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

Full text of "Standardization of potassium permanganate solution by sodium oxalate"

yr% 



STANDARDIZATION OF POTASSIUM PERMANGANATE 
SOLUTION BY SODIUM OXALATE 



By R. S. McBride 



CONTENTS 

Page 

I. Introductory Part 612 

1. Object of the research 612 

2. Considerations affecting the choice of a standard 612 

3. Reasons for the choice of sodium oxalate 613 

4. Normal course of the reaction 614 

II. Experimental Part 616 

1. Reagents employed 616 

2. Weighing of oxalate and permanganate used 617 

3. Effect of conditions upon the result of a titration. 618 

(a) End-point corrections 619 

(6) Rate of addition of the permanganate - 62 1 

(c) Volume of the solution 624 

(d) Acidity and temperature 624 

( e) Air access — oxidizing effect 628 

(/) Presence of added manganous sulphate 63 1 

III. Discussion and Conclusions 632 

1. Review of possible errors 632 

(a) Loss of oxalic acid by volatilization 633 

(b) Decomposition of oxalic acid by water 633 

(c) Decomposition of oxalic acid by sulphuric acid 633 

(d) Oxidation of oxalic acid by the air 634 

(e) Incomplete oxidation of the oxalic acid 636 

(/) Loss of oxygen 637 

(g) Incomplete reduction of the permanganate 639 

(k) Impurities in the oxalate 639 

(*) Abnormal oxidation products 639 

2. Summary and conclusions 640 

3 . Method recommended for use 641 

4. Accuracy and precision attainable : 641 

611 



612 Bulletin of the Bureau of Standards ivoi.s 

I. INTRODUCTORY PART 
1. OBJECT OF THE RESEARCH 

The standardization of potassium permanganate solutions has 
been the subject of much study and an excessive amount of con- 
troversy. There have been far too many standards proposed for 
this work to permit one not familiar with the subject to select the 
best; and, in fact, it is doubtful whether or not any one of those 
standards proposed can in all senses be considered the best. 
However, the work of the Bureau of Standards has demanded 
that some substance be selected for this use which could be 
employed with a certainty of a reasonably correct result. It was 
desired, if possible, that the standard selected should serve a 
threefold purpose, viz: First, as a primary standard of oxidimetry ; 
second, as a working standard for regular use in our own labora- 
tories; and third, as a substance which could be distributed by 
the Bureau with a guarantee both as to its purity and as to its 
reducing value when used under specified conditions. 

Although the voluminous literature relating to the standardiza- 
tion of potassium permanganate solutions has been examined 
with considerable care, it is not thought desirable to give a history 
of the subject, or even a bibliography. None of the theories con- 
sidered here are new; but it is hoped that the experimental facts 
presented will be of value as a guide to the proper use of sodium 
oxalate as a standard. 

2. CONSIDERATIONS AFFECTING THE CHOICE OF A STANDARD 

It is seldom that a single measure can be used both as a primary 
reference standard and as a regular working standard, but for some 
volumetric work this is possible. A substance to serve any such 
double function must be such that the following conditions will 
be fulfilled: 

(a) Reasonable ease of preparation and accurate reproduci- 
bility of material must be assured. 

(6) The purity must be determinable with sufficient accuracy, 
and the purified material must be stable under ordinary conditions 
of the laboratory. 



McBride] Standardization of Potassium Permanganate 613 

(c) The use of the material in regular work must demand neither 
complex apparatus nor difficult manipulations. 

{d) Such precision must be obtainable when it is used with 
ordinary care that one, or at the most a very few determinations 
suffice for the fixing of the value of a standard solution. 

(e) The accuracy obtained under ordinary conditions of its use 
in standardization must be at least as great as that required in the 
use of the solution to be standardized. 

3. REASONS FOR THE CHOICE OF SODIUM OXALATE 

In the above six respects it appeared that sodium oxalate was 
probably best suited to our needs, and a detailed study of this 
standard was undertaken. 

The methods of preparation and testing of sodium oxalate have 
been carefully studied by Sorensen. 1 In continuation of such study, 
Mr. J. B. Tuttle and Dr. William Blum, of this laboratory, have 
carried out several important lines of work and, at the request of 
the Bureau of Standards, several firms of manufacturing chemists 
have improved their methods of preparation of sodium oxalate on 
a large scale. At this point it is sufficient to state that all of this 
work indicates that it can be prepared in a suitable form at reason- 
able expense ; that it is reproducible ; that its purity can be tested 
readily; and that once purified it is satisfactorily stable under 
ordinary conditions. 2 

The discussion of the other three criteria as to the value of 
sodium oxalate, viz, convenience, precision, and accuracy, forms 
the subject of the present article. 

In advance of the general discussion it is not amiss to state the 
conclusions drawn as to these three points. It appears that when 
the conditions for the use of sodium oxalate have been defined they 
may be conformed to easily and no unusual apparatus or com- 
plex procedure is necessary. The accuracy obtainable is, within 
the limits of our present knowledge, sufficient for even the most 
refined work (see p. 641), and the precision or agreement of dupli- 
cates is satisfactory. 

1 Zs. anal. Chem., 36, p. 639, 42, p. 333, and p. 512. 

2 See Blum, J. Amer. Chem. Soc, 34, p. 123; 1912. 



614 Bulletin of the Bureau of Standards [Voi.8 

Along with such striking advantages we find certain disadvan- 
tages, but none of these appear serious. The more important are 
as follows: 

(a) The largest sample which can be used ordinarily (for 50 cc 
of N/10 KMn0 4 ) is only 0.3 gram. This necessitates an accurate 
weighing in order to gain a high degree of accuracy. It is not 
desirable to use a standard stock solution, unless freshly prepared. 
If such solution is kept for a considerable length of time, it acts 
upon the glass of its container ; it is also slowly decomposed by the 
action of light. 

(6) The detection of small amounts of allied organic compounds 
in the sodium oxalate is rather difficult, except by comparison of 
the reducing value with that of other samples of known purity. 

(c) The initial drying of the oxalate is subject to slight uncer- 
tainty, but once dried it is practically nonhygroscopic. 

4. NORMAL COURSE OF THE REACTION 

In 1 866 Harcourt and Bsson 3 concluded from a study of the 
speeds of reaction under various conditions, that the steps of the 
reaction are as follows : 

I. 2 Mn(OH) 7 +5H 2 C 2 4 =2Mn(OH) 2 +ioC0 2 +ioH 2 (very slow) 

II. 3Mn(OH) 2 + 2 Mn(OH) 7 =5Mn(OH) 4 (very fast) 

III. Mn(OH) 4 +H 2 C 2 4 =Mn(OH) 2 +2C0 2 +2H 2 (fast, but less so than II.) 

Among the more recent articles presenting either experimental 
or theoretical evidence on this subject, the more important are 
those by Schillow 4 and by Skrabal. 5 

From measurements of the speeds of reaction, the former 
presents the following system as representing the steps of the 
reaction : 

I. Mn(OH) 7 +2H 2 C 2 4 =Mn(OH) 3 +4C0 2 +4H 2 (very slow) 

II. Mn(OK) 3 .2H 2 C 2 4 +Mn(OH) 7 =2Mn(OH) 3 +4C0 2 +4H 2 (measured) 

III. Mn(OH) 3 +2H 2 C 2 4 =Mn(OH) 3 .2H 2 C 2 4 (practically instantaneous) 

The second of these can be divided into two parts : 

II,a. Mn(OH) 3 .2H 2 C 2 4 +Mn(OH) 7 =Mn(OH) 6 +Mn(OH) 4 . 2 H 2 C 2 4 (measured) 
II,b. Mn(OH) 4 .2H 2 C a 4 +Mn(OH) 6 =Mn(OH) 3 + 4 C0 2 +4H 2 (practically instanta- 
neous) 

» Philos. Trans., 156, p. 193; 1866. * Ber., 86, p. 2735; 1903. 6 Zs. anorg. Chem., 42, p. 1; 1904. 



McBride) Standardization of Potassium Permanganate 615 

However, this author qualifies the proposed explanation by the 
following statements: 

"This scheme is considered only as an approximate picture of 
the phenomenon and is true particularly for mean concentration 
of hydrogen ions and low temperatures (0-2 5 °). Under other 
conditions, side reactions and disturbances enter in which can be 
partially observed or foreseen. * * * 

"In addition to the reactions given above, the following two 
also require consideration: 

IV. 2Mn(OH) 3 +H 2 C 2 4 =2Mn(OH) 2 + 2 C0 2 +2H 2 
V. 4 Mn(OH) 2 +Mn(OH) 7 =5Mn(OH) 3 " 

These last reactions, however, are not assigned a definite role in 
the general scheme. 

After an extended series of experiments on the speed of the 
reaction under various conditions, Skrabal (loc. cit.) advances the 
following scheme as representing the course of the reaction. 

Incubation period: 

(1) H 2 C 2 4 +KMn0 4 =Mir--+C0 2 (measurable) 

(2) H 2 C 2 4 +Mn-"=Mn(OH) 2 +C0 2 (practically instantaneous) 

Induction period: 

(3) Mn(OH) 2 +KMn0 4 =Mn- (less rapidly) 

(4) Mn""+H 2 C 2 4 =Mn(OH) 2 +C0 2 (practically instantaneous) 

(5) Mn— +H 2 C 2 4 =Mn(OH) 3 .H 2 C 2 4 (practically instantaneous) 

(6) Mn ,, *=Mn(OH) 2 +Mn(OH) 4 (practically instantaneous) 

End period: 

i(7) Mn(OH) 3 .H 2 C 2 4 =Mn--- (measurable) 
(8) H 2 C 2 4 +Mn--'=Mn(OH) 2 +C0 2 (practically instantaneous) 



"i; 



[(10) H 2 C 2 4 +Mn'* , =Mn(OH) 2 +C0 2 (practically instantaneous) 

Since this last system of reactions is based upon elaborate 
experimental work, it can be accepted as probably best repre- 
senting the normal course of the reaction. For the present con- 
siderations any one of the systems would serve equally well to 
explain the observed facts. 

The variations from a normal course will be considered after the 
experimental part of the present work has been described. 



6i6 Bulletin of the Bureau of Standards [Voi.8 

II. EXPERIMENTAL PART 
1. REAGENTS EMPLOYED 

The water and sulphuric acid employed were frequently tested 
for reducing matter and in each case were shown to be free from 
appreciable amounts of such impurities. The manganous sulphate 
was similarly tested for reducing and oxidizing influence and 
shown to be satisfactory. To purify the air used for the tests of 
Tables VI and VII, it was passed through cotton wool to remove 
dust and grease, bubbled through a solution of potassium hydroxide 
to take out any acid vapors present, and then through solutions of 
chromic acid and potassium permanganate. The purified air had 
no detectable reducing effect upon dilute solutions of permanganate 
under the conditions of titration. 

The carbon dioxide used for the tests of Table VI was taken 
from a cylinder of the commercial liquid and passed through 
water, chromic acid, and permanganate solution before use. An 
analysis showed the presence of about 3 per cent of methane; 
but this gas seemed to have only a very slight reducing action on 
the permanganate in dilute solution, so that the results of the 
series in which it was used can be regarded as but slightly less 
reliable than the series in Table VII. For this latter group of 
tests, pure carbon dioxide was made from acid and soda. This 
source gave a gas which had no detectable reducing action under 
the conditions of its use. 

The potassium permanganate used for most of the work was a 
sample of good quality which had been made up in normal solution 
for over six months before filtration and dilution to tenth normal 
strength for use. The diluted solution was filtered frequently 
through asbestos to insure freedom from precipitated manganese 
dioxide. For the series of tests reported in Table IV, b, a second 
permanganate was used. In this case the strong solution was 
boiled for a few minutes, cooled, filtered, and diluted to tenth 
normal strength. For the series of Table VII still a different 
permanganate was employed, this stock being prepared in the 
same manner as the main solution. About 40 grams of solution 
were used for each titration. 



McBride] Standardization of Potassium Permanganate 617 

The sodium oxalate used for all of the experiments except those 
of Table IV, b, was a sample specially purified in this laboratory 
by Mr. J. B. Tuttle. It is sufficient to state here that all tests 
indicated a total impurity of not over 0.05 per cent. For the 
tests of Table IV, b, another sample of sodium oxalate, made by 
the Mallinckrodt Chemical Works especially for our use, was 
employed. This material was shown by test to be as pure as 
that prepared in our own laboratory. 

2. WEIGHING OF OXALATE AND PERMANGANATE USED 

Since in all of the work described in the following part of this 
article only comparative values were required, the absolute amount 
of oxalate employed in any one test was not of importance as long 
as the relative amounts present in the experiments of a series were 
accurately known. Therefore it was found desirable to use about 
20-gram portions of a N/5 stock solution of the oxalate for each 
test. Each such sample was weighed from a burette to the nearest 
5 mg; the relative wieght of each sample was thus determined to 
better than 1 part in 2000 in less time and with greater certainty 
than would have been possible by weighing out the dried powder. 

In order to prevent any change in the strength of the stock solu- 
tion from affecting the conclusions drawn from the results of any 
series of titrations, only those values obtained during a period of 
a few days are compared with each other. This plan avoided any 
uncertainty due either to slow decomposition of the oxalate in 
solution occurring through action on the glass of the container, 
or oxidation by the air through the aid of light, or to change in 
oxidizing value of the permanganate solution employed. 

The measurement of the permanganate solution was accom- 
plished by the use of a weight burette. For this work this form 
of instrument possesses the following advantages: (a) Correction 
for the temperature changes which affect the volume of the solu- 
tion is not necessary; (b) completeness or uniformity of running 
down of the solution from the burette walls is unessential; and 
(c) the solution can be weighed readily to 0.0 1 g (1 part in 5000 
on a 50-g sample), whereas measurement to 0.0 1 cc is exceedingly 



618 Bulletin of the Bureau of Standards [Voi.s 

uncertain. The freedom from errors due to temperature changes 
is of great importance in this work, as the burette is often very 
appreciably warmed by the steam rising from the hot liquid under- 
going titration. 

The burette used was made from a 50 cc cylindrical separatory 
funnel by drawing down the stem to the form of an ordinary 
burette tip. For exact work it has been found desirable to have 
this tip so drawn down that it will deliver about 10 cc per minute 
or 0.03 cc per drop. 

3. EFFECT OF CONDITIONS UPON THE RESULT OF A TITRATION 

The ordinary procedure for the use of sodium oxalate in the 
standardization of potassium permanganate, is as follows: Dis- 
solve about a quarter of a gram of the oxalate in 250 cc of water, 
acidify with sulphuric acid, warm to 70 °, and titrate to the first 
permanent pink. 

It was desirable to determine the effect of the variation of the 
following conditions upon the result obtained, viz, temperature, 
acidity, volume of solution, rate of addition of the permanganate, 
access of air, presence of added manganous sulphate and in con- 
nection with these, the corrections necessary upon the apparent 
end points. In order to accomplish such determination, the 
factors were varied one at a time, noting the difference, if any, 
produced upon the apparent value of the permanganate. The 
results of the titrations are reported as the ratios of the oxalate 
used to the permanganate used multiplied by 10 -2 . These values 
are proportional to, and, indeed, numerically almost equal to, the 
iron value of the permanganate. Therefore, for purposes of dis- 
cussion the values are treated as if they were the iron value of the 
permanganate, expressed in grams of iron per gram of solution. 
It should be noted that an increase in the iron value represents a 
decrease in the permanganate consumed, and vice versa. A 
variation of 0.0 1 cc in the amount of permanganate used in a 
titration is approximately equivalent to a change of one unit in 
the last expressed figure of the iron value. 



McBride] Standardization of Potassium Permanganate 619 

(a) END-POINT CORRECTIONS 

In a recent article Dr. W. C. Bray 6 has suggested the necessity 
for the correction of the apparent end point obtained in the 
reaction of oxalic acid and permanganate, and has recommended 
that the correction be determined as follows: After reaching the 
end point the solution is cooled to room temperature, potassium 
iodide solution is added, and the iodine liberated is at once titrated 
with dilute (N/50) sodium thiosulphate. 

In order to test the necessity and the accuracy of this method 
under the various conditions of titration, we examined experi- 
mentally the following points: 

(a) Does the equivalent in oxidizing power of the permanganate 
excess remain in the solution long enough to allow the necessary 
cooling before the titration with thiosulphate? 

(b) Does the thiosulphate titration give the total permanganate 
excess used to produce the end point, or does it indicate only the 
permanganate which remains in the solution as such ? 

(c) Does the depth of the pink color at the end of the titration 
show how great an excess of permanganate has been added in 
order to produce that end point ? 

Before applying the method of correction proposed by Bray, it 
is necessary to cool the solution to room temperature. Therefore, 
if, as is desirable, the permanganate end point is obtained while the 
solution is hot, some time must elapse during cooling and it is 
possible that a loss of oxidizing (or iodine liberating) power would 
occur, which loss would render the correction as subsequently 
determined valueless. This point was tested by determining the 
residual oxidizing power of solutions to which had been added 
known amounts of permanganate at various temperatures (30-90 °), 
with varying acidities (2-10 per cent by volume H 2 S0 4 ) and with 
the addition of various amounts of manganous sulphate (up to 1 g 
MnS0 4 .4H 2 0). These experiments showed that no appreciable 
loss occurred within a period of one-half hour, if only a small 

8 J. Amer. Chem. Soc, 32, p. 1304; 1910. 



620 Bulletin of the Bureau of Standards [Vol. 8 

amount of permanganate (0.02-0.20 cc of N/io solution) is added 
to any such solution. Moreover this condition was found to exist 
even when the permanganate was wholly decolorized by reaction 
with the manganous salt. The full oxidizing power was then 
retained in the solution, probably in the tri- or tetra-valent 
manganese salts (Mn*" or Mn""). The lapse of 5 to 10 min- 
utes necessary for cooling the solution to room temperature, is 
thus shown to be without influence ; and it is evident that the pro- 
cedure described determines not only the permanganate remaining 
in the solution as such, but also that which was not completely 
reduced to the manganous condition. 

The procedure described by Bray has been accepted on the 
above basis as giving a determination of the real excess of per- 
manganate which was added to produce the end point. However, 
the question as to the difference between this end-point correction 
and that determined by the depth of the pink color at the end 
point, must be considered on a different basis. 

For each of the titrations made in this investigation, two values 
were obtained, the one using the color method, the other the 
iodine method for correction of the total permanganate added. 
In several of the following tables both of these values are reported. 
By inspection of these data (Tables II, III, and V) it will be 
apparent that, in general, the color correction and the iodine 
correction differ by no more than 0.02 cc, except when the end 
point was reached with the solution at temperatures below 35 , 
or when the permanganate solution was added rapidly just at the 
end point. This was true in all but 6 of over 200 tests; and even 
in these the difference between the two corrections was not more 
than 0.04 cc. It is therefore certain that if the end point is 
approached slowly in a solution above 6o°, the depth of color will 
be proportional to the total excess of permanganate added, i. e., 
no permanganate will be decolorized without at the same time 
being completely reduced to the manganous condition. How- 
ever, at lower temperatures, in the presence of much sulphuric 
acid (more than 5 per cent by volume), and particularly with 



McBride] Standardization of Potassium Permanganate 621 

rapid addition of permanganate or with insufficient stirring just 
preceeding the end point, the amount of permanganate decolor- 
ized but not wholly reduced to the manganous condition, easily 
may be as much as 0.1 cc. 

(b) RATE OF ADDITION OF THE PERMANGANATE 

The rate of the addition of the permanganate solution in the 
titration of iron in solutions containing chlorides has been shown 
to have considerable effect upon the result obtained. 7 The influ- 
ence of this factor upon the oxalate-permanganate reaction was, 
therefore, studied. 

The influence of varying the amount of permanganate added 
before the quick color change of rapid reduction began was first 
tested. This factor has an influence only at low temperatures 
(below 40-50 ), as at any higher temperature the reduction pro- 
ceeds rapidly from the very start. Tests made at 30 in 5 per 
cent sulphuric-acid solution gave results as follows : 

Amount of KMn0 4 added before 

rapid decolorization was evident. . 3 cc 10 cc 25 cc 40 cc 
Results obtained (strength of 

KMn0 4 ) o. 005936 o. 005936 o. 005932 o. 005929 

37 35 33 

34 37 

These results show that the addition of large amounts of per- 
manganate before rapid oxidation of the oxalate begins tends to 
give lower values for the permanganate solution, i. e., to cause a 
larger consumption of permanganate. The influence of this factor 
is small, but it is of some significance, as will be apparent from the 
later discussion. 

7 A. Skrabal: Zs. anal. Chem., 42, p. 359; 1903. A. ' N. Friend: J. Chem. Soc. (London), 95, p. 1228; 
1909. Jones and Jeffrey, Analyst. 34, p. 306; 1909. 



622 Bulletin of the Bureau of Standards [Voi.8 

TABLE I 
Effect of Rate of Addition of Permanganate During Main Part of Titration 



Experiment No. 


Acidity, volume 
per cent H 2 S04 


Temperature 


Time for titration 
(minutes) 


Result, corrected iron 
value 


118 






5 


0. 005945 


120 






5 


45 


119 


5 


80° 


8 27 


46 


125 






9 120 


56 


129 






» 120 


56 


121 






[ 7 


0. 005942 


123 
122 


[ 5 


30° 


6 
»37 


42 
44 


124 




_ 


[ »120 


44 


230 






0.5 


0. 005592 


231 






0.5 


92 


206 






5 


92 


223 


2 


30° 


4 


93 


207 






5 


91 


225 






8 60 


94 


226 






8 60 


95 


232 






0.5 


0. 005589 


233 






0.5 


88 


208 
209 


2 


90° 


5 
5 


98 
5600 


227 






8 30 


00 


229 






830 


5598 



8 Dropwise with stirring. 

9 Intermittent, without continuous stirring. 

After the initial reduction of 3-10 cc of permanganate the rate 
of addition of the main portion of the permanganate was varied, 
giving the results listed in Table I. These experiments indicate 
that except on long standing at a high temperature (after starting 
the reaction), or with a very rapid addition of permanganate to 
a weakly acid solution at a high temperature, the rate of the addi- 
tion of the permanganate during the main part of the titration 
has no influence. The high values obtained in experiments 125 
and 129 are probably due to oxidation of the oxalic acid by the 
air, and the low values of experiments 232 and 233 show a loss of 
oxygen, because of the excessive rate of the permanganate addi- 
tion. 



McBride] Standardization of Potassium Permanganate 



623 



The effect of the rate of the addition of the permanganate just 
at the end point, has already been suggested (p. 620), but the 
following data will make this point clearer. In each pair of tests 
the only variable was the rate of the addition of the permanganate 
solution just before the end point. Only the two experiments in 
each group may be compared. (See Table II.) 

In the first part of Table II some rather extreme cases (experi- 
ments done at 30 ) have been chosen to show how large the error 
of the apparent end point may be, without any influence upon the 
corrected value. In cases where much smaller corrections were 
necessary (temperature 6o° and above) the agreement of uncor- 
rected values was better, regardless of the speed at which the end 
point was approached. The latter part of Table II shows this 
agreement of the uncorrected values at the higher temperature. 
Nevertheless, it is not desirable to approach the end point so 
rapidly as to make a correction larger than 0.1 cc necessary. 

TABLE II 
Effect of Rate of Addition of Permanganate at the End Point 

[In the first of each pair of experiments the end point was approached slowly, in the second rapidly] 





Acidity, 
volume 
per cent 
H2SO4 


Color blank 


Iodine blank 


Value of permanganate 


Temperature 


Uncorrected 


Color 
corrected 


Iodine 
corrected 


30° 
30° 
30° 
30° 
30° 
60° 
60° 
83° 
96° 


2 
5 
5 

5 

10 
2 
5 
2 
2 


f 0.01 
1 .25 
J .03 
1 .15 
J .05 
1 .08 
J .04 

J .06 
1 .08 
[ .05 
1 .03 
J .05 
1 .05 
J .06 
1 .06 
f .05 
I -2 


0.03 
.34 
.09 
.39 
.085 
.19 
.08 
.51 
.11 
.14 
.05 
.05 
• 07 
.06 
.07 
.06 
.04 
.15 


0. 005584 

37 

. 005929 

5894 

. 005030 

17 

. 005923 

5871 

.005636 

33 
. 005642 

45 
.005638 

40 
. 005642 

45 
. 005647 

33 


0.005586 
71 

.005933 
12 

.005034 
26 

.005929 

. 005642 

43 
. 005649 

49 
. 005646 

44 
. 005650 

52 
.005652 

61 


0. 005589 

87 
.005942 

42 
. 005037 

38 
. 005935 

37 
. 005648 

50 
.005649 

51 
. 005649 

46 
. 005651 

52 
.005651 

54 



73764 — 13- 



250 cc 


500 cc 


700 cc 


1000 cc 


0. 005139 


0. 005136 


0. 005133 


0. 005127 


39 


34 


32 


29 


37 









624 Bulletin of the Bureau of Standards [Voi.s 

(C) VOLUME OF THE SOLUTION 

The volume of the solution in which the titration is made might 
affect the rate of the reaction and, therefore, possibly the result 
of the titration. Tests were made of this point by varying the 
initial volume in a series of titrations made at 90 in solutions 
containing 5 per cent by volume of sulphuric acid. The results 
were as follows : 

Initial volume 50 cc 

Values obtained o. 005139 

36 

By increasing the initial volume beyond 250 cc, the resulting 
value for the permanganate solution is markedly decreased. That 
this is due to the change in the initial concentration of the oxalate 
is indicated by the results obtained in two titrations in which the 
initial volume was 1000 cc but the amount of oxalate used was 
increased threefold, thus making its initial concentration about 
that ordinarily resulting in an initial volume of 350 cc. The 
results thus obtained were 0.005136 and 0.005136. 

At 30 the initial rate of the permanganate reduction is so much. 
less than at 90 , that it is not possible to note the effect of initial 
volume free from this influence. A few results were obtained 
as follows : 

Initial volume 50 cc 250 cc 1000 cc 

Values obtained o. 005943 o. 005942 o. 005945 

45 42 

In these experiments no decided effect of volume is to be found. 

It should be noted that in each of these cases the change in 
volume had no effect upon the concentration of the sulphuric acid 
present. 

(d) ACIDITY AND TEMPERATURE 

The effects of acidity and of temperature were investigated by 
two series of experiments, the results of which are given in Tables 
III and IV. These data show that either higher temperature or 
less concentration of sulphuric acid tends to give higher values 
for the permanganate, i. e., to reduce the amount of permanganate 
required. These effects are not large and, indeed, the results 



McBride] Standardization of Potassium Permanganate 



625 



obtained in some cases show duplicates as discordant as experi- 
ments carried out under somewhat different conditions; neverthe- 
less, the tendency in the two directions is rather convincing that 
these factors have a real influence upon the results obtained. The 
importance of even such slight variations is considerable, as will 
be apparent from the discussion later. 

It will be seen from the data of Table IV that the first series 
exhibits a larger apparent influence of temperature than the 
second series reported. This is due to the greater care exercised 
in the latter series to stir the solution very vigorously and to add 
the permanganate more slowly, particularly at the beginning and 
end of each titration. However, the first series shows better the 
variation which may be expected if only ordinary titration pre- 
cautions are observed. 

TABLE III 
Effect of Acidity and Temperature 





Acidity, 
volume per 
cent H 2 SO* 


Color blank 


Iodine blank 


Value of permanganate 


Temperature 


Uncorrected 


Color 
corrected 


Iodine 
corrected 




2 


0.04 


0.06 


0. 005129 


0. 005135 


0. 005137 




2 


.04 


.05 


33 


33 


39 


30° 


5 


.04 


.05 


27 


32 


34 




5 


.03 


.035 


27 


31 


31 




10 


.05 


.10 


18 


25 


32 




10 


.02 


.09 


21 


24 


33 




2 


.02 


.03 


. 005134 


.005136 


. 005138 




2 


.03 


.04 


34 


38 


40 




5 


.03 


.04 


30 


34 


35 


45° 


5 


.03 


.025 


30 


37 


- 36 




10 


.01 


.03 


35 


36 


39 




10 


.03 


.03 


31 


35 


35 




10 


.02 


.035 


37 


40 


42 




2 


.02 


.02 


.005129 


.005132 


.005132 




2 


.03 


.03 


36 


40 


40 




2 


.035 


.035 


33 


37 


37 




2 


.04 


.05 


35 


41 


42 


60° 


5 


.04 


.06 


26 


32 


46 




5 


.05 


.05 


32 


39 


39 




10 


.05 


.05 


33 


40 


40 




10 


.03 


.05 


33 


37 


40 




10 


.02 


.04 


34 


38 


40 



626 Bulletin of the Bureau of Standards 

TABLE III— Continued 
Effect of Acidity and Temperature — Continued 



[Vol.8 



1 


Acidity, 
volume per 
cent H2SO4 


Color blank 


Iodine blank 


Value of permanganate 


Temperature 


Uncorrected 


Color 
corrected 


Iodine 
corrected 




2 


0.03 


0.02 


0. 005141 


0. 005144 


0.005143 




2 


.03 


.03 


37 


41 41 


75° 


5 


.04 


.025 


37 


42 40 


5 


.03 


.03 


39 


43 43 




10 


.05 


.06 


29 


36 37 




10 


.06 


.04 


34 


43 40 




2 


.05 


.045 


.005138 


. 005145 . 005144 




2 


.02 


.03 


36 


39 40 




2 


.04 


.05 


36 


42 43 


SO* 


5 


.03 


.05 


38 


42 


44 




5 


.04 


.04 


39 


44 


44 




10 


.03 


.03 


34 


38 


38 




10 


.03 


.03 


36 


40 


40 




2 


.04 


.03 


. C05142 


. 005146 


.005144 




2 


.02 


.03 


40 


43 


43 


86° 


j 


.03 


.035 


38 


42 


43 




5 


.02 


.02 


37 


40 


40 




10 


.02 


.02 


38 


41 


41 




10 


.03 


.045 


35 


39 


41 




2 


.02 


.03 


. 005143 


. 005146 


. Q05148 




2 


.02 


.03 


46 


48 


49 




5 


.035 


.04 


39 


43 


44 


92" 


5 


.05 


.05 


32 


39 


39 




5 


.05 


.06 


30 


37 


39 




5 


.06 


.06 


31 


39 


39 




10 


.05 


.05 


32 


38 


38 




10 


.06 


.06 


30 


38 


38 



McBrid*} Standardization of Potassium Permanganate 

TABLE IV 
Effect of Acidity and Temperature 

(a) SERIES 1, REARRANGEMENT OF DATA OF TABLE III 
[Each value reported is the average of at least two determinations] 



627 





Temperature 


Acidity, volume per cent H2SO4 
















2 per cent 




5 per cent 


10 per cent 




30° 


0. 005138 




0.005133 


0. 005132 




45° 


39 




35 


38 




60° 


38 




42 


40 




75° 


42 




42 


38 




80° 


42 




44 


39 




86° 


44 




42 


41 




92° 


48 




40 


38 



(b) SERIES 2 



30° 


0. 005646 


0. 005646 


0. 005649 


60° 


50 


48 


48 


80° 




50 




85° 


51 


50 


44 


92° 




50 




95° 


53 


52 ' 


46 



In addition to the data of Tables III and IV, those in other 
tables will show the same influence of temperature and acidity as 
are here indicated. Taking these data all together, it is certain 
that, though small, these influences are appreciable. 

The time in the titration at which the tempertaure has an 
influence upon the result is evident from the tests of Table V. 
These data show that, except in its influence upon the size of the 
end-point correction, the temperature of the solution after the first 
few cc of the permanganate have been decolorized is without 
appreciable effect. The significance will be apparent when con- 
sidered in connection with the results of the tests given in Table 
VIII. 



628 



Bulletin of the Bureau of Standards 

TABLE V 
Time at which Temperature has Effect 



[Vol.8 





[All titrations started in 250 cc oi 


5 per cent by volume sulphuric acid] 














Value of permanganate 


Temperature 


: (a) Initial, (b) during 




Color 
blank 


Iodine 






main part 


of titration, (c) at end 










point 










Uncorrected 


Color 
corrected 


Iodine 
corrected 


(a) 
90° 


(6) 

90° 


(e) 

90° 


{ 


0.05 


0.06 


0. 005130 


0. 005137 


0. 005139 








r 


.06 


.06 


31 


39 


39 








J 


.03 


.07 


29 


33 


38 


90° 


90° 


30° 


[ 


















t 


.03 


.065 


30 


34 


39 








.03 


.09 


25 


?9 


37 


90° 


30° 


30° 




















.04 


.12 


19 


24 


35 










.03 


.05 


. 005126 


.005130 


. 005133 


30° 


30° 


30° 


..05 


.08 


24 


32 


36 










.04 


.085 


22 


27 


32 








[ 


.05 


.06 


25 31 


33 










.07 


.09 


25 34 


35 


30° 


30° 


90° 


I 


















.05 


.05 


25 30 


30 








I 


.04 


.05 


29 34 


34 


30° 


90° 


90° 


{ 


.03 


.03 


29 34 


34 








I 


.03 


.03 


31 35 


35 



(e) AIR ACCESS— OXIDIZING EFFECT 

It has been suggested 10 that during the course of the reaction 
atmospheric oxygen, through the carrying action of the manga- 
nous salt, might cause the oxidation of appreciable amounts of 
oxalic acid. To test this point, a preliminary series of experiments 
was made as to the results obtained on the removal of the larger 
part of the air from the titration vessel. To accomplish this, the 
solution to be titrated was made up in a flask with recently boiled 
water and a small amount of sodium carbonate was added just 
before starting the titration. The carbon dioxide evolved carried 
out the bulk of the air, and the flask was kept stoppered and the 
permanganate run in through a small funnel, which passed through 
the stopper. From the results it appeared that no oxidizing 
action of the air was to be feared, but to further test this point 
the following method was employed : 



10 Who first suggested this is not known; Schroder has tested this point recently (Zs. offentliche Chem., 
16, p. 270; 1910). See p. 634 for discussion of his data. 



McBride) Standardization of Potassium Permanganate 



629 



The solution to be titrated was placed in a glass-stoppered flask 
of special form (see Fig. 1), and a strong current of air or carbon 
dioxide, as desired, was bubbled through the liquid before com- 
mencing the titration and during its progress. When carbon 
dioxide was used, the solution was heated to boiling and then 
cooled to the desired temperature in a stream of the gas, thus 
insuring the absence of all but the last traces of air. 



•=^7 




The two series of tests made in this manner are summarized in 
Tables VI and VII. 

TABLE VI 

Effect of Access of Air on Titration 

[All titrations started in 250 cc of 5 per cent by volume sulphuric acid] 



Temperature 



Titration in flask with air 
bubbled through solution 



0. 005041 
40 
42 
42 
42 
41 



Titration in flask with car- 
bon dioxide « bubbled 
through solution 



Titration in beaker 
open to air, as in ordi- 
nary procedure 



0. 005037 
38 
40 
43 

47 
51 



0. 005040 
39 
41 
40 
41 
44 



11 The carbon dioxide used in the experiments of Table VI was found to contain 3 per cent of methane. 
See p. 616. 



630 Bulletin of the Bureau oj Standards [Vol. 8 

TABLE VII 
Effect of Access of Air on Titration 

[All titrations started in 250 cc of 2 per cent by volume sulphuric acid] 



Temperature 



Titration in flask with air 
bubbled through solution 



Titration in flask with car- ! Titration in baaker 



bon dioxide 
through solution 


bubbled 


open to air, as in ordi- 
nary procedure 


0. 005592 




0. 005591 


93 




92 


99 




97 


99 




97 


5600 




98 


01 




5632 



30° 
60° 
90° 



0. 005591 
91 
94 
94 
94 
95 



The data of Tables VI and VII show that no large error comes 
from atmospheric oxidation of the oxalate during titration. The 
largest difference between results obtained using air and using 
carbon dioxide are only i part in 500 (0.2 per cent), and the 
differences were usually less than one-half this amount. More- 
over, if carried out in a beaker, as ordinarily done, the values 
obtained are not appreciably different from those where air is 
wholly excluded (cf . last two columns of Table VII) ; and even 
when air is bubbled through the solution during a titration at 90 
the oxidation is scarcely appreciable. No such chance for atmos- 
pheric action is met with under the ordinary conditions of titra- 
tion, since when following the usual procedure the carbon dioxide 
formed during the reaction prevents much oxygen from the air 
remaining in the solution during the reaction. 

One other point tested was the effect of time elapsing before the 
beginning of the titration. Four samples v/ere made up to 250 cc 
with 5 per cent by volume sulphuric acid and heated to 90 . Two 
were titrated at once, the other two after standing an hour at 90 °. 
The four results agreed within 1 in 5500, showing that even under 
these rather severe conditions oxalic acid is sufficiently stable and 
nonvolatile to prevent loss during a titration from oxidation (in 
the absence of manganous salts, at least) from decomposition by 
the sulphuric acid, or from volatilization with the steam. 



McBride] Standardization of Potassium Permanganate 63 1 

As a further proof that atmospheric oxidation is not an impor- 
tant factor in the oxalate titration, the effect of the rate of titra- 
tion.may be cited. In the data of Table I the titrations covering 
a period up to one hour show almost no tendency to give higher 
values. This is contrary to what might be expected if oxidation 
were taking place, as the oxidation occurring would naturally be, 
at least roughly, proportional to the time elapsed, and such is not 
the case. Experiments 125 and 129 are the only exceptions met 
with. (See Table I.) 

(f) PRESENCE OF ADDED MAKGANOUS SULPHATE 

Since the presence of added manganous sulphate would prob- 
ably influence the initial rate of the reaction, its effect upon the 
result of the titration was tested. A series of tests (all but the 
last two at 30 ) was carried out with the addition of manganous 
sulphate at different points in the course of the reaction. One cc 
of the solution of manganous sulphate added was equivalent to 
the manganous salt formed by the reduction of 10 cc of the per- 
manganate solution. The results of these experiments are given 
in Table VIII. 

It appears from these data that unless the MnS0 4 be added before 
the beginning of the titration it has little effect upon the result 
obtained. However, if even a small amount is added before start- 
ing the reaction at 30 , the value obtained is the same as is ordinarily 
obtained at 90 without the addition of manganous salt. Addition 
before titration at 90 ° has no influence. 



632 



Bulletin of the Bureau of Standards 

TABLE VIII 
Effect of Added Manganous Sulphate 



[Vol.8 



[All but last two tests at 30 , last two at 90 ; all started in 250 cc of 5 per cent by volume sulphuric acid] 


Amount MnSC>4 so- 
lution added 


| 
Time of addition 


Value of KMnO* 
found 


None. 




0. 005576 
77 


1 cc , 


[Before start of titration 


77 

79 

[ . 005582 
1 




1 81 


20 cc 


[After decolorization of 10 cc of KMnO< j 

[just before end point : • 


.005583 


Ice 

20 cc 


83 
. 005578 

77 
. 005578 


1 cc 


80 
. 005579 




81 


20 cc 


J 

i 


. 005578 


None 


I 78 
. 005582 




|l 

1 



III. DISCUSSION AND CONCLUSIONS 

1. REVIEW OF POSSIBLE ERRORS 

As the standardization procedure consists in solution of a 
weighed sample in dilute sulphuric acid and direct titration with 
permanganate, any error occurring must be due either to error in 
weighing of oxalate or measuring of the permanganate or to a 
variation of the reaction itself from the form usually given, viz: 

5H 2 C 2 4 +2KMn0 4 +3H 2 S0 4 =K 2 S0 4 +2MnS0 4 +ioC0 2 +8H 2 

Since, as already indicated (p. 61 7) , there is no necessity for an error 
in either the weighing of the oxalate or the measuring of the per- 
manganate greater than 1 part in 2000, only the irregularities of 
the reaction need be considered at length. 

The variations of the reaction from its normal course may tend 
to cause the use of an excess of either permanganate or oxalate, 



McBride] Standardization of Potassium Permanganate 633 

or both sorts of influences may operate at the same time, in equal 
or unequal degree. There are nine possibilities of such irregularity ; 
the first five of these would cause too small a consumption of per- 
manganate, the next two an excessive use of it, and the last two 
might cause either of these effects. These possible sources of error 
are : (a) Loss of oxalic acid by volatilization from the solution ; (b) 
decomposition of oxalic acid by water ; (c) decomposition of oxalic 
acid by sulphuric acid; (d) oxidation of oxalic acid by contact of 
the solution with the air; (e) presence of oxalic acid unoxidized 
at the end point ; ( / ) liberation of oxygen during the reaction ; 
(g) presence at the end point of unreduced permanganate or other 
compounds of manganese higher than manganous ; (h) presence of 
impurities in the oxalate, either of greater or less reducing power 
power than the oxalate ; (i) formation of other products of oxida- 
tion than carbon dioxide and water. 

(a) LOSS OF OXALIC ACID BY VOLATILIZATION 

The fact that oxalic acid is readily volatile alone suggests the 
danger of loss from its solution by vaporization with steam. 
However, from the dilute solutions employed for titration pur- 
poses, no such losses occur, as has already been shown. The 
results obtained after an hour's standing at 90 in 5 per cent sul- 
phuric acid solution showed (p. 630) that no loss of oxalic acid had 
occurred. Similarly when titration extended over a period of one 
hour (Table I, p. 622) or when carbon dioxide or air was bubbled 
through the solution (Tables VI and VII, pp. 629-630) no such 
higher values of the permanganate were obtained as would have 
resulted had volatilization of the oxalic acid occurred. From these 
results it is apparent that no appreciable vaporization of oxalic 
acid will take place during the course of an ordinary titration 
which lasts, at the most, only 5 to 10 minutes. 

(b), (c) DECOMPOSITION OF OXALIC ACID BY V/ATER OR BY SULPHURIC ACID 

The same data which have shown that no volatilization occurs 
also prove that the water and sulphuric acid had not caused a 
decomposition of the oxalic acid. These possible errors can, 



634 Bulletin of the Bureau of Standards [Vd. 8 

therefore, be considered improbable under any ordinary conditions 
of titration. 

The experiments reported by Caries 12 from which he was led 
to state that at ioo° oxalic acid in sulution is slowly decomposed 
with the formation of carbon dioxide and formic acid are not 
confirmed by our experiments. If such decomposition does take 
place, it is so slow as to have no appreciable influence upon the 
results of a titration made at 90 ° or below. 

(d) OXIDATION OF THE OXALIC ACID BY THE AIR 

Schroeder 13 has shown that in the presence of manganous sul- 
phate oxalic acid solutions are oxidized by the air at an appreciable 
speed. This author presents a large amount of experimental data, 
from which he draws very decided conclusions. Those which are 
of importance in the present discussion are as follows (numbered 
as in original) : 

(3) In the oxidation of oxalic acid by potassium permanganate, an error (too little 
KMn0 4 used) enters through the presence of atmospheric oxygen, which, error is 
greater in the presence of manganous sulphate or more especially titanium dioxide. 

(5) By rapid titration in a stream of oxygen this error of No. 3 disappears, the 
oxygen acting on the oxalic acid to give hydrogen peroxide, which requires a cor- 
responding amount of permanganate. 

(6) In the presence of much manganous sulphate, with oxalic acid of high con- 
centration, especially in the presence of titanium dioxide, this ibydrogen peroxide 
of No. 5 disappears, very rapidly if hot, and the error of No. 3 is then evident, as the 
compensating action of the hydrogen peroxide is not possible. 

(7) Rapid titration at 50 with 30 cc of 1 : 1 sulphuric acid in 200 ec of water gives 
results which agree with the iodine method of standardization. 

Schroeder therefore recommends that the greater part of the 
permanganate be rapidly run into a strongly acidified solution of 
the oxalate and then the titration finished more slowly. Addition 
of manganous sulphate is not recommended. 

The experimental data presented by Schroeder which are of 
importance to this discussion are those in his tables numbered 
3, 4, 5, 6, 10, and n. These data, with very few exceptions, 
seem to warrant the following conclusions (not expressed by 
Schroeder, but drawn from his data) : 

12 Bull. Soc. Chjm., 14, pp. 142-144; 1870. 13 Zs. offentliche Chem., 16, p. 270; 1910. 



McBride] Standardization of Potassium Permanganate 635 

(a) High temperature tends to cause a smaller permanganate 
consumption. (Tables 3, 4, 5, and 6.) 

(b) High, acidity tends to cause a high permanganate consump- 
tion. (Tables 3, 4, 5, 10, and 11.) 

(c) Slow titration (10 minutes) requires less permanganate than 
fast (1 to i}4 minutes) titration. 

(d) Presence of added manganous sulphate (1 to 5 g) tends to 
cause less permanganate to be used, even with rapid titration. 
(Tables 5 and 6.) 

(e) Titration in an air current required slightly more perman- 
ganate than in a current of carbon dioxide, when made either 
rapidly or slowly or in the presence of large or small amounts of 
acid. 

It must be noted that in his experiments the stirring was very 
ineffective, since Schroeder often refers to his titration as "fre- 
quently whirled" (umgeschwenkt) and "whirled from time to 
time"; and no end-point corrections were made even when the 
titration was conducted rapidly, at low temperature, in the pres- 
ence of large amounts of acid and manganous salt. 

These conclusions as to the influence of conditions are confirmed, 
in general, by the data obtained in our own work, but it does not 
seem that the conclusions expressed by Schroeder (p. 634) neces- 
sarily follow from the data which he presents. In fact, from his 
own experimental results, we would infer, as stated by us in our 
conclusion (e) (second preceding paragraph), that the error due 
to atmospheric oxidation claimed by him (see (3) p. 634) does not 
exist. Although it is admitted that on long standing at high 
temperature oxalic acid in solution in the presence of much man- 
ganous sulphate is oxidized by the air, such combination of con- 
ditions is not met in the standardization of permanganate, and 
such oxidation does not take place to any such extent as claimed 
by Schroeder under the conditions which are properly employed. 
On the above basis we believe that Schroeder's conclusion (3) is 
not warranted, even by his own data. 

If, as seems necessary, the explanation that loss of oxygen and 
not atmospheric oxidation is responsible for the variation of the 



636 Bulletin of the Bureau 0} Standards 1 Vol. <? 

permanganate values obtained under different conditions of 
titration, then the conclusions (5) and (6) expressed by Schroeder 
are not necessary for the explanation of the experimental data. 

The fact that the use of the conditions chosen by Schroeder gave 
values which agreed well with those obtained by the iodine method 
of permanganate standardization (see (7) p. 634) would not prove 
that these conditions are correct, since no evidence is presented to 
show that the iodine standardization was carried out under con- 
ditions giving correct values. As a matter of fact, the iodine 
standardization is even more largely affected by conditions than 
is the oxalate method, and none of these influences have been 
systematically studied. 

As has already been stated (p. 630), the experimental data 
obtained by the author do not indicate that any appreciable 
atmospheric oxidation of oxalic acid occurs during the course of 
a standardization. Moreover, Schroeder 's data warrant the con- 
clusion that when titrations are made in a stream of air, no less 
(and perhaps even more) permanganate is required than if made 
in a stream of carbon dioxide. 

Therefore it appears certain that the oxidizing effect of the air 
during the permanganate-oxalate reaction will not have an 
appreciable influence upon the result obtained from such standard- 
ization. 

(c) INCOMPLETE OXIDATION OF THE OXALIC ACID 

In so far as our present knowledge goes there is no evidence to 
indicate any incompleteness in the oxidation of the oxalic acid. 
Since one of the products of the reaction, the carbon dioxide, is 
almost wholly removed from the sphere of action, there is little if 
any tendency to come to an equilibrium before the whole of the 
oxalate is oxidized. The reaction: 

Mn- (or MnOJ + H 2 C 2 4 =Mn'" +H 2 + C0 2 

can not be reversible, under the conditions of titration, and 
unless it be reversible unoxidized oxalic acid would hardly remain 



McBride] Standardization of Potassium Permanganate 637 

in the presence of an excess of permanganate. Moreover, if no 
free oxalic acid remains the complex Mn(OH) 3 .H 2 C 2 4 could not 
be stable, since it would be completely decomposed according to 
the reaction : 

Mn(OH) 3 .xH 2 C 2 4 = Mn(OH) 3 +xH 2 C 2 4 

Since the H 2 C 2 4 formed by this last reaction is destroyed by the 
KMn0 4 excess as fast as formed, the reaction would continue to 
the complete decomposition of the complex. The fact that the 
solutions at the end of a titration contain an excess of perman- 
ganate and do not lose oxidizing power, even after standing for 
half an hour or more at 90 °, indicates that the oxidation of the 
oxalate must have been complete during the titration. 

(f) LOSS OF OXYGEN 

In a recent article Sarkar and Dutta 14 claim that unlimited 
amounts of permanganate can be reduced by small amounts of 
organic material at high temperatures (85 to 90 °) and high acidity 
by the following cycle of reactions : 

(a) 5H 2 C 2 4 + 2KMn0 4 + 3H 2 S0 4 = ioC0 2 + K 2 S0 4 + 2MnS0 4 + 8H 2 

(b) 3 MnS0 4 + 2KMn0 4 + 2 H 2 = 5lvin0 3 + K 2 S0 4 + 2 H 2 S0 4 

(c) 2Mn0 2 + 2H 2 S0 4 = 2MnS0 4 + 2H 2 + 2 

(d) 3MnS0 4 + 2KMn0 4 + 2H 2 = 5Mn0 2 + K 2 S0 4 + 2H 2 SC>4 

Similarly other authors have claimed that oxygen is evolved from 
the interaction of sulphuric and permanganic acids. 15 

In reply to Sarkar and Dutta's claim Skrabal 16 has proposed 
the following scheme as explaining the evolution of oxygen when 

14 Zs. anorg. Chem., 67, pp. 225-233; 1910. 

15 Hirtz and Meyer: Ber., 29, pp. 2828-2830; 1896. Gooch and Banner: Am. J. ScL, [3] 44, pp. 301-310; 
1892. Jones: J. Chem. Soc, 33, p. 95; 1878. 

16 Zs. anorg. Chem., 68, pp. 48-51; 1910. 



638 Bulletin of the Bureau of Standards [Voi.8 

permanganate and oxalic acid react. (Compare with this author's 
vScheme for the normal reaction given on p. 615) : 



I. Incubation period'. 




Mn VII ->Mn n + 2 


(1) 


II. Induction period: 




f->Mn n +Mn IV 


(2) 


Mn VII +Mn n ->Mn m or 




|->Mn n + 2 


(3) 


III. End period: 




Mn IV ->Mn n + 2 


(4) 



Whether one wishes to accept this last scheme or that of Sarkar 
and Dutta, or even some entirely different explanation, the fact 
remains that oxygen may be evolved during the course of these 
reactions when they take place under certain conditions. Such 
oxygen loss would be expected under either of the following condi- 
tions, viz, an increased concentration of acid or an increased con- 
centration of HMn0 4 or Mn0 2 (and possibly also of Mn 2 3 ) in the 
solution. The first of these conditions is produced by direct 
increase in amount of acid added, and the result has been shown to 
be an increased permanganate consumption. (See Tables III and 
IV.) The presence of HMn0 4 , Mn0 2 , and perhaps Mn 2 3 , one or 
all, is caused by any one of the following influences : Low tempera- 
ture, which decreases the rate of their reduction to the manga - 
nous condition. Large bulk of solution, which reduces the oxalate 
concentration, and therefore the rate of their reduction to manga- 
nous condition. Rapid addition of permanganate (especially at 
the start or just before the end point) or insufficient stirring, both 
of which allow the unreduced or partially reduced manganese to 
accumulate throughout or in parts of the solution. 

Each of these conditions has been shown to cause an increase 
in the permanganate used for the titration. On the other hand, 
addition of manganous sulphate, which increases the rapidity of 
the reduction of the manganese, causes a decreased consumption 
of permanganate. (See Table VIII.) 

To briefly summarize these points: The loss of small amounts 
of oxygen, which certainly does occur under some conditions, 



McBride) Standardization of Potassium Permanganate 639 

would account for the effect noted in all of the experiments 
described where change of temperature, of acidity, of volume, of 
rate of addition of the permanganate, of rate of stirring, or the 
addition of manganous sulphate has an appreciable influence upon 
the result of the standardization. Since with proper precautions 
these variations can be reduced, indeed almost eliminated, over 
considerable ranges of temperature, acidity, and volume of solution, 
there is good reason to believe that the loss of oxygen is almost, 
if not wholly, prevented. In this connection, it must again be 
emphasized that vigorous stirring throughout the titration, and 
slow addition of the permanganate at the beginning and at the 
end, are essential if correct results are to be had. 

(g) INCOMPLETE REDUCTION OF THE PERMANGANATE 

It has been shown in the discussion of the end-point corrections 
(pp. 620-621) that in addition to permanganate there is often 
manganic (Mn'") or tetravalent manganese (Mn*"*) salt present 
at the end point. However, the error caused by their presence is 
corrected by the end point "blank" described above, and there- 
fore need not be considered as influencing the results already 
given. In the ordinary titration, such error would be serious only 
if improper conditions of titration were chosen. The conditions 
recommended below entirely eliminate this source of error. 

(h) IMPURITIES IN THE OXALATE 

The question of purity of the soidum oxalate, water, and acid, 
used in the titration, must, of course, be considered; but there is 
no need for appreciable error to arise due to this source if the sim- 
ple tests referred to above under "reagents used" are applied. 
As especially affecting the present discussion this source of error 
has been eliminated even beyond the accuracy of the tests for 
purity of the oxalate, since only comparative values were employed. 

(i) ABNORMAL OXIDATION PRODUCTS 

Several abnormal products of the oxidation can be imagined, for 
example , carbon monoxide, hydrogen, hydrogen peroxide, and 
oxides of carbon higher than carbon dioxide. It is not at all 

73764 — 13 3 



640 Bulletin of the Bureau of Standards [Vol. 8 

probable that either hydrogen or carbon monoxide should result 
from such a reaction as that under consideration, since the former 
would be liberated in a nascent state very easily oxidized and the 
latter could be formed only by a decomposition of the oxalate 
molecule in a way never observed except under the action of a 
strong dehydrating agent, e. g., concentrated sulphuric acid. 

The possibility of the formation of hydrogen peroxide or higher 
carbon oxides has been discussed by Schroeder. 17 This could 
lead to error under only the first of the following two conditions, 
viz, that they subsequently decomposed, setting free oxygen, or 
that they remained intact at the end point. The former condition 
has already been discussed (p. 637) ; the latter possibility, that they 
remain unacted upon to the end of the titration, is both improb- 
able and immaterial, since, when the end-point correction is 
employed, the excess of oxygen which they might contain would 
undoubtedly be corrected for by the "blank" as determined with 
potassium iodide. 

2. SUMMARY AND CONCLUSIONS 

As is indicated in the discussion immediately preceding, there 
appear to be only two sources of variation from the normal course 
of the reaction which are at all probable, viz, loss of oxygen from 
the solution or oxidation of part of the oxalic acid by atmospheric 
action. Although either of these theories explains a large part of 
the experimental facts, only the former can be held in the light of 
the experiments of Table VII and the other facts discussed in con- 
nection with this table. 

Therefore, if the main source of error is due to oxygen losses, the 
higher values obtained in the various series are more nearly correct. 
This conclusion has been accepted as a working basis for the rec- 
ommendations regarding choice of conditions for titration. Indeed 
such choice can not be far in error, since the greatest discrepancies 
noted under wide ranges of conditions is not over 0.2 per cent (not 
including errors due to long standing, inefficient stirring, and 
excessively rapid permanganate addition just at the end point). 

17 Zs. offentlicheChem., 16, p. 27c: 1910. 



McBride) Standardization of Potassium Permanganate 641 

3. METHOD RECOMMENDED FOR USE 

Although it can be seen that under quite varied conditions of 
standardization, the values obtained for the permanganate will 
vary less than 1 part in 1000, it seems desirable to fix more defi- 
nitely the conditions of titration, both in order to make the results 
obtained slightly more precise and also to insure the use of those 
conditions under which the difficulty of proper operation is a 
minimum. For this purpose the following detailed method of 
operation is recommended : 

In a 400-cc beaker, dissolve 0.25-0.3 g of sodium oxalate in 200 
to 250 cc of hot water (80 to 90 ) and add 10 cc of (1 : 1) sulphuric 
acid. Titrate at once with N/ 10 KMn0 4 solution, stirring the liquid 
vigorously and continuously. The permanganate must not be added 
more rapidly than 10 to 15 cc per minute and the last yito 1 cc 
must be added dropwise with particular care to allow each drop to 
be fully decolorized before the next is introduced. The excess of 
permanganate used to cause an end-point color must be estimated 
by matching the color in another beaker containing the same bulk 
of acid and hot water. The solution should not be below 6o° by 
the time the end point is reached ; more rapid cooling may be pre- 
vented by allowing the beaker to stand on a small asbestos-covered 
hot plate during the titration. The use of a small thermometer as 
stirring rod is most convenient in these titrations, as the variation 
of temperature is then easily observed. 

4. ACCURACY AND PRECISION ATTAINABLE 

The agreement of duplicates in the numerous tables given above 
shows the precision which can be expected in the use of sodium 
oxalate when the conditions or titration are regulated. It appears 
that agreement of duplicates to 1 part in 2000 should be regularly 
obtained if weight burettes are used. Failure to obtain this pre- 
cision should be at once taken to mean that some condition has not 
been regulated with sufficient care. 

Although the absolute accuracy of the values obtained under 
different conditions has not yet been studied, it seems probable 



642 Bulletin of the Bureau of Standards [Vols 

that if the method recommended is followed the error will not 
exceed 0.1 per cent and probably will be less than 0.05 per cent. 

It is hoped to undertake a comparison of this method with other 
oxidimetric standards, particularly iron, silver, and iodine, in 
order to check its value. However, such work involves one uncer- 
tainty which renders the best results which could be obtained 
probably no more conclusive than those now at hand. The diffi- 
culty is that of end points, for in each case mentioned the uncer- 
tainty in the blank correction required for an end point is about of 
the same order as that of the uncertainty in the oxalate values, 
namely, 0.1 per cent (usually equivalent to 0.05 cc of N/10 KMnOJ . 
We therefore, at the present time, would assume no greater abso- 
lute accuracy for the values obtainable than one-tenth of 1 per cent. 

Washington , June 1 , 1 9 1 2 .