STUDIES OF THE CHEMICAL

DETERMINATIONS OF VITAMIN 81

Thai: for the Degree of M. S.
MlCHiGAN STATE COLLEGE

Lorraine A. Rauls

I940

 

 

 

STUDIES OF THE
CHEMICAL DETERMINATIONS

or VITLMIN Bl
by

Lorraine-A. Reuls

A THESIS
Submitted to the Graduate School of Michigan
State College of Agriculture and Applied
Science in partial fulfilment of the
requirements for the degree of

MASTER OF SCIEECE

Department of Chemistry

1940

Acknowledgement

Grateful acknowledgement is hereby
given to Dr. C. A. Hoppert whose friendly
encouragement and advice have made this

work interesting and possible.

Table 9: Contents_

Page
Introduction _____________ 1
History _______________ 2
Object ———————————————— 6
Principle —————————————— 6
Experimental Procedure ———————— 7
Preparation of Zeolite ------- 7
Adsorption Tube (diagram) ————— 8
Hydrolysis of Sample -------- ‘8
Adsorption and Elution —————— 9
Determination of Thiamine ————— 10
Tables — Data ____________ 14
Graph, Standard Curve
Discussion ______________ 19

Conclusions ————————————— 23

_ 4 A .
Inuryouction

Thiamin KYQTOCfllOFlue (Vitnnin Bl) is a WmiLB crvs,alline powoer
having an emoirical for ula oi C_15’ ON oCl? and a melting point of
245VC. its synthesis in 1937 by R. R. hilliams (l) wzs the culmination
of a half century of effort to identify and isolate the factor responsi~
ole for beriberi, a diserse evidenced in its acute stages by a genera-
lized edema, or peripheral neuritis or enltrrement of the heart. As
early as 1884 Takcki, a Japanese doctor, recornizod that beriberi was
amenable to dietary treatment but it mas not until l900 to lQlU that the
'n of this oisouae beczne acceitec to any great e tent. In
l9ll bunk (2) succeeded in extracting from rice coiisninfs a crystalline
material which cured polyneuritis in pigeous and three yenrs later pub—
lished his book ”Die Vittnine" wtich first gave that name to the acces—
sory food factors. This publication provided a distinct impetus to
investigations in t1e dietary field and many methods for isolation of
the polyneuritic factor (Vitamin Bl) were propcsed and tried. fiObt
notable among these was the use of Pullers earth as an adsorbent by
Seidell (3) and quinine sulfate elution by Killiazxxs (A). Ethan-{nile
Jansen and Donath in 19:6 (5) and Ohdake in i932 (h) succeeded in iso—
lating enough crystalline material to prooose a tentative formula for
this factor. In leo niiliams proooseo a revised formula which accur—
ately denoted the molecular structure of the crystalline substance and
one year later announced tne synthesis of a compound, tniamin hydro—

Chloride, having tne sane biological activity and onenic;l pro erties

as the natural occurring polyneuritic rector.

iii-£29.21

Until 1934 all assays of vitamin Bl potency were made with
pigeons or rats. However, in that year Kinnersly and Peters (7) pub-
lished the first chemical test for this vitamin. They had found that
diazotized sulfanilic acid in the presence of a buffer and formalde-
hyde formed a pink colored compound with vitamin Bl‘ This test was
applicable only qualitatively in fractionation experiments.

In the same year W. H. Schopfer (8) proposed a quantitative test
using mold growth. As early as 1932 he believed that some "vitamin"
influenced bacterial growth on media containing maltose. He later iden-
tified one of the factors as vitamin B1 and then developed the method.

It consisted essentially of growing the mold Phycomyces Blakesleeanus

 

on a synthetic media to which had been added varying amounts of the
vitamin. The weight of the mycelium developed varied directly with the
amount of vitamin Bl in the substrate. This method was applicable only
within narrow limits, being dependent upon the possibility of growth by
the mold. It was Schopfer's theory that the organism used vitamin B1
to synthesize its own growth factor.

The following year Peters (9) found that the purest vitamin Bl
preparations available could be converted by oxidation in aqueous solu—
tion into substances showing an intense skyublue fluorescence. Berger,
Bergel and Todd (10) then found that this fluorescence was due to a
pale yellow sulfur containing compound (m.p. 2210) having all the re—
corded properties of "Thiochrome" (ClZHlAONAS) which was first found by
Kuhn (11) and his colleagues. The fluorescence of this compound was

also used as a qualitative test for vitamin Bl'

In 1936 the first quantitative chemical methods for the esti-
mation of vitamin B1 were published. One of these was a gravimetric
procedure proposed by Naiman (12). He found that bismuth iodide in
potassium iodide formed an orange-red precipitate with thiamin
chloride which could be filtered, dried and weighed. The weight of
the precipitate was found to be proportional to the amount of vitamin
Bl present.

Far more important than this, however, were several modifications
of previously developed qualitative methods which made possible their
application to quantitative determinations. The less important of
these were colorimetric procedures consisting of two modifications of
the Kinnersly Peters color producing reaction (13, 14) and the use of
a new color producing reagent described in a preliminary report by
Prebluda and McCollum (15). The other line of attack was the quantita—
tive measurement of fluorescence produced after the oxidation of thiamin
chloride to thiochrome. Jansen (16) developed the latter method which
has up to the present time found the widest application of any method.
He found that the thiochrome produced upon the oxidation of thiamin
chloride with potassium ferricyanide could be extracted by isobutanol,
in which the fluorescence of thiochrome is much brighter than in water
solution. By visual comparison of this fluorescence with that of a
series of standards it was possible to determine the amount of vitamin
Bl originally present.

It should be remembered that up until this time vitamin B1 was

obtained solely from natural sources. Its concentration and purification

were accomplished almost entirely through adsorption on Fullers earth,
Lloyds reagent or acid clays followed by elution with alkalis, aqueous
alcoholic hydrochloric acid, quinine sulfate, pyridine or the acid salts
of the nitrogen bases. All of these methods were slow and recovery was
not quantitative. For that reason, the discovery by Cerecedo and
Hennessy (17) that thiamin could be quantitatively adsorbed on and
eluted from synthetic zeolites provided a definite step toward more
accurate analysis. It made possible the separation of thiamin chloride
from some impurities which might interfere with any of the quantitative
determinations previously proposed. Whitehorn (18) in 1923 had shown
that organic bases of moderate strength are removed from solution by
passing through a bed of permutit, one of the synthetic zeolites. Fur-
thermore he showed that the organic base could subsequently be removed
from the bed by a salt solution such as potassium chloride. The appli-
cation of this knowledge by Cerecedo and Hennessy to the problem of
purificatinn of vitamin Bl was of importance because permutit adsorption
has several distinct advantages over that of Fullers earth or acid cla's.
One of the most important is the use of neutral salt solutions for re»
covery of the vitamin rather than an alkali in which thiamin chloride

is relatively unstable. Further advantages are the known composition

of the adsorbing agent, the controlled size of the particles, the possi-
bility of modifying the composition of the zeolite and the possibility
of regeneration of the adsorbent. Through the use of this agent
Cerecedo and co-workers were able to prepare crystalline vitamin B1

from yeast (17) rice (l9) and wheat bran (20).

Another quantitative method which is gradually gaining wider
acceptance was also proposed in 1937. Schultz, Atkin, and Frey (21)
observed that crystalline vitamin B1 exerted a pronounced influence
on the rate of alcoholic fermentation. By using carefully controlled
conditions it is possible to detect as litthe as one microgram of
vitamin B1 through the measurement of the volume of gas produced by
yeast in a certain period of time.

Since 1937, three colorimetric methods have been proposed. A
Japanese worker (22) converted thiamin chloride to thiochrome with
potassium ferricyanide. The resultant ferrocyanide was then converted
to prussian blue. This method involves a notable weakness in its lack
of specificity for any reducing substance might change ferricyanide to
ferrocyanide. Villela and Leal (23) found that treating thiamin chloride
with ammonium molybdate in 3 N sulfuric acid and aminonaphthosulfonic
acid resulted in an intense blue color. Since this is the same proce-
dure used in the Fiske and Subharow determination of inorganic phos—
phorus, it is necessary to run two determinations, one on the original
sample and one after the destruction of organic matter (including thiamin)
with a sulfuric acid — nitric acid mixture. The difference between the
two determinations would then represent the color due to thiamin. It
appears that any organic phosphorous present would make this method in—
accurate. A third colorimetric method was one advanced by Melnick and
Field (24) using the reagent introduced by Prebluda and McCollum (25).

It is this method and the thiochrome method of Cerecedo and Hennessy

(26) which have been investigated for this thesis.

Object

The object of these experiments was to study the effectiveness
and applicability of the chemical methods for determining thiamin
chloride. Two of these, the thiochrome method and a colorimetric
method were studied because they have merited the widest application

and attention.

Principle

Thiamin pyrophosphate (cocarboxylase) a naturally occurring
form of thiamin is converted to free thiamin by enzymatic hydrolysis.
The thiamin solution is then purified by adsorption on and elution
from a zeolite. This is followed by the determination of the total
thiamin (free + phosphoralated) by one of two methods, thiochrome or

colorimetric.

Experimental Procedure

 

Preparation of the Zeolite

 

Permutit was originally used in this investigation but was dis-
carded when it became apparent that it would not give satisfactory re-
sults. Decalso (30—40 mesh), a sodium aluminum silicate produced by
the Permutit Company of New York was finally adopted. A preliminary
treatment of this compound is necessary to remove any excess alkali
and to replace the sodium with potassium. Two methods have been pro—
posed to accomplish these ends.

Aethod A - The zeolite is placed in the adsorption tube and
fifteen ml of hot 2 percent acetic acid are passed through it. This
is followed by fifteen ml of hot 25 percent potassium chloride solution.
This alternate washing is repeated four times after which one hundred
mls of boiling water are passed through the zeolite bed. It is then
ready for use.

Methgdgfi - Decalso is suspended in water and while being
thoroughly stirred is adjusted to a pH of 4.5 with concentrated sulfuric
acid. This pH is maintained by the further addition of sulfuric acid
for fifteen minutes. The suspension is allowed to settle and the liquid
is decanted. After repeating the acid wash three times, the zeolite
is washed five times with water, three times with 95 percent ethyl al-
cohol, once in acetone and three times in ethyl ether. It is dried in
a flat pan at 37°C for three days. The dried material is placed in the
adsorption tube and washed with 30 mls of 25 percent potassium chloride

solution followed by 500 mls of hot distilled water. The Decalso bed

is then ready for use.

Adsorption Tube

 

 

 

 

\dad4adflk
03‘“!
¢n0f63+ ‘0“,
urban
r ~(—— 541.3"
- HJsorf+';' bed
1‘.
Vaumn
Puflf

_flydrolysis of Sample

A sample which will contain approximately 150 gamma of total
thiamin is refluxed for fifteen minutes with twenty volumes of 2 per-
cent acetic acid. After cooling, one tenth volume of l N NaOH and one
tenth volume of freshly prepared 6 percent Takadiastase (Parke-Davis)
are added. The mixture is incubated at 3800 for one and one half hours,
brought to a boiling point, cooled and filtered. The total filtrate
or an aliquot thereof is then ready for passage through the adsorption

tube.

Adsorption and Elution

 

A volume of thiamin chloride solution containing not more than
200 gamma of the vitamin is introduced into the adsorption tube at room
temperature. If the concentration is low, the solution is permitted to
run through the Decalso bed at a fairly rapid rate (10 ml/minute). If
the concentration is higher, the rate of flow should be such that the
solution is passing through the bed for a minimum of 7 minutes. When
this is completed, the adsorption tube is heated by steam and the Decalso
bed washed with 30 ml of distilled water. This wash serves two purposes.
It removes any impurities which might have been left in the tube and
also helps heat the Decalso bed preparatory to elution. Immediately after
the last of the wash water has been sucked out, the stopcock is closed
and while steam is still passing through the outer jacket, 25 percent
potassium chloride solution adjusted to pH 2.0 with sulfuric acid is
added to the tube. This is then permitted to pass through the Decalso
bed at such a rate that 20 mls of eluate are collected in 6 to 9 minutes.
The potassium chloride solution is added to the tube in 10, 5, 3, and 2
m1 portians to facilitate complete elution. The eluate is cooled and is
then ready for a quantitative analysis of the thiamin chloride content by
either the fluorometric (thiochrome) or colorimetric procedure. The
Decalso bed is washed with 500 m1 of hot distilled water using the con-
stant level siphon arrangement (see diagram) to prepare it for the next

sample.

b,termination of thiaain

 

luorimetric. A voi.1e of potSSSIIm cklorice elm lzte contain-

a. 1
ing not more than 20 gamma oi thiamin chlorine is diluted to 5 mls. 3
ml of 15 percent sodium ’"ULOVlGV are added follo'ed 3m,d1at y by

0.05 to 0.2 ml 01' 1 percer nt potassium ierrocv"r3fl . The thiochrome is
extrac ted at once with 20 mls of 'sobutanol by shafing one minute, cen—
trifuging and separating into the two layers. The isobutanol is treated
with anhydrous sodium sulfate until it is perfectly clear. The fluores-
cence of the isobutanol solution is compared either visually or in a
fluorometer with st: ndarEs of known thiamin chloride solution wuicn have
been similarly treated. For visual comparison a General Electric Vapor

Lamp producing black light was used.

B. Colorimetric. The color—producing reagent, dissatized p-amino—
acetophenone, was prepared according to the directions of ”rebluda and
McCollum (25) as follows:

Solution A — 3.18 gm of p—ixiniacec phenone (Eastman noaax com—
pany, No. 631) are dissolved in 45 11 of CQHQPHJPJ+HN nyerecnieric acid
(37 percent) and made up to a final volume of 500 ml in a volumetric
flask with distilled ate er. Precautions are tm.k n to kee3-3 the solution
in a glass-stoppered flask and to avoid the ptes ence oi strong light
when not in use. The activity of this solution is unaltered alter st and—
ing for six months.

Solution B — 22.5 gm of C.B. sodium nitrite are dissOIVed in distilled
water and ma me up to a final volume of 503 ml in a glass stoppered volu-

“1

metric f1: Hi. This rea

V
4

1'4

ant begins to deteriorate after standing for a

month. It is best preserved at reir ige3 ation texnerntures.

11

Solution 0 - 20 gm of C.P. sodium hydroxide are dissolved in
600 ml of distilled water; 28.8 gm of C.P. sodium bicarbonate are then
added and the solution made up to a final volume of one liter with dis-
tilled water.

Diazotization and Final Reagent — The diazotization reaction is
carried out in an ice bath with use of one part by volume each of Solu-
tion A and of Solution B. This reaction requires a period of about 10
minutes. The mixture is agitated by means of an electric stirrer. At
the end of this period 4 parts of Solution B are added to the resultant
mixture. The solution is stirred and maintained at a temperature of
0 - 5°C for at least 20 minutes. When the reaction is completed, the
diazotized solution remains satisfactory for use at least 12 hours at
refrigeration temperatures. In making the final reagent for reaction
with the vitamin, add 20 ml of freshly diazotized p— aminoacetophenone
to a flask containing 275 ml of Solution C. This mixture is stirred
for a period of 5 to 10 minutes after which the reagent is ready to use.

Preparation of Standards. Since a photelometer is used in this

 

work it is necessary to prepare a reference standard curve from pure
thiamin hydrochloride (Merck). A stock solution containing 100 gamma

of thiamin chloride per ml is prepared. This solution, when acidified
to pH 2.0 with sulfuric acid, is stable for at least 2 months at refri-
geration temperatures. In a series of 50 ml centrifuge tubes are placed
0.2, 0.4, 0.6, 0.8 and 1.0 ml of the stock thiamin solution. Each is
diluted to 5 ml with potassium chloride solution (25 percent) which has

passed through the adsorption tube when no thiamin was present. To

each is added 5 ml of an alcohol-phenol solution containing 5 mgs of
phenol per ml and two drops of thymol blue. While this mixture is be—
ing stirred with a fine stream of nitrogen, N sodium hydroxide is added
drop by drop until the first faint blue color appears and this is followed
immediately by the addition of 2 volumes (10 ml) of the diazo reagent.
After thorough mixing, the solution is allowed to stand from 15 to 24
hours at room temperature. The red compound is extracted with 4 ml of
redistilled xylene (3.9. 134° — 1380) by adding the xylene directly to
the tube containing the reaction mixture and agitating by means of a
stream of nitrogen for 1% minutes. The layers are allowed to separate
and the red color in the xylene layer is determined with the photelo—

meter.

 

Standardization of the Photelometer. The photelometer was stan—
dardized using a green filter having maximum transmission between 520 —
580 mu. The standard xylene solutions prepared above contain the color
produced by 5, 10, 15, 20 and 25 gamma of thiamin chloride per ml of
xylene. These solutions were read in the photelometer and the gamma
per ml xylene (ordinate) plotted against percent transmission (abscissa).
A straight line results which was used as a standard curve for the thiamin
diazotate.

Preparation of Sample. An aliquote of potassium chloride eluate
containing 30 to 70 gamma of thiamin was used for the actual.determina-
tion. It was treated the same way as previously described for the stan—

dard solutions. In some cases a heavy precipitate was formed in the

reaction mixture upon standing overnight. When this occurred, the reaction

13'

mixture plus the xylene was treated with 5 ml of 20 percent sulfuric

acid and thoroughly shaken. Ten ml of 15 percent sodium hydroxide were
then added, the mixture again shaken vigorously and allowed to separate.
The color in the xylene layer was then read in the photelometer and the

amount of thiamin present calculated from the standard curve.

TABLE A

Zeolite - Permutit

Preliminary Preparation — Method B

14

 

Exp. : 8Wx>: Vol. of : Adsorp— : Elution : 6' : % Re- : Remarks
No. : tube : soln. tion : : Found :covery :
: : : x : : :

l 100 10 Hot Hot 30 30

2 100 10 Hot Hot 37 37

3 200 20 Hot Hot 56 28

4 250 25 Hot Hot 155 62

5 600 10 Cold Hot 504 84 Color in 2nd 10

ml eluate.

6 600 10 Cold Hot 522 87

7 600 10 Cold Hot 558 93

8 75 10 Cold 35 47

9 100 10 Cold 57 57
10 125 10 Cold Hot 73 58
11 150 6 Cold 60 40
12 175 7 Cold 40 23
13 200 8 Cold Hot 60 3O
14 75 25 Warm Hot 55 73
15 100 25 Varm Hot 67 67
16 125 25 Warm Hot 66 53

 

Preliminary Preparation — Method A

Zeolite

TABLE;§

Decalso

l5

 

 

Exp. : pH of : r to :Vol. of:Adsorp—:E1ution: r' : % Re—:
No. :Adsorp—z tube : soln. : tion : Found :covery:
: tion : z : z : :
17 = 7.0 3 150 : 100 3 Hot 3 Hot 3 92 3 61 3
18 7.0 Filtrate Hot Hot None Adsorption complete
from #1
19 7.0 100 100 Hot Hot 59 59
20 4.5 200 100 Hot Hot 192 96 Color in 2nd 10 ml
of eluate
21 4.6 200 100 Hot Hot 176 88
22 5.2 200 100 Hot Hot 158 79
23 7.5 100 50 Hot Hot 91 91
24 7.0 100 50 Hot Hot 89 89
25 4.2 100 50 Hot Hot 78 78
26 3.8 200 25 Hot Hot 180 90
27 4.0 100 25 Hot Hot 87 87
28 2.6 100 25 Hot Hot 90 9O
29 7.4 100 25 Hot Hot 78 78
30 7.3 200 25 Hot Hot 164 82
31 5.0 100 25 Hot Hot 78 78
32 6.3 100 25 Hot Hot 90 9O '
33 6.9 150 25 Hot Hot 132 88
34 6.2 200 25 Hot Hot 168 84
35 6.4 100 25 Cold Hot 88 88
36 7.0 150 25 Cold Hot 132 88
37 6.2 200 25 Cold Hot 180 90

 

 

16

 

 

 

Zeolite e Decalso
Preliminary Treatment — Method B
3 pH of : J to :Vol.of:4dsorp~: : a” : % Rem:
:AdSOTPrt tube : soln.: tion :Elution: Found :covery:
tion : - g g : :
7.3 100 25 Hot Hot 64 64
7.2 150 25 Hot Hot 102 68
7.2 200 2 Hot Hot 144 72
4.2 200 10 Hot Hot 144 72
4.1 'Filtrate Hot Hot None Adsorption complete
from 41
4.0 200 25 Hot Hot 156 78
4.1 200 25 Hot Hot 144 72
4.0 200 10 Hot Hot 136 68
4.0 Filtrate Hot Hot None
from 45
7.0 200 2 Cold Hot 160 80
7.1 Filtrate Cold Hot None
from 47
7.0 200 2 Cold Hot 172 86
7.0 200 2 Cold Hot 120 60 Elution rate too fast
7.0 200 2 Cold Hot 144 2 K01 not adjusted to pHZJ
7.1 200 2 Cold Hot 154 77
4.0 200 2 Cold Hot 184 2
4.1 200 2 Cold Hot 174 87
4.1 200 2 Cold Hot 178 89
3.9 200 2 Cold Hot 188 94
3.7 200 2 Cold Hot 188 94
3.7 200 2 Cold Hot 176 88
3.9 200 2 Cold Hot 192 96
3.7 200 2 Cold Hot 184 92
3.8 200 2 Cold Hot 184 92
3.8 200 2 Cold Hot 184 92
7.2 200 2 Cold Hot 192 96
3.7 200 2 Cold Hot 184 92
7.0 200 2 Cold Hot 184 92
7.0 150 1.5 Cold Hot 129 86
7.1 100 1.0 Cold Hot 84 84
7.0 200 2 Cold Hot 184 92
7.1 150 1.5 Cold Hot 132 88
7.2 100 1.0 Cold Hot 84 84
7.0 200 2 Cold Hot 192 96
3.7 150 1.5 144 96
3.5 100 1.0 90 9‘0

 

 

 

 

 

 

 

 

 

TABLE D

Photelometer Readings of Standards

 

 

 

Gamma of

thiamin/m1 xylene Low : High Average
5.0 65 66 65.7

10.0 45 48.5 46.5

12.5 37 40.0 38.3

15.0 29.5 33.2 31.3

17.5 26.0 28.0 27.0

20.0 22.0 24.1 23.4

25.0 16.0 18.9 17.4

TLBLE E

Progressive Percent Recovery of Thiamin

 

Experiment :

Tube Number

 

 

Number : 1 x 2 : 3
38 ° 64 ° 68 72
41 72 — 78
44 72 68 -
47 80 - 86
5O 60* 72 77
53 92 87 89
56 94 94 88
59 96 92 92
62 2 96 92
65 92 86 84
68 92 88 84
71 96 96 9O

 

* Elution too fast

17

18

 

 

 

 

 

TABLE F
Effect of Temperature of Adsorption
Experi— : pH of : 5' : Volume : Adsorp— : : J’ : % Re-
ment : Adsorp— : thiamin : of : tion : Elation : thiaminzcovery
number : tion : to tube :solution : : : found :
: z : : x z : :
32 6.3 100 25 Hot Hot 90 90
35 6.4 100 25 Cold* Hot 88 88
33 6.9 150 25 Hot Hot 132 88
36 7.0 150 25 Cold Hot 132 88
34 6.2 200 25 Hot Hot 168 84
37 6.2 200 25 Cold Hot 180 90
x Hot - 90° - 95° C
* Cold— 22° — 25° C
TABLEJQ
Effect of pH of Adsorption
pH of Adsorption : 4.0 : 3.9 : 3.9 : 3.8 : 7.0 : 7.0 7.0
% Recovery : 92 z 94 : 96 : 92 z 92 : 92 96

 

-‘"l "I U Q. A.

CAT. NO. 1 2342

Gamma

NO ILVHLN3ONOO

of Thaamnn

 

per m1

100

80

60 70

40 50

"PHOTELO METER" READING

30

20

10

19

Discussion

 

Adsorption and Elution

 

It is apparent that the success of either the fluorometric or
colorimetric method depends to a great extent upon the efficiency of
the adsorption and elution process which precedes the actual determina-
tion of thiamin chloride. Therefore, when early in this study per-
centage recovery of pure thiamin was consistently low, an attempt was
made to discover how the results might be improved. The factors of
pH, time, temperature and volume of solution were varied one by one
without any consistent success. A study of the data, however, reveals
one seemingly significant fact. It will be noted (Table E) that the
percentage recovery figures for any zeolite bed become progressively
higher as it is used, almost irrespective of the other conditions in-
volved. This would seem to indicate that the preliminary treatment of
the zeolite does not leave it in a condition most suitable for complete
recovery of thiamin. It is my opinion that with either method of pre—
liminary treatment, the excess alkali (sodium) is not completely removed.
If this were true the alkali concentrated at the particle surface might
destroy some of the thiamin adsorbed and thus account for the low re-
covery. It might also account for the progressively higher results from
any single zeolite bed since one might expect the excess alkali to even—
tually be washed away by repeated use of the tube.

With that belief in mind, selection of data from zeolite beds
which were in the same condition makes it possible to Show the influence

of some of the other factors involved. The temperature of adsorption

20

used by other workers varied between room temperature (220 - 250 C) and
steam temperature (900 - 950 C). Both of these extremes were tried. No
evidence of any difference could be found (Table F). Cold adsorption
was finally adopted because of the theoretical considerations involved.

Many attempts were made to find what difference if any the pH of
the zeolite bed and of the thiamin solution might make. Alkaline con—
ditions were excluded because of the instability of thiamin in alkalies.
Table G shows that other conditions being kept the same, adsorption is
equally efficient at pH 4.0 or pH 7.0.

Elution was judged complete when the addition of more potassium
chloride to the adsorption tube showed no evidence of thiamin. Twenty
m1 of potassium chloride was found to completely elute up to 250 gamma

of thiamin in a period of 5 minutes.

Fluorometric Method.
Thiamin or its hydrochloride upon treatment with potassium ferro-
cyanide in the presence of an alkali yields thiochrome, a compound which

gives a sky - blue fluorescence in ultra violet light.

NzC-NHZ H

 

I I H c-s N:C-N=C-S
HBC-C c-c-N' I K3Fe(CN)° I / H I I
nu H §\C=C-CZH40H ; HBO—C C-C—N C-C2H4OH
N-CH cu NaOH II II H ‘c‘
50143 N-C (5H3
H

Thiamin Thiochrome

The fluorescence of the thiochrome is best read in a fluorometer, an
instrument using ultra violet light at 3685 2 which is changed by thio—
chrome to the visible range between 4300 3 and 5000 3. The fluorescent
light is measured at right angles to the incident light by means of a
photocell, galvanometer arrangement. Such an instrument was not avail-
able so an attempt was made to compare the sample with standards vis—
ually. It was found that isobutanol will not quantitatively extract
one gamma of thiochrome per ml. The fluorescence from that amount of
thiochrome is too slight to read accurately by eye so the method was

abandoned in favor of the colorimetric method.

Colorimetric method.
Thiamin hydrochloride reacts with the diazo compound of p—amino—
acetophenone to form a red, stable, water-insoluble, xylene soluble

compound having a maximum absorption at 520 millimicrons.

C1
I
Mb NEN N=N®H N=N-mm
I I I I
C C C C
\. /’ Q» /’ <6 ‘6
c/ ‘0 H01 C C New C c c’ c
II I -———-) II I ——-—-> II I ————-—) II I
C C NaNO C C C C
‘0’ 2 \c’ \08 \c/
I I I I
C=O 0:0 0:0 0:0
I I I I
CH3 CH3 CH3 CH3
p—aminoaceto- diazonium ' diazo

phenone

N::N ONa NHZHC1
I I CH3
/c\\ N= 0 I
C I I H [c = c — CEHAOH
II I + H30- -—C c- c N I Nazca
\ /c II II H I‘ H‘s NaOH 3 i
c/ N— c (31
I
0:0
I
CH3
CH3
N =0 - N112 |
I I H CzC-CZHOH
H c - c c — c — N’ I 4
3 r‘e 4
II II H (51 C—s
N - c I
N 2:
/
\
C/ \
II I
Red C
\CI
Coufiound I
0:0
I
CH3

The reagent does react with some other compounds to yield colored sub-
stances but none of them are extractable by xylene.

The use of the photelometer makes it unnecessary to prepare a
fresh standard with each determination, but necessitates working with
small samples because of the intensity of the color developed. For that
reason special care must be taken to avoid excess alkali (pH 7.5 to 8.5
required for reaction) when adding sodium hydroxide to the reaction mix-
ture and also to insure complete extraction of the thiamin compound by
thorough mixing with xylene. When these precautions are observed the

method is very satisfactory.

23

Conclusions

 

1. Adsorption and elution take place best on Decalso which has

been given extended acid treatment.

2. The Fluorometric method of analysis is rapid and accurate

but requires a fluorometer for accurate reading of results.

3. The colorimetric method of analysis is complex and requires
more time than the above but is accurate and applicable to research

problems.

Bibliography

 

Cline, J. K., Williams, R. R. and Finkelstein, J.
Synthesis of Vitamin Bl'
Jour. Amer. Chem. Soc. 59, 1052 — 1054 (1937).
Funk, C.
Chemical Nature of Substance Which Cures Polyneuritis in
Birds Induced by Diet of Polished Rice.
J. Physiol. 43, 395 — 400, (1911).
Seidell, A.
Further Progress Towards the Isolation of the Antineuritic
Vitamin (Vitamin B) from Brewers Yeast.
J. Biol. Chem. 82, 633 (1929).
Williams, R. R., Waterman, R. E. and Keresztesy, J. C.
Larger Yields of Crystalline Antineuritic Vitamin.
Jour. Amer. Chem. Soc. 56, 1187 — 1191 (1934).
Jansen, B. C. P. and Donath, W. F.
Isolation of Anti—beriberi Vitamin.
Mededel v.d. dienst. d. volksgezondh in Neder lausch
Indie 16, 186 (1927).
Ohdake, 3.
Isolation of 0 ,zanin, "Antineuritic Vitamin" from Rice Polish.
Bul. Agric. Chem. Soc. Japan 8, 179 (1932).
Kinnersly, H. W. and Peters, R. A.
The Formaldehyde-azo Test for Vitamin 31‘

Biochem. . 28, 667 - 670 (1934).

10.

11.

12.

13.

14.

15.

16.

Schopfer, W. H.
Plant Test for Vitamin Bl.
Z. Vitaminforsch, 4, 67 — 75 (1935).’
Peters, R. A.
Vitamin Bl and Blue Fluorescent Compounds.
Nature 135, 107 (1935).
Barger, G., Bergel, F. and Todd, A. R.
A Crystalline Fluorescent Dehydrogeneration Product from Vitamin Bl.
Nature 136, 259 (1935)-
Kuhn, R.
Z. physiol. Chem. 234, 196 (1935).
Naiman, B.
A Reagent for Vitamin B1‘
Science 85, 290 (1936).
Deviatnin, V. A.
A Chemical Method for Determination of Vitamin Bl.
Compt. rend. acad. sci. U.R.S.S. (N.S.) 4, 67 — 71 (1936).
Panshina - Trufanova
Colorimetric Determination of Vitamin Bl'
. Biokhimiga 1, 597 - 602 (1936).
Prebluda, H. J. and McCollum, E. V.
A Chemical Reagent for the Detection and Estimation of Vitamin Bl.
Science 84, 488 (1936).
Jansen, B. C. P.
Uhemical Determination of Aneurin (Bl) by Thiochrome Method.

Rec. Trav. chim 55, 1046 _ 1052 (1936).

l7. Cerecedo, L. R. and Hennessy, D. J.
The Use of Synthetic Zeolites in the Isolation of Vitamin Bl’
I Experiments with Rice Polishings.
Jour. Amer. Chem. Soc. 59, 1617 - 1619 (1937).
18. Whitehorn, J. C.
"Permutit" as a Reagent for Amines.
J. Biol. Chem. 56, 751 (1923).
19. Cerecedo, L. R. and Kaszuba, F. J.
The Use of Synthetic Zeolites in Isolation of Vitamin Bl.
II Experiments with Brewers Yeast.
Jour. Amer. Chem. Soc. 59, 1619 - 1621 (1937).
20. Cerecedo, L. R. and Thornton, J. J.
The Use of Synthetic Zeolites in Isolation of Vitamin B1.
III Experiments with Wheat Germ.
Jour. Amer. Chem. Soc. 59, 1621 — 1622 (1937).
21. ochultz, R., Atkin, L. and Frey, C. N.
A Fermentation Test for Vitamin Bl.
Jour. Amer. Chem. Soc. 59, 948 — 949 (1937).
22. Tauber, H.
Colorimetric Method for Determining Vitamin Bl'
,Mikrochim. Acta 3, 108 — 109 (1938).
23. Villela, G. G., and Leal, A. 1.
A New Color Reaction of Vitamin B1(Aneurin, Thiamin).

Science 90, 179 (1939).

27

24. Melnick, D. and Field, H.
Chemical Determination of Vitamin B1.
J. Biol. Chem. 127, 505 - 514 (1939).

25. Prebluda, H. J. and McCollum, E. V.
A Chemical Reagent for Thiamine.
J. Biol. Chem. 127, 495 - 503 (1939).

26. Hennessy, D. J. and Cerecedo, L. R.

i The Determination of Free and Phosphorylated Thiamin by a

Modified Thiochrome Assay.

Jour. Amer. Chem. Soc. 61, 179 a 183 (1939).

 

 

 

CHEMISTRY DEPTo

’1612.015 3245
Heals

1"“ r -'\Ifi\, V " 'I‘~.1F"
Chlamibi .LX L)... 1 .

T612.015 L’R245 132665

Rauls

Studies of the chemical deans
minations of vitamin Bl‘

 

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\O\\\\\ 1111111 “111 111

 

33129