Arnold Bennett.

Chemical news and journal of industrial science online

. (page 59 of 88)
Online LibraryArnold BennettChemical news and journal of industrial science → online text (page 59 of 88)
Font size
QR-code for this ebook

minute : —

After a discussion in the Sedion on the continuance of
a committee to confer with others with a view to founding
a National Chemical Society, Dr. Albert B. Prescott sug«
gested as a proper funAion of such a society the publica-
tion of an index to periodical chemical literature. He
deprecated the multiplication of periodicals, and main-
tained the utility and importance of an index to the oon«
tents of those already in existence.

RespedfuUy submitted,

H. Carrington Bolton, Chairman.

F. W. Clarkb.

Albert R. Lbbds.

Alexis A. Julxbn.

John W. Lanqley (in Europe).

Albert B. Prescott.


By W. E. STONB.^

(Concladed from p. 207).

B.^Quantitativi Reduction o/Fehling's Solution
by the Pentaglucoses,
Both arabinose and xylose reduce Fehling's solution
strongly, but their quantitative relation to it has not
previously been studied, with the exception of a brief
mention by Bauer of some limited determinations for
arabinose, from which he concluded that it possessed in a
slight degree greater reducing power than dextrose
{Landwirthschaftliche Versuch'Stationen, xxxvi., 304).

In view of the common occurrence of the pentaglucoses,
as already pointed out, it is evident that in the quantiu-
tive examination of any material for carbohydrates by
hydrolytic methods, i .«., by inversion with acids and titra*
tion of the glucose thus formed with Fehling's solution,
we may be dealing not only with the true glucoses derived
from starch, cane sugar, &c., but also with more or less of
one or both of the pentaglucoses, since we know that the
gums from which the latter are derived easily undergo
hydrolytic change by the adion of acids. Such
analytical results may, therefore, be erroneous from two
causes : first, from the assumption that the reduAion of
Fehling's solution is due solely to the presence of the true
glucoses; or, second, from the assumption that the penta-
glucoses, when present, reduce Fehling's solution in the
same degree as the true glucoses.

Take, for example, such a material as brewers* grains,

* Cofltribationi from the Chemical Laboratory of Pardae UoiTtr-
sity. From tbs Amermn ChemvMl JouHi4t, Vol. aiii., lib. £

Digitized by


Concerning thi Ptniaglucosei^


in which it might be desired to ascertain the thoroughness
of malting and fermentation by determining the residaal
carbohydrate. Such material might contain no trace of
starch or sugar, and yet might nve abundant indications
of the presence of these substances after inversion,
because it contains so large an amount of the pentaglu-
cose bodies that it is possible to isolate from it 2 percent,
or more, of the crystallised pentaglucotes {Liebig^s Ann.
Chtm,, ccxlix., 241).

It seemed important, therefore, to know the relations
existing between these bodies and the true glucoses, and
I therefore determined the comparative values of arabin-
ose and dextrose in dilute solutions toward Fehling's

X. Arabinou, — The material used was prepared by me
from cherry gum, and was pure, as was shown by its
specific rotation ( [a] o « 104*1"). From this material, solu-
tions of I, }. }, and ^ per cent strength were made. The
Febling*s solution was prepared according to the standard
formula, viz., 34*639 grms. of hydrous copper sulphate
dissolved to a volume of 500 c.c, and 60 grms. sodium
hydrate wfth 173 grms. sodium potassium tartrate dis-
solved to a volume of 500 c.c, the two solutions being
preserved separately, and mixed in equal volumes just be-
fore using. The determinations were made gravimetrically
«s follows : —

70 c.c. of the Fehling's solution were brought to boiling,
and exaAly 25 c.c. of the arabinose solution quickly run
in from a full pipette, after which the whole was kept
over the flame exaaiy four minutes, of which the first
fifty to sixty seconds were consumed in bringing the liquid
to boiling again. The time of boiling and filtering was
observed carefully. The precipitated cuprous oxide was
colle^ed on a Soxhlet asoestos filter with the aid of a
pump, washed with boiling water, alcohol, and ether, re-
duced to metallic copper in a current of dry hydrogen over
a low flame, and, after cooling in hydrogen, weighed. The
following are the results obtained : —

X p£r ant Solution. — 25 c.c, containing 0*250 grm.
arabinose, precipitated respedively : —

X. 0*4850 grm. copper.

2. 0-4874 „

3. 0*4860 „

An average of 0*48623 grm. copper, or for o*oox grm.
arabinose 0*00x945 grm. copper.

}/#r cent Solution. — 25 c.c, containing 0*1875 grm.
arabinose, precipitated respedively :—

X. o 3621 grm. copper.

2. 0*3631 „

3. 0*3640 „

An average of 0*36187 grm. copper, or for o'oox grm.
arabinose 0*001929 grm. copper.

I fer ant Solution.^2$ c.c, containing 0*125 grm.
arabinose, precipitated respedively : —

X. 0*2446 grm. copper«

2. 02430 „

3. 02479 „

An average of 0*2448 grm. copper, or for 0*00 x grm. arabi-
nose 0*001958 grm. copper.

Jrffir ant 6otution.^2S c.c, containing 0*0625 grm. of
araoinose, precipitated respectively :—

X. 0*1264 grm. copper.

2. 0*1248 „

3. 0*1250 „

•An average of 0*1254 grm. copper, or for o*oox grm. arabi-
nose 0*002 grm. copper.

From this it appears that in solutions of x per cent, or
less, strength, x m.grm. of arabinose precipitates x'9~2
in.grms. of copper.

2. Xylou. — The material used was prepared from wheat
lUawi and was of the desired puiity, m showo b^ Its

specific rotation ([a]D>"x8'4i*).* The determlnatiolks
were made under conditions as nearly as possible identical
with those for arabinose, with the following results :— >

X pit ant Solution.-^2^ c.c, containing 0*250 grm.
xylose, precipitated respedively : —

X. 0*4686 grm. copper.

2. 04640 „

3. 0*4667 „

An average of 0*4664 grm. copper, or for o*oox grm. xylose
0*001864 grm. copper.

} ptf ant Solution.^2s cc, containing o*x875 grm
xylose, precipitated respeAively : —

X. 0*3444 grm, copper.
a. 03446 „

3. 0*3^62 „

An average of 0*3451 grm. copper, or for o*oox grm. xylose
0*001841 grm. copper.

k'pir ant 10/ti^ion.— 25 c.c, containing 0*135 grm.
xylose, precipitated respedively :—

X. 0*2378 grm. copper.

2. 0*2366 „ „

3. 0*2381 „ „

An average of 0*2375 grm. copper, or for o'oox grm.
xylose 0*0019 grm. copper.

kpir ant io/ii/»oii.— 25 cc, containing 0*0625 8^1.
xylose, precipitated respedively : —

X. o*X235 grm. copper.

2. 0*1222 „ „

3. 0-X2I7 „ „

An average of o*x224 grm. copper, or for o'oox grm.
xylose 0*00x959 grm. copper.

Under the stated conditions, therefore, x m.grm. of
xylose precipitates x*86— 1*959 m.grms. of copper.

Dextrose, the most strongly reducing glucose hereto-
fore studied, precipitates, tinder similar conditions, x*8—
x*9 m.grms. copper, or slightly less than the pentagla*
coses. Stated in another way, the amounts of the
different sugars in x per cent solution required to precipi-
Ute the entire copper from x c.c of Fehling*s solution
are: —

Arabinose 0*004523 grm.t

Xylose 0*004617 „

Dextrose 0*004753 »•

Levulose 0*005x44 „

Invert sugar o'00494X „

Galadose 0*0051x0 „

Milk sugar 0*006757 „

Maltose 0*007780 „

From this it appears that the two pentaglucoses are the
most strongly reducing sugars yet studied, and this is
particularly true of arabinose. Both diffier from the value
for invert sugar sufficiently seriously to effiea the accuracy
of analyses in which they are concerned but not recog:

C. — FermintahiHty of th$ Pintaglucosis.
In connexion with the history of arabinose, mention, is
made b^ various investigators of its behaviour toward
fermenting agents, in which it is regarded as not subje^
to alcoholic fermentation. In some earlier studies of this
body by Prof. ToUens and myself this view was substan*
tiated, and it was further shown that this was a specific
property of arabinose, in contradistinaion from the true
glucoses, gaUaose, levulose, and dextrose, which are all
fermentable, some confliding views regarding galadose
being at that time cleared up.§

* This nnofaally fine preparation was kindly farniahed me by Dr
E. W. Allen, of Waahiof ton, D.C. who prepared it himaelf, and to
whom my thanka are dae.

•f Tollent, Handbuch der KohUnkydratet p. 284.

% H. Oet {B§r. d. Chtm, Ge*., 23, 3006), notes the redodnf effeA of
arabinose on a solution of poiasainm-copper carbonate, and finds its
▼alae lower than that of dextrose, hot higher than that of galaaoae,
toward thia reafcnt.

I nn* Chem,jLUbif), 349, aj; ; sad B$f, d. Chm, On.f ax, i^y^

Digitized by



Concerning the Pentaglucoies.


1 May 8, xSgt.

In his original paper, in which the discovery of xylose
was: announced, Koch described the same as non-fermen-
table, which was somewhat remarkable if it were a true
glucose, as was at that time supposed. Its identity as a .
pentaglucose being afterwards established, it was of
importance to show that it is, like arabinose, stridly non-
fermentable, thus fixing this as a charaAer of the
C5HZ0O5 group. In order to determine the non-
fermentability of a given substance it is very essential
that such a fermentation tent shall be made under
conditions favourable to the requirements and a^ivity of
the yeast plant used. No better example of this is known
than that of galactose, which has been reported non-
fermentable by some experimenters, who seemed to have
worked with poorly nourished yeast or that deficient in
vitality. Where there has been due regard for these
conditions, galaAose will produce nearly theoretical
results of fermentation, although k is plain that it is less
rapidly and easily subjeA to the influence of yeast than
its isomers, levulose and dextrose. The method used for
xylose was the same as that formerly applied to arabinose.
Its special feature consists in supplying the yeast used
with a nutrient extrad, made by boiling 100 c.c. of the
semi-fluid brewery veast with xoo c.c of water, and
filtering off the dark brown extras resulting. Such a
solution contains the natural constituents of the yeast,
and, as a nutrient, seems to give better results than
artificially prepared solutions of salts.

In the study of xylose two sets of experiments were
performed. In one the carbon dioxide produced was
colle&ed, and in the other the alcohol. The former were
conduced in eudiometer tubes which were first filled to
within 15 c.c. of their capacity with mercury ; then were
added 5 c.c. of distilled water, 5 c.c. of the above-
described nutrient solution, 5 c.c. of a mixture of 5 grms.
of Fleischmann's compressed yeast with 50 c.c. of water,
and lastly the weighed xylose. The tube was then
inverted in a mercury trough, and the gaseous produds
of fermentation were allowed to colledt in the upper end.
A second tube filled in exadly the same way, but con-
taining cane sugar instead of xylose, was arranged for
comparison, and as an index of the adivity of the yeast.

For the first experiment were taken 0*139 &^*
xylose and 0*1566 grm. cane sugar. Within a quarter of
an hour lively fermentation had begun in the cane-sugar
solution, and after ten hours seemed pradically complete.
In the xylose solution no adion was observed. After six
days, no further adion being noticeable, the necessary
readings were taken for calculating the amount of gas
colleded (including that absorbed by the liouid) from the
cane sugar. Less than 0*5 c.c. of gas had colleded on
the xylose solution, too little to indicate any appreciable
fermentative adion. The gas produced from the cane
■ugar was wholly absorbed by potassium hydrate solution,
and was therefore carbon dioxide. Reducing the tem-
perature and pressure to o* and 760 m.m., the volume of
the undissolved gas from cane sugar was 31*07 c.c, and
of that absorbed in the solution, 13*16 c c, or a total of
44*33 c.c, equal to 787736 ro.grms., or 50*49 per cent of
the sugar used.

In the second experiment were taken 0*1526 grm. of
cane sugar and 0x572 grm. xylose under the same
conditions as before. In this case also the cane sugar
underwent adive fermentation, but no adion was observed
in the xylose solution. After seven days, measurements
of the gas from the cane sugar, which was found to be
carbon dioxide, gave a total of 78*506 m.grms., or 51*5
per cent.

The yield of alcohol was next determined in a third
experiment, x grm. each of xylose and cane sugar was
placed in flasks of about 200 cc. capacity, with 35 c.c. of
water, 5 cc of the nutrient fluid, and xo cc of the
yeast mixture. The flasks were closed with a water-
valve. As in the other cases, the cane sugar showed
adive fermentation, but no adion was deteded in the
xylose. After five days both tolutiont were neutralised

with sodium carbonate, and after slight dilution, 50 cc*
distilled off and the alcohol determined in the distillates
by taking their specific gravities with a Sprengel's
specific-gravity flask.

50 cc of distillate from the cane sugar solution had
the specific gravity 0*9982, equal to 0*47* grm. alcohol*
or 47*2 per cent.

50 c.c. of distillate from the xylose solution had the
specific gravity 0*9999, showing no appreciable amoont of
alcohol present.

Summing up the results of these experiments, we
have : —

Carbon-dioxida. AlcohoL
Per cent. Per cent.


. Cane sugar
f Xylose
\ Cane sugar
( Xylose
^* ( Cane sugar
Calculated for QitHtfiu




From this it appears that under those condition! which
produced almost theoretical fermentation of cane toou-,
xylose did not ferment in the least. It is therefore, like
its isomer arabinose, non-fermentable, and thit becooica
a specific charaderistic of the pentaglucoses.

It seems, therefore, that the pentaglucotet are of
abundant occurrence in nature in the form of giim-Uke
constituents of vegetable tissues ; that these bodies are
easily converted into the pentaglucoses themselves, and
in such form are liable to be confused with the tme
glucoses, from which, however, they have quite different
analytical, and as far as we know economic, values. In
addition to the furfurol readion, non- fermentability seema
to be a further distinguishing property of the penta-

In view of the importance of this group, a method by
which they may be determined quantitatively in the
presence of the true glucoses seems desirable. Snch a
method has lately been proposed by Giintber and ToUeos

iBtr. d, Chtm, G^i., 23, 1751). Simultaneously with them
have been engaged upon the same subjed with tome-
what similar results, a more extended report of which is
for the present reserved.




Until recently our knowledge of the volumetric com*
position of water depended on the results of Humboldt
and Gay-Lu8sac. They presented their memoir to the
Academy of Sciences at Paris. The memoir was printed
in full in the Journal de Physique, vol. Ix., p. 229, and
translated in Gilbert's AnnaUn dtr Pkyuk^yoU xx., p. 38,
1805. Chaptal and Berthollet made a report on the
memoir to the Academy, which is contained in the
Annates de Chimie d* et Physique,^ Humboldt and Gay*
Lussac made twelve experiments with an excesa of
hydrogen, and twelve with an excess of oxygen. They
determined the amount of nitrogen in their oxygen 1^
absorption with an alkaline sulphide, and with this oxygen
determined the amount of nitrogen in the hydrogen. The
mean error of the measurement of the residue after explo*
sion with an excess of hydrogen was one part in five
hundred, and in the experiments with excess of oxygen*
one part in two hundred and fifty. From the experiments
, with excess of hydrogen they deduce 1*9989 as the
measure of the ratio sought ; from the experiments with
excess of oxygen, thev infer that both series together
justify the conclusion that one hundred volumes of oxygen

• Vol.mi.,p.a39,i8ci9(

Digitized by


CtelttcAL iliwa,!
M«y8,i89i. r

Volumetric Composition o/^ Water.


combine with very nearly two hundred volumes of hydro-
gen. They do not compute the ratio from the experiments
with excess of oxygen : it would be 1*982. Since I began
experiments on the matter, Scott has published several
statements of his results. In his first paper/ he gives
the results of twenty^one experiments. He gives two sets
of values of the ratio sought ; one computed on the
aaaiimption that the impurities found in the residue after
esploeton were originally distributed proportionally be-
tween the two gases, and the other on the assumption
that all the impurity was contained m the oxygen. From
the whole twenty-one experiments he gets the two values
2*9857 and 1*9941 respedively ; excluding two experiments
in which the impurity was very great, he gets X'9897 and
2*9959 ; from the best four experiments he gets the values
1*9938 and 1*9964; from the six best, he gets the values
2*9938 and X'9967. The mean error of a determination
was one part in two hundred and fifty on the first assump-
tion, and one part in five hundred on the second. Reje^-
ing the two worst experiments, the mean errors become
one part in five hundred and one part in seven hundred
and fifty parts. He gives the value 1*994 as the most
probable value of the ratio sought ; but from a considera-
tion of the same experiments, Youngf judges that the
value of the ratio is between 1*996 and X'998, and may
perhaps be taken as 1*997.

In the autumn of 1887, Scott ^ stated that he had then
made over thirty experiments, and gave the most probable
value of the ratio as 1*996 to 1*997. In the spring of 1888,
Scott § published four other experiments with a new and
larger apparatus; their mean is 1*997; ^^^ gases used
sometimes contained as little impurity as one part in
fifteen thousand. In the autumn of 1888, Scott II stated

* Proceedings R. 5., vol. xlii , j^. 398.
t Nature, vol. xxxvit., p- 390. 1888.
X Br, Assoc, Trans,, 1(87. p. 668.
f Nature, toI. xxxvii., p. 439, z888.
f Br, Assoc, Trans. 1888, p. 631

that the volume of hydrogen required seemed to decreaia
when it was evolved continuously from the same appa*
ratus ; and that the variation showed some impurity at
present undeteded. He published the results of four
experiments giving values varying from a*ooi down to

My own experiments on this matter are a part of one
of my processes for determining the ratio of the atomic
weights of oxygen and hydrogen. In this determination
the ratio of the densities of the two gases under ordinary
conditions is one faAor, and the ratio of their combining
volumes under the same conditions is a second fador.
Since it is difficult to free hydrogen from nitrogen, I
hoped to obtain it free from every other impurity, and to
determine the amount of nitrogen contained in it, so that
I could compute a numerical corredion to the observed
density. I have now finished the determination of the
jcombining volumes of the gases unleas some as yet unde-
teded error should necessitate further investigation. I
have been able to reduce the mean error of a determination
to less than one half of that which I ventured to predid
early in the volumetric studies preliminary to the adual
determination.* As the degree of accuracy which I hope
it will be found that I have attained is very considerably
greater than in determinations of the same kind by others,
I have thought it needful to give a somewhat minute
account of the details of the work, in order that those
interested in the matter may better judge what degree of
confidence mao fairly be reposed in the result, or may be
in a position to suggest improvements or corrediont
needed in my processes. Though some parts of the work
go back for many years, yet, thoroughly agreeing with the
expression of Ostwald that he undertakes a heavy respon*
sibility who publishes values of constants, I have made
public no figures obtained till I have done the best that I
know how to do.

* jim. Chem, Joumaft vol. %, p. aj, i888»

Digitized by



VolufHeMc Composition of WaM.

1 May8,t89l.

Pfiparation of Pun Hydrogin.^Tht preparation of
pare hydrogen has been difficalt. I tried long to obtain
It by the a^ion of dilute acids on zinc. It is in this way
not difficult to obtain a gas free from arsenic and sulphur
(or chlorine), and easy to obtain it free from oxygen by
passing the hydrogen over heated copper; but two diffi-
culties remain, one of which is serious. The amount of
hydrogen which can be obtained from a given weight of
roatenals is not always enough to sweep out all the
nitrogen present in the apparatus or contained in the
liquids used and still leave much of the gas to be utilised.
Perhaps, by construding the apparatus so that it can be
repeatedly exhausted, this difficulty could be overcome.
But a difficulty which I have not yet surmounted proceeds
from impurities found in every sample of zinc which I
have heretofore obtained. The metal contains gases
which were absorbed by it during metallurgic processes ;
of such gases, carbon dioxide is but one ; this coold be
easily removed, but there are present other gates which
contain carbon; until they shall have been sufficiently
investigated it is not certain whether they can be removed
by absorption.

It is to be noted that this last difficultv is by no means
removed by amalgamating the zinc, and using it as one
pole of a voltaic element or of a decomposing cell. The
nest zinc I have found gave when so used hydrogen which
after combustion in no very long time caused a precipitate
in lime water. The gas also contained nitrogen which
came from the zinc employed ;* I therefore abandoned
for the present the use of zinc. Dr. W. H. Burton kindly
distilled in a vacuum for me some kilogrm^. of so-called
perfeAly pure zinc ; with which produA I shall some time
resume the preparation of pure hydrogen.

Having abandoned, perhaps too hastily, the attempt to

fet pure hydrogen from zinc, I resorted to ele^rolysis.
he decomposition of an alkaline hydroxide seemed pro-
mising; by it one would exped to get nothing but hydro-
fen, oxygen, hydrogen dioxide, and ozone. From the
ecomposition of absolutely pure potassium or sodium
hydroxides, no doubt this pleasing ideal might be realised.
But two decomposing cells which I construAed for the
purpose and filled with so-called pure potassium hydroxide
yielded hydrogen containing carbon. This might possibly
come from organic matter adhering to the interior of the
cells, although I was at that time especially on my guard
against carlK>n ; but it was more probable that it came
from the simultaneous eledrolysis of an alkaline carbon-
ate. I therefore made a new decomposing cell most
carefully of clean glass and platinum, cleaned it from
organic matters derived from glass blowing manipulations
by long immersion in chromic acid, but filled it with
potassium hydroxide to which I purposely added potas-
sium carbonate. The eledrolysis of this solution yielded
hydrogen which instantly clouded lime water on combus-
tion. It was therefore plain that the alkaline solution
submitted to eledrolysis must not contain any caibonatr.
This might no doubt be attained by using barium hydrox-
ide, either alone, or added to potassium hydroxide to
remove carbon dioxide. But since the so-called pure
potassium hydroxide is purified by solution in alcohol, it
18 by no means certain that it may not contain carbon
other than that existing in an alkaline carbonate. The
matter, therefore, began to assume such an aspeA that I
resorted to the use of sulphuric acid.

Decomposing Cell for 25 Litres an Hour.—l have at
commanr*. through the courtesy of the East Cleveland
Railroad Company, by day and night an elearomotive
force of five hundred volts. It was obviously proper to
vse the current '^rom thi^ «ource by passing it in succession
through many small decomposing cells, rather than
through one large cell. It was convenient to have at
command a current of hydrogen up to twenty-five litres
an hour. To obtain this from a single cell would reouire
a current of some fifty amperes, consuming thirty-three

* Amer* Ckem. Journal, vol. sii., p. ^tfr, 1890.

horse-power. But if a current of one-thirtieth this araoant
is passed through thirty cells in succession, the sftine
amount of hydrogen is obtained from one horse-power. I
therefore sealed thirty decomposing cells to two delivery
tubes, so that when the eledric current is passed through

Online LibraryArnold BennettChemical news and journal of industrial science → online text (page 59 of 88)