the relative interference effects of these 



. }:• ''n< 



-5- 



groups on the rates of racemization of biphenyls 



I 



One other study on the Cannizzaro reaction deserves mention. 
Weissberger and Haase concluded from a study of the reaction that 
a relationship exists between the rate of Cannizzaro reaction of 
aldehydes, RCHO, and the dissociation constants of the corresponding 
acids, RCOOH. With ortho-substituted aldehydes the reaction is 
slower than would be expected from dissociation constajits, thus in- 
dicating the operation of another factor, presumably steric hindrance 

Addition of HCI\[ and Stability of Resulting Cyanohydrins : 
H e 1 1 e r'2 reported that in the reaction of KCl on benzaldehyde" 
derivatives the substituents have the power to hinder or facilitate 
addition. With o-nitrobenzaldehyde the addition talces place only 
V'jlth the aid of bisulfite in concentrated solution. 

Lapworth and Manoke "^ have studied the rates of dissociation 
the cyanohydrins of substituted benzaldehydes . 



O 



ClOH ^ ( VcHO + HCH 



They report that the effect of ortho substituents examined (other 
than OH) is such as to depress the dissociation constants of the 
cyanohydrins, and that this is in direct contrast to the result to 
be expected from the classical "space-occupation" hypothesis used so 
frequently to account for the other effects associated with ortho 
substitution. 

Hantzsch's Condensation ; Hinkel, Ay ling and Morgan have 
studied the effects of various substituents in the Hantzsch conden- 
sation: 

2 CHsCOCH^COOCoHs ] C2H5OOC-C" C-COOC2H5 

)l II +3 H2O 



+ NII3 + CgHsCHO \ CH3-C .C-CH3 

They make the following general! z at ion't v^ith each substituent the 
yield is lowest for the ortho derivative and, while the difference 
between the ortho and para compounds is small, there is definite 
evidence of a steric or ortho effect. 

The greatest hindrance v;as observed with 2-methyl-, 2,4,6-tri-^ 
methyl-, and 3, 5-dinitro-2, 4,6-trimeth3^1bcnzaldehyde . In the latter 
two cases less than 1% yields were obtained and in the 2-methyl- 
benzaldehyde only 8.5^0 as contrasted with 74% for benzaldelx/de . It 
is interesting to note that methyl groups, regardless of position, 
reduced the yields. 

Grignard Re acti on : Reich reports that 2,6-dichlorobcnzaldehydo 
gave excellent yields of the carbine 1 when treated with phony Imagne- 
siujn bromide. Lock^ also reports failure to observe any hindrance 
when pontachlorobcnzaldehydc v;as treated with CH3MgI or CglisMgl. 
Both gave the corresponding carbinol in good yields despite the great 
insolubility of the aldehyde in ether. 



-6- 

Formation of Anil : Hantzsch has shown that bcnzaldohydcs will 
not react v;ith cliortho- substituted aniline derivatives. Reich^ , 
hov;cver, has shov\Qi that diortho- substituted benzaldehydes will react 
amoothly with aniline . 

110 2 HO 2 

ITO2-/ VcHO + H2N-/ \ — > NO2-/ \cE=Iiy \ 

NO 2 NO2 

This reaction v;as also successful with 2 ,6-dichloro- and 2,4,6-tri- 
methylbenzaldchydes . Gattermann' ® has used this reaction to separate 
the mesityl aldehyde from the reaction mixture in which this aldehyde 
was prepared.^ 

Bibliography ; 

1. Pischer and Giebe, Ber. , 31, 545 (1898). 

2. Lock, ibid., 72, 300 (1939T. 

3. Bamberger andTIlger, Aim., 371, 327 (1909). 

4. Meyer and Beer, Monatsh. , 3T7~649 (1913). 

5. Reich, Salzman and Kawa, Bull. soc. chim., (4) 21, 217 (1917). 

6. Lock, Bock and Schmidt, Monatsh., 64, 393 (1934T7 

7. Lock and Schmidt, J. prakt . Ohem.,'T2) 140, 229 (1934). 

8. Lock, Ber., 62, 1177 (1929); ibid., 61, 2234 (1928),- Monatsh., 

62, 178 (1933); ibid., 55, 307 (1930); ibid., 64, 341 (1934); 
iFid., 67, 320 (1936); ibid., 68, 51 (1936); Ber., 66, 1527, 
1759 (1933); ibid., 68, I505 (T935); ibid., 69, 2253 (1936). 

9. Pranzen and Greve , J. prakt. Chem., (2) 80, 368^X1909) . 

10. Gilman, "Organic Chemistry", John Wiley and Sons, New York, Vol. I 

p. 282. 

11. V/eissbergcr and Haase, J. Chem. Soc, 1934 , 535. 

12. Heller, Ber., 46, 280 (1913). 

13. Lapvrorth and Ivlanske , J. Chem. Soc, 1928 , 2533. 

14. Hinkel, Ayling and Morgan, ibid., 1931 , 1835* 

15. Hantzsch, Ber., 23, 2776 (1890). 

16. Gatterraann, Ann., 347, 374 (1906). 



Reported by R. V. Lindscy, Jr. 
April 12, 1939 



i 



RELATION OP BASICITY AND SOLUBILITY TO THE TOXICITY OP MIIHES 

Kindler -- Chemische Staatsinstitut der Universit^t Hamburg 

Kindler, in considering amines as agents for the specific 
treatment of diseases caused by protozoa, has made a system.atic 
survey of the toxicity of practically all available types of organic 
nitrogen compounds. 

A method for measuring the toxicity was desired, so Kindler and 
his coworkers devised the following method. One cc . of a water solu- 
tion of the amine (or it s hydrochloride) was mixed with an equal 
volum.e of culture solution containing paramecia. The time required 
for 90/0 or more of the organisms to be killed was then measured. 
The toxicity num.ber (Til) is the reciprocal of the concentration of 
the original amine solution in milligrams, multiplied by the recipro- 
cal of the time in minutes. To obtain reproducible results the 
strength of the solution must be adjusted so that this time lies be- 
tween one and ten minutes. The culture solution V\fas adjusted so that 
one cc. contained a constant number of paramecia. Since the resis- 
tance varies with different cultures, the TK of quinine was measured 
with each series and the TW of the other amines calculated relative 
to quinine = 1000 and this was called the relative toxicity nui.iber 
(RTN) . This admits comparisons between series. 

It was found in preliminary tests that the RTIT with paramecia 
compared very nicely with the RTN for pathogenic protozoa, thus 
allov/ing the convenient substitution of paramecia for the pathogens. 
The RTN in this resport are the values for paramecia only. 

Hydrolysis of salts is an important factor in toxicity. It is 
shown in Table 1 that free amines are Eiuch more active than their 
hydrochlorides. Table 2 shoves that the weaker the amine, the more 
hydrolysis will occur, therefore the more toxic the salt will be. 



p-H0C6H4CH2CHoira2 60 D-H0C6H4.CH2CH2N(CH3)2 400 

— -HCl 1 ^ — -HCl 1 





Tabl 


3 1 




60 

1 




£-H0Ce 


5H4. 


CQH5CH2lfH2 


1000 
•HGl 2 




Table 2 


p 


pH 
8.72 
5.11 
4.99 




1-100 

60 

70 
100 


Xl 



RTN on dilutions of 

HCl salt of: pH 1-100 1-500 1-2000 1-4000 

p-PC6H4CH2N(C2H5)2 8.72 60 50 25 — 

p-BrCeH4CH2N(C2H5)2 5.11 70 200 650 1000 

p-IC6H4CH2N(C2H5)2 4.99 100 400 1100 I5OO 

The rest of this pajjer deals with solubility relationships. 
Kindler postulates that the more soluble a substance in the lipoidal 
material of the cell, the more toxic it will be. The following 
generalizations and tables are offered to support this theory. 

1. In homolgous series water solubility decreases with incrcas- 



-2- 



ing chain length and the toxicity sharply increases (Table 3). 

Table 3 

CeHgCHsNHR 



40 

75 

3000 
19000 

35000 



R 


R2NH 


C6H5CHENH2 


CH3 


■— 


6 


C2H5 


— 


30 


n-C3H7 


— 


130 


n-C^Hg 


25 


500 


n-CyH.s 


13000 


20000 


n-CgHi 9 


— 


50000 


n-C, 1H23 


— 


— 



. 2. Amines with straight chains are less soluble in water than 
their branched isomers (Tables 4 and 5)» 



R in C6H5CH2IIHR No 
CH3 ( CH2 ) 9 

(CH3)2CH(CH2)3CH(CH3)(CH2)2 
CH3(CH2) ( I 
CH3(CH2)8CH(CH3)CH2 



Table 4 



of atoms 


RTN 


17 


30000 


17 


16000 


19 


40000 


19 


10000 



C6H5ira(CH2)9CH3 

C6H5lT(CH2CH2CH2CH3)2 

C6H51TH(CH2)mCH3 

CqHs ( CH2 ) 3N( CH2CH2CH2CH3 ) 2 
C6H5CH2NH(CH2)7CH3 

C6H5(CH2)3N(CH2CH2CH3)2 



Table 5 



Of C 
17 

17 
19 

19 
15 
15 



atoms 



RTN 
30000 

500 

40000 

4000 

13000 

300 



3. Di-aryl- aliphatic amines are much less soluble in v/ater 
than the primary amine and therefore are more toxic to paramecia 
(Table 6) . 

Table 6 



C6H5(CH2)nNH2 
C6H5(CH2)^K( 0113)2 
(C6H5(CH2Tn)2M 





RTN when n = 




1 


2 3 


4 


2 


20 80 


150 


2 


— 50 


— 


300 


600 — 


— 



4. Amines of the benzene series dissolve to a greater extent 
in v;ater than the corresponding naphthalenes (the members of Table 7 
are chosen from about 20 compounds)*. 



Table 7 



R = an alkyl group 
C6H5CHRCH2NH2 
P-C,oH7CHRCHoNH2 



RTN 

1000 - 15000 

2000 - 33000 



) mmi 






! ! 



-3- 

5. The hydrochlorides of tertiary cjiiines with two methyls 
on the N atom arc less toxic than the corresponding primary amine 
salts (Table 8) . 

Table 8 
CH2R RTN when X = 

R in C6H5CHCH2X-HC1 NHg ^(^3)2 

3,4-(CH30)2C6H3 1000 800 

p-CH3C6H^- 10000 3200 

l-CeHs-O-CeH^ 15000 4500 

6. The toxicity of the phenyl, cyclohexyl, Gn.d n-hexyl groups 
give increasing toxicity to the amino in the order given (Table 9). 

Table 9 

R in ti-R-C6H4.-CH(CH3)NH2 RTN 

CqHs 4000 

CqH, , 10000 

n-C6H,3 20000 

7. Naturally occurring alkaloids v;ith a few exceptions are 
relatively non-toxic to paramecia (Table 10). 





Table 


10 




Alkaloid 






RTi^r 


Atropine 






10 


Heroin 






10 


Morphine 






10 


Apomcr phine 






330 


Lobelin 






1000 


Quinine 






1000 



Those amines shown to be highly toxic to protoxoa by this 
siorvey arc to be thoroughly tested in warm-blooded aminals for 
toxicity to the hosts. It is hoped some efficient synthetic 
therapeutics will be discovered in this manner. 

Some of Kindler' s preparative methods are briefly outlined 
below: 

1. Reduction of omides catalytically or clcctrolytically at a lead 
cathode 

a. RCOM2 > RCH2ira2 65^ 

b. RCOIIR2 > RCH2NH2 

2. Reduction of nitriles 

a. ArCN -S ?? ^^^ ' ArCH2NH2 

b. RCHO + HON — » RGH(0H)C1T ^^2^ ) RCH(OAc)CN -ik_±_A^'t . 

RCH2CH2NH2 



I 



•-' i!?''^ 



-4- 

c. ArCOCl + HC — > ArCOC¥ ^^P '^\ ArCH( 0H)CH2NH2 

d. ArCHsCN + (C H5)2im J^? ^ ^^1 ^ ( ArCHsCHa) 2NH in 50% 

yield; some ArCH2CH2N( C2H5) 2 ; remainder primary am.ine 

e. 2 ArCHgCN -^2 — _ > ( ArCH2CIi2 )2M 90?^ or above 

11 2 "^ u 4. c at • 

3. Hypohalogen degradation of amides 

£-CH30C6H4.CH2CH2C0NH2 ^^q^ > £-CH30C6H4.CH2CH2NH2 90% 

4. Ar'CHpCOOH + HpNCHsCHsAr ^°^^ ■ > Ar' CHgCOMCHgCHpAr ^ - ^nzene ^ ^ 

^ 222 tetralin ^ ^ ^ POCI3 

CH2 
^E2 




!;H2Ar' 



Bibliography; 



Kindler, Arch. Pharai., 277, 14 (1939); ibid., 276, 107 (1938); 

ibid., 272, 811 (1934); ibid., 266, 19 (T928). 
Kindler et al . , Arch. Pharm., 277, 25'TT939); ibid., 273, 478 (1935); 

ibid., 272, 60, 236 (193477 ibid., 271, 439 (1933T; ibid., 270, 

340, 352, 410 (1932); ibid., 269, 581 (1931); ibid., 265, 389 

(1929). 
Kindler, Peschke and Brandt, Ber. , 68, 2241 (1935). 



Reported by J. H. McCracken 
April 12, 1939 



■ru 



I 



THE USE OP METALLIC COBTPLEXES IN THE 
DETEK^IIUTION OP COIJFIGUEATIOK 

T.V/. J. Taylor -- The I^son Perrins Laboratory, Oxford 



The metals of G-roup VIII and adjacent groups of the periodic 
table form complexes with certain oximes, hydrazones, and oxim.e- 
hydrazones of 1,2-diketones . Complex formation, in some cases, can 
be used to determine the spatial arrangement of the groups in these 
compounds, since only those with favorable conf iguia tions will foim 
a complex. In vievj of this Hieber and Sonnekalb distinguished 
between the isomers of benzilphenylosazone by stxi dying the complexes 
formed with stannic chloride. The methods has been extended to the 
other types of compounds mentioned. 

Oximes : Salicylaldoxime forms two complexes v/ith copper, 
which are shown by I and II. The product obtained depends upon 
experimental conditions . Hence the oxime must have the anti config- 
uration. 



.CHv OH 





6u 



II 



The monoximes of 1,2-diketones have been shovm to form 
complexes provided their structure is given by III. An example is 
the complex formed by the action of three molecules of a-benzil- 
monoxime with cobalt, IV. 





III 



-q— c- 
II II 

N-OH 



^i/W=C-C6H. 
Co. 



^1 V 



IV 



The mechanism appears to be the follov/ing: 

+++ 



R_C — C-R 

II II + 

NOH 



R_C C-R 

- II II 

0^ ,N0H 



--C-Cells 



R-C. 

II 





\.< 



C-R 

II 

N-^0 



+ 



H 



+ 



The (3 -oxime cannot form a complex because the position of its hydrox- 
yl group prohibits the preliminary stage. 

a-Benzildioximc, V, gives a scries of complexes, VI, while 
the =-isomcr gives none. 

OH 0^ 
C«H^-C — C-C^Hr, C«Hc-C=C J^N=C-i 



■^6^5 



II H V 1 ^^l^r 1 VI 

HON NOH C«H^-C=Nr ^N^C-CgHs 



C6H5-C=:N^ 



'0 HCf 



■iOG 



'0 IK I . 



:;.?>r:;YJ.i 



T I " 



/ 



, n<- 



-2- 



Hydrazones : The complex formed by the action of benzilmono- 
hydrazone upon nickel acetate has the composition R3M2 where 
R = C6H5C0C(C6H5)='MH-. The complex is amorphous, slowly attacked 
by hydrogen sulfide, unaffeciE d by potassium cyanide, and is decom- 
posed by nitric acid to give benzil or benzoylketazine depending 
upon the conditions used. The comples presents no true analogies 
to those of the oximes and is much less stable than the latter. 

An unusual reaction takes place v;hen the nickel complex of 
benzilmonohydrazone is prepared in boiling acetone solution. A new 
complex is formed, called the acine complex, that contains two benzil- 
monohydrazone residues, one acetone residue, and a nickel atom. This 
is the only hydrazone complex that was obtained in crystalline form. 
Of the four possible structures, VII and VIII are preferable for 
stereochemical reasons. 



VII 



CsHs-C C-CsHs 

n II 



3J2 



II 



II 

•C-C5H5 




'^-^6^5 



VIII 



NiT C(CH,) 



V _ ir"^— -V 



0' l\l=F 
I I 

'-'6^5-0 C-C5H5 



V2 



IX 



CeHs 



^m c(CH3 2 
II 






X 







^ 






rk 



^M^ ;c(CH3)2 



0' ^11^ 

II 






6 -^^5 



Poimation of the azine complex does not clear up the stereochemical 
problem since IX or X might be possible in viev; of the unusual 
behavior of camphorquinone monohydrazone as explained below. The 
reaction shows that the hydrazone complex must contain reactive 
imino groups. This is probably the cause of its indeterminate and 
unstable nature . ^ 

Tertiary -bu-bylglyaxal monohydrazone also forms a complex with 
nickel acetate with the formula R3Fi2. It is more unstable than 
those jast described. 



The p -monohydrazone of camphorqi inone gave an indication of 
complex formation, while the a-isomer did not. Since the configura- 
tion of the former has been shown to be XI, the complex presumably 
has the structure XII. 



XI 



R-C- 
II 



H2N-N 



• C-R 
II 



R 
^'^^ 

XII H I 

NH 
"^Fi^ 






-ii- 



■ — • • ».'. f 



-•qv-.. , ... .. 

u = 1 s .{ /♦ .>.■ *.: ■ . i, t'.'f .v; 



;;.rD; 






•. :'.: }''. j 



:-'cii.i 



V : V 



(•!- %-,--i :•;. 



-3- 

This behavior is unexpected since p-benzilmonoxime forms no complexes 
and is known to have a configuration corresponding to XI. 

Salicylaldohydohydrazone, of v;hich only one form is knovm, 
fonns complexes with copper, nickel, cobalt, and iron. However, 
these complexes present no true analogy to the corresponding oximc 
complexes, and no deduction as to the configuration of the hydrazone 
can be made from their behavior. This is probably due to the presence 
of two nitrogen atoms in the hydrazone group, either of which may 
coordinate with a metal atom so that the complex fonned is of the 
chain type. 

In general, the attempts to detennine the stereochemical 
configuration of the hydrazones by means of metallic complexes were 
not successful because of the instability and difficulty of purifi- 
cation of the products. This is in contrast to the oxime complexes 
which have shown to be useful in this way. 

Oxime -hydrazones ; Theoretically, benzil oxime -h^ydrazone has 
four possible configurations as shown below. Only two of these are 
known. 

CsHs-C-C-CeHs GeE^-G -C-CeHs GqEs-G C-GeEs CeHg-G -C-GeHg 

H0¥ MH2 NOH H2OT WOH 'ME2 HON E^M^ 

XIII XIV XV XVI 

a-Benzilmonoxime reacts with hydrazine to give a compound that is 
assigned the structure shov/n by XIII. This compound combines with 
several metals to form complexes, of v\rhich the most characteristic 
is the magenta ferrous compound XVII. 

C rH c: - C ■ C - C f5 He 

II II 

Of- Us, ^N-NH2 
H2 0-»Pefe2 XVII 

HpN-N-^ N-^O 

II It 

GeE^-G C-CqHs 

The oxime-hydrazonc prepared from (3-benzilmonoximc docs not form 
such complexes, and is assigned configuration XIV. The configura- 
tion of the oxime group is knoi'^m from earlier complex studies, and 
since the physical properties of the oximc -hydrazones vary in the 
same manner as the a- and p-dioximes the above assignment is thought 
to be correct. The p-dioxime forms no complexes, but the a-dioximc 
forms very stable complexes with nickel and palladium. This evidence 
does not exclude formula XV as a possibility for the p- oxime -hydrazone 
The two unknown oxime-hydrazones are very probably the ajnphi-compoundr' 
XV and XVI. 



..;1/:r;',-r.;, 



. '~' . ft 



:5^^:. in:. 









■■'l 



-4- 



Bibliography ; 

Taylor, Callow and Francis, J. Chom. Soc, 1939, 257- 
Taylor and Baker, "The Organic Chemistry of Nitrogen", 

Claredon Press, Oxford, 1937, p. 193. 
Hieber and Sonnekalb, Ann., 436 , 86 (1927). 
Ephraim, Bcr. , 63, 1928 (193^77 



The 



Reported by V/m. H. Sharkey 
April 19, 1939 



SjO^I"^' 



" l'.» ' 



ORGANIC C0R1P0UHDS CONTAINING SELEIIEUM 
MJD TELLURIUIVI 

Behaghel -- University of Giessen 
Morgan -- University of Birmingham 

Several different types of organic compounds containing Se and 
Te have been prepared "but no very general study has been made on 
them. Genarally the sajne type compounds may be made from Se and Te 
as can be made from and S. However, the basicity of the compounds 
increases in going from to S. The latest work has been done on 
the organo-sekeniuLi and tellurium halides. 

The Ic^iphatic seleno- and telluro-m.er cap tans are relatively 
unimportant and have not been studied very extensively. Shaw and 
Reid prepared C2H5SeH and tried the reactions given below, v\rhich 
show it to react practically the same as the -SH and -OH compounds. 

(C,H3Se.),C(CH3.)2 < ^^^^ , ^^^^Q 

_(H) I 
CgHsBr + Na2Se2 > CgHg-Se-Se-CgHs > CgHsSeH 




BfT--^.^^ /HBr - (0) - air 



i- 



CgHgMgBr -^^ C2H5SeMgBr Jife-» C2H5SeH "C2H5SeBr3 

lTa(C2H5)S04. + NaHSe > CgHsSeH 40-50^^ 

These mercaptans show an increase in b.p. in going from S to Te, as 
v\fell as bad odors, all sol. in o^-^ganic solvents. 

Other aliphatic compounds are the seleno- and telluro-ketones , 
which are made as belov;. They do not have the disagreeable odors 
of the compounds above, but are not as stable as the oxygen and 
sulfur derivatives . 

(CH3)2C=0 + E^Te --^ > (CH3)2C=Te 

^'^ A reflux ^ "^ 

The aromatic compounds to be discussed are mainly the Aryl-SeH 
type and their derivatives and horaologues. The best methods for 
preparation are given in equations A and B, the latter by the better, 

Br ^.-^eCN 

+ KSeCN > \ \ (A) 

.NHo ..-^-^NgCl ..<.^^,. ^..^-^eCN 




O ^ O 



u 



These selenocyanates cannot be directly converted to the selenophcnol 
due to the ease of oxidation of the selenophcnol. 



OSeCN ^^^^SeOH 

+ H2O — > 1 J + HCNO 



.^ 



-2- 



a 

a 



OSe02H ^^-v^SeH 



+ 



^.-'^SeOH 



^'^^Sc-Sc 



O 



+ HoO 



However, indirectly one can proceed to the selenomercaptans and 
obtain a very good yield. The reactions are shovm in the following: 
scheme (R = alley 1 or aryl) : 



VI (0) kJ IV 





^^^0^ Br, ^ /^eBrJ^ (^'"^^^Jieal 

K^ I CHCI3 k^ Br2(xs)k^ III 

fl in CHCI3 

/ HBffI NaOH 

H2O / AggO xs 



^^--^ScBr 
Br k^ XI 



1-R 
XII 

+ HBr 



O 



SeOAc 



.HAc 
"llaAc 




eOH 



VIII 



O 



Se02H 



V 



Bromine is used only as an example here and chlorine gives the same 
reactions, its compounds being more stable than the bromine deriva- 
tives. The iodine derivatives axe all quite unstable. The dinitro 
and trinitro derivatives become sp electronegative that only the 
monohalide can be prepared, and the trihalide decomposes very 
rapidly. 

The acids are not as imxportant in the Se and To series as in the 
S series. The sclenonic RSCO3H can best be made by heating CqEq with 
H2Se04. HovvFCVer, it is unstable. 

The tellurium compounds behave in the same manner as the Se so 
no chart will be included. 

These reactions show that the basicity increases from to S as 
evidenced by the formation of chlorides. Also the C6K5S2O3H will 
give a salt with nitric acid having the formula CeH5Se(02)0N02 . These 
compounds have not been studied as much as S derivatives due to their 
odors and also because they are very poisonous. 



•r" 



X- 



^'N- 



'v.,. 



J 






f 



^' 






■.X' 



I k.^ 



:;-A'. ; 



'^/ 



*>^^ ,• 






-3- 



Bibliography : 



Behaghel et al., Ber., 65, 812 (1932),' ibid., 66, 708 (1933); ibid., 
6J7, 105 (1934); ibid., 68, 1540 (1935); ibid., 72, 582 (1939). 

Morgan et al . , J. Chem. Soc, 1925 , 1757; ibid., 193^ 2599; ibid., 
1929 , 2214. 

G-a^thv/aite, Kenyon and Phillips , ibid., 1928 , 2280, 2293, 2880. 

Poster et al , , J. Am. Chem. Soc, 50, 1182 (1928); Rec. trav chim., 

53, 405 (3934). 
Shaw and Reid, ibid., 48, 520 (1926). 



Reported by P. E. -^urney 
April 19, 1939 



THE CHEMISTRY MD STRUCTURE OF LldT IN 

Hibbert -- McG-ill University, Montreal 

Preudenberg -- Chem. Institut d. Universit^t, Heidelberg 

I. Isolation of Lignin ; Lignin has not been isolated unchanged. 
Whatever method is employed, a lignin preparation is obtained which 
is no longer identical with the natural lignin. The various methods 
for the isolation of lignin may be conveniently divided into two 
classes: 1) those that depend on the removal by hydrolysis of Ihe 
cellulose and other components, leaving the lignin as an insoluble 
residue, and 2) those that depend on the removal of lignin from the 
substances with which it is associated. 

Methods of Class 1 ; 

1. PQason Sulfuric Acid Method. Ground wood is gelatinized 
with 66^ sulfuric acid. Product is diluted with water, filtered, and 
washed; it is then heated on the steam bath with 0,^% HCl for 12 hrs., 
and finally washed and dried. 

2. Willst^tter Fuming HCl method; Wood is hydrolyzed v/ith H^l 
producing a product called "Willst^tter lignin". 

3. Freudenberg' s "Cuproxam" Lignin. Wood is treated ivith 
Schweitzer's reagent followed by treatment with dilute acids until 
residue has the same methoxyl content as "Willst^itter lignin". 

Met hods of Class 2 : 

T. Sulfite Method. The delignif ication is b.rought about by 
heating wood vn.th acid sulfites under pressure. Lignin is obtained 
as water soluble sulfonic acids. 

2. Separation by Alcoholysis. Lignin is heated with various 
hydro lytic compounds, e.g., ethanol , butanol , ethylene glycol, phenol, 
etc , , to form a product containing the alkyl or ar^^l group in combina- 
tion with the lisnin. 



■o-^ 



II, Constituent G-roups in Lignin : The presence of methoxyl and 
hydroxyl groups ilas been definitely established in lignin. Alcoholic 
as v^ell as phenolic hydroxyl groups seem to be present. 

Freudenberg obtained formaldehyde upon distilling Urban' s 
lignin with 12/o HCl and accordingly advanced the laypothesis that the 
formaldehyde arises from a methylene dioxide grouping (-0CH20^ in the 
lignin. Hibbert disagrees with thj.3 hypothesis. 

Klason has postulated the presence of the ethylenic bond to 
account for the formation of rather stable lignin sulfonic acids. 
Hibbert disproved thi^ by obtaining negative results when fully 
metliylated spruce lignin was 3ubj3;:;ted to the action of hydrogen in 
presence of catalysts under conditions in v^^hich reduction of open 
chain ethylenic linkages readily occurs. 

The evidence as to presence of carbonyl is not conclusive. 

III. Reactions of Lignin : 1. Acylati on. Acetyl content of 
acctylated lignm varies considerably' \vitli the source of lignin. 

2. Alkyl at io n. Alky lat ion readily occurs with usual agents. 

3. Halogena^ETon . Substitution but not addition of halogens 
occurs. Little is Icnown about the character 01 the products. 

4. Nitration . Lignin is nitrated very rea.dil^'- leading to 
the conclusion that lignin is of phenolic constitution. 

5* Sulfonation . Li^jiin dissolves in sulfurous acid and acid 
sulfites to form lignin sulfonic acids. Melandcr found two fractions 
of lignin sulfonic acids in sulfite liquor. One could be precipitatoc 



O ■■jV^t-.• 



■ j.;Of; 



■:• i: .rr> 



J '^i^J; 



^TOl. ';>.:: 






-2- 

with NaCl, called a-lignin-S-acid; the other was not precipitated 
with NaCl and called p-lignin-S-acid. Sulfonation proceeds in two 
stages. In the first stage an insoluble lignin sulfonic acid is 
foimed, which in the presence of the cooking liquor is converted into 
soluble lignin sulfonic acids. Prieser shov\/ed that lignin already 
completely freed from polysacchsirides by treatment v;ith mineral acids 
undergoes autocondensation under the influence of the acid and can no 
longer be dissolved by sulfonation. 

^* Oxidation . When lignin is subjected to oxidation even undei 
mild conditions , complete disruption of the molecule occurs giving 
simple degradation products. Pennanganate oxidation yields simple 
organic acids, as formic, acetic, oxalic, etc. Oxidation of methy- 
lated lignin yielded anisic acid and of ethylated lignin yielded 
p-ethoxybenzoic acid. These results indicate the presence of the 
xollov/ing group in the molecule of lignin: ^-HOCeH^.-^- . 

7. Zinc Dust Distillation . Philips distilled alkali lignin 
from com cobs with Zn dust in an atmosphere of H2 at less than 400*^0. 
The following products were obtained: catechol, guaiacol, 1-n-propyl- 
3-methoxy-4-hydroxybenzene, and anisic acid upon permanganate oxid. 

8. Dry Distillation . Dry-distilled alkali lignin i2rom com 
cobs at 25 nan. m presence of COg yielded same products as in 7 and 
in addition, phenol, o-cresol, and 1- vinyl- 3-niethoxy-4-hy droxy- 
benzene. Results of reactions 7 and 8 indicate the following tvw 
nuclei in the structure of lignin: OCH- 

CH30/ Vc-c-c HO-/ yc-c-c 



,^c-c-c no-Q. 



IV. Possible Structures of Lignin : The literature concerning 
the structure of lignin is now extensive and a large number of 
constitutional formulas have been proposed. All these formulas are 
more or less speculative in character, although evidence of a frag- 
mentary character can be mustered for their support. 

A. Hibbert ' s Structure : Hibbert subjected spruce wood meal 
and maple wood meal to the action of HCl and ethanol to obtain an 
ethanol lignin. By precipitating the ethanolysis mixture with HgO 
he obtained three fractions: (1) crude water soluble oils; (2) water 
insoluble ethanol lignin; and (3) ethoxylated lignin left in wood 
residue, Fraction 1 v;as investigated chiefly and found to contain 
some readily distillable phenols and aldehydes; in addition it con- 
tained some non- distillable products which could be converted to 
distillable products by pressure hydrogenation. Prom fraction 1 of 
spruce wood a-ethoxypropiovanillone (lA) was obtained. Prom maple 
wood a-ethoxypropiosyringone (IIA) was also obtained, v/ith aldehyde 

'^^^' nn w OGH3 

C06HCH3 ^^^^^ ) HO-/ yCOCHCHa HO-/ V-COCHCH3 

OCH3 I OCH3 lA OCH3 II 



HCl 



C2H5OH 



OCH3 OCH3' 

) — i OH ) — v OC2H5 

HO-/ \-6HCH2CHO HO-/ VCOCHCH3 

OCH3 III OCH3 IIA 



■•liirl: 



I 



. n-':i:;'n;-i. 






i 



0- 



.1 ;. 



-3- 



The following aromatic fission products were obtained from 
spruce and maple li,5nin sulfonic acids: vaiiillin, acetovanillone , 
guaiacol, syringic aldehyde, acetosyringone , pyrogallol 1,3-dimethyl 



ether, 
the 



It 



following 



appears that lignin is a simple condensation polymer of 
bwo structural units : OCH3 

C-C HO-/ Vc-C-C 



HO 



O- 



•U' 



OCH3 OCH3 

this connection it can be pointed out that ultra-violet spectra 
and lignin derivatives indicated a benzcnoid structure for 
lignin resembling the spectra of conifcryl alcohol and related com- 
pounds. Hibbert postulates the existence of the following typos of 
structujres in lignin in addition to the types I, II, and III: 



In 
lignin 



01 



HO 



OCH-, 



OH 



•CH-CHo^ N ( 



6h 



IV 



OCH- 



HO 



O 

OCH, 



OH^GH=CH OH J V 

0^ .0— ( VOH V 

^CH=CH \ ( 

OCH3 



/ \ ^^ 

HO/ V6~CHCH3 



CH3CHC0 



6h 



VI 



-> trimer 



HO 




OH , 
/CHg-iH^Q 



0. 



6^gH-cH. 



/ 



OCH, ' OH 




OH 



HO-/ VC. 

^ / CH-CH^j 

CH,0 6e I 




OH VII 
OCH3 

OH VIII 
OCH3 



Until quite recently there has been a general tendency to 
ascribe a highly com.plex structure to li^iin, but the opinion is 
growing that in the wood it is a simple substance of low molecular 
weight and which, due to its unstable character, very 'readily under- 
goes condensation-polymerization reactions yielding the more complex 
materials isolated in the form of ""extracted" lignins . Hibbert sug- 
gests that structural typesi, II, III, IV, V, and VI probably exist 
in natural lignins, while in extracted lignin, types VII and VIII 
exist as reversible and irreversible polymers. Staudinger's molecula: 
weight determinations on solid lignins show that they have relatively'" 
low molecular v/eights and their whole behavior indicates that they 
are built up on a different principle than cellulose and other 
similar high molecular weight polymers, 

Hibbert obtained a-ethoxypropioveratrone (XI) from methylated 
ethanol lignin. The final steps of the synthesis of XI were carried 

OC2H5 
-> CH3O-/ \-COCHCH3 

^ 'C^HgOH ) — ^ 
■COCHCH3 +gci OCH3 



out in two different ways 

OCOCPI3 
CH30-y yC0CHCH3 

OCH3 IX 



(A) CpHgOH + HCl 




CH3O 



OCH3 



XI 



OH 



X 



Step A gives a yield of 80^o while step B gives 12/o, a large amount of 
a polymer being formed. It is highly improbable that the a-liydroxyl 



ii :\i.&i::::^(yyu. 



-\ 



v.. 



\^J 



r , 
"' -4 



> 



— -~\ 



I 



-4- 



group in a building unit such as X exists free in the wood but is 
bound up, possibly as a glucoside. In the extraction process with 
HCl and ethanol an ether interchange probably occurs giving the 
ethoxy derivative. Hibbert shov;ed that on warming I (related to X) 
with dilute H2SO4 it was converted into a light brovm condensation 
product nearly identical with Preudenberg' s "cuproxam''' lignin. 

Possible Synthesis of Lignin Buidling Units in ITatu re : Methyl 
glyoxal formed during plant metabolism is postulated by Hibbert as 
the key substance in the formation of lignin. It is knovm to form a 
dimer of unknown structure. Possibly it is the cyclic dikctone XII 
formed by an intcrmolecular aldol condensation of two moles of the 
methyl glyoxal. Lignin building units IV and V arc then formed by 
condensation with guaiacol and syringone. 

9K 

/;h=ch^ 



^CHs-CH^ 



2 GH3COCHO-* 0=C 

XII OH ' ' 



C=0 



o=c 



-H20 



o 



IV HO 

CH3O 




2 HO 
CH3O 

^OH 
H^CH2-6h^0H/— N _H,o 

tJ VoH — ^ HO 
^CH-CH2'^ \_/ 

OH OCK3 CH3O 



^CH=CH^ 



2 HO 



.0=0 



o 




CH30 



gH,CH=Cfi^OH 



,/ 



OH 



CH=CH' \ ( 

OCH3 



Building units I and II could be formed from vanillin and CE3CHO 
through a "carbo^Q^lase" enzyme synthesis similar to ITeuberg's syn- 
thesis of a-hydroxypropiophenone by the addition of benzaldehydc to 
a solution of glucose under vigorous yeast fermentation. 



OCHO 
+ CH3CHO 

OCH^ 



OCHCOCH3 
6h ^ 

OCE^ 



H 




9=5-CH3 

OH OH - 



HO 



OCH- 



OCO9HCH3 
on 



OCH- 



I . 



B. Frcudcnberg' s Structure of Lignin : Preudenberg regards 
lignin as being made up of low molecular weight units which are 
derivatives of 3?4-(0H)2C6H3-Pr. Primary lignin in young wood prob- 
ably exists as an admixture of structural units XIII and XIV which 
are five benzene units in length. A post mortem condition or the 
action of chemical agents cause formation of 3- "dimensional structures 
XV and XVI of high molecular weight. 



OHoCHO 



/ 

CH. 



a/un2>^nu 
OH y ^ 
I2-0 V-/ 



n-cocH3 

XIII 



CH30 

/ "V-/ ^-i VCH-O— 

CH2-O \ / 

CH3O XIV 



0- 



/ V /CH2-CHOH 

0/ ycH y — V 



CH^-O 



CH,0 



:H-9-CH3 

I OH 







/ 



CH2-O 



o 



)H 



.CH2CH 



XV 



CH-0— XVI 



CH3O 






m*' 






v.— 






-5- 

C. Hilpcrt's "^JGws on Lignin ; Hilpert's views arc drastically 
opposed to those of HiblDert's and -bVcudenberg' s . He claims that 
"native" lignin actually does not exist in the wood but is formed as 
a product of the resinif ication of especially sensitive methylated 
carbohydrates when they are subjected to extraction methods. Most 
of the experimental evidence seems to disprove this viewpoint. 



Bibliography : 

Philips, Chem. Rev., 14, 103 (1934). 

Hibbert, J. Am. Chem. Toe, 59, 2447 (1937); ibid., 58, 340, 345 

(1936); ibid., 61, 509, 516 (1939); ibid., 60, 2274, 2813, 

2815 (1938). 
Hibbert, Can. J. Reseeorch, I3B, 61, 78, 88, 103 (1935); ibid., 14B^ 

12 (1936); ibid., 15B7"38, 404 (1937); ibid., 16B, 54, 69 

(1938). 
Hibbert, Ber. , 70, 560 (1937); ibid., 71, 734, 746 (1938). 
Freudenberg, ibid., 69, 1415 (1936); ibid., 70, 500 (1937); ibid., 

71, 2500 (1938T; Ann., 518, 62 (193577 Papier Pabr. , 36, 34 

TT938). — 

Staudinger, Ber., 69, 1729 (1936). 
Prieso, ibid., 71j~T303 (1938). 
llitikin and Orlova, ibid., 69, 2434 (1936). 
Eilpert, ibid., 70, 413, 560 (1937). 



Reported by L. A. Patterson 
April 26, 1939 



ALIPHATIC DIAZO COMPOimDS AND THEIR REACTIONS 
WITH CAJRBONYL DERIVATIVES 

The first aliphatic diazo compound was prepared in 1883 by 
Curtius v;ho obtained eth^l diazoacetate , CH2K2^00C2H5 , by the action 
of nitrous acid on the ester of glycine hydrochloride. Diazoraethane 
v/as prepared by von Pechmaiin in 1894 by treating N-methyl-iT-nitroso- 
benzamide, C6H5C0N(W0)CH3 or methylnitrosourethane , CH3N(N0)C00C2H5 , 
with allcali. 

I. Preparation of Diazo Hydrocarbons : The diazo hydrocarbons 
cire prepared today by several methods: 

(a) Prom nitrosourethanes : 

CH3-N-COOC2H5 + alkali — > CH2l^2 

(b) Prom the amine salt: 

CH3lvTH2-HCl -^.22iL> CH3NHCONH2 + KCl 
CH3iraC0NH2 ^'^ 2 > CH3N(W0)C0NH2 + HgO 
CH3N(1T0)C01JH2 ^^^^ ' ) CH2N2 + CO2 + liHj 



(c) Prom mesityl oxide: 



iTHCHs 



(CH3)2C=CHC0CH3 + CH3m2 —^ (CH3 ) 2C-CH2COCH3 ™^P > 

(CH3)2C-CIl2C0CH3 ^^^^^3^7 ^ CH2lf2 + ( CH3 ) 2C=:CHC0CH3 

CH3-il-K0 

^ BO^i + H2O 

(d) Prom hydrazine: 

H2MH2 + CHCI3 + 3 KOH — * CH2W2 + 3 KCl + 3 H. 

Substituted diazomethanes may be prepared according to the method of 
Staudinger by oxidation of the hydrazones of ketones: 

(CeH5)2C=NNH2 ^^^ > (C6H5)2C=H=K 85-98^ 

II. Reactions with Carbonyl Comioounds ; 

(a) Aldehydes : Aldehydes, both aliphatic and aromatic, behave 
toward diazomethane as though theu contained an active hydrogen atom, 
and give methyl ketones: 

n-CsH,3CH0 ^^S^ 2 > n-CsH,3C0CH3 + Ng 

CeHsCPIO ^^"-^^^^ > C6H5COCH3 + N2 

The homologs of diazomethane are even more reactive and accordingly 
almost quantitative yields of propio- , butyro-, and valerophenones 
Inave been obtained from benzaldehyde and the appropriate diazo com- 
pound. 



;J'- 






-2- 

Amdt has shown that the reaction wi"tti aldehydes is not 
always clean-cut, for in some cases the ketone is formed only as a 
by-prodact, the principal product being the ethylenic oxides. For 
example m- and £-nitrobenzaldehydes give the corresponding aceto- 
phenones"lDut the o-isomer gives o-nitrophenylethylone oxide as the 
main product: 

^2 



[ \ + CH2N2 — > I i + 



Piperonal reacts with diazomethane to give safrole oxide along with 
piperonyl acetone and acetopiperone : 

\0K^ CH2N2 ^o^V^ ^oK^ ^oK^ 

In general it can be stated that aldehydes and diazomethane 
yield a m^ixture of methyl ketone and ethylenic oxide, the relative 
amounts of each depending upon the particular aldehyde involved. 

(b) Ketones: Ketones are much less reactive toward diazo- 
methane than are aldeliydes. Under the proper conditions, however, 
they will react to give a mixture of a higher ketone and an ethylenic 
oxide. 

CH3COCH3 + CH2N2 -^ CH3COCH2CH3 + (CH3)2C-CH2 

'0 
C6H5COCH3 + CH2N2 — > C6H5CH2COCH3 + CeHs-C--- CH2 

CH3^ O'^ 

It has been shov/n that alcohol or water accentuates the dipolar 
condition of the carbonyl group and so aids the reaction. One or the 
other is usually added to the reaction mixture for this purpose. 

Groups which activate the carbonyl favor the formation of the 
ethylenic oxide. Thus Adamson and Kenner obtained an 86^ yield of 
a-phonyl-a-chloromethylethylenc oxide from oa-chloroacetophenone and 
diazomethane in other and methyl alcohol: 

CH2CI 

CeHsCOCHaCl + CH2N2 — > CeHs-t.— CH2 

(/ 

(c) Cyclic Ketones : Cyclic ketones undergo similar reactions 
but in this case ring enlargement takes place. For example, free 
diazomethane reacts with cyclohexanone v\fith the formation of eyclo- 
heptanone and pentamethylene-ethylene oxide along with a trace of 
cyclO(yctanone : 

A 

(CH2)5C0 + GE2^2 — * (CH2)2C0 + (CH2)5=C-CH2 + (CH2)7C0 

Cyclopentanone yields the same products as cyclohexanone. Presumably 
cyclohexanone is first formed and reacts further with the diazomethane 

(CH2)4C0 ^^^^'^^ ) (CH2)5C0 ^^^^^^ > (CH2)6C0 

Cycloheptanone gives a fair yield of cyclodctanone and some hexamethy- 
lene-ethylene oxide. Cyclodctanone is unaffected either by diazo- 



( 
i 



1 






'.^'.^.is^-'^ 



-3- 

methane or by its more active homologs such as diazojethane."' 

However, with the other cyclic ketones mentioned the homologs 
of diazomethane yield substituted cyclic ketones. Thus, cyclohexanone 
when treated with diazoethane and diazo-n-octane yields 2-methyl- and 
2-heptylcycloheptanone respectively. 

2-Methylcyclohexanone yields with diazomethane a mixture of 
2- and 3-methylcycloheptanones and an equal quantity of ethylene 
oxide derivatives: 

9H3 CH3 

^^» CHCH3 pXJ TT ^,,<^CH«^ _„.^ CH 

(CH2)4 I ^ ^ > (CH2)4 J^Hg + (CH2)4. i^=0 

A further extension of this ring enlargement process made by 
Adamson and Kenner involves the substitution of a diazo compound 
having a functional group for the diazolriydro carbon. In this way 
2-6-carbethoxybutylcycloheptanone was prepared: 

(CH2)5C0 + Et02C-N-(CH2)5C02Et > (^2)5 | 

10 ^---CH(CH2)4C02Et 

(d) Diketones: Aliphatic diketones such as diacctyl and 
acetonylacetone are not affected by diazomethane but benzil reacts 
to give a,[3-dihydroxystilbene methylene ether: 

CsHsCOCOCsHs + CH2N2 > CeHs-C^-C-CsHs 

III. Mechanism : The mechanism of the reactions of the alipha- 
tic diazo compounds naturally entails a consideration of their stiuc- 
ture. Curtius originally used the cyclic structure which was accepte 
for many years until Angeli suggested the linear formula R2^'=^^ — N* 
Opinion was divided on the subject since work of either a physical 01 
chemical nature led to ambiguous results. 

It became obvious that the Angeli foimula containing the 
pentacovalent nitrogen atom could not be correct and Langmuir pro- 
posed the structure R2C=N=iN. Another modification of Angeli ' s for- 
mula, in which the middle nitrogen atom donates two electrons to the 
carbon atom, has been considered: R2C't--N=-lT. 

Electron diffraction studies by Boersch have shovm conclusive- 
ly that the chain foimula is the corro-ct one, but the low dipole 
moment of the diazo group excludes either of the two chain foimulas 
alone. The only difference between the two lies in the distribution 
of electrons and so the tendency today is to consider diazo compound? 
as a resonance mixture of the two electronic structures: 

R2C=N=^N fc=> R2C-«-NSN 

Either of these two structures might be expected to undergo 1,3- 
addition in the following manner: 

R2C=¥==^N + XY > R2C-N=N-Y 

X 



■•-•?.-ii/•J^.■■ 



"J 



i J '■ 



• "PJi 



-4- 

Applying this to the reaction with carbonyls, the mechanism may b^ 
considered as taking place in the following manner: 



For aldehydes : 



R-C=0 
H 



^^ 2 ^2 , 



^0 ^ 

R-CH II 



Similarly for ketones: 



R. 



0=0 



R 



y 






.^5^ R-CH 



R^ ^0- 


R^ "^CH. 




ROH-OH- 



* ROOCH3 
RCH2CHO 



< 



Ro — CHo 



ROOOH2R 



Bibliography : 

Adamson and Kenner, J. Ohem. Soc, 1939 , 181. 

Smith, Ohem. Rev., 23, 193 (1938). 

Amdt and Amende, Z. rjigew Ohem,, 43, 444 (1930) 

Curtius, Ber., 16B, 2230 (1883); J. prakt . Ohem. 

von Pechmann, Ber., 27B, 1888 (1894). 

Staudinger, ibid., A-W 7 1928 (1916). 

Amdt, Z. angew. Ohem., 40, 1099 (1927). 

Langmuir, J. Am. Ohem. Soc, 41, 1546 (1919). 

Boersch, Monatsh. , 65, 331 (1^5). 

Meerwein and Bumeleit, Ber., 61B, 1840 (1928). 



(2) 59, 107 (1389) 



Reported by 1/V. H. Rieger 
April 26, 1939 



■ iVl 



SOME RECENT ADVANCES IN CHEMILUMINESCENCE 

I. Introduction : Chemiluminescence may be defined as the 
light given off as a direct result of a chemical reaction. This 
discussion v/ill be limited to those types of chemiluminescence which 
are produced by oxidation reactions in solution. 

A. Early Work ; 

' 1. liioluminescence : Marine animals, fireflies, etc., 
have all been shown to produce chemilumiscence by the action of 
atmospheric O2 on a substance called "luciferin" in the presence of 
an enzyme lucif erase. 

2. Yellow phosphorus in air. 

3. Pyrogallol 

a. Enzymatic (potato juice + air) 

b. HCOH + KOH + H2O2 . 

B. Lophine (triphenylglyoxaline , I ) : Lophine produces a good 
light when oxidized either by H2O2 + J^aUlO + NaOH^ or by H2O2 + 
K3Ee(CN)6 + NaOH. Radziszev\fki^ , who discovered the luminescent 
properties of lophine prepared it in the following two ways: 



3 r^ + 2 NH3 > (/^CH=N)2CH-/~^ 

a 



Op (heat in air) 



C— N-H 
lophine (I) || C-/ \ + H2O 



O' 



// 

■N 



C6H5-C=0 
G6H5-C=0 



+ CgHsCHO + 2 NH3 > I + 3 H2O 



Using these two methods many derivatives gave been prepared^. 
The one that gave the best chemiluminescence was the tri-£-methoxy 
derivative II. lophine, being easily prepared, is often satisfactory 

for demonstrations^, although 
CH3OC5H4-C — N-H luminol is still unequaled for this 



CH3OC6H4 



II ): — C6H4OCH3 II purpose 



C. Luminol (3-aminophthalhydrazide , III) : Luminol, discovered 
by Albrecht,"^ can be prepared as followsT^ 

NOo NO2 NH2 

G^^g (NH2NH2)2H2S04 ^ r^^^^H (NHjpS ^ f^^^NH jjj^ 
C^O CH3C00Na Sv>c4^^ \>C=0 



In recent years many similar compounds have been prepared and 
studied. Wegler^ reports that the alkaline diazo solution of luminol 
gives brighter luminescence than luminol itself. 

Drew and PearmanQ after preparing and studying many derivative 
of phthalaz-l,4-dione conclude "that substitution of immobile groups 
for two of the enolizable H atoms removes the luminescence and that 



i.^ :".•■- 



<1 



V..^ 



\<^ 






'■itfV if1rifrVitT'-'"-mi I... 



/*••■-:. 



•-v.y^' 



-2- 



the substitution even for one of them greatly diminishes, if it does 
not entirely remove that property". 

Drew and -Grarwood^ were able to isolate a peroxide (CQHe04.N3Na, 
the Na salt of 5-amino-l, 4-dihydroxy-2,3-dihydrophthalazine peroxide) 
by the action of HgO^ on a basic solution of 5-£^inophthalaz-l,4- 
dione. This peroxide gave chemiluminescence in a water solution in 
the presence of hemoglobin or copper salts. 

II. Some Recently Discovered Chemilumine scent Reactions : 

A. Uitr'aconic anhydride : it has been observed that 
the decomposition products of a large number of organic substances 
are chemilumine scent when oxidized. Of the many examples observed 
only in the case of citric acid have pure substances been isolated 
from the decomposition products and tested for chemiluminescence. 
Of the substances isolated only citraconic anhydride (IV) gave 
chemili;miincsoence . 



CH2COOPI 

6{0H)C00H 

6H2COOH 



heat 



CH2COOH 
C-COOH 



•COOH 



citric acid 



CHCOOH 
aconitic acid 

i 



CHCOOH 



9H3 

C — GpO 



IV 



itaconic 
acid 



CH. 
II ^ 

C-COOH 

I 

CH2COOE 



citraconic acid citraconic anhydride 

CH2 
II ^ 

> C 0=0 itaconic anhydride 

CHo-Cf=0 



B. Biacridylium salts : G-len, Glen and Petsch, and 
Decker and Petsch' '^^ have shown that N,N' -dimethylbiacrydylium dini- 
trate (V) will give a very bright chemiluminescence when allowed to 
react with HgOg in an alkaline solution. A few drops of OsO^. solu- 
tion is reported to increase greatly the brightness of the light at 
the expense of its duration. These workers report that this is the 
brightest case of chemiluminescence known, even brighter than luminol 
N,1T' -Dimothylbiacrydylium dinitrate may be prepared as follows 

CH3 QH- 

Nt..^^ (CH3SO4.) 



'¥• 




^ 



( CH3 ) 2 SO 



^ 




alkaline 
K3Pe(CN)6 




9?3 



(1^03) 



'^ 




^. 



HNO- 




6g- 



(NO3) 




Zn, CH-^COOH 



(OH) 



C. Complex magnesium compounds : Helberger,'^ and 
Helberger and Hever''^ have recently reported that magnesium phthalO' 
cyanine VI and many magnesium complexes of the porphyrins VII were 
chemilumine scent when added to boiling tctralin. They showed that 



■■-».';'^ 



vr....>^. .." 



>.A 



I ;?■ ^ I- 

t 

f 



i nil 



.•<> 



-3- 



this was due to the presence of tetralin peroxide in commercial 
tetralin. 




VI 



Magnesium 
phthal o cy anine 



I 5 






\ I 



VII 



CH=^/Nx^CH 



g complex of 
porphyrin 
nucleus ' ^ 



I 6 



Rothemund' has shown that both 
chlorophyll a and chlorophyll b 
are chemi luminescent under similar conditions. He also worked v/ith 
derivatives of the porphyrins. 

It has long been known that aromatic G-rignard reagents will 
produce chemi luminescence on oxidation.'"^ The brightest one to be 
reported is £-chlorophenylmagnesiuin bromide. 

III. The Use of Hemoglobin or Hemin as an Oxidizing Agent ; For 
most organic chemiluminescent reactions it is necessary for best 
results to have tv/o oxidizing agents present. One of these is a 
peroxide such as H2O2 , Na202, .KaB03 , benzoyl peroxide, etc. The 
other m.ay be any one of a large number of substances such as K3Pe(CN)g 
KaClO, etc., whj ch will be reduced hj H2O2 in basic solution to 
liberate ox^-^gen; ^ ' One of "the most interesting substances that has 
been used as an oxidizing agent is hemoglobin or hemin (VIII). 



CE2CH2COOH 




I CH2CH2COOH 



I 6 



Hemin'" (VIII) 




'CH2CH^ 



IX 



01 



Many workers have reported the use 
of this substance. Some have attrib- 
uted its influence to a catalytic 
effect '^'2° but it has been found 2' 
that substances such as iron, iron 

compounds, Mn02 , etc. which 
liberate oxygen from H2O2 by purely 
catalytic action do not produce any 
chemi luminescence with lophine . 

Thiebert and Pfeiffer^s have found 
that m.any other iron complexes can 
be used instead of hemin. The best 
v/as salicylaldehyde-eth^lenediimine 
f errichloride (IX). 

A very interesting article appearec 
in 1937 by Specht^^ suggesting what ±z 
probably the first practical use of 
chemi lujnine s c enc e . Specht suggests 
that a solution composed of luminol , 
H2O2, 1-Ia2003 and H2O could be used 
for the detection and identification 
of blood stains in criminal investi- 
gations. A solution of the above 
ingredients can be applied in the 
dark with an atomizer to an area that 



^!V 



X- 






•■■> ;?„.''■-''■•" 



-4- 

is suspected of being blood stained, and if any trace of blood is 
present a glow will be produced. This method is capable of detecting 
and identifying blood specks that are either too small or so changed 
in appearance as to be overlooked in an ordinary examination. The 
older the blood stains the better the method works. 

Numerous tests showed that the following substances give no 
luminescence with this solution: saliva, urine, pus and other body 
secretions, milk, coffee spots, starch, organic or inorganic dyes, 
fabrics, leather, skin, fungus, oils, waxes, earth, stone, wood, 
metal rust, grass, or leaves. 

Since Specht's article appeared two exceptions have been 
noted, namely certain samples of cinders, and the cores of cabbage 
and lettuce heads in the region of the cambium layer, 

Bibliog phy ; 

1. Ward and Bennett, Trans. Illinois State Acad. Sci., 30, 198 (1937), 

2. Cottman, Moffett and Moffett, Proc. Indiana Acad. ScIT, 47, 124 

(1938). -~ 

3. Radziszewki, Bcr. , 10, 70 (1877); ibid., 15, 1493 (1882). 

4. Cottman, J. Chem. E^cation, 14, 236 (1937T. 

5. Albrecht, Z. physik. Chem., 13^, 321 (1928). 

6. Huntress, Stanley and Parker, J. Chem. Education, 11, 142 (1934). 

7. Wegler. J. prakt. Chem., 148, 135 (1937). 

8. Drew and Pearman, J. Chem. Soc. , 1937 ? 26. 

9. Drew and Garwood, ibid., 1938 , 79T: 

10. Glen, Z. angew. Chem., 4771T0 (1934). 

11. Gleu and Petsch, ibid,, 48, 57 (1935). 

12. Decker and Petsch, J. prakt. Chem., 143, 211 (1935). 

13. Helberger, Naturwissenschaften, 26, 31b (1938). 

14. Helberger and Hever, Ber. , 72B, TT (1939). 

15. Gilman, "Organic Chemistry", John Wiley and Sons, Inc., Nev; York, 

1938, Vol. 2, p. 1647. 

16. Rothemund, J. Am. Chem. Soc, 60, 2005 (1938). 

17. 'Evans and Dufford, J. Am. Chem. Soc, 47, 295 (1925). 

18. Dietz, J. Chem. Education, 12, 208 (195^). 

19. Wegler, J. prakt. Chem., 14^7 135 (1937). 

20. GlcTi and Pfannsteil, J. p5Falct . Chem., I46, 137 (1936). 

21. Cottman, Thesis, M.A. Butler Universit3rTi935) . 

22. Thiebert and Pfeiff er, Ber., 71B, 1399 (1938), 

23. Specht, Z. angew. Chem., 50, tSS (1937). 



Reported by R. B. Moffett 
May 4, 1939 



ACTION OP ORGMIC ACID CHLORIDES UPON ALIPHATIC ETHYLEITIC 
HYDROCARBONS IN THE PRESENCE OP STANNIC CHLORIDE 

Colonge and Mostafavi -- Ecole de Chimie Industrielle 
et Paculte des Sciences de Lyon 

The addition of acid chlorides to the d:hylenic double bond 
was reported by Kondal^off^ a half -century ago. He pointed out the 
role of zinc chloride as a catalyst in the addition reaction between 
trimethylethylene and acetyl chloride to produce a (3-chloroketone 
which could then lose-one molecule of hydrogen chloride, leading to 
the formation of an a-unsaturated ketone. In 1898, Blanc^ reacted 
acetyl chloride with l,l,2-trimethylcyclopentene-2. He employed 
AICI3 as catalyst, thinking to apply the method of Priedel-Ciafts to 
aliphatic hydrocarbons. Ten years later, Krapiwin'^ treated certain 
olefins with acetyl chloride or bromide in the presence of large 
amounts of aluminum bromide or chloride at a low temperature. He 
reporlE d 20-40^ yields of certain unsaturated ketones and observed the 
transitory formation of p-chloroketones. He stated that tetramethyl- 
ethylene did not react with acid chlorides to produce ketonic products. 
According to him, the absence of a hydrogen atom on the carbon atom 
of the double bond prevents the normal reaction. Still later, Darzens^ 
reported that cyclohexene adds acetyl chloride in the presence of 
certain metallic chlorides, of which stannic chloride is by far the 
most efficient catalyst. He extended the reaction to the homologs of 
acetyl chloride. 

In the years since Darzens' work there have been numerous 
applications of this method which the authors call the Krapiwin- 
Darzens method. Most of these v^orkers, thinking to apply the procedure 
of Prie del -Crafts in which AICI3 is more a reactant than a catalyst, - 
used large quantities of the metallic chloride and worked at tempera- 
tures below 0°C, 

In repeating the work of Kondalcoff , the authors have shov^fn 
that stannic chloride is a much more active catalyst for the reaction 
than is zinc chloride. Stannic chloride possess the added advantage 
of being soluble in organic acid chlorides and also in mixtures of 
olefins and acid chlorides. 

Influence of Nature of Catalyst on Yield ; The reaction was 
carried out with propionyl chloride and trimethylethylene in the 
presence of various metallic chlorides. The yields of unsaturated 
ketone follow: 

Catalyst Yield 

HgCl2 " ^0 

ZnCl2 16 

AICI3 13 

TiCl4 40 

SnGl4 60 

Mechanism of the Reaction : Two mechanisms may be postulated: 

I 

-C- -C-Cl -C 

(1) 11 + RCOCl > I > II + HCl 

-CH -CH-COR -C-COR 



.^lEviilE;; -^.1:1^ Dl,jJi{)/ 



•.»^»L. — — J, Vk ; 



.'5.*3S »!= 



-•Ji^ 






. _ oi:- iiT-iv; :: ...... 



:.<^ifijt 



n% 'N :-. J- ;•-.- 



- hf 



1. J / 









. c. J-- .!. ^ I,' .L no 



_i 



a-cYi.: 



~ r 



c •-■-■■ .■ .- 



fifirioea owT rnoj:^-:^.;; .ji -a':- r.; ~-r 



-■}■ 



.wo -HO- 






no- 



-2- 

-C- -C- -C-Cl 

(2) il + RCOCl > Ij + HCl -— > I 

-CH -C-COR -CH-COR 

Mechanism (1) is believed to represent the true picture of 
what occurs, because, contrary to Krapiwin's findings, the authors 
were able to add. acetyl chloride to te tram ethyl ethyl en c , obtaining 
a product which could be explained only by an initial addition. 

Constitution of Products ; An acid chloride mcy add to an 
unsymmctrical ethylenic hydrocarbon in two possible ways, Experimenta] 
results of this and of former investigations have proved that the 
chlorine atom fixes itself upon the more highly substituted carbon 
atom and the acyl radical upon the carbon atom carrying the hydrogen 
in compounds of the type RR,C=CHR2. 

Although the addition of the acid chloride furnishes only a 
single p-chloroketone, the less of h;>^drogen chloride may, by contrast, 
yieid either-the a- or the p-unsaturatcd ketone. For example, 
4,5-dimethyl-5-chlorohexanone-3 "by losing hydrogen chloride may yield 
two different ketones: 

CI - H __-.————' CH3-(!j=2=C-C0CH2CH3 

CH3-6 C-COCH2CH3 "^IIIIIIIl 

6H3 6H3 ~~~ -^CH2==C — CH-COCH2CH3 

CH3 6H3 

Which isomer is obtained is determined by whether the hydrogen atom 
which--is eliminated on losing hydrogen chloride is on the a- or on 
the TT-carbon atom with respect to the carbonyl group. It is generally 
accepted that the hydrogen on the a- carbon is the more mobile, althougj 
in some cases the two isomeric unsaturated ketones are formed simul- 
taneously. 

The simultaneous formation of the a- and p-unsaturated ketones 
by loss of hydrogen chloride from the p-chloroketone seems to depend 
upon the nature of the hydrocarbon chain. That is, v;hereas the 
products resulting from the addition of acid chlorides to trimcthyl- 
ethylenc are always mixtures of the two forms, that fomied when 
2-methylpropene reaets with propionyl chloride is a pure compound, 
5 -methyl- 4-hexene- 3- one : 

CH3 
CH3-.6=CH-C0CH2CH3 

One may deduce from the above that the presence of a hydrocarbon 
residue on the a- carbon with relation to the carbonyl group favors 
the formation of the p-unsaturated isomer along v;ith the a-isomer. 

Experimental Results : The following table gives the yields 
of unsaturated ketones after removal of liCl from the p-chloroketones : 

Hydrocarbon 

Heptene-1 - 

2 -Methyl -1-propene 

2-Methyl-2-butene 
If ti 



Acid chloride 


Yield 


Acetyl 


not det'd. 


Propionyl 


30^ 


Acetyl 


53 


Propionyl 


60 


Isobutyiyl 


40 


Pivaloyl 


37 



■-■'v: 



II 



r'f: 



fCf.';' 






.:;.;..' 11 J.O 



,-'• .'•;.. 



7 <;''■'■' 



i 

Y"^^ 



R-f ?.*-?-■ -f 






^f >■:■-•/( 



.U ,v;iu:^|, 



;-vo;:;-:.T '?:;.'" 






■JC)i TOff 



'. > *' 






"fi'". .vrf ••^ .•■■'^r ?■: 






-3- 

2 -Methyl -2-butene Diethyl acetyl 52 



II It 



Benzoyl 40 

2-Mcthyl-2-hexcne Acetyl 65 

" " Propionyl 50 

2,5-Diraothyl-2-butene Acetyl 30 

1, 2 -Dichloro ethylene " 

3-Chloro-l-propcnc " 

In fliionco of Stru.cture of Reactants Upon Yield : Olefins of the 
type RCH=CH2 react energetically with the acid chloride, but v/ith the 
T'.uthors' technique it is impossible to isolate the ketone because it 
condenses vrlth itself under the influence of the reactants. 

More highly substituted olefins of the types R2C=CH2 , R2C=CHR, 
and R2C=CR2, do not give rise to the secondary reaction noted above as 
taking place with mono-substituted ethylenes. 

The influence of the structure of the acid chloride on the 
yield of ketone appears to be rather feeble. In effect, for a given 
olefin, such as trimeth;>^lethylcne , the yield of unsaturated ketone 
rcjiges betivoen fifty and sixty per cent when the acid chloride is of 
the type RCHgCOCl. These yields ccce lowered to forty per cent when 
the acid chlorides are branched, as R2CPICOCI and R3C-COCI. 

Bibliography : 

1. Colonge and Mostafavi, Bull. soc. chim., (5) 6, 335 (1939). 

2. Kondrl^off, ibid., (3) 7, 576 (1892). 

3. Blanc, ibid., (3) 19, 703 (1898). 

4. Krapiwin, Bull. Soc. Imp. Wat. Moscow, 1908, 1-176. 

5. Darzens, Corapt. rend., 150, 707 (1910); 151, 758 (1910). 

6. Wiclond and Bettag, Ber., 55, 2246 (192277" 

7. Stevens, J. Am. Chem. Soc, £6, 450 (1934). 

8. Worris and Couch, ibid., 42, ?329 (1920). 

9. Calloway, Chem. Rev., 17, 327 (1935). 

10. Kroeger, et al . , J. Org. Chem., 1, I63 (1936). 



Reported by H. E. Conde 
May 3, 1939 






i^'- 



<xa '.rx: v; 



^-;" r 



^■•. 



r ? ■ • 



•-•<•• 



V-i 



=;^-^oO ,,. 



PEGAITIKE - VASICINE 

Fifty years ago D. Hooper discovered an alkaloid, vasicine, in 
the leaves of the Himalayan plant, "Adhatoda vasica (L.) Nees". 
Adhatoda is used widely in India as fish poison, insecticide, and for 
the relief of asthma. Peganine is obtained from the mother liquors 
resulting from the extraction of hazmine and harmaline from the plant 
"Peganum haimala". It was proved to be the same as vasicine. 

The compound has the following properties: 

1. Poimula, C,,H,201l2. M.P. 212°. 

2. Monobasic and foims a methiodide. 

3. POCI3 replaces an OH with CI to form chlorovasicine. 

4. Zerewitinoff determination indicates one active hydrogen. 

5. Acetic anhydride forms two acetyl derivatives. 

6. KMn04 oxidizes it to an acid," the methyl ester of which 
gave anthranilic acid and glycine with NaOH. Sp^th suggested 4 pos- 
sible structures for the methyl ester. 







II t) 

.c. _ c 



a^]^-C0CH3 r"^ \h f^^ ""iJI-CH- 

^^^C-CH3 l^ t-CH2C00CH3 l^ C-COC 



^3 
.C-COOCH3 



III 



II 



I 

jj f^^^N-CHsCOOCH; 



La 6h 

The acid was decarboxylated with Cu-bronze to give 3-methyl- 
4-k:eto-3,4-dihydroquinazoline, a compound already synthesized. There- 
fore it was given the structure IV. Sp^th also synthesized the acid 
as follows: 



II ^=0 + NH2CH2C00CH3 — » LJ 



MCH2C00CH3 



'W "^ NH2 

H 

VI 



HCOOH 



For the alkaloid SpSth and Nikowitz suggested the following possible 
structures, 

^H2-CH0H OH 



aCH2 CH I ^6 

1 CH, . p~/>_iH, f^ 



V 



XVII XVIII IX 

.CHp CHo _ OH. 



CH2 CH— CH2 



H2 ^^^ ^CHp OH 

N-CHpCOCH, r T l[-CHo-C= 



OH 
X XI XII 



;h 



2 - v^ — CH2 







o 



<fi9~ 



J 



o 



^i^i 



i'J. 



^■v ^.^ 



-2- 



Sp^th preferred IX but the fact that the alkaloid cannot be 
catalytically reduced questioned this structure. Hanford, Liang and 
Adams prepared compound XVI which was not identical with the com- 
pound C| 1H14.N2 obtained from vasicine with reduction by sodium and 



alcohol. 
XIII 



=0 



NH2CH2CH=CH2 



H. 



a1-CH2CH=CH2 



Na + 



a, 
1 





C-NH-CH2CH=CH2 



XIV 



HCOOH 



NH 



XVI 



C2H50H 






N-CH2CH=CH2 



XV 



+ 



CO. 



Reynolds and Robinson also synthesized IX by reacting allyl 
iodide with quinazoline and found it was not identical with peganine. 

Two questions now arise: 1) Does peganine consist of a quin- 
azoline ring with a side chain or a tricyclic ring system? 2) Is 
the oxygen in position 4 of the acid resulting from the YMn.0^ treat- 
ment originally present in peganine or does it appear with oxidation? 

In order to prove whether airing system was present desoxy- 
vasicine (XIX) was prepared simultaneously by Hanford, Liang and - 
Adams, and by Sp^th, Kuffner and Platzer. Desoxyvasicine, pegene-9, 
is obtained from chlorovasicine v;ith Zn + HAc. SpMth used the 
following synthetic procedure. 



CH2CI 




XIX 



H2NCH2CH2CH2COOCH3 



a 



8 
CH2 

ill 

9 



■ CH. 



6h' 



CHf 

3 "^ 



POCI3 



a 



1 






CH2^ 

o=d c 
NO2 CH2 

Zn + HCl 
CH. 



2 XVII 






■OE 



'CH< 



6h 



2 XVIII 



An outline of a second synthesis of XIX by Adams et al . 
another by Sp^th, is as follows: 



and 



^H 4h2 

0=C-CfH2 
UO2 

XX 



CH2 CH20C5H5 




E5 _ CH2 



CHsBr 



W^ "^CH^ ^ XXII 



NaOH 



XIX 



n 



-^ 






J 



o 



o 



>i 



a«-j- 



^ 



-3- 



(Sp^th) 



CH. 



%H- 



0- 



•CH. 



0=C 



CHf 



XIX 



The final step in the determination of the structure of pega- 
nine was the placing of the hydroxyl group. To deteirnine this, 
Morris, Hanford and Adams studied the derivative of vasicine obtained 
v;ith H2O2. In proving the structure of this oxidation product desox^^- 
vasicine was oxidized to XXIII whose structure was proved by a 
synthesis similar to that above. Phenoxybutyryl chloride reacted v/it} 
o-aminobonzaraidc to yield the corresponding amide. The ring was 
closed by heat, the phenoxyl group replaced by Br with HBr, and 
finally a ring closure with alkali gave XXIII. V/hen XXIII v/as 
treated with Pb(Ac)4. followed by hydrolysis XXIV was obtained which 



XXIII 



a 






-CH2 

CHo 



CH^ 



a 







;h 



N — 
jj^ CH2 



2 XXIV 



v;as identical with the oxidation product of vasicine 
ture of vasicine was deduced as being XXVIII. 



and the struc- 



A synthesis of peganine by Sp^th cinched the correct stnicture 



a 



N-CH2CH2CHOE 
XXV 



HCl 



NH^ 



heat 



HOOC 



OH 



CHp 

t 

.CH 



NO2C6H4CI 



2 
XXVI 



a' 



CH2V 
o=(i CH2 

OH XXVII 



Zn + HCl 
-H2O 




XXVI I : 



Later Sp^th gave the follov/ing simple synthesis of peganine: 



a 



CH2NH; 



0- 



• CH. 



XXVIII 



+ I i . 

0=C^ .CH2 
mip CH^ 

6h 

Sp^th resolved peganine by means of tartaric acid. With 
d-tartaric acid he isolated the 1-form and from the mother liquors 
likewise the d-form v/ith 1-tartaric acid. He isolated the 1-form of 
vasicine in tHe plant "AdEatoda vasicin". "" 

Finally work has been done on the biogenesis of vasicine by 
Sch6pf and Oe'chler. They-imagined that the quinazoline ring resulted 
from the interaction of o-aminobcnzaldehydc and a-hydroxy-?"- amino - 
butyric aldehyde followed by isomerization and shift of two hydrogen 
atoms, a -Hydroxy- f-aminobutyraldehydc is not laiown but a synthesis 



':h 







■ ^ ■ X") 



.i.i''' , 



{ il 



.^ 



(.!::'♦ 







N 



r' 



-4- 

of desoxyvasicine by the use of if-aininobutyraldehyde in the form of 
the dicthylacetal makes the above hypothesis reasonable. The 
f-aminobutyraldehydc condenses v;ith o-aminobenzaldehyde to form the 
base XXIX in which under the action of palladium and hydrogen two 
liydrogen atoms shift to give desoxyvasicine. 

pTT OH" 

XXIX 



Bibliography : 

Spilth, Monatsh. , 72, 115 (1938). Porty-two references are given 
in this ar^cle. 




Reported by J. Harkema 
May 10, 1939 



- r 




THE TAUTOMERISM OP £-HYDROXYAZO COPilPOUNrS 

H. Shingu — Laboratory of G. Kita, Tokyo 

The introduction of a hydroxy 1 group into aQ aromatic azo com- 
pound gives rise to the possibility of tautomerism. Thus benzeneazo- 
a-naphthol may be transformed to the monophenylliydrazone of a -naphtho- 
quinone, and this reaction would appear to be general. Although the 
hydroxyazo compounds have been studied from many aspects, until recent- 
ly their constitution in many cases has not been satisfactorily estab- 
lished. The coupling reaction used in their preparation and the 
directing influence of the hydroxy group on bromination or nitration 
indicate the structure I, whereas condensation with hydrazine or 
addition of dienos appear to favor the structure II. 



/^\N=Ny VoH fc=^ ^^A-NH-n/ Vo \ / 



N-Ar III 
I / \ II / \ 0-^^ 

azoid V / quinoid 

In the case of the o-hydroxyazo compounds most of the evidence 
indicates the hydrazone strm.cturo III {3a). The compounds are sluggish 
to alkali and phenylisocyanate; their absorption spectra suj^port the 
quinoid structure (2). The evidence of infra-red absorption spectra 
(4) indicates hydrogen bridging and the chelate structure is preferred. 

The case of the ^.-hydroxyazo compounds is not as clear cut. The 
oxidation of p ,p' -dihydroxyazobenzene by silver oxide to the corres- 
ponding quinone azine indicates that it has the azoid structure (3a). 

HO-/^^-N=¥-/ VoH > o=( Vn-¥=/ Vo 

Many p- hydroxyazo compounds in contrast to the £- are alkali soluble 
and react with phenylisocyanate. 

However, Lauer and Miller (6) using the Diels -Alder reaction 
showed that certain of them, such, as 2,4-dinitrobenzenoazophenol, reac-i 
as the quinone -hydrazone (at least two nitro groups in Ar were 
required) . 


R^y^ /CH=CH R,^/\^CH^CH 

+ CH2 I — It 1 JH2 H (R = H, CH3, 
^CH=CH \j^>^6h^CH 2t) 

I-iraAr K-NHAr 

Kuhn and BSr (5) concluded from their study of the absorption 
spectra of benzeneazo-a-naphthol that the reaction I *=> II took place 
depending on the solvent. The absorption spectrum in pyridine was 
different from that in acetic acid. Bergmann and Weizmann (1) con- 
firmed Kuhn's conclusions on the basis of dielectric measurements of 
£-hydroxyazo compounds . 

The body of this report will deal chiefly with the work of 




-rrnf 



v) 




T,lf, 



\J 




•^=«o 



-■f^ V* 



\3=/ >C=/ 



-• 



0--0" 



ocTtr 



;j ji 



lOr^ 



-^IIo. 



5i 



J-rs. 




-2- 



Shingu (8), who has conducted extensive absorption spectra studies and 
has applied the electron and chromophore theories to their interpre- 
tation. 

Bcnzeneazoanthranol ; - This author prepared benzencazoanthranol, 
to which on the basis or spectral and other evidence a quinone- mono- 
phenylhydrazone structure may be assigned. The benzeneazoanthrcjiol 
made from anthranol and diazotized aniline is identical with the 
anthraquinone-monophenylhydrazone made through condensation of meso- 
dibromo an throne and phenylhydrazine . 

Methylation with dimethyl sulfate or methyl iodide and alkali 
yields the N-methyl derivative. Diazomcthane fails to raethylate. The 
benzenoazoanthranol is split by acids into anthraquinone and phenyl- 
hydrazine . 

In respect to the stability of the azoid and quinoid structures, 
benzeneazo-a-naphthol is intermediate between £-hydroxyazobenzenc, 
v/hich has the azoid structure, and benzeneazooiithrcjiol, which has the 
quinoid structure, a fact which parallels the stability of the keto 
forms of phenol, a-naphthol and anthranol. This is corroborated by a 
consideration of the difference between the azoid and hydrazone- 
quinoid total bond energies, as calculated according to Pauling (7). 



Coupling component 



phenol 



a-naphthol 



rjithrrjiol 



Approximate stmc- 
tural formula 



N=N- N-NH- 





II=N- 



]^T-M- 



^ c^ 



Difference of the 
total bond energies 
( azoid - quinoid) 



-35 Cal 



-6 Cal. 



+23 Cal. 



The chief band of benzenoazoanthrajiol , occurring 
unaltered by the solvent (pyridine favors the azoid cjid 
the quinoid form) . A comprxison of the absorption bcjid 
the N-methyl and 0-acyl derivatives in Pig. 1 of knovjn 
the parent substance shows that the benzenoazoanthranol 
ogous to that of the N-methyl derivative, and according 
may be formulated as quinoid. The displacement of the 
360 A to the red may bo attributed to the bathy chromic 
substituent methyl. In contrast the 0-benzoyl and 0-ac 
have entirely different absorption curves. 



t 4500 A, is 
acetic acid :' 
spectra of 
structure with 
curve is cjial- 
ly its structure 
curve some 
effect of the 
etyl derivative: 



Comparison of the influence of the solvent demonstrates (Fig. 2) 
the distinction between the two types of absorption b.nnds. The 
absorption bands of the N-mcthylphenylhydrazono of anthraquinone and 
of the parent substance are displaced in a polar solvent such as 
alcohol toward the red, whereas those of the 0-benzoyl derivative are 
shifted to the shorter waves. 



Influence of substituents on the azoid-quinoid tautomerism ; - In 
the terminology of Ingold and Kobinson (3b) tine influence of substi- 
tuents on the azo-hydrazone tautomerism may be discussed on the basis 
of inductive and mesomeric effects according to the following electron 
shift scheme, which is analogous to the keto-enol tautomerism. 



'jnnt 



a) ' 



-3- 
H. 



g?/^,T=^. ^ oV^l^-^N- 



The so-called negative groups (terminology of Robinson) such as 
NOg , COOH and halogens, have an electron pulling effect and the so- 
called positive groups such as CH3 , CH3O, and ITH2, an electron repel- 
ling effect. Hence the latter will favor the azo structure and the 
former the hydrazone structure. The order in which various substi- 
tuents favor the hydrazone structure in a compound such as benzeneazo- 
a-naphthol (v/here substitution occurs in the benzene ring) in general 
is according to the series: 

o,£-N02 > o,£-COOH > ra-ITOg, m-COOH ... m-X > o,£-X ... 

m-OR > a-CH3, (H) > o,£-CH3 > o,p-OR 

Benzeneazophenol derivatives:- The azoid structure of this com- 
pound is so stable that it is unchanged by any substituent. 

Benzeneazoanthranol derivatives:- The introduction of a substi- 
tuent favoring the azo structLire does not alter in any way the quinoid 
structure of this molecule. Both the parent substance and its p- 
mcthoxj'' derivative have similar absorption spectra. 

Benzeneazo-a-naphthol derivatives:- Not only substituents but 
also the solvent have marked influence on the constitution of this 
compound. Pig 3 shows two absorption bands in the visible (ethanol as 
solvent). Kuhn and BSr (5) have correlated the band of longer wave 
length (H-band) with the hydrazone and that of shorter wave length 
(A-band) with the azoid structTire , which is confirmed by the absorp- 
tion spectra of the derivatives. In Pig. 4 the curve for the p-methyl 
not only corircsponds much more to that for the 0-mcthyl than it docs 
to that for the K-methyl derivative, but also to the curve for 
£-hydroxyazobenzene . 

« 

Introduction of the m.ethyl group in the ^-position strengthens 
the absorption band of shorter wave length (A-band) and weakens 
absorption at the H-band. The height of the H-band for the o-metliyl 
derivative is lowest and for the m-derivative highest, and conversely 
the A-band is heightened in the order £>P>2- 

A consideration of the effects of other substituents based on the 
absoiption spectra of various benzencazo-a-naphthol derivatives loads 
to the following summary. In alcohol as a solvent the hydrazone 
structure is favored according to the following order. The values 
given are for the ratio of the extinction coefficients of the tv>/o 
bands 6H/^A, which may be taken as an approximation for the proportion 
of hydra z one -quino id to the azoid foim. 

T)-IT02 > o-COOH > p-COOH > 0-WO2 > m-NOs > m-Cl > m-COOH > 0-CH3O, 
"c« " Oo C30 2.3 1.9 1.5 1.15 0.9 

m-CH30 > p-Cl > £-Br > 0-C2H5O > m-CH3 > (H) > o-Cl > £-CH3 > 
■"0.89 U.84 0.8 0.73 0.71 0.63 0.5 0.4 

2,4,6-tribromo > 0-CH3 > £-CH3 
0-^ U.?3 0.22 



>/'T'Ar* 




\=x/ 



-« < h 



jx £i'- 



oo 



-4- 

In benzene as a solvent: 

0-C2H5O > £-Br > (H) > 2-CH3 > 2-CH30 

Except for the special cases of o-alkoxy rjid o-halogen derivatives 
the conclusions based on the electron theory are confirmed. 

Chromophoric consi derations i - In general the auxochromic effect 
of substituents on the chromophoric groups considered (azoid and 
quinoid) differ for the unionized or horaopolar form, and the ionized 
form or heteroj)olar anion. 

The homopolar form is considered first. The chief band of 
p-h^rdroxyazobenzene , analogous to the A-band above, is not markedly 
influenced by the introduction of a -methyl group in the benzene 
nucleus, but a nitro group in the ^-position to the azoid group has 
a pronounced bathychromic effect, which is trivial for the m- and 0-NO2 
compounds. This Sc?jne auxochrDmic effect is noted in azobcnzone, tHat 
is, the bathychromic effect of IIO2 is in the order p» o>m> (H) (?ig. 
5). " " 

In the case of a stable quinoid structure as found in benzencazo- 
anthranol, the auxochromic effect of substituents is reversed. Posi- 
tive groups as CH3 and CH3O have a bathychromic effect and the negative 
ITO2 group a ' liyp so chromic effect in the order (H)> £>p>m in the 
bathychromic direction (Pig. 6). " 

The chromophoric evidence for the benzenazo-a-phenol derivatives 
is less satisf actary. In the case of the H-band, CH3 ajid CH3O groups 
in the p-position have a bathychromic effect, and the HO2 group a 
hypsochromic . For the A— band the CH3 group has a hypsochromic and 
the halogens a considerable batlriy chromic effect. For th^s compound it 
is noteworthy that the same substituents which deviated from predic- 
tions based on electron theory have an an.omalous auxochromic effect 
also, namely, the o-alkoxy tind o-halogcn derivatives. 

Auxochromic effect of substituents in colored anions:- Y/hcn a 
chromophoric homopolar compound is converted to a heteropolax through 
salt formation, on important color change takes place. The auxo- 
chromic groups v/ork in a different manner on the ionized chromophore 
than on the homopolar form. For instance the NO2 group has a bath-y- 
chromic effect on the heteropolra: form of bcnzencazo-a-naphthol in 
the order p » o>m >(H) . The effects of v.arious substituents on the 
colored anions may be summarized as follows: 

1. In the £-position a bathychromic influence is evidenced in an 
order which corresponds to the so-called negativity, but this is not 
true for the m- and £-positions. 

P-IIO2 > COOH > CI > Br > (H) > CH3 > CH3O 

2. The auxochromic effect depends on the position as follows: 

R bathychromic hyperchromic 

ITO2 £>X > m > (H) p ;» m > o,>(H) 

COOH £>;>TH)>o,m o;>p>m>(H) 

CI p>m>o>THT £>m>o>(H) 

CH3 (H)>p,m>o m>TH)>p>o 

CH-.0 o>m,rHy>"'p TH)>o,m^£ 



-.T-.rf 



-i-.- 



-fo 



-5- 

3. Bathychromic influence has the following order: 
2,4-dinitro > P-KO2 > 0-NO2 > m-NOs > £-Cl > m-Cl > 2-COOH > ±-Cl ;> 
£-Br > 0-CH3O > m-CHsO, 0-C2H5O > (H) > £-CH3 > m-CH3 > o-,m-COOH > 
P-CH3O > 2,4,6-tribromo > 0-CH3. 

Bibliography : 

1. Bergmann and Weizmann, Trans. Paraday Soc. , 32, I3I8 (1936). 

2. Buraway and Markowitsch, Anji., 521, 298 (193^. 

3. Gilman, Organic Chemistry, Joym Wiley pjid Sons, New York, 1938; 
(a) p. 671; (b) v. 1616. 

4. Hendricks and Wulf, J. Am. Chem. Soc, 58, 199 (1936). 

5. Kuhn and Bftr, Ann., 516, 143 (1935). 

6. Lauer and Miller, J."Im. Chem. Soc, 57, 520 (1935). 

7. Pauling, J. Chem. Phys . , 1, 607 (193377 

8. Shingu, Sci. Papers Inst. Phys. Che. Research (Tokyo), 35, 78 (1938) 



Reported by S. J. Circle 
May 17, 1939 




€0h 



6 



A 



alcohol 
benzene 




tSOO Zooo tsoo ^OOQ ^^ 5^^ soo *f$0 4^o SSQ 

Fig.l. l.Benzeneazoanthranol. Fig.S. I'.Benzeneazoanthranol. 

2.N-metliyl derivative. 2.N-methyl derivative, 

3.* ^. O~benzoyl derivative. S.O-benzoyl derivative. 

4, 0-acetyl derivative, 
(all in alcohol) 



benzeneazo- 

c<-naphthol in 
alcohol 





Zooo -^ L> 2'SOO 

Fig.S. p-methyl benzeneazo-o<:^- 
naphthol in various solvents, 
1. alcohol 9. acetic acid 
3. benzene 4. alcohol-KOH 




HOOO 2S'00 3000 3SOO 

Fig. 4. l.N-raethyl benzeneazo- c<:_ 
naphthol (in benzene). 
S.O-methyl derivative in benzene. 
S.grmethyl derivative in benzene. 
4,Benzeneazophenol in alcohol. 
- 5 . /Cf-naphthol in hexane. 



10 



Fig. 5. 
p-deriva- f^ 
tives and ^ 
the N- £ 
methyl ^^ 
derivative 
of benzene- 
azoanthran- 
ol(in al- ^ 
cohol) . 



Sub:itance P 



Fig.S. 




./o^^ 



SDO ^SO ^OO 

The three isomeric anthra- 



¥SO 



quinone-momonitrophenylhydrazones 
(in alcohol). ^oc 



•vJb- 



I n \ 



I. 



: \V- 






ooc 



.^ ; 




i.-.:'. •^. 




—- i 



**s? ■ «. 



-ts'» 



(3-KETO BASES 



Mannich -- Pharm. Inst. Univ. of Berlin 



Sh^er and Tollens (1) obtained from the reaction of acetophenone, 
formaldehyde, and arnmonium chloride a base to which they assigned the 
formula {061150001120112)3^. Steam distillation of the hydrochloride of 
the base split off vinyl phenyl ketone. 

Mannich and coworkers (2) repeated the same experiment several 
years later and found the particular condensation to be more complica- 
ted than indicated by Sh^er and Tollens. Meanwhile, Mannich had 
successfully applied the reaction to a variety of ketones sind amines. 



A typical general equation demonstrating the preparation of 
P-keto bases follows: 



ROOOH2R' + HCHO + R2"NH2'^C1 



RC0CHR'0H2NR2*'H"^C1" 



H2O 



Aliphatic p-keto bases were obtained through the reaction of 
aliphatic ketones with formaldehyde and ammonium salts, or more 



readily, from the hydrochloride 



For example, 
{OH3OOOH 



20112)3 



CH3000HoOHoN(OH 



when 



-2' 

the 



'2 
salts 



of methyl amine or dimethyl amine (3) 
gave with acetone, respectively, 
l^ (isolated as the oxime); (0H30O0H20H2)2N0H3 J and 

foimed 



these three amines 



3)2. Other less readily defined products were 
of ammonia and methyl amine v;cre employed. 



It was found that acyclic ketones also condense with formalde- 
hyde and amine salts to form the expected (3-keto bases (4). Oyclo- 
hexanone yielded stable salts which decomposed at their melting points 
to the original amine salt and 2-mcthylenecyclohexanone. a-Tetralono 
under the same conditions produced compound I, which, after conversion 
to the free amine, was reduced to the hydroxy compound which could be 
benzoylated. It could be dehydrated to the dihydro compound and then 
reduced to 1,2, 3, 4-tGtrahydro- 3- (dime thy laminomethyl) naphthalene. 



a'^OHC 



■CH0H2MVIe2 -p 




V R = H 
Va R = OH3 



Acetoacetic ester was condensed with l-(dimcthylaminomethyl)- 
2-cyclohexanone, II (5). 2 -Methylene cyclohexanone was probably 
formed primarily, thus explaining the reaction of acetoacetic ester; 
and, since compound IV and not III was isolated, ring closure had 
evidently taken place. Substitution of compound I for II gave com- 
pounds V and Va when acetoacetic ester and methyl acetoacetic ester, 
respectively, wore used, 

COOEt 



'Sx^O II 



OHpOOOEt 
OOOE3 



OOH29H 
g=o III 
0H3 



CD' 



OOOEt 



IV 



Mannich (6) carried out the same condensation with the replace- 



o 



.f<^ 



.•»- 






. >o o;;-. 



-2- 



ment of II by l-dimethylaiiiino-3-"butanone. Compound VI was obtained 
along with some of VII, but boiling VI with dilute sulfuric acid 
converted it to VII. 



CH. 



,CH2MC2 



o=c 



^ I 



CH2C00Et 



/CH2\ 



.CH 



CH. 



CH-COOEt 



\ 



COCH- 



CH. 



VI 



CH 
I 

o=c 



2\ 

CH. 



^CH^ 



C-CH3 
VII 



Repetition of this reaction, using malonic ester instead of 
acetoacetic ester, provided a satisfacroty yield of the expected 
diethyl 3-kotobutylmalonatc, which on further treatment with sodium 
ethoxide gave dihydrorcsorcinol through inner condensation rjid loss 
of a carbethoxy group . 

Robinson (7) has been studying s5"ntho3ei3 of substonces related to 
sterols. He and his associates synthesized certain octaloncs and 
ketotetrahydrohydrindcnos which arc of the cjigular methyl substituted 
type. In most of his work quaternary ejmnonium salts prepared by 
Mfinnich's reaction were used as a source of the ketones. The follow- 
ing equations illustrate these results: 



^^^.Xs>'CH2Fi:t2MeI CH2C00St - 



Mr 







COCH- 



CH. 



,CH2l^t2MeI 



0=C 



•CH. 



CH— CH. 



+ 0=C. ^CH2 
CH2 




CH2^ COOEt 
CH 
COCH-, 








>.CH3_ ^ 



A large number of p-keto bases have been prepared from aliphatic- 
aromatic ketones, f ormaldoliyde, and secondary amines, as both the 
ketone and amine were varied widely (8). Acotophenone, for example, 
reacted with formaldehyde and dimethyl amine, afforded ur-dimcthylcjnlne 
hydrochloride in good yields, according to the equation: 



C6K5COCH3 + HCHO + 



(CH3)2HH2'^C1" 



C6H5C0CH2CH2lT(CH3)2H Cl' VII: 



Aqueous solutions on boiling decomposed to give the simple amine 
hydrochloride cjid the a ,p -unsaturated ketone. Super>jeated sto.am or 
dry distillation in vacuo produced the same effect. In the case of 
vinyl phenyl ketones, the yields of the unsaturated ketone wore de- 
creased due to polymerization. Reduction of the unsaturated ketones 
actually afforded a means of metlwlating aromatic- aliphatic ketones. 
For example, propioajiisone was prepared from p-methoxyacetophcnonc . 

Compounds such as C6H5C0CH2CH2l^"C5H, q , j3-pipcri dine ethyl phenyl 
ketone, wore described as local anesthetics. Reduction of the free 
p-keto bases to p-amino alcohols caused a loss of anesthetic propertici 
but the benzoylated alcohol produced marked ajiesthesia. 



11. 



o 



t 



^ 



^ 



-3- 

Rcichert (9) used the method of Mannich for the preparation of 
1,3-^^i^ino ketones. With these he condensed nitromethane , and, in the 
case of oy-dimethylaminopropiophenone (IX), throe products were identi- 
fied (X, XI, XII). 

C5H5COCH2CH2CH2ITO2 (C6H5C0CH2CH2)2CHN02 (C6H5C0CH2CH2)3CR02 

X XI XII 

It is of interest to note that substances of the type C6H5C = CH 
react with formaldehyde and secondary ajnines, according to the equa- 
tion (10) : 

CsHsC^CH + HCHO + MR2 — > CgHsC = C-CH2NR2 + H2O 

XIII 

The l-phenyl-3-diallcylamino-l-propyne (XIII) was obtained in good 
yields. Reduction gave the corresponding propane. Addition of cold 
dilute sulfuric acid to XIII produced IX, which is the sajne aanine as 
VIII. 

Recently, Mannich (11) treated p-keto ajnines with formaldehyde 
in order to obtain aminohydroxy ketones and aminopolyhydroxy ketones. 
To avoid complications v/hich can arise through the reaction of foimal- 
dehyde with a primary or secondary ajiiino group, ajnino ketones v^ith a 
"tertiary nitrogen were employed. Prom the condensation of 1-dimcthyl- 
3-butanone (XIV) with formaldehyde, a part of XIV was recovered along 
with XY and a basic mixture vdiich could not be cleared up. But, when 
the reaction mixture va s acidified and reduced, a mixture of bases 
was isolated, which, upon fractional distillation, gave some recovered 
XIV, compound XVI, a mixture of liquid diastercoisomeric dihydroxy 
bases (XVII), and a mi:cture of trihydroxy compounds, one of which 
probably has the structure XVIII. 

OH 
CH3C0CH2CH2mie2 CHsCOCIU CE2Me2 ) 2 CH3CHCH( ^2^03) 2 XVI 

XIV XV OH CHsOH 

CH3CH-dECH2mi0 2 XVII 
OH CH2OH 
HOCH2CH2CH-CHCH2MC2 XVIII 

Bibliography ; 

1. ShHfer and Tollens, Ber. , 39, 2181 (1906). 

2. Mannich and AbdulLoh, Ber., 68B, 11 3 (1935). 

3. Mannich ct al . , Arch. Pharm., 255, 261 (1917); ibid., 264, 65, 164 

(1926); ibid., 26S, 589 (1927TT ibid., 265, 684 (192^17 

4. Mannich et al., Ber., 53B, 1874 (1920); ■35r3lar Pham., 265, 251 

(1927); ibid., 265, 5^ (1927); ibid., 275, 54 (1937)": 

5. Mannich, Koch andlFrkowsky , Ber., 7 OB, 3^^(1937). 

6. Mannich and Poumeau, ibid., TIB, 2TW (1938). 

7. Robinson ct al., J. Chem. Sec, 1937 , 53; ibid., 1938, 1097. 

8. Mannich ct al., Ber., 55B, 356, 5^57 3510 (1922) ;"lrch. ,Phajmi. , 

276, 206 (1938) . 

9. ReTcEert and Posemann, ibid., 275, 67 (1937). 

10. Mannich and Chang, Ber., 66B, TO (1933). 

11. Mannich and Salzmann, ibicHT 72B, 499 (1939). 

Reported by J. H. Burckhalter 
May 17, 1939 



r-D'^. 



ISATOGENS AND ISO-ISATOGENS 



Ruggli -- University of Basel, Switzerland 



In 1882 A. von Baeyer in his work on the structure of indigo 
discovered and named the first isatogen which was the ethyl ester of 
isatogenic acid (IV). P. Pfeiffer discovered a simple synthesis for 
isatogens and prepared a large number of these compounds. The simplest 
member in this class, isatogen (I), is unlmown. It is isomeric with 
isatin (II) and possibly an equilibrium exists between the two, with 
the equilibrium displaced far to the right. 



I 







a 






=0 





II 



aC=:C-C00C2H5 
N02 

III 



f^^ — ?=0 



H2S0, 



r^^A-^ 



4^ 




i-COOCoHc 



IV 



Structure and Reactions of Isatogens :- When isatogens are treated 
with zinc and acetic acid they are converted into indoxyls. Prom this 
it v;as concluded that they were cyclic compounds. They contain two 
reactive groups, a carbonyl group and the nitrone grouping =1T— >0. 
When isatogens react with hydroxylamine some oxime formation 'takes 
place on the carbonyl group and some on the nitrone group. Pfeiffer 
pointed out that the isatogens were the first definitely laiown meta - 
quinoid compounds. It has been possible to form the quinhydrono 
analogue by adding phenylindoxyl (the hydro quinonc analogue) to phenyl - 
isatogen (the quinonc analogue). In this way a highly colored sub- 
stance (phenyliBatogen) and an almost colorless substance (phenyl- 
indoxyl) form a black crystalline compound which is easily separated 
into its components . 

The addition reactions of the isatogens can most easily be shown 
by means of a diagram using 6-nitro-2-phonylisatogen (V) as the 
starting material. 



°^K^.^ 



VII 



C ' 

W ^0C0CH3 
OCOCH3 




heat 



HCl 
C2H5OH 

V i VI 






C«H 



e-'-^s 



6e 



0C2H5 



CH3COC1 



heat 



As 



02^-^ 



r 



■CO 
LCgHe 



!./ 



ice 



VIII OCOCH3 



O2W 



\A 



IX 



•CO 
l/CgHs 

C 
N'^ ^Cl 
6COCH3 



/\ 



CH,OH 



-CO 

^^ ^N^ OCH3 
X 6COCH3 



XI 



O2N 






2-N02C6Hi).ira2 



— CO 

c 

IJT^ ^NHC6H4N02 
OCOCH^ 



■VM^orytAZi -^■- 



-^ 



v\. 




i 



t 



•n I f ) 



-2- 



The addition of acetyl chloride to V yields IX. This compoimd 
is very reactive. With traces of water it forms isatogen. It reacts 
with methyl alcohol to give compound X and with p-nitroaniline to give 
compound XI. With a more basic compound the isatogen V is formed 
immediately. Tlie ether acetates cannot be made by the direct addition 
of the alkyl acetates to the nitrone. It is interesting to note that 
no carbonyl group could be detected in compound VII nor in XI. 

L. I. Smith points out that most of the reactions of isatogens 
which have been studied in detail can be interpreted logically as 
1,3-addition. There is also a conjugated system in the quinoid 
structure of the isatogens: 0=(J-C=N->0. This would be capable of 
1,5-addition. The further study of these compounds should be very 
interesting. 

Preparation ;- The first isatogen was made by the action of H2SO4. 
on the ethyl ester of o-nitropropiolic acid (III) to give IV. 

Pfeiffers's synthesis consisted of exposing a -pyridine solution 
of the halogen derivatives of o-nitrostilbene or o-nitrotolanes to 
sunlight . 

Recently Ruggli found a nev/ method for the preparation of isa- 
togens. He dissolved the o-nitrotolanes in chloroform using nitroso- 
benzenc as a catalyst. Ho allowed tho mixture to stand for 19 days 
and obtained a very pure product in 39% yield. 

Ruggli' s latest synthesis consists in reducing £-nitrobenzil with 
Rancy nickel. He found that a solution of o-nitrobenzil readily ab- 
sorbed four atoms of liydrogen. If he stopped tho reaction at this 
point he obtained a 34^ yield of 2-phcnylisatogen. 

Mechanism of Isatogen Formation ;- All that can be definitely said 
about the foimation of isatogens is that in Pfeiffer's method three 
steps are involved: 1) HX is eliminated from the halogcnatod stilbcne 
2) the oxygen migrates to the carbon j 3) the ring closes. 

It- has been definitely shown that only the o-nitrotolanes form 
isatogens. Pyridine, and quinolinc to a lesser degree, seem to be 
specific catalysts for this ring closure. 

Iso-isatogen :- In 1919 Ruggli found that when the deep red 
6-nitro-2-phcnyiisatogen was treated with alcoholic HCl there was 
obtained a light yellow isomer. To this he tentaliively assigned the 
structure XII. Recently he- has studied the formation of an analogous 
isomer from 2-phcnyl-6-carbethoxyisatogen using alcoholic H2SO4.. 




/C-^sHs 



XII 





/-CsHs 



O2IT 



XIV 



■CO 



+/-C6H5 



Ruggli gave the name iso-isatogen to those compounds which arc isomer- 
ic with the quinoid stnicturc. 

The iso-isatogens differ from tho isatogens in many ways. 

1. They do not shov/ any phenolic properties and hence structure 
XIII is ruled out. 

2. They arc practically insoluble in water which rules out the 
salt -like structure XIV. 



.A 



-3- 



thc corresponding indoxyls 
solution of KI in acetone 



3. They form only one oximc whereas the quinoid isatogen foims 
tv/o oximes (the C- and the N-oximc) . 

4. They react v\fith phcnylhydrazinc to give compound XV while 
the isatogens are reduced to indoxyls. 

5 . They do not f onii quinhy drones with 

6. They do not liberate iodine from a 
v\?hile the quinoid form does. 

7. They have the same composition cxid. the same molecular weight 
as the isatogen from v;hich they are formed. 

8. On stereochemical considerations XVI cm be ruled out. 

9. \Vhcn iso-jsatogen is heated above its melting point in acetic 
acid for five minutes it rearranges to the quinoid structure. 

The evidence presented above shows that all obvious structures 
/or the iso-isatogens , other thari the threc-membered ring, must be 
abandoned since they conflict v^ith some of the experimental data. 



/\ C=MHC qEs 

02N\y\ ^C-CsHs XV 







XVI 



i 



Bibliography; 



Acta, 4, 626, 637 (1921); 6, 594 (1923); 7, 689 

938 (1:927); 14, 1256 (1931); 15, 856 (1952); 
17, 1328 (1937T; 19, 5 (1936); 20, 250 (1937); 
, 22; 134, 140, 147T1939). — 
■ Ber., 15, 5211882) . 
ibid., 32, 1 (1919) • 
Chem. RevT, 23. 193 (1938). 



Ruggli, Helv. Chim. 
898 (1924); 10 
16, 69 (1933TT 
21, 33 (1938); 

Baoyer 

Raggli 

Smith , 



Reported by P. C. Dietz 
Mo^^ 24, 1939 



n:^ 



IIECHAl^ISM OP ICETONE P0miA.TI01T_ PROM OAHBOXYLIC ACIDS 

Weunhoeff er and Paschke -- Breslau 

The formation of ketones from carboxylic acids or their salts by 
thermal decomposition has been surmised to take place in various ways : 

1. (CHsCOOjgCa -^ CH3COCH3 + CaC03 

2. (CH3C00)2Ca — > CaO + (CH3C0)20 —^ CH3COCH3 + CO2 

3. (CH3C00)2Mg — > CH3COCH3 + MgO + CO2 

P^ ^000^ .CH2, 

4. H00C(CH2)4C00H ^ )J^^. > (CH2)4 ^e ~* (CH2)2 )^=0 + PeCC 

Ba(0H)2 ^ ^000'' <Oni^ 

It can be said in general that acids v/ith basic catalysts or salts of 
acids form intermediate salts which decompose into ketones and metal 
oxide or carbonate depending upon the basicity of the oxide. Catalytic 
transformations, those involving carbo^cylic acids and basic catalysts, 
take place at lower temperatures and give better yields than acid or 
salt thermal decompositions. Tho difference would lie in the reaction 
mechanism. 

Only salts of carboxylic acids which have a h;>'"drogen atom alpha 
to the carboxyl group can give ketones upon themal decomposition. 
Calciiom trimethylacetate yields no hexamethylacetone . Parmer and 
Kracovski have obtained no cyclic ketone from a,a,a' ,a' -tetramcthyl- 
adipic acid, Diisopropyl. ketone ?;as obtained from calcium isobutyrate 

To account for the above facts a tv/o-stop mechanism for the 
decomposition v;as proposed and shown to be correct. 

1, Acid or salt — > p-kcto acid or salt 
{CH3C00)2Ca -^ CH3C0CH2C00-Ca0H 

2. Decarboxylation 

CH3C0CH2C00-Ca0H —3. CH3COCH3 + CO2 + CaO 

There might be some doubt as to the application of this mechanism 
to the formation of mcjx^'^-m.emborcd cyclic ketones. Ruzicka has assrmicd 
that an intrpniolecular reaction loads directly to the cyclic ketone. 
The fact that monocyclic ketones have been obtained by the thermal 
decomposition of dibasic acids rjid their ojihydrides weakens the claim 
that the metal ion in the case of salts, thorium for cxojnplc, tends 
to bring into close approach the ends of long chains so that closure 
can take place. 

Carothers, basing his conclusions on analogous reactions of his 
polycjihydrides and polycotcrs, has claimed that by the thermal decom- 
position of a salt a linerjr polyketone is first produced which 
separates into simple ketones . 

-R-CO-R-CO-R-CO-R-CO-R-CO- > r1:!=0 



.rii^ 'fim\ ^iiiMSsr- ■-■•^•qi^v -:j' wsr^M^'sz: 



; nv:?ei-if f-v-' ^ ■'■'.■ 



7' 'V. 



urfl' 



■-■'•ij^uu 



-2- 

To support Ms claim ho presented as evidence the thennal decomposi- 
tion of thorium octadecanedioate ; first was produced a white solid 
polymeric ketone of consistent formula which upon molecular distilla- 
tion was trcjisformod into cycloheptadcconone» No mechanism was 
suggested to account for the breaking of the linlcages in the chain; 
it was also thought unlikely a cyclic ketone would polymerize to a 
polyketone. 

The proposed intermediate basic p-keto salt of Neunhoeffer could 
undergo enolization cjid loss of v^ater to give a salt incapable of 
forming a ketone unless acted upon by water or acid. 

1. CHaCOCHpCOO-CaOH -liis^., CH3CH-CH2-C=0 

I I 

0— Ca— 

The reversal of this proposed reaction would account for better ketone 
yields in the presence of excess acid cjid steam. V/ork by Vavon ojid 
Apiche, Vogel, and Ardagh support this. Evidence for the two-step 
mechanism follows: 

2. Ba adipate (dried 110°) 290-430 ^ cyclopentanone {Q5%) 

3. Ba adipate (very dry) ^^ > cyclopentone + some cyclopenta- 

none + BaC03 + C (residue) 

4. Ba adipate. 260° ^ cyclopentrjione + H2O 
Adipic acid Pure BaC03 residue 



B 



n ri 



dipate + BaO ^'^ — > cyclopentone (41^); 47% C in residue 

Note that basic conditions lower the reaction tcmpGratur- 
cjid favor the formation of side products. 

CH2-C=0 

6. j JDH-COO'^ > cyclopontene; BaC03 ^residue + C 

CH2-CH2 2 

This reaction corresponds to reaction 5 where the BaO 
favors a supposed reaction 1 unfavorable to ketone 
formation. 

Since it v;as found impratical to isolate from the decomposition 
of barium adipate salts of cyclopentrjione-o-carboxylic acid, detection 
was done by color comparisons. Small amounts of barium adipate with 
BaO in small flasks were heated in a copper block for 1,5 minutes to 
450°. After being made acid, ferric chloride was added to produce a 
green color: blue from the reaction with cyclopentanone-o-carboxylic 
acid and yellov; from excess ferric chloride. Fully agreeing color 
shades appeared when synthetic barium cyclop entanonc-o-carboxylate 
was treated similojrly. The barium salt of phenylpropTonic-o-crrboxylit 
acid gave ari excellent ferric chloride raxction. Also by means of 
favorable basic conditions the intennediate p-keto acid foimation was 
shown. 



-'•.w 



- 1 tn<" 



• •( ;1f T '-. o Crr^ ■''■ ^? ?> e» ' . . ■ r« >* 



C/ ■♦<■•? -^<" 



n--TJTTi,-J.«". 



.^-TvY-^TF-.- '^rrr; 



O - 



-3- 

p^0H,-CH,-C=O KOH,NaOH,H,0^ pfVcoOBaOH .fY'^^H, 

k^COO-Ba-^. 270-280" k^C^O k>"C^O ' 

2 hours I ^ 

while hot Removed 



K 



a^CHg J^r 1 "^H-COOH PeCl3 - violet 
C^O ^0 +\x^C^O 
heat 

It was shown that under suitable conditions basic constituents 
of salts could be dispensed with. The monoethyl ester of adipic acid 
was trojisformed by thermal decomposition into the ethyl ester of 
cyclopentanone-£-carboxylic acid. Adipic acid distilled from Jena 
glass and quartz gave quantitative -yields of cyclopentanone. Phcnyl- 
propionic-o-carboxylic acid gave a-hydrindone almost quantitatively at 
300 . Sebacic and azelaic acids gave as low yields of cyclic ketones 
as the salts did. In the case of acetic acid basic catalysts and high 
temDerature formed the yields. The yield of acetone varied with 
temperature from 1.% at 290° to 12^ at 340 . Above 340 the yields fell 
off. 

It was indicated that the foimation of p-keto acids and their 
salts by the thennal decomposition of carboxylic acids or their salts 
is a reversible reaction. The decarboxylation of (3-keto acids could 
be expected to be an equilibrium reaction in which the equilibrium 
has been shifted far toward decarboxylation. Acetic acid was obtained 
from acetone, steam and CO2 by heating for several days at 350°. 
Adipic acid was prepared from cyclopentanone, v/ater and CO2 by heating 
at 330° for a day. The reversibility of the thermal decomposition 
reaction was indicated. 

To summarize, the mechanism of -ketone fonmtion from carboxylic 
acids involves the foimation of a (3-keto acid and subsequent decarboxy- 
lation. Basic catalysts are not entirely necessary for the transfonna- 
tion to proceed and the reactions are reversible. 

Bibliography : 

Neunhoeffer and Pas dike , Ber. , 72, 919 (1939). 

Bamberger, ibid., 43, 3517 (191TJT. 

Krfeig, Z. angew. UEem,, 37, 657 (1924). 

Farmer and Kracovski, J. 'CHem. Soc, 129, 680 (1927). 

Popov, Ber., 6, 1255 (1873). 

Vavon and ADiche, Bull. soc. chim., (4) 43, 667 (1928). 

Vogel, J. Chem. Soc, 1929, 127- 

Ardagh et al., Ind. EngT'Uhem. , 16, 1133 (1924). 

Aschan, Ber., 45, I6O3 (1912). 

Neunhoeffer anT'Kfilbel, ibid., 68, 255 (1935). 

Ruzicka et al., Helv. Chim. Acta, 11, 496, 670 (1928). 

Carothers and Hill, J. Am. Chem. Soc, 55, 5043 (1933). 

Ann. Repts. Progress Chem., 23, 112 (19^). 

Reported by J, P. McPherson 
May 24, 1939 









•OsC;-.. 



•^- 






-L^ 



r-^^Ij^ 



• -*>. 



i<J^^J ; ^ i A 



n-^^'-r^^-f^-r:^ ,1 .T, "i 



..V 



MOLECULAR DISSYMP/IETRY DUE TO RESTRICTED ROTATION 
IN THE BENZENE SERIES: AN OPTICALLY ACTIVE ETHYLENIC DERIVATIVE 

Mills -- Cambridge University 



Molecular dissymmetry arising from the restriction of rotation 
about a single bond has been shown to occur in peri -substitution 
derivatives of naphthalene (1) and later in c ertain ortho substituted 
derivatives of benzene (2). 

The derivatives of naphthalene and benzene in v/hich this effect 
was demonstrated contained a grouping XYZ which could rotate about the 
axis AB when the adjacent positions (o- or peri-) were occupied by 
hydrogen atoms but was confined between more or less narrow limits of 
rotation when a large substituent, such as -803" or— N( 0113)3 was 
introduced into one of those positions. 

The types of structure most obviously 
suitable for the rotating complex arc the 
substituted amino group I and the substituted 
vinyl group II. 

The compound III was prepared by Mills ajid 
Kelham (2) and resolved with active brucinc. 
The brucine salt showed mutarotation in chloro- 
form solution. In this case -NCH3(Ac) is the 
rotating group and -SO3H the obstructing group. 
Apparently the grouping -COOCH3 is not large - 
enough to cause restricted rotation since the 
analogous compound IV could not be resolved. 




R, ^R' 

N^ 



R' 



Ac-K- 



R' 



II 



O 



CH3 
SO3E 



Ac-N-CH- 



III 



OCOOCH3 



IV 



The compounds V, VI and VII which contain a substituted vinyl 
group were investigated (3)» They were found to be non- re solvable, 



CH- 



CH=6-C00H 
02Ny^C!H3 



H,C 



'NO- 



CH- 



CH=C ( CH3 ) 2 
H3C>^\.CH3 

'Xj^NHa 
CH3 



VI 



CH3 
CH=6-C00H 
Bry.'-^X^r 

VII 



V 



presumable because of the presence of an a-hydrogen atom in the 
[3 ,p-disubstituted vinyl groups (CH=CRR') employed. With a hydrogen 
atom in this position the vinyl-benzene link caji probably be distorted 
so as to allov; the =CRR' group to pass over the ortho substituents . 

Restricted rotation was observed in VIII (4), v\fhich contains the 
p ,p-dimethyl-a-isopropylvinyl group as the rotating complex, and the 
cationic group '''N( 0113)3 ^-^ the obstructing ortho substituent. It was 
resolved into stable, optically active_components by means of d-bromo- 
camphor sulfonic acid, and the d- and l^iodides obtained showed 



I An^' 



I 



III 



-2- 



molecular rotations, [M] 545 i > of +55 and -58°. The optically active 
salts v/ere quite stable; they retained their optical activity after 
boiling the ao^ueous solutions for eight hours. This resistance to 
racemization is in marked contrast to the stability of the compounds 
containing a substituted amino group as the rotating complex and is 
in accordance with the large overlap of the groups as indicated by 
the atomic dimensions concerned. 

Compound IX, which differs from VIII in containing the methyl- 
ethylvinyl group instead of the dimethylis op ropyl vinyl group as the 

rotating complex, was found to be 



CH.3 

Co Her -C = C-H 




■W(CH3) 



3 ^3 



IX 



X 



Synthesis of VIII: 

2 (CH3)2CHMgCl 



COOCH3 

N(CH3) 




non-resolvable, presumably on ar^.ccunt 
of the insufficient overlapping of thr 
subs tituent groups . 



(rS^:^CH)2C-0H 




(CH3)2CHC=C( 0113)2 
VIII 




5(CH3)3 
+ 



Kal 



(CH3)2S04 



(CH3)2CHC=C(CH3)2 

>-N^(CH3)2 



XII 



Bibliography; 



1. Mills and Elliott, J. Chem. Soc, 1928 , 1292. 

2. Mills and Kelham, ibid., 1937, 274. 



3. Maxwell and Adams, J. Am. i)hem. Soc, 52, 2959 (1930). 

4. Mills and Dazeley, J. Chem. Soc, 1939, 460. 



THE NUCLEAR METHYLATION OP PHENOLS 



Caldv/ell -- Temple University 



The nuclear methylation of certain phenols is of interest in the 
preparation of intermediates for the synthesis of certain compounds 
having physiological action, notably psGudocumohydroquinone which is 
an intermediate in the synthesis of a-tocopherol . The method involves 
the preparation of the "Mannich" base by the action of formaldehyde 
and dimethylaraine on the appropriate phenol, followed by hydrogenoly- 
sis with copper chromite as catalyst. In the case of phenol itself: 



OH 





OH 



CHoO 



( CH3 ) 2HH 



O 



CHpNCCH,) 



OH 



3/2 



H. 



A 

177 atm. ,160° V^ 



CH- 



J .\ 



o 



V 



X 



1 , ,. 



:i 



i-'HB 



'^.X 



o 



^^2^ 







-3- 

The yield of the Mannich reaction is practically quantitative but is 
only 30^ on the hydrogenolysis. This is explained by the fact that 
the liberated base poisons the catalyst. If the phenol is acetylated 
prior to the hydros nolysis the yield is improved, being 60-10% in 
some cases. The method is applicable in the methylation of cresols, 
xylenols, and other phenols and polyhydroxybenzenes, and naphthalenes. 
The application to aliphatic compounds has not yet been investigated. 

Synthesis of pseudocumohydroquinone (VII):- Several syntheses of 
this compound have been reported in the literature (2,3,4). The 
method involves the methylation of sym-xylenol (I) to form 2,3,5-tri- 
methylphenol (III). Reduction of a coupling product and oxidation 
of the amine (V) to the quinono (VI), followed by reduction, gives the 
desired product: 

OH OH OH 

(CH3)2]m^ >v.CH2lT(CH3)2 jg^ r^^^^ NaS03C6H^Np'' 
CH3 CH2O CH3^*\^CH3 CH3''\^CH3 

I II III 

OH OH 0. OH 

.CH3 x>>Y^^^ 



01' 



|-^CH3 Na^S^O^ ^ rV^^ PeCl. ^ fAr^^ 

5^^vA;H3 CH3^\^CH3 CH3^\^Jx:] 

N=N-C«H.,SO,lTa feo 



K^^ 



CH3^\,.'^H3 CH3'N^CH3 CH3^X^H3 CH3''\^CH- 

!6H4S03lTa teg OH 

IV V VI VII 

It was thought that pscudocumohydroquinone could be prepared 
by the direct methylation of hydro quinone , i.e. by treating hydro- 
quinone with three equivalents of formaldehyde ajid dime thy lamin e , 
followed by hydrogenolysis of the Mannich base. Instead, only tivo 
methyl groups were intriduced, forming 2,5-dimethylhydroquinone (X). 

OH OH 

(CH3)2HH^ j-^CH2N(CH3)2 h^ . /^^^^ 

CH2O (CH3)2NCH2^>-^ CH3'\^ 

OH OH 

VIII IX X 

Morpholine and piperidine may be used instead of dime thy lamine , 
and the base hydrogenated to the corresponding methyl derivative. For 
example, the base obtained by treating o-crcsol with formalin and 
morpholine on hydrogenolysis yields vic-xylenol; 3,5r6-trime,thylphcnol 
yields 2,3j5 )6-tetramethylphenol. 

The orientation in all cases is practically completely ortho to 
the hydroxy group. This IS also tine in the hydrogenolysis of the 
bases formed by treating phenols with piperidine and f onnaldehydc: 
sym-xylenol yields 2,3,5-trimethylphenol; and a- and (3-naphthols yield 
the p- and a-methyl derivatives respectively. The orientation in the 
addition of piperidine as reported by- Auv/ers and Lomerowski (5), who 








' (-"si 41V,' ^ -J^' J 




• 


5. 


• 


HC 
cHCy.^ 


VI 

.odr - 


• 





- 6 


. 







-4- 



statod that the orientation is para v;hcnovcr this position is open, 
was shown to bo incorrect by hydrogenolysis of the bases and identifi- 
cation of the methyl derivatives so obtained. 

Bibliography ; 



Caldwell and Thompson, J. Am. Chem. Soc, 61, 765 (1939). 

Baumann, Ber. , 1152 (1885). 

Nietzki and Schneider, ibid., 27, 1430 (1894). 

Smith, J. Am. Chem. Soc, 56, Tr3 (1934). 



Prepared by F 
May, 1939 



Ri chtor 



'A- 



THE FORMATION OP QUATERITARY A]\'ITviONIUM SALTS PHOM DIHALOGENO- 
PARAPPINS, ETC., IE AQUEOUS ACETONE SOLUTION 

Davis, (Miss) Evans, and Hulbert -- Tatem Laboratories, 
University College, Cardiff 

According to Sclimidt and Litters chied (1), methylene dibromide 
and di-iodide react with trim ethyl amine in alcoholic solution at 
room temperature thus : CH2X2 + NMe3 = XCH2NMe3 X~ ; at 100° excess 
of the amine reacts with methylene di-iodide to give the same product, 
together with formaldehyde and tetramethylammoni\:ijii iodide. Trimethyl- 
p-bromoethylammonium bromide is obtained from ethylene dibromide and 
aqueous or alcoholic trimethylaraine below 50°, but with excess of 
alcoholic amine at 100°, a variety of products, including the mono- 
and the di-ammonium salt, is obtained (2). Lucius (3), however, 
obtained triethyl-p-bromoethylajomonium bromide and etliylenebistriethyl. 
ammonium bromide from the interaction of triothylamino and ethylene 
dibromide. The mono- and di-ammonium salts v/ere also prepared from 
the interaction of trimethyl- or triothylamino and trimcth^'-lene 
dibromide. These results suggested that the formation of the di- 
ammonium salts from methylene and etliylenc dihalidcs is at"bonded with 
difficulty. 

The reactions bctiveen a number of organic dihalidcs and tri- 
mcthylamine have now been studied kinetically, in order to find out 
the extent of the diaramonium salt formation and to investigate the 
factors affecting the formation of quaternary ajimionium salts. 

The principal results obtained regarding the reactivities of 
organic halides towards trim ethyl amino arc summarized below, the 
reaction of only one of the halogen atoms (^) being considered. 

C2H5CI > CICH2CI > CICII2CH2CI 

Cn3Br > BrCH2Br > BrCH2CH2& 

BrCH2CH2CH2^ > BtGE^OEz^ 

BrCH2CH2Br > Br" (IMc3CH2CH2Br) 

Br"(hlc3CH2CH2CH2Br) > Br' (Me3CH2CH2Br) 

BrCH2GH2CH2Br = Br" (NIvIe3CH2CH2CH2Br) 

Two factors (4) affect the velocity of formation of an ammonium 
salt: (a) an.ionisation of the halogen atom of the organic halide, 
facilitated by an electron recession to the halogen; and (b) the 
co-ordination of the base by means of its unshared electrons to tho 
methylene group of the organic hoJ-ide, facilitated by an electron 
recession from the halogen. These are depicted in I. Although the 

Hal 

at 

I ^CHof-^tNRj 
I 
R 



-2- 

formation of the salt requires the completion of the cycle, it ccn be 
considered that the initiation of the cycle depends on either of the 
factors (a) or (b). Substituents in the organic halide detciTnine the 
probability of initiation by (a) or (b). 

The halogens have complex polar effects, -I, +M. In the methylene 
dihalides the +M effect, which increases the ease of anionisation of 
chlorine, may operate, Cl^CH^Cl, as v/ell as the inductive attractive 
effect, C1-CH2->C1. In the etliylenc dihalides, however, the fozroer 
effect is excluded because of the intercalation of the saturated 
carbon atom. The greater speed of reaction with trimethylcjnine of 
methylene di chloride than of ethylene dichloride (the dibromides are 
of almost equal reactivity) is, therefore, to be expected. 

On comparing ethylene and trimcthylene dibromides it will be 
seen that removal of one bromine atom from the other reacting bromine, 
with a consequent diminution of the attraction (-1) of the one for 
the other, _^increases reactivity. Substitution of the strongly attrac- 
tive group Mcj for Br further decreases reactivity in the ethylene 
derivatives. It will be noted that, v;hereas the first halogen in 
ethylene dibromide is replaced much more readily than the second in 
the monoammonium salt produced, yet in trimcthylene dibromide the 
diammonium salt is produced as readily as the monoammonium salt. This 
is due to the greater damping out of the polar effect of the halogen 
or the Me3 group as it is moved fcj^lics: away from the reaction center. 
Electronic effects (-1 in these cases) can be transmitted through only 
two saturated carbon atoms at the most, and it v/ould be expected, 
therefore, that the attractive electronic effect of BrCHgCHg- would be 
less than that of "^]Me3CH2CH2- , whereas the attractive effects of 
BrCHgCHgCHg- and '''Me3CPl2CH2CH2- would be approximately equal. 

Consideration of these results indicates that initiation of the 
electron cycle leading to ammonium- salt formation depends on factor 
(a). 

Vinyl bromide end trans - a, p-diio do ethylene are both highly 
unreactive towards ter"biary bases, v/hereas the related compounds, 
allyl bromide and 1, 4-dibromo-2-butenc , are very reactive. Toward 
dimethylaniline , allyl iodide is more reactive than methyl iodide (5). 
The vinyl group is considered to have an intrinsic attraction for 
electrons (-1), as v;cll as a possibility when it is a conjugate 
system of an electronic displacement, +T (6). The non-activity of 
the first two compounds, therefore, is due to electron movements (II) 
retarding anionization of the halogen. Phase III is less important. 
With allyl bromide (and 1, 4-dibromo-2-butene) either of the movements 
IV or V could take place. In IV there would have to be a change in 
the mechanism of the initiation of the cycle leading to the formation 
of the calt , to factor (b). 

Ix Br, Br 

HC^-:ffi1c3 HC^:Me3 II CH III 

lU (li 11^ 

CHI CHo CHo 



'I 



Mo, 
CH2=CH^CH2-Br 



-3- 



IV 



^r* 



Mo, 
I ^ 



CH2=CH-CH2-Br 



V 



The effect of alkyl groups on the speed of the reactions betvvcen 
^-substituted benzyl halides and pyridine (7) and of p- alky 1 dim ethyl - 
anilines with methyl iodido (8) where storic effects can be deemed 
unimportant, is similar to that found with the reaction between alkyl 
halides and trimcthylamine (Me>Et>Pr). Theoretical considerations 
advanced for the former examples could, therefore, be applied to the 
simple alkyl halides. An explanation frequently given for this order 
of the alkyl groups in the reactions of the alkyl halides depends on 
the steric effects of the groups (9). Although the present work has 
not decisively shown the absence of steric effects in the reactions 
being considered, yet it has been dbmonstrated that the experimental 
results can be adequately explained by considering the operation of 
polar effects alone. 

Energy of Activation and Probability Factor:- Values of E and 
log PZ derived from the Arrhenius equation are given in Table I. 
These values are probably not highly accurate, but they give some 
useful indications. 

Table I 

Formation of Monoaramonium Salts from Trimcthylamine and Dibromides 



(1) CH2Br2 

(2) BrCH2CH2Br 

(3) BrCH2CH2CH2Br 



E, kg.-cals. 
16.1 
12.7 
14.2 



Formation of Diammonium Salts from Bromo- substituted Monoammonium Salts 



(4) BrCH2CH2B'[e3 Br" 

(5) BrCH2CH2CH2M,(IC3 Br" 



18.3 
12.5 



log PZ 

7.3 

5.0 
6.8 


tui 


ted ] 


8, 
5^ 


.0 
.5 



Winkler and Hinshclwood (10) showed that, in the reaction of 
trimethylamine v/ith alkyl halides, the order of increasing E rjid. of 
decreasing rate of reaction was MeHal, EtHal, Pr^Hal. The increase 
in E may be attributed to the increasing strength of the C-Hal bond, 
which might be expected if the initiating factor is a, and the total 
electronic effect of the alkyl group is in the order Me>Et >P2t, On 
the other hand, the increase in E might bo due to steric hindrance, but 
the results in Table VI do not support this view. The increase in E 
from (1) or (2) to (4) can be attributed to strengthening of the 
C-Hal bond in (4) by substitution of "^03 for Br. The greater simila5r- 
ity of the E values for (3) C'jid (5) is to be expected, for now the 
substituonts Br and"*TM03 have less effect on the strength of the C-Hal 
bond. Comparing (1) and (2), the fall in E is to be expected from the 
weakening of the electronic effect of the one halogen on the oiher in 
(2). A similar explanation serves for the decrease in E in passing 
from (4) to (5). 



-- -^H ::•■:{ 'J^.HO 



n 



■o 



*3iiffS nr'Af f'e ^-'i.t/o'f 



■ :iJ:^^n\rr tr 



r-l'- 



vc 



-4- 



Bibliography ; 



1. Schmidt and Littershied, Ann., 337, 67 (1904). 

2. Schmidt and Kleine, ibid., 337, "BT (1904). 

3. Lucius, Arch. Pharm., 245, "^ (1907). 

4. Baker, J. Chem. Soc, T^2 , 1148. 

5. Preston and Jones, ibiTr7"101, 1930 (1912). 

6. Dippy, ibid., 1937 , 1008. 

7. Baker and Nathan, ibid., 1933 , 1844. 

8. Davies and Hulbert , ibid.7 T^38 , 1865; J. Soc. Chem. Ind., 57, 349 

(1938). -^ 

9. Vavon, Bull, soc. chim., 49, 979 (1931). 

10. Winkler and Hinshelwood, T7 Chem. Soc, 1935, 1147- 



Prepared by R. W. Kell 
May, 1939 



r- SUBSTITUTION IN THE RESORCINOL ITUCLEUS 



Shah -- Ismail College, Bombay 



The Grattermann reaction, v/hen applied to resorcinol or its mono- 
or dimethyl others, gives an almost quantitative yield of 2,4-di- 
hydroxybenzaldehyde. Hov/ever, when applied to methyl-p-resorcylate 
by Shell and Laiwalla under the usual conditions, viz., using HCN with 
aluminum chloride in benzene solution, or with zinc chloride in ether 
solution, the reaction failed. In the presence of aluminum chloride 
the reaction proceeded smoothly when dry ether was used as a solvent, 
r^d a 63% yield of the T- substitution product, methyl 2,6-dihydroxy- 
3-fo2:Tnylbenzoate, was obtciined. 



HOy^^OH 



The designations used by Shah, 
referring to the positions on the resor- 
cinol nucleus, are shown. 



The product of the above reaction vjas identified by its reduction 
(Clemmensen) to methyl 2,6-dihydroxy-m-toluate, v;hich, on partial 
methylation with sodium methoxide ajid methyl iodide , gave the known 
methyl 2-hydroxy-6-methoxy-m-toluato. It gave the characteristic 
o-hydroxyaldehyde reaction when condensed with ethyl malonate in the 
presence of piperidine. The product was ethyl 5-hydroxy-6-carbometh- 
oxy- couma>rin-3-carboxylate, as indicated by its insolubility in 
aqueous alkali and the red color given with ECCI3. 

CH3 CH3 

HOv^^^OH H0sx>^sv^CH3 



HOy-'^OH 



COOMe 



ZnCl'I 



ether) 



CHO 

HOy^^OH 



COOMo 




COOMe 




OOMe 



II 



OH 
.^ MoOOCy^^-v.^ 



■C-COOEt 
.6=0 



Other reactions of the product II arc: 

a. V/ith methyl iodide and K2CO3 — > methyl 2-hydroxy- 
4-methoxy-3-f ormylbonzoato . 

b. V/ith methyl sulfate and NaOH — > methyl 2,4-dimethoxy- 
3-f ormylbenzoate. This compound gave the expected product upon reduc- 
tion with zinc and HCl, v\rhich product could also be prepared by methy- 
lation of the above methyl 2-hydroxy-6-methoxy-m-toluatc. 

c. Prolonged hydrolysis with cold, dilute alkali gave 
2,4-dihydroxy-3-f ormylbenzoic acid. This, upon heating in water at 
100° , gave -f-resorcylaldehyde . 



CHO 
HOy>^OH 

^N^COOMe 



dil. 
HaOH . 

cold 



CHO 

HOy-^OH 



COOH 



The latter dissolved in dilute alkali 

OH 

rS-^^^*C-COOEt 
CHO ^^ k,,^^0^^"° 
H.O HO,^OR ^^ 

\-^ Nao5^^^^====:^ 

HO- 
yellow 



100' 




■T Yil ypiT^'T^TTr 



ri ^• 



"O 



r'^ - 



.^. 



^ 



7r ■-•■«■ 



nun 



,,, 



c 



^ 



<^ 



O^ 



-2- 

with a deep yellow color (probably due to a tautomeric quinoid form) , 
and was reduced to If-methylresorcinol. A Knoevenagel condensation 
with ethyl malonate gave ethyl 5-hydroxycoumarin-3-carboxylate. 

Similar substitution takes place v/hen (3-resacetophenone is 
formylated by the Gatteraiann reaction and the structure of the product 
is indicated by the same general methods. Likevaae, if a group is 
substituted in the 5-position of methyl p-resorcylate, the 3-position 
is foirmylated. 

The explanation for the predominance of T- substitution - this 
being the first example of the particular type discussed above - is 
given by Shsli as being due to a fixation of bonds in the Kekule for- 
mula by chelation bet^'/een the carbomethoxy and the o-hydroxy groups 
in methyl p-resorcylato, requiring the presence of a double bond 
between the carbon atoms bearing those two groups. 

HOvj^^'^^OH Very few instances of direct substitution in the 

L |[ '^^ T-position have been recorded and there always seems 
^''vAlI^OMe to be a greater quantity of p -substitution. Baker and 
Carruthers have shov;n that, in p-substitutcd resorci- 
nols, the position taken by the incoming group depends 
upon the group already substituted in the ring, and, in all cases 
observed up to that time, the predominating reaction v;as to give the 
symmetrically substituted ring. This led to the idea that there was 
no fixation of the bonds in the ring as a result of chelation, and the 
normal directing influences of the groups are free to manifest them- 
selves. Mills and Nixon have pointed out that, when such phenolic 
compounds react, the substitution tokes place on the carbon atom 
attached by a double bond to the carbon carrying the hydroxyl. 

Other methods of substituting in the T-position of resorcinol 
have been advanced. Shah calls attention to a "highly involved" syn- 
thesis reported by Limaye in which he demcthylates 2,6-dimcthoxy- 
bcnzaldehyde using aluminum chloride in benzene . 

A few of the older syntheses for the introduction of a carboxyl 
group into the T-position involved such methods as heating resorcinol 
v/ith ammonium carbonate in a sealed tube, and heating with potassium 
bicarbonate in glycerol. The yields were all low, most of the substi- 
tution being directed to the p-position. Later, Hautliner, working on 
the basis of an observation of Lobry de Bruyn, heated m-dinitrobcnzene 
with potassium cyanide in methyl alcohol. The resulting compound 
contained a nitre-, a methoxyl- cjid a cyano-group in the proper 
relationship to afford a method of synthesis of ccxbo::cyresorcinol. By 
heating this substance with methyl alcohol and KOH, he obtained the 
2,6-dimethoxybenzonitrile , which on hydrolysis, gave the corresponding 
acid. The methyl groups were then removed with aluminum chloride in 
chlorobenzene . 

CN CN COOH COOH 

OgNv-^-x^NOg CH30.>>-yN[02 CHjOy-K^OCHj CH3 0v,>-v^0CH3 HO^^OH 

Wittig cjid Pockels have very recently employed phenyl lithium as 
a reagent for substituting in the T-position of resorcinol methyl 




■"O 



^ f J 



,. 



O 



-^OT 



dqiJC-i^ IV- 



110 



-Tf. 



<^ 



-3- 



ethers . 
synthcs 
ration 
the re?. 



Lithium goes into the ring and the product may be used for 
is of other compounds. Those investigators mention the prepa- 



of the corresponding acid, using 
ction Yilth benzophenone 



CO 



2> 



ojid of a tritajiol from 



CH3O 




OCH- 



+ 




Li 
CII30y.--S^CH3 



CO 



COOH 

CH3 0>^>^0CH3 








C5H5COC5H5 



AICI3 



OC0QIX.i 
OH 



OH 
^sKs-C-CgHs 
CH3 0y"^0CH3 



The preparation of T-substituted resorcinols is best accomplished 
by en indirect method. Though nitration of resorcinol is unsuccessful, 
by first sulfonating to obtain the disulfonic acid derivative (symmet- 
rical) cjid then nitrating, a product is obtained, v/hich, on hydrolysis, 
results in the 2-nitrorGsorcinol. 



HO 



y-^OH 



fumin g 
H2SO 



HO 
4 HO3S 




H 



O.^H 



HUO 



3-* 



HO3S 



NO. 



HOy^OH 



HpO- , 



SO3H heat 



HO2 

HOy^^OH 



dimethyl 
sulfate 



1^0. 



CH3 0y>v,^0CH3 



Kauffmpjin has prepared the dimethoyl ether of the nitro- compound 
and has converted this to a number of other derivatives. For example, 
he reduced the nitro group, diazotizcd the resulting ojiiine and replacec 
the diazonium group with Br, I, and CH. It is worthy of mention that 
all of these reactions are greatly influenced by steric effects which 
do not seem to act v/hen the free h^'^droxyl groups are present. 



1938, 1828; ibid., 1939 , 132; ibid., 



. Bibliography ; 

Shah ajid Laiwalla, J. Chem. Soc. 

1939, 300. 

Gatteraann, Ber., 3I, 1765 (1898); ibid., 32, 278 (1899). 
Limaye, Brit. Chem. Abst., 1937 > 258. 
Baker end Carruthors, J. Chem. Soc, 1937, 479. 
Mills ajid Nixon, ibid., 1930 , 2510. 
Mau timer, J. prakt . Chem.7rT21, 259 (1929). 
Wittig and Pockels , Ber., 727"89 (1939). 
Kauffmann and dePay, ibid., 37, 725 (1904) • 

Kauffmann .and Pranck, ibid. ,"3^, 2724 (1906); ibid., 40, 3999 (1907). 
Bacycr, Ann., 372, 127 (1910). 

Prepared by J. W. Shackleton 
Moy, 1939 



~r5 



T ' ^ 



LJ 



u 



kJ 



■J 



RAMM SPECTRA. IIT ORGMIC CHEMISTRY 

Kohlrausch, K.W.P. 

Origin of Raman Spectra :- Raman spectra owe their origin to the 
interaction of a light quantiJin hv and a molecule. In this effect each 
mode of oscillation in the molecule which interacts v;ith the light 
quantum results in the absorption of energy from this quantum and its 
consequent rescattering with less than its initial energy. Since each 
type of vibration or rotation gives rise to a Raman line, in a diatomic 
molecule only one line is probable. These vibrations are analogous to 
a simple mechanical system consisting of two masses held together witl.i. 
a spring. The frequency with which they vibrate is represented by 



= ii7¥ 



where P is the force per unit displacement ,/< is the reduced mass, and 

V is the frequency in wave numbers per centimeter, v is found on the 

photographic plate as the shift in frequency of the scattered light 
v;ith reference to the incident light. 

A system of 3 atoms of the type AXg may exist in a linear or non- 
linear form. The modes of vibration of the non-linear form is illus- 
trated in Pig. 1. Here the asymmetric oscillation is indicated by (A), 
j^ the symmetric by (B) and the 

o — > o deformation motion by (C). Per 

/ \ /i\ t/'^ \t "^^® linear model the shifts may 

-X o--*>o X o—'^o o o be calculated and are in reason- 

/" / A\ /g\ /Q\ able accord with the experimen- 



tal observations. 



Pig. 



It has already been indicated that the magnitude of Raman shifts 
are a function of the force exerted between the atoms, the type of 
motion, and the relative masses of atoms. As the masses decrease the 
Raman shifts increase. Therefore the shifts corresponding to the 
linear or symmetric oscillation of the C-H bond are of large magnitude 
They vary from 2800 cm~ to 3400 cm" ' , the rather wide deviations 
being due to variations in the force constant with constitution. Be- 
tween 1100 and 1460 cm"' lie the deformation oscillations of the C-H 
bond. The C-C linlcagos present a very complex problem. The lines 
which lie between 800 and 1100 cm" ' are generally attributed to this 
linkage. The total number of lines which occur v;ill depend on the 
molecular configuration and the interaction of valence forces. 

Charateristic Prequency : - The characteristic frequency is the 
frequency obtained from the oscillation of a C-X bond, which is the 
linkage of the carbon chain and a substituont. It is designated as 
W| . The symbol Wq is used to designate Eigenfrequencies which refer 
to the oscillations of an isolated bond. In tabic I are listed w, 
values for some tertiary -butyl derivatives. 

Tabic I 

(CH3)3C- -H -OH -CH3 -SH -CI -Br -I 

796 746 730 587 570 515 484 436 



w 



-2- 

If the substitucnt X is heavy compared to CH3 and the C-X (v/q) 
value is small compared to the C-C (wq) frdqucncy, we have the case 
where most of the shifts become independoht of X. The exception is 
the W| frequency. If one considers the mass of X becoming infinitely 
heavy, then the force between the C and X grows infinitely small; 
finally the case is reached where the (0113)30- is bound to an unyield- 
ing wall by an infinitely weak spring. In this manner we may obtain 
some information as to the vibrations of the radical by itself. 

The situation is similar if the Wq value for the substituents is 
appreciably higher than that of the 0-C bond, as is the case with 
=:0-H, -OH, -M, -SH, -0=0, -C=1T. Por these more complex substituc:r:-s 
several valence frequencies are found. In the divalent cajrbonyl ajid 
the tctravalent ethylene groups the situation is remarkably complex. 
The carbonyl frequency depends only slightly on the branching of the 
attached carbon chains. It varies considerably in the different 
functional groups which contain it. The shift lies approximately at 
1654 cm"' for acids, 1710 cm"' for methyl ketones, 1720 for aldehydes, 
1735 for esters, and at 1795 for acid chlorides, following is a list 
of some shifts attributed to variously substituted ethylenes. 

H2C=CH2 1623 cm-' RHC=0H2 1647 cm"' RHO=CHR' 1658 cm"' (trans) 

RgC^CR'^ 1678 cm"' RHC=OHR' 1674 cm"' (cis) 

The Raman spectra for 0=0 linkages fall into the region I6OO- 
1700 and are more or loss completely separated from any lines due to 
any other typo of oscillation. These lines are sharp and can be 
measured with accuracy. Therefore the effects of mass and of binding 
force caji be determined v^rith considerable exactness. This opens up 
a more or less endless avenue in the investigation of constitutional 
problems where ethylenic linkages are involved. It may be stated 
categorically that any straight chain and most cyclic compounds having 
the structure -0=6- show a Raman shift in the 1600 region. If the 
structure is -0=0=0- this is not true. If the stracture is a conjugat- 
ed one 5 the characteristic 1600 line is lowered by about 20 cm"' and 
its intensity is increased. 

i s - Tj^ans I s omeri sm : - It has been found that with reference to 
the 0=0 oscillation, the trans compound has a shift greater by 15 cm." ' 
than the cis compound. It is quite possible to determine the composi- 
tion of mixtures containing cis ajid trans f onus , ajid to follow the 
change in composition on distillation by its spectra. 

Free Rotation :- 1. In going from l,l-dichloroetha2ie to 1,2-di- 
chl ore ethane the number of lines between O-I6OO increases from 
approximately 11 to 17 although the number of ch^\in members is the 
same. This seems to indicate an increase in the number of molecular 
f orms . 

2. The sharpness and the brightness of the lines has not changed. 
It should be remembered that in the cis ojid t rans forms of 1,2-dichlo- 
roethylene two distinct and separatenSaman spectra were found. If we 
are to assume all possible configurations for 1,2-dichloroethane, we 
should have bands for the Raman spectra instead- of lines. 

3. In going from 1,2-dichloroethane to 1,2-dibromoethane , the 
number of lines again increases. 



• y: 



•3- 



4. There are depolarized frequency lines. These are only pos- 
sible in the spectra of a four-membered, open chain if both cis and 
trans forms are present. 

Conclusions: There are definitely semi-stable cis and trans 
foznns present in 1,2-dichloroethane. The presence of interm.odiate 
foiras are not necessary to explain the spectra obtained. Since no 
band spectra is obtained, their presence is unlikely. The relative 
amounts of the cis and trans fonns can be calculated from intensity 



and polarization measurements . 
exists only in the trans form. 



In the solid state, 1,2-dichloroethane 



T automerism :- The demonstration of keto-enol tautomcrisra was one 
of the earliest applications of the Raman effect. The classical work 
was done on the acctoacetatos . More recently kotimid-enamine isomer- 
ism has been investigated in this manner. 

Ring Strain ; - Belov./ arc given some frequency values for some 
linkages in saturated cyclic compounds. 



radical 


vv (C-0) 

1640 


w(C-H) 
5080 


w (C: 


=0) in YCOR 


(J2H3 




1718 


C3H5 


1138 


3011 




1719 


C4H7 


950 


2920 




1722 


O5H9 


890 


2873 




1724 


Oe^i I 


800 


2854 




— 


^7^\ 3 


729 


2858 




— 


CsH.s 


700 


2857 




— 



V/lien we remember that the frequency is proportional to the square root 
of the binding force , v./e sec that with increase in ring members the 
C-C and C-H binding forces decrease whereas the C=0 binding force in 
the external shain increases. 



Disubstituted Benzenes; - The fact that para derivatives have less 
lines for a given range than the corresponding ortho and meta deriva- 
tives indicates that the para derivatives have a greater symmetry. 

In closing it might be v/cll to remark that the number of lines 
found are large and increase v\;ith the complexity of the compound. Thu 
it is evident why only relatively simple molecules have been studied. 

The magnitude of the shifts is not the only consideration in- 
volved in matching and follov/ing the changes in frequency with the 
changes in the compounds. Two other parameters which may be invoked 
arc the intensities and the degree of polarization. In Raman spectra 
the symmetrical oscillations arc the most intense and most strongly 
polarized. 

Bibliography : 

Kohlrausch, Ber. , 71A, 171 (1938). 
Hibbcn, Chem. Rev., 18, 1 (1936). 

Prepared by A. W. Anderson 
May, 1939