Biochemical Society (Great Britain).

The biochemical journal online

. (page 14 of 62)
Online LibraryBiochemical Society (Great Britain)The biochemical journal → online text (page 14 of 62)
Font size
QR-code for this ebook


from 35"* to 95** in 120 ± 2 sees. Owing to variations in the gas pressure
I found it necessary repeatedly to check the heating power. I tried several
methods of obtaining a constant pressure, and finally adopted the manometer
shown in Fig. 1. Once the apparatus is assembled an estimation can be
performed without the delay of having to determine the heating power.

On an adjustable ring stand about 4 cm. above the top of the burner
place an unperforated sheet of asbestos gauze. Turn on the tap B to its full
extent. Tighten the screw A till the pressure is reduced about one-third.
Allow the gauze to get thoroughly heated before making a test. In a 200 cc.
Erlenmeyer flask of Jena glass and of about 6 cm. basal diameter place 60 cc.
of distilled water. The flask is fitted with a 2-hole rubber stopper carrying
a thermometer so graduated that the stem above S*** is visible above the
upper edge of the stopper. The lower end of the thermometer should be
about 2 mm. from the bottom of the flask. By means of a stop watch note
the time for the temperature to rise from 35° to 95°. If the time is less
than 120 sees, tighten the screw A and repeat the experiment until the
desired heating power is obtained. Note the reading of the manometer.
This must be observed fix)m time to time and the screw adjusted to maintain
it at the correct level as the pressure of the gas supply varies. The height
of the ring and the thickness of the asbestos should be such that the pressure



Digitized by



Google



S. W. COLE 137

is well under the minimum supplied to the laboratory and yet sufficient to
prevent any risk of the flame striking back. On making a fresh set of
observations within a week it is only necessaiy fully to open the tap B and
to adjust the screw A till the previously recorded pressure is obtained.
With longer intervals it is desirable to make a fresh observation of the
heating power owing to the possibility of the evapomtion of some of the fluid
in the manometer tube.

n



Fig. 1. Apparatus for maintaining a standard heating power. The manometer tube contains a
dilate solution of eosin. It also contains a globule of mercury which nearly fiUs the bottom
of the tube. This prevents the rapid oscillations of pressure due apparently to the explosions
of local gas engines.



Filtering apparatus.

I have found it most convenient to use the apparatus shown in Fig. 2.

-4 is a Jena flask of 200 cc. capacity. If one is chosen with a perfectly
flat bottom no trouble with fractures under pressure need be feared. The
tube B is an ordinary calcium chloride tube.. The lower end should project for



Digitized by



Google



138



S. W. COLE



at least 3 cm. below the. lower edge of the stopper to prevent loss by splashing
during filtration. The filtering mat is made of glass wool, asbestos, powdered
pumice and asbestos added in that order. After a test the cuprous oxide on
the mat is dissolved in nitric acid diluted with an equal volume of water, and
then thoroughly washed. I have used one tube for more than 100 estimations
without having to add any more asbestos. The thickness of the mat should
be such that the filtrate comes through under pressure in a steady stream.




Fig. 2. Apparatus for the filtration of the reduced copper.



Method of analysis.

Into a 200 cc. Erlenmeyer flask measure 20 cc. of the standardised
copper sulphate solution, 20 cc. of the alkaline tartrate and 20 cc. of the
sugar solution (which must contain between 6 and 250 mg. of anhydrous
lactose). Fit the two-holed rubber stopper firmly into the neck of the flask,
adjust the thermometer so that its lower end is 2 mm. fix^m the bottom of
the flask and place on the heated gauze, note the time when the mercury
indicates a temperature of 95°. Allow the heating to continue for exactly
20 sees, beyond this. Remove the flask by gripping the rubber stopper
and swill it for a second or two under the tap or in a bowl of water. The
lowering of the temperature practically stops the reduction. Filter the hot
fluid at once using the stem of the thermometer as a stirring rod. Wash
the flask twice with about 7 ca of distilled water. Cool the filtrate by



Digitized by



Google



S. W. COLE



139



holding the flask under the tap. Add exactly 4 cc. of strong sulphuric acid,
insert a thermometer and cool to 20°. Add 6'6 to 7 cc. of the saturated
solution of potassium iodide, washing the stem of the thermometer with this
solution. Titrate at once with the standardised solution of sodium thiosul-
phate as described above, using soluble starch as an indicator when near the
end point.

Ths reducing power of anhydrous lactose.

I give in table I the mean values I have obtained of the amounts of
copper reduced by varying amounts of lactose. Duplicates agree to about
0'4 mg. Cu.

The stock solutions of lactose were prepared from the hydrate several
times recrystallised. The values agreed exactly with those determined
polarimetrically at the Cambridge University Chemical Laboratory by the
kindness of Prof. Pope.







TABTiR I.






Anhydroas






Anhydrous






lactose,


Copper,


Lactose


lactose.


Copper,


Lactose.


mg.


mg.


Copper


mg.


mg.


Copper


3-9


3-3


1-182


70


88-2


0-793


5


4-4


1136


75


94


0-798


8


7-4


1-081


80


100-9


0-793


10


9-8


1020


86


107-3


0-792


12


13-3


0-902


90


113-6


0-792


15


16-6


0-903


95


120-1


0-791


20


24-1


0-830


100


127


0-788


25


30-5


0-820


120


152-8


0-786


30


36-9


0-813


126


160-6


0-785


35


43-2


0-810


130


166-1


0-783


40


49*9


0-802


140


179-4


0-780


45


56-1


0-802


150


191-6


0-783


50


62-5


0-800


175


223


0-785


55


69


0-805


200


256-6


0-782


67


71-3


0-800


240


307-3


0-781


60.


75-2


0-798


250


320-7


0-780


65


81-3


0-799









I have plotted these values and find that they are practically linear,
except from 1 to about 20 mg. lactose. I have therefore constructed the
curve shown in Fig. 3, by means of which the weight of anhydrous lactose
present can be obtained from the amount of copper that it has reduced.



Bioch. Ym



10



Digitized by



Google



140 S. W. COLE

Example.

20 cc. of a solution of lactose were treated with 20 cc. of copper sulphate
containing 348*8 nig. Cu.



230 160



220 150



210 140



270 200 130



260 100 120



250 180 110



240 170 100



230 160 90



^ 220 150 80

I

^ 210 140 70



120


130


140


150


160


170


180


100




200


220


230


240


250


260


270


280


200


300


320


330


340


350


360


370


380


390


400



100 110

200 210

300 310

mg. copper

Fig. 3. Carve showing amount of oopper reduced by lactose anhydride.

The copper in the filtrate liberated iodine which required 22*41 cc. of a
thiosulphate, 1 cc. of which = 1 2*784 mg. Cu.
Cu in filtrate is 22*41 x 12*784= 286*4 mg.
Cu reduced is 348*8 - 286*4 = 62*4 mg.



Digitized by



Google



S. W. COLE 141



o o o o o o o

BO d r- O a> CO '^

O O Q O O O O

wO N. <0 lO <^ CO 0<





OO






^ <o






CN OO






OO






CO lO






Ol CO






o o






S5






CO






r- CO






<N CO


i




o o


p




S8


CO




o o







t- CO


1




o o


g




® 9


•m




*- CO


o

1




o o


§




j^ a>






1




o o


'S




CD 00






»- d


1










o




o o


■^




lO r^






T- o.


i




o o






^ ©






T- Ol






o o






CO »o






— d


1




oo


1*




<N -^




'- <N


t2) .


o o




a


CD in






o o






^ o












asooni8


•8ui





10—2



Digitized by



Google



142 S. W. COLE

On reference to the curve in Fig. 3 this is seen to correspond to 50 mg, of
anhydrous lactose.

The copper values above 26 mg. Cu can be converted to anhydrous
lactose by the use of the following empirical formula, deduced from the curve.

mg. anhydrous lactose = 1*25 + 0*778 x mg. Cu.

Thus in the above example

Lactose = 125 + 0*778 x 62*4 = 4971 mg.

This agrees fairly well with the value deduced from the curve, i.e. 50 mg.

In Fig. 4 I give the corresponding values for glucose, taking the values
obtained by Peters, except that I have not given the values above 180 mg.
glucose. Peters states that 200 mg. glucose reduce 349*6 mg. copper. I find
that even 192 mg. glucose completely reduce the whole of the copper present
in the strongest of my solutions (352*9 mg.). In the range of the curve that
I give I have obtained results in very close agreement with his.

It is of some interest to note that in the case of lactose the sugar/copper
ratio decreases as the amount of sugar increases. The ratio lies between
0*81 and 0*78 between 30 and 250 mg. of lactose.

In the case of glucose the ratio falls to 0*522 at 25 mg. and keeps fairly
constant to about 110 mg. It then rises again.

This is not in agreement with the usual text-book statements that " the
greater the excess of the copper, the greater is the amount of copper reduced
by a certain amount of sugar."

The steady fall in the case of lactose seems to indicate that there is
a slight hydrolysis of the sugar by the alkali, so that the amount of copper
reduced by the disaccharide is relatively greater when it is present in higher
concentrations. I have previously had indications of the same phenomenon
when attempting to estimate lactose by the application of Benedict s method.



BEFEBENCES.



Peters (1912, 1), J. Amer, Chetn. Soc. S4, 422.
(1912, 2), J, Amer, Chem, Soc, M, 928.



Digitized by



Google



XX.- A VOLUMETRIC METHOD FOR THE
ESTIMATION OF ETHEREAL AND INOR-
GANIC SULPHATES IN URINE/

By otto ROSENHEIM AND JACK CECIL DRUMMOND.

From the Physiological Laboratory ^ Kings College, London.

{Received March 5th, 19 H,)

Ever since the discovery by Baumann of the existence in urine of
ethereal sulphates in addition to inorganic sulphates, their estimation has
been carried out by weighing them as BaSOi on the principle of Baumann's
original method, as modified in some more or less important details by
subsequent investigators ^

Until quite recently it has been assumed that the estimation of sulphuric
acid as barium sulphate is one of the simplest and most trustworthy
methods of analytical chemistry. The large amount of research which has
been devoted by analytical chemists to this question [Hulett and Duschak,
1904; van 't Kruys, 1910; Allen and Johnston, 1910; Jarvinen, 1913, etc.]
has, however, abundantly proved that this assumption is erroneous. It has
been conclusively shown that even in comparatively simple solutions the
method is liable to yield inaccurate results. Without going into detail it
may be stated that correct results can only be obtained by a careful adjust-
ment of the conditions, and Allen and Johnston, indeed, attribute the many
good results, which are obtainable by this method, to the happy neutralisa-
tion of inaccuracies.

If we consider that all these difficulties are met with when dealing
with simple' salt solutions, we are led to assume that they are still greater
when we are dealing with sulphate estimations in such a complex organic
fluid as urine. This was indeed recognised by O. Folin [1906] when he
said: "The investigations described in this paper are the outcome of a
conviction that the published records of sulphate and sulphur determinations
in urine, including many of my own, are intolerably unreliable."

* The methods proposed by B. v. Lengyel (estimation as strontium salphate) and by Freand
(titration with bariam aoetate) do not seem to have acquired more than academical interest.



Digitized by



Google



144 O. ROSENHEIM AND J. C. DRUMMOND

Folin has overcome by his careful work the difficulties inherent in the
method, but there remains always the fact that, even at its best, the method
is a very tedious one, requiring a great deal of time and experience if accurate
results are desired.

We have worked out a volumetric method for the estimation of ethereal
and inorganic sulphates in urine, which is based on the well-known in-
solubility, of the sulphate of the base benzidine (p-diamino-diphenyl
NHj,CeH4.C6H4.NH,). The method seems to be free from the objections
of the barium method and has the advantage of being very rapid, whilst
the results are at least as trustworthy as those obtained by the Baumann-
Folin method, which we have adopted as a standard.

The principle of the method depends on the fact that insoluble benzidine
sulphate is precipitated from solutions of sulphates on the addition of a
soluble benzidine salt. As benzidine is a weak base, its salts with strong
acids are readily dissociated and the sulphuric acid contained in benzidine
sulphate may be quantitatively titrated with standard alkali solutions, using
phenolphthalein as an indicator. The method was introduced in this form
into inorganic analysis by Baschig [1903] and its reliability was confirmed
by V. Knorre [1905, 1910] and by Friedheim and Nydegger [1907]. Jarvinen
[1913] came to the conclusion that the benzidine method gives the most
trustworthy results and is preferable to the old barium method in inorganic
analysis.

Before the benzidine method can be applied to urine it is necessary to
investigate if it fulfils the essential conditions which are demanded. Any
reagent which is intended to replace barium chloride for the estimation of
ethereal and inorganic sulphates in urine must fulfil three conditions.
(1) It must precipitate in the cold only the inorganic sulphates, free from
ethereal sulphates. (2) It must not precipitate any other urinary con-
stituents. (3) The precipitation of the total sulphates after hydrolysis
(« inorganic + ethereal sulphates) must be complete.

With regard to the first condition we have tested the behaviour of
benzidine solutions towards several ethereal sulphates which we prepared
sjmthetically according to Baumann [1878]. It was found that aqueous
solutions of the potassium salts of phenylsulphuric acid, jp-cresylsulphuric
acid and resorcinolsulphuric acid behave towards benzidine hydrochloride
exactly in the same way as towards barium chloride, i.e. the organically
bound sulphuric acid is not precipitated in the cold, but readily after boiling
for a short time with dilute hydrochloric acid. We have further tested the
behaviour of the ethereal sulphates contained in normal urine towards



Digitized by



Google



0. ROSENHEIM AND J. C. DRUMMOND 146

benzidine solutions in the following way. 200 ec. of urine were freed from
inorganic sulphates by precipitation with bitrium chloride in the usual way.
The filtrate which contained only the ethereal sulphates remained perfectly
clear after the addition of a solution of benzidine hydrochloride. We must
therefore conclude that under the ordinary conditions of urinary analysis the
ethereal sulphates are not precipitated by benzidine.

With regard to the second condition we have found that the addition of
a benzidine solution to normal urine results at once in the production of a
bulky cream coloured precipitate, which on examination is seen to be partially
crystalline and partially amorphous. It can easily be shown that this precipi-
tate contains benzidine phosphate besides benzidine sulphate. This tsLct
alone would seem a priori to condemn the method. (See also K. Spiro
[Neubauer-Huppert, 1910], which is the only reference to the subject we
came across in the literature.) We found, however, that urine which is
faintly acidified with dilute hydrochloric acid, gives a crystalline precipitate
which rapidly settles. This precipitate consists entirely of crystals of benzi-
dine sulphate. It was tested qualitatively with negative results for phosphoric
and uric acids, and the absence of other acids than sulphuric acid was proved
by the close agreement of control analyses in which sulphuric acid determina-
tions were carried out in the same sample by the barium method and by
direct titration. The experiment described above, in which it was shown that
the addition of benzidine solution to urine freed from inorganic sulphates
produces no further precipitate, may also be quoted as evidence that sulphates
are the only urinary constituents which are precipitated from acid urine by
benzidine.

With regard to the completeness of the precipitation of sulphuric acid
by benzidine in the presence of the other constituents of urine, we consider
this proved by experiments in which added sulphuric acid was quantitatively
recovered, and further by the fact that after the removal of the benzidine
precipitate no further precipitate was produced by the addition of barium
chloride.

Benzidine seems therefore to fulfil the conditions laid down above and
we felt justified in proceeding to the actual quantitative estimations. It now
became necessary to carry out a series of experiments in order to investigate
(1) the influence of increasing quantities of hydrochloric acid on the accuracy
of the result, (2) the minimum time necessary for complete precipitation, and
(3) the quantity of benzidine solution required for complete precipitation.

The results of these experiments showed that excess of hydrochloric acid
prevents the precipitation of benzidine sulphate and that a quantity sufficient



Digitized by



Google



146 0. ROSENHEIM AND J, C. DRUMMOND

to produce an acid reaction to Congo red paper gives accurate results.
Further it was found that the precipitation is complete after five minutes
and that one hour's standing does not influence the result. Even a large
excess of benzidine solution does not affect the accuracy of the result.

The ethereal and inorganic sulphates in urine may obviously be estimated
in several ways by means of benzidine. (1) Inorganic and total sulphates
may be estimated directly and the amount of ethereal sulphates be calculated
by the difference between the two. (2) Total and ethereal sulphates may be
estimated directly and inorganic sulphates calculated by difference. (3) In-
organic and ethereal sulphates may be estimated separately.

As the direct estimation of ethereal sulphates, owing to the small amount
present, requires a relatively large volume of urine, which is not always
available, we have estimated them indirectly by the difference between the
total and inorganic sulphates, a procedure which is analogous to that employed
by Folin in the barium chloride method.

In our experience the practical application of the benzidine method to
urine does not meet with any difficulties if the precautions are observed,
which have already been described by Raschig and by Jarvinen in the case
of inorganic analysis. They refer to care in filtration and washing of the
precipitate. It is also important that an excess of free hydrochloric acid be
avoided. Partly for this reason and partly in order to avoid the excessive
formation of dark pigments which might interfere with the end point of
titration, we have found it advisable to reduce the amount of hydrochloric
acid in the hydrolysis of the ethereal sulphates to the minimum necessary for
complete hydrolysis.

It is evident that the benzidine method can also be employed for the
estimation of the neutral sulphur, by subtracting the quantity of total
sulphates from that of the total sulphur. The latter may be estimated by
applying the benzidine method to the final solution of mineral salts, as
obtained by Wolf and Oesterberg's modification [1910] of Benedict s method
[1909]. The presence of copper in this solution does not in any way interfere
with the accuracy of the results.

By working with O'Ol N KOH solutions and methyl-red as an indicator
it seems also possible to use much smaller quantities of urine and to convert
the method into a micro-analytical method. Such a method would be of
use in metabolic experiments on small animals and we are still engaged with
experiments in this direction.



Digitized by



Google



0. ROSENHEIM AND J, C. DRUMMOND 147

Experimental.

The preparation of the benzidine solution, 4 g. benzidine (Eahlbaum) are
rubbed into a iinie paste with about 10 cc. of water and transferred with
about 500 ca of water into a 2 litre flask. 5 cc. of concentrated HCl (spec,
grav. 1*19) are added and the solution made up to 2 litres with distilled
water. 150 cc. of this solution, which keeps indefinitely, are sufficient to
precipitate 0*1 g. H2SO4.

Influence of oddity, time and volume. The following series of experiments
was carried out on a large sample of mixed urine. 25 cc. of urine were
taken in each case. An estimation of inorganic sulphates by the barium
method gave 0-104 Vo SO3.

Influence of increasing quantities of HCl (1 : 4),





Benzidine








HOI,


solution,


Time of


01 N KOH,




cc.


cc.


standing


CO.


SO,Vo


1


100


10 mins.


6-45


0-108


2






6-60


0104


3






6-30


0-101


4






600


0-096


5






5-60


0089


7






4-60


0074


10






2-20


0085


15






none






This table shows clearly that an excess of hydrochloric acid prevents the
formation of the benzidine sulphate precipitate. The quantity of hydro-
chloric acid necessary to produce an acid reaction with Congo red paper
was 0-8 cc. In all the subsequent experiments we used therefore 1-2 cc.
HCl (1:4).

Influence of time of standing.





Benzidine










HCl,


eolation.


Time of


01 N KOH,




cc.


cc.


standing


CO.


80,0/^


2


100


5


mins.


6-40


0102






' 10





6-45


0103






15


,,


6*40


0102






20


M


6-60


0-107






30





6-45


0103






45


U


. 6-46


0108






60


It


6-50


0104



These results prove that precipitation is complete after five minutes and
that the result is not affected by one hour's standing. The short time



Digitized by



Google



148 O, ROSENHEIM AND J. C. DRUMMOND

required for precipitation is a distinct advantage as against the lengthy
period usually required before barium sulphate precipitates can be filtered.

Influence of volume of benzidine soltUion.





Benzidine








HCl,


solntion,


Time of


01 N KOH,




cc.


cc.


standing


CO.


so,%


2


70


5 mins.


6-50


0104


)f


100


>♦


6-40


0102


»i


160


,,


6-60


0104


»»


180


»»


6-50


0104





200


»»


6*50


0104



The final result is evidently not influenced by an excessive quantity
of the reagent and 100 cc. of the standard benzidine solution are sufficient
to precipitate the whole of the sulphates contained in 25 cc. of normal urine.

An experiment may be mentioned here in which a known quantity of
sulphate was purposely added to the urine. 2 cc. of a dilute sulphuric acid
required 10*9 cc. 01 N KOH for neutralisation. This solution of sulphate,
containing 53'4 mg. HaSOi, was added to 25 cc. urine which was previously
found to contain 31 '9 mg. H^04. In two experiments the final benzi-
dine precipitates required 174 and 17*5 cc. O'l N KOH = 85*3 mg. and
85-4mg. H^04.

H2SO4 added H,S04 foand

(1) 53-4 mg. 53-4 mg.

(2) 53-4 mg. 53-0 mg.

This experiment not only proves that the organic constituents of the
urine do not interfere with the precipitation of sulphates by benzidine, but
also that the precipitation is quantitative.

As an outcome of these preliminary experiments we have finally adopted
the following procedure :

1. The estimation of inorganic sulphates.

25 cc. of urine are measured into a 250 cc. Erlenmeyer flask and acidified
with dilute hydrochloric acid (1 : 4) until the reaction is distinctly acid to
Congo red paper. Usually 1-2 cc. of dilute acid were added. 100 cc. of the
benzidine solution (see above) are then run in and the precipitate, which
foi-ms in a few seconds, allowed to settle for ten minutes. The precipitate
is filtered under pressure and washed with 10-20 cc. of water, saturated with
benzidine sulphate. The precipitate and filter paper are transferred into
the original precipitation fltisk with about 50 cc. water and titrated hot with



Digitized by



Google



(). ROSENHEIM AND J. C. DRUMMOND 149

0-1 N KOH, after the addition of a few drops of a saturated alcoholic solution
of phenolphthalein. 1 cc. 0*1 N KOH corresponds to 49 mg, H3SO4.



2. The estimation of toted sulphates {inorganic and ethereal).

The hydrolysis of the ethereal sulphates may be carried out in the usual



Online LibraryBiochemical Society (Great Britain)The biochemical journal → online text (page 14 of 62)