LEM. ‘~ ' mi
'ILJ- .5 ”Xi
g. I o f"
‘f-nr-fxnf a1 -. ~ .w
L554? 7: . »‘..r_'-J

I
It!»

‘lu‘jv-

.

:5 H? j
" Jz'L

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8/01 c:/CIRCIDateDue.p65-p.15

 

 

SILICA GEL

A Thesis
Submitted to the Faculty
of
Michigan State College
In Partial m1f111ment of tin
Requirements for the Degree
of
Doctor of PhilosoPhy

Department of Chemistry

By

BonJamin Lavi Smite

1926

ill?! :1. 2‘1 " '.' V R '1' Vittoria;
* \7- ’ -
A. MK (0

a. 4. I f y.“
__> (o to {,3

AMO'WZN T

I wish at this time to express an; appreciation of the help
snd.kind1y suggestion given to me by Doctor D. T.IEwing during the pro-

gress of this work.

THE FAQTOTE INFIIIENQING THE AQTIVITI 0F SQIQ

There are many factors which tend to effect the porosity of silica
gel, which may be modified by the method of preparation. The object of
this investigation has been to determine a few of these factors with the
idea of develOping a method of preparation that would avoid as many of the
defects of the other methods as possible.

The two methods in general use today are namely,» the method
develOped by Patrick!“ which consists essentially of decunposing sodium
silicate by means of hydrochloric acid and washing the fresh undried gel
free of its acids and chlorides and then drying it; and the method orig-
inally reported by Briggs”, in 1921 of allowing the mashed gel to dry
to a rigid solid at 390° c. and then washing it in boiling water. The
high temperatures used by Briggs”, and later by Falls and Firth“) have
been found to be unnecessary for the production of an active gel.

holiness“ investigating methods to improve the mode of washing
the acids and salts fran silica 391, found that a gel dried to a rigid
solid at temperatures little above that of the average laboratory gave
an exceptionally active gel. He holds that this is partly due, at least,
to the support given the gel structure by the chloride solution. The
criticism of the Patrick”) method is that the capillaries of the soft
fresh gel are especially liable to collapse and injury during handling
am washing.

_ The gels of colloidal silicic acid, or the so-called soluble
silica, obtained either by the decomposition of an alkaline silicate in

the presence of an excess of the acid or by the hydrolysis of a silicon

-1-

compound have been the object of study of many chemists. The history of

this research has been given by Fruhlingw)

in 1895 and later continued
by waiden(5’ in 1910. As early as 1773 Baume't7) demonstrated the exis-
tence of What he called a soluble form of silica obtained when a solution
of liquor silicium was treated with an excess of acid.

(8)

The use of diatomaceous earth or Kieselghur as a decoloriz-
ing agent is of great antiquity. This substance is found in marine de-
posits in man places, being made up of the siliceous skeletons of diatoms.
the silica probably in a colloidal fem having been derived from the water
in Which the diatoms lived. A good quality of Kieselghur should absorb at
least four or five times its weight of water. It was because of this liquid

(9) decided upon its use

holding capacity of diatomaceous earth that Nobel
as a vehicle for nitroglycerine in dynamite since it will hold as much as
75% nitroglycerine and still remain dry enough to mould into sticks.

‘10) prepared gels of silicic acid by two dif-

In 1830 Berzelius
ferent methods. In one he treated silicon tetra fluoride with water
which upon evaporation set to a gel and in the other he treated crystal-
lized boric acid with silicon fluoride, removing the fluorides and borates
by the use of a large excess of amonia. In 1846 Bbelmanun hydrolyzed
ethyl ortho silicate forming a gel. Although these early workers only
vaguely guessed the colloidal nature of soluble silica their contribu-
tions to the knowledge of its chemical properties are invaluable.

Various methods have been deve10ped of preparing the hydrogel
of silica by means of electrolyzing a solution of an alkaline silicate.
lacqueraluz’ recorded the formation of gelatinous silica about the pos-

itive pole when an electric current was passed through a solution of

potassium silicate. Kroger?“ working with a cell in which the anode

-2-

was separated from the cathode by means of a porous cylinder, found that
with a 1.5% solution of sodium silicate a clear hydro sol of silicic
acid was fanned in the anode chamber, Which did not gel within four weeks,
electrolyzing a 6% solution gave a hydro sol Which set almcst instantan-
eously, while with a 30% solution a dense gelatinous fans of silica was
deposited on the anode. Spencer and Proud“) prepared wimt they tensed
ortho silicic acid by electrolyzing a 50% solution cf water glass using
a very heavy anode current density.

The history of silica gel however preperly dates from the work

(15) in 1861. He was the first to show that the clear

of Thanas Graham
limpid solution obtained by the decomPosition in the presence of an ex-
cess of acid of an alkaline silicate solution by an acid according to the
following idealized reaction:
111323103 . 2 sci .- £25103 . 2 Bacl

(since water glass is not such a simple compound but a complex mixture)
would not diffuse throng the pores of an animal nor a vegetable membrane.

A hydrosol of silicic acid used by Grahamus’ in his work was
obtained by mixing with constant stirring equal volumes of a ten per cent
solution of sodium silicate and also a ten per cent solution of hydro--
chlorio acid. This colloidal solution was freed from its admixture of
chlorides and acid by dialysis. Graham'sum heap dialyser consisted
of a glass cylinder open at both ends and so arranged that a piece of
animal or vegetable membrane could be stretched across one of the Open
ends and. held in place by means of a metallic heap. The solution to be
dialyzed was placed on the cylinder and the apparatus suspended in such

a way that the membrane was in contact with the surface of water. Grahamus)

found that after dialyzing for four days, the water being changed oc-
casionally as the electrolytes diffused into it, the solution in the di-
alyzer was not rendered turbid by the addition of reagent silver nitrate.
The dialyzed hydrolsol of silicic acid fails to show the usual Tyndall‘l’n
cone observed when a beam of light is focused,ina colloidal solution in-
dicating that at least in this condition the molecular aggregations are
not sufficiently large to scatter light.

Although the majority of chanists have used hydrochloric acid
to effect the decomposition of the alkaline silicates in the preparation
of silica gel, the use of many other acids has been reported. Plessyua)
in 1855 using acetic acid prepared a very interesting gel which he termed

(19i-

artificial hydrOphane. Monier reports the use of oxalic acid, while

Meunierwo) found that sulphuric acid decomposed sodium silicate to form

21) has prepared a useful table in which

a firm translucent gel. Holmes‘
he gives the concentration of a number of acids both organic and inorganic,
Which, When mixed with equal volumes of sodium silicate solution of a cer-
tain density and Nazo : 8102 ratio, results in a gel which will set in any

22
( ) studies of the relationship existing between

desired time. Holmes'
syneresis and vibrations of gels are of interest.

Of the three common inorganic acids -- hydrochloric, nitric,
and sulphuric -- hydrochloric reacting with sodium silicate gives a gel
Which can be purified of its alkaline salt and excess acid while the use
of nitric acid results in a gel which it is almost impossible to free
from the nitrate ion. The gel formed by sulphuric acid stands midway

between: the two extremes according to Gmelin-Krautszz)

There are many factors which influence the rate of setting of

silica gel. Grahamue)

found that the addition of powdered graphite
caused a dialyzed sol to set to a gel within two hours While carbon di-
oxide caused it to become turbid. Flemingw“ working on the rate of
setting stated that the hydrogen and hydroxyl ions have a catalytic effect
upon the coagulation of the gel. He has summarized his findings by say-
ing that as the concentration of the hydrogen ion‘.increases the rate of
setting is at first retarded and then accelerated, while in the case of
the hydroxyl ion concentration the rate of setting is at first accelerated

(21)

with increasing concentration and then retarded. Holmes found that the

delwdrating influence of the unionized acid molecule had a greater effect

than either the hydrogen or deroxyl ion concentration.
(15)

found that he was able to obtain a chlorine free sol
(25)

Graham
of silicic acid by four days of dialysis while in the work of Jordis
it took marly three weeks to remove all traces of chlorine and from four
to six weeks before the sodium had disappeared. Zsignomly and Beyerws’
working with an improved form of dialyses confirmed Graham'sum original
findings. However, as a means of producing a satisfactory silicic gel in
quantities sufficient for an extended research problem, dialysis is much
too slow.

It has been demonstrated that an active gel can be produced by
allowing the colloidal sol of silicic acid to set to a solid gel While
still containing the sodium chloride formed as a by-product of the re-
action between sodium silicate and hydrochloric acid along with the excess
acid and later washing the gel free of these electrolytes. The effect of

the hydrochloric acid and sodium chloride upon the porosity of the gel is

probably that of increasing the size of capillaries.

 

 

Winder and Suleimanw“ have shown that at ordinary temperatures
gelatinous silica is soluble to an appreciable extent in dilute hydro-

28)

chloric acid. J. von Bennnelen‘ found that the dried gel adsorbed hydro-

chloric acid from dilute solutions and is of the opinion that the acid re-

(29)

places the water of structure of the gel. Jordis has pointed out that

the early observation on the solubility of gelatinous silica, such as that

(16) Who found that 100 grams of water at ordinary temperatures

of Graham
dissolved 0.021 grams of silica, did not take into effect the influence
of traces of acid and also of the alkalies derived from the glass utensils.

Jordis(29’

has shown that when firm gelatinous silica 361 still
containing the by-products of the reactions between sodium silicate and
hydrochloric acid is washed in cold water, the wash waters soon becane free
of both 01' or lila" ions. However if this same gel is again digested using
water at the temperature of 100° 0., it was found that the wash waters show
strong evidences of chlorine and sodium. If the digestion is continued at
this temperature, the amount of alkaline salt in the water does not decrease
but reaches a minimum value and continues there in spite of the duration of
the digestion. Jordis and Karnstenwo) are of the Opinion that some com-
pound is formed between the hydrochloric acid and the silica, as are also
Mylius and Groschuffsn’ Water freed of all alkalies and acids has been
shown by Lenher and non-111‘”) to act as a solvent upon the dry gel,
solubility depending directly upon fineness of grinding, temperature, and
pressure. The water first converts the powdered silica into a gel and
later dissolves it. Mellorwz) has shown that sodium chloride in solu-

tion acts to increase the solubility of gelatinous silica.

Silicic acid at ordinary temperatures is one of the most feeble

inorganic acids known and it is probable that if it was isolated it would
very readily dissociate into silica and water. However the large number
of naturally occuring silicates gives evidence of its existence. Many of
the earlier chemists have recorded their efforts to determine whether there

existed a definite hydrate of silica. h‘belmanul’

obtained a gel by the
action of moist air upon silicic ether which when air dried appeared as
a glassy, amorphous substance which had the canposition of 28102 3320.

Fromm“ found that if the same gel was dried in vacuo that the molecular

)

conIposition was 35102 21-120. Doveri‘35 prepared a gel by decanposing an

alkaline silicate with hydrochloric acid which when dried in vacuo over

(11)

concentrated sulphuric acid corresponded to Ebelman's hydrate. Fuchsws)

obtained two hydrates of silica, one with a canposition of 38102 H20 and

(37)

the other 45102 n o. Grahamus’ and also Roberts found that if the

2.
clear transparent gel obtained by allowing a dialyzed sol of silicic acid
to gelatinize and drying the gel so obtained in vacuo at 15° 0. was further
dried over concentrated sulphuric acid, it corresponded to the formula of

meta-silicic acid $102 320.

J. van Bermelenwe) has made an exhaustive study of the physical
prOperties of the hydro gel of silicic acid. In an investigation of the
equilibrium conditions existing between the dehydrated gel and the water
vapor in which it was in contact, he found that the dehydration curve was
not reversible, but that rehydration takes a shorter course to reach the
saturation curve representing the vapor pressure of water at that partic-
ular temperature. He also found that the nature of the rehydration curve

varied with the age and method of preparing the gel. According to von

Bemelenise) the hydrogel of silicic acid is of an unstable canposition

of the general cmnposition 8102131120 in which the value of n is determined
by the previous history of the gel. Jordiswg) concludes frun von Bam-
melen'sz’a) work that the amount of water in a gel at equilibrium is pro-
portional to the partial pressure of the water in the surrounding atmcs-
phere. Zsignondymo’ explained the hysteresis in the dehydration curve

on the assunption that after the primary dehydration there is a film of
adsorbed air on the surface of the capillaries which the water can not wet.

Carnelley and Walkerul)

working with inorganic oxides found in
dehydrating a sample of pure silica obtained by the decomposition of an
alkaline silicate by means of hydrochloric acid that there was a terrace
on the dehydration or time-temperature curve Just below 100° 0. here was
not however a sufficiently abrupt change in the rate at which the water
vapor was given off to account for the existence of any definite hydrate.
J. von namelemwa’ later showed that this terrace on the dehydration curve
‘ was characteristic of other colloidal gels.

Hel lor (42)

has sunmarized this extensive work on the hydrates of
silicic acid by stating that the majority of these hydrates are but arbit-
rarily selected states in a continuous series between the sol of silicic
acid and dehydrated silica and are represented by arbitrarily selected
points on a continuous drying curve. In this connection it is interest-
ing to note that Faber and OlsenM’z) working at the Edgewood Arsenal re-
search laboratory, Maryland, U.8.A., found that silica gel could be used
repeatedly without loss of efficiency for the adsorption of many different

vapors, such as water, benzene, nitrous oxide, and nitric acid, provided

that the temwrature of reactivation did not exceed 250° C.

The history of the influence of physical factors upon the gels
of silicic acid is not comPlete without some reference to the work of
Tschermak‘?“ Tschermak(44) has preposed a system of classifying the
different silicic acids based on the natural silicate from Which these
are obtained by decomposition with hydrochloric acid. He has shown that
these various gels showed a lag in their dehydration at different tauper-
atures and assumed that it was due to a change from the evaporation of

a. 7o
absorbed water to a dehydration of the loss of water of structure and

(45)

represented a definite acid. Ragga has shown that the rate of de-

“M depended upon the tanperature of de-

hydration observed by Tschermak
hydration and that the hysteresis in the dehydration curve could notbe
interpreted as a transition from absorbed water to a loss of water of

(46)

structure. It has been further shown that the A. Suide reaction of

methylene blue towards Tschermak'smu acids was due to the changes in
the hydrogel brought about by the gradual loss of water.

Grahamum found that it required a very small quantity of
alkali to neutralize a dialyzed sol of silicic acid and concluded from

(47) from

this that it must have a very large molecular weight. Sabeneff
boiling point and freezing point determinations on the soliof silicic acid
calculated that the value of n in the formula (31°2’n must lie sanewhere
between 800 and 1000 or that the molecular weight was over 40,000. Since
the colloidal 801 of silicic acid is not a homogeneous solution but a homo-
geneous suspension, the laws of osmotic pressure applying to dilute solu-
tions can not be applied in the case of silicic acid.

me specific gravity of the hydrogel as detemined by different

workers has been found to vary from 1.50 and 2.390. These values are of

-9-

little worth however since it has been impossible to determine the state
of matter in which the water is that is held on the surface of the capil-
laries. Anderson!“I found that the specific gravity of the gel thoroughly
permeated with water was 1.5, that of the dry gel was 1.980, and that of

the gel substance was 2.048. Earl and Urban(49)

found that a gel dried
at 25° 0. had a specific gravity of 2.465, dried at 300° 0.2.390, while
When dried at 1000° c. the specific gravity had fallen to 2.271. When
the gel was further digested in concentrated hydrochloric acid and after
evaporation ignited to a White heat, the dehydrated silica had a specific
Gravity of 2.627. Ruth‘s“ found that a gel in which the adsorbed water
had been replaced by ammonia and evacuated at a tauperature of 250° 0.
had a specific gravity of 2.210.

Physically the gel of silicic acid is apparently made up of a
cellular or net-like structure penetrated by innunerable capillaries.
J. van Malena“ has studied this structure microscOpioally and applied
the term "wabenstrulttur" to the condition that he found. He stated that
this structure was due to two solutions of markedly different concentra-
tions, the more concentrated forming the walls of a mass of cells which

(51)

enclosed the more dilute solution. Butschli found tint this net-lilac

structure appeared at certain stages in dehydration of the gel and later

(51)

disappeared. The foam-like structure mentioned by hutschli has been

shown to be due to refraction phenomenon. The gels prepared by Butschliwn
and also by von Malena“ appeared at a certain stage in dehydration
to be Opaque and as the dehydration progressed further to become clear.
Biltz and Geibelw‘?" were among the first to study the physical structure

of the gel by means of the ultra microscope. The observations of 281g-

.10..

53)
mondy‘ upon the physical structure of the gel agreed with those of

the'earlier workers concerning the macro structure, but he has shown
from his ultra microscOpic investigations the pores of a gel dried slow-
1y into a compact mass were amicroscOpic in size ranging from 3.65 we to

(51)

5 uu. The coarse structure observed by hutschli was shoWn to be due
to an accumulation of water in the larger extra cellular spaces. Zsig-
mondy‘sz’ found that by saturating a gel with benzene the trmultra micro-
scepic structure was revealed, since the index of refraction of the benzene
correSponded so closely with that of the gel structure.

Anderson ‘48 ’

studying the problem of pore size from the angle
of the lowering of vapor pressure due to the presence of a large number
of extremely fine capillaries obtained values which agreed with those of
Zsignondy(§3) Bachmansa’ observing the process of gelatinization of a
hydro sol of silicic acid ultra microscOpically, found that the amplitude
of the Brownian movement decreased coincident with an increase in the size
of the molecular aggregations due to gel formation. During the process of
aging, the cellular structure of the gel becomes less distinct and appears
finally as an ultra microscOpically homogeneous systan.

According to von Weimarnws) a gel is formed when the degree of
momentary super saturation of the precipitate is excessively high, the de-
gree of diapersion being a direct function of this super saturation. Pappada
and Sadowskiwé) have distinguished between coagulation and gelatinization,

arguing that since in gelatinization there is a smaller cencentration of

the silicic acid consequently there is more time allowed for the orienta-

tion of these extremely fine granules. La Chatelierw“ holds that at no

time is the silica in a dissolved state, but that it is held in a gel
structure by the extreme hydration of the particles. J. van namelenwe)
states that the water holding capacity of the dehydrated silica gel is
decreased by its formation from a concentrated solution of silicic acid

in which rapid coagulation takes place,by rapid prolonged drying and,
under adverse conditions, by age.

This extensive review of the literature relating to silica gel
has not been made with the idea of covering the entire field, but has been
done with the idea of bringing out if possible those facts which have been
shown to have a direct effect on the physical properties of the final de-
hydrated product. Credit should be given to the excellent bibliographies

found in Mellor' s (59’

Canprehensive Treatise on Inorganic and Physical
Ghenistry and also in Gmelin-Kraut's‘so’ Handbuoh der anorganische Chanie.

The ten silica gel is of course a misncmer, as it is now gener-
ally applied to the hard, glass-like substance derived by dehydrating the
gelatinous precipitate of silicic acid. The factors influencing the ad-
sorptive activity of this form of silicic acid can be grouped into three
general classes.

1. Mode of preparation.

2. node of freeing the hydrosol of the by-products of the decom-
position of an alkaline silicate or the by-products of hydrolysis.

3. node of drying which also includes the temperature of final
activication.

J. van Melensse) Holmes?“ and others have shown that the
most active gel results from a hydrosol sufficiently dilute of silicic

acid Which requires an appreciable time to gelatinize after mixing the

acid and silicate solutions together, in order that in the orientation of
the molecules of silicic acid there can be an orderly aggreation into
colloidal particles.

As a means of preparing a quantity of silica gel necessary for
any extended research, hydrolysis of a silioan canpound may be dismissed
due to the difficulties encountered. Hydrolysis as a mode of preparation
has however a real value, since by the use of a cornpound such as silicon
tetra fluoride, hydroflurio acid, the by-product of the reaction,

SiF4 9 41120 3 823103 t 4H? + H,0
is volatile am can be easily removed from the resulting hydrosol and a
very pure gel can be secured.

The method used in preparing any quantity of silica gel is he'-
ever in general practice the decomposition of an alkaline silicate, usually
water glass, by means of an acid. is has been shown, the gel resulting
from the use of hydrochloric acid for this purpose can be more readily
freed of the electrolytes than the gels made from other mineral acids.

There are two general methods used for freeing the hydrosol of
silicic acid of its electrolytes. One consists of dialyzing the acid sol
of colloidal silica until the electrolytes have all been removed by dif-
fusion and subsequent gelatinization of the pure sol. This method results
in an extremely compact, fine-pared gel, but is difficult when applied to
quantity production. The other method is that in general use by the major-
ity of workers at the present time. It consists of allowing the acid hydro-

sol of silicic acid to gelatinize and then washing it free of its acid and

sodium chloride.

-13..

 

.

5:11
U. $-
is g
cf the
in”:

302081?

tially
Stilts.

time i

form c
of the

uble c

by was

 

If the hydrogel is washed free of its electrolytes while it is
still in a soft Jelly-like condition, as described in Patrick's”, first
U. 8. Patent, there is a great possibility of injuring the activity of
‘ the gel due to partial collapse of the capillary spaces and obstruction
of the ultra microscOpic pores of the external surface. On the other
hand, if such a gel in its original condition is allowed to dry until the
concentration of the sodium chloride solution in the capillary spaces
reaches the point where crystals are deposited, these spaces are eviden-
tially injured, possibly by the splitting force, and an inactive gel re-
sults. As will be shown, there is a definite degree of drying at Which
time if washing is instigated an exceptionally active gel results.

Probably the most important single factor in producing an active
form.of silica gel is the method.of drying, and this effect is independent
of the condition of the gel; that is, Whether it still contains its sol-
uble chlorides and hydrochloric acid or whether these have been removed

by washing. Rapid drying invariably results in an inactive gel.

The water glass used in this work was a'wateraWhite product
having an original density of 1.4042 with a ratio of N320 to $102 of
lit 3.46. There was a trace of iron present, but it was not considered
to be large enough to materially affect the results. water glass con-
taining approximately the same distribution ratio of Nazo and 3102 has

been found by Holmes to give the most satisfactory results.

-14-

TABLE;
Msi tion of Water Glass
Silica (8102)you...".oo....o.........o.27.50%
Sodium oxide (Reach..................... 7.49%
Moisture.................................57.22%

Specific gravity......................... 1.4042

The water glass was diluted to a density of 1.18 by adding it
with agitation by means of compressed airjto the water? rather than by
adding the water to the silicate, the resulting diluted solution being
colorless and free from traces of coagulated particles. is the same
apparatus was used in diluting the water glass solutions as was used in
mixing the silicate with the acid solution, it will be described here.

A large mouthed bottle having a capacity of at least twice the volume of
gel solution desired, so as to avoid loss due to the excessive foaming,
was fitted with a stapper throng: Which a funnel and four glass tubes were
passed. These tubes were long enough to reach nearly to the bottom of

the bottle and had their lower ends bent in such a way so as to produce

a whirling effect in the solution when compressed air was forced through
them. A vent large enough to allow of the free escape of the air should
be provided in order that there may be no back pressure on the funnel
through which the water glass is added. The funnel should be so orien-
tated that the falling silicate solution does not strike the compressed

air tubes. By using this method for diluting the water glass solutions
the ill effects due to the presence of dissolved carbon dioxide in the

water‘are eliminatei, since in this way the water is considred to be a

-15..

9......

potential acid due to the 002. The density of the diluted water glass,
1.18, was determined experimentally as being nearly as high as could be
used with an equal volmne of 3.8 hydrochloric acid to give uniformly
good results.

The method of preparing the gels used in these preliminary in-
vestigations was as follows. The colloidal solution of silica obtained
by mixing equal volmnes of water glass and hydrochloric acid, as pre-
viously described, was poured into shallow crystallizing dishes and
allowed to stand until the gel was firm enough to handle. large vari-
ations in temperature were avoided, the average being 18.5° c. As soon
as the gel had reached this condition of firmness, it was renoved from
the dishes, and after being broken into inch cubes was placed upon cheese
cloth racks in a drying room with an average temperature of 45° 0. Drying
was continued until the smaller pieces of gel showed a tendency to split
due to the fonnation of sodium chloride crystals in the capillary spaces.
The larger lmnps of gel at this stage of dryness were rigid pieces with
a glassy surface when broken,having a water content of between 55% and
65%.

The gel in this condition was then washed by decantation until
the wash water did not show a trace of cpalescence due to chlorides when
tested with reagent silver nitrate solution. After washing the gel was
further dried at the same temperature, 45° 0., until it had reached the
point where there was no further change in volume of the gel pieces due
to dehydration. The water content of an active gel in this condition
was found to be 14%. Final activation of the gel was obtained by placing

it in a U-tube submerged in an oil bath and drawing a slow stream of pre-

-16-

heated air through it free from.the usual laboratory fumes and dried by
means of Gaelz. The temperature of the oil bath was gradually raised
from40o to 225° 0., the time required for securing this temperature being
never less than two hours. Activation was continued after this temperature
was reached, the time depending upon the amount of gel, until the water
content had fallen to between 6 and 7%. The gel in this active condition
was protected from.fumes by'being kept in glass stOppered bottles.

The adsorptive capacities of the various gels used in this work
were detennined dynamically. Air saturated with benzene vapors was passed
over a weighed sample of the gel and the sample weighed again When it had
reached saturation. Benzene was selected because of its relatively low
boiling point.

The adsorption train used for this purpose was set up as follows.
Air drawn from out-of-doors by'means of a suction-blower was dried by pass-
ing through a calcium chloride tower. As the rotar of the suction-blower
moved in oil it was necessary to provide a trap for oil vapor‘whidh con-
sisted simply of a flask suhmsrged in a freezing mixture.

As a final precaution against the possibility of any oil parti-
cles being carried over by the air stream, it was passed through a tube
lightly packed with cotton wool and then through a calcium chloride dry-
ing tower. The rate of flow was determined by passing the air through a
flowmeter. This was provided with a trap to catch the mercury with.Which
the flowmeter was filled in case of an.accident to the blower. The rate
of flow of air used throughout the work was an average of thirty cubic
centimeters of air per gran of gel.

The metered air'Was saturated with benzene vapors by being

-17-

\

passed throug) saturators so constructed that there was the minimum of
contact between the benzene-saturated air and the rubber connections.
The saturators were chletely submerged in a constant temperature water

bath maintained at 12.50 1: 1° 0. After leaving the saturators, the air

passed through a spray catcher filled with glass wool and into the gel

tube.

The gel tube consisted of a glass U-tube, with short side arms,
fitted with perforated ground glass stoppers. The U-tubes used held ap-
proximately tfle‘ngggams of gel. In order to avoid the possibility of
there occurring a condensation of the benzene vapors in the gel tube,
it was maintained in a water bath at a constant temperature of 25°; 0.50 c.

A number of preliminary runs were made with this apparatus to
detemine its stability. With an air stream (320 cc. per minute) satur-
ated with ben2ene vapors at 12.5° 0., a sample of silica gel obtained from
the Silica Gel corporation Showed a capacity to adsorb 27.81% of its own
weight of the vapor. thrperiment showed that there was 1473 by weight of

benzene in the air stream.

W
Per cent of Benzene Vapors by Weight in Air Stream

(320 c.c. per minute) saturated at 12.5° C.

 

 

 

Time in Weight 320 cc of Air Weight of Per cent by Weight
at 25° C. and 747 m/m ‘ of Benzene in
Minutes jressure Begene A_ir AveraQ
1 0.3897 gr. 0.0570 gr. 14.57% 14.54%
2 0.7795 gr. 001325 gr. 14052:;

-18..

These results were obtained by placing a weighed gel tube
filled with an active gel in the adsorption train and passing air satur-
ated with.benzene vapors at 12.5° 0. through it at the rate of 320 c. e.
per minute until the first traces of benzene could be detected by smell.
It was found that for a fifteen gram sample of gel it required less than

three minutes.

$é2£§.1£i
Benzene Adsorbed by Comercial Silica Gel
Air Stream (320 c.c. per minute) saturated with

Benzene Vapors at 12.50 C.

 

tuber Temperature Temperature of Grams of 0 H6
Adsorbed p31?
of E of Saturator idsomtion Bath Gram of Gel
1 12.5° c. 25.0° C. .2788
2 12.20 00 25030 Co .2765
s 12.50 c. 25.0° C. .2790

When a gel of silicic acid of such a concentration that it re-
quired at least an hour's time to set is observed under a beam of ligxt
incident to the surface of the gel at an acute angle, bright refractive
lines are seen to fans slowly becoming feathery with much branching, as
setting proceeds. These lines have been cemented on by Roberts(.37 work-
ing in Graham'sus’ laboratory, Who thought that they were due to a fungus
growth. This seems doubtful, as they form not only on the surface but in

the body of the setting gel, and reform almost immediately when distorted

by ratatione

.19..

A series of gels was made, using equal volumes of sodiun
silicate solution having a specific gravity of 1.18 and derochloric
acid, varying only the concentration of the acid. The process of dry-
ing, washing, and activication was identical throughout.

These gels were compared as to their capacity to adsorb benzene
vapors at 25° 0. from an air stream (320 c.c. per minute) saturated at

o
12.5 0. Table 1v gives the sunmary of this work.

TABLE I1
Adsorptive capacity of different gels made with varying concen-
trations of hydrochloric acid.

Benzene adsorption at 25° C. from an air stream (320 c. e. per

minute) saturated at 12.50 0.

 

Normality of Time of set Condition of Grams of 0 I15
Adsorbed r
3071 in Minutes Feather Hirines gtram of Ge}:
10.0 muately Absent 0915
7.5 15 Absent 0.18
5.0 45 Short unbranched 0.22
3.9 105 Feathery 0.245
3.0 500 Feathery 0.243
200 720 Paathery 0018

U was found that a gel formed by decanposing a sodim silicate
solution having a specific gravity of 1.18 by means of 5.9 N hydrochloric

acid had the highest adsorptive capacity.

or SYNEEESI mourm AQTIETY OF

W

 

Grahamué) was the first to state that the dominating quality
of colloids is the tendency of their particles to aggregate and contract.
In the gel state this contraction or syneresis still proceeds, causing
separation of the water from the gel structure. Schwarz and Stowener‘so)
have shown that this contraction is not dependent upon evaporation, but
will continue even if the gel is suhnerged in water.

A series of gels was made using the same concentration of
hydrochloric acid (3.9 normal) and sodium silicate solutions (1.18 speci-
fic gravity) throughout. The following conditions were investigated:

(1) The effect of drying a gel rapidly, as soon as set, either by means
of a higier tennperature 40 - 50° 0. or by the rapid removal of moisture
at room teznperature effected by an air blast from an electric fan. (2)
The effect of allowing syneresis to comPIete itself, in one case in which

evaporation at room temperature was prevented and in the other in which

there was no check on evaporation.

-21..

MEL!
EFFECT 01? 311333813 UPON THE BENZBNS ADSORPTIVE CAPACITY
OF A can.
Air stream (520 c.c. per minute} saturated at 12.50 C. with
benzene vapors.

Temperature of Adsorption Bath 25° 0.

 

nethod of Drying Time Grams of 0636 Adsorbed
2311113 fi gperation Repaired“ 331-er pg; 39L
40° - 50° As soon as set 50 minutes 0.14
Air blast 18°C. As soon as set 30 minutes 0.12
Air blast 1860. Syneresis complete 5 days 0.46
Air blast 13°c. Dried to glassy 8 days 0.33

solid

It has been found that drying a gel rapidly before syneresis
had started should be avoided, as there is apparently a mechanical in-
Jury to the gel structure due to the rapid removal of water resulting in
a gel which when activated had relatively low porosity and mediocre ad-
sorptive capacity for benzene vapors. It is thought that syneresis
effects to a certain degree the final orientation of the colloidal
particles. If this is allowed to proceed to commotion, a gel of super-
ior activity is formed. An arbitrarily selected point was chosen in the
progress of the syneresis of a silicic acid gel which was defined as that
at which syneresis was complete. This was said to be comPIete when a gel,
which had been covered to avoid evaporation from the time it was mixed,

failed to squeeze out any water during a twenty-four hour period. Under

-22-

usual laboratory conditions a gel (made from mixing equal volumes of
sodium silicate (specific gravity 1.18) and 3.9 N hydrochloric acid and
allowed to set in a shallow crystallizzing dish) required five or six days

to reach this point.

wagg;nc

The object of washing a gel of silicic acid should be to remove
as much of the soluble electrolytes as possible without injuring the phys-
ical structure of the gel. A series of gels was made using equal volumes
of 3.8 N hydrochloric acid and a sodium silicate solution having a speci-
fic gravity of 1.18. They were allowed to stand covered at room tauper-
ature until syneresis was canplete, and then washed free of chlorides.
Further drying was done at ordinary temperature. The gels were activated

at the usual temperature.

ELLE.
ZEFEECT OR THE use or HOT WATER IN WASHING A.ohi UPON THE
mm: VAPOR ADSORPTIVE CAPACITY

Air stream (320 c.c. per minute) saturated with benzene at 12.5°C.

Temperature of adsorption hath 2.50 c.

 

Method of Teaperature' Time Grams of 0 H6 Ad-
sorbed r
Washipg of Water Hours gram of gel.
Refluxed 90° 0. 12 0.300
Running Distilled 10° c. 24 0.4.62
320

-23..

Althougi no quantitative analysis was made, it was found that
washing could be stepped with no loss of activity at the point where no
further precipitate of silver chloride formed but while there was still
enougi chlorides present to give a translucent Opalescence. Refluxing
With hot water removed the chlorides rapidly but a loss in'adsorptive
activity for benzene vapors resulted if carried to the point at Which there

was no further trace of chlorides observable.

mas.

The porosity of a gel of silicic acid depends largely upon the
method in which the soft gelatinous gel is dehydrated. A freshly pre-
pared gel in which syneresis has not commenced, if placed in the blast of
an electric fan at room temperature, soon shows signs of rapid dehydra-
tion. The surface of the gel becomes covered with a multitude of fine
checks and if this drying is continued the entire gel breaks up into fine
pieces. These pieces, if washed free of chlorides and activated, showed
a very low adsorptive capacity for benzene vapors.

Table VII shows the results of drying a series of gels in various
ways. The gels were made using the standard concentrations (3.8 normal
hydrochloric acid and an equal volume of sodium silicate solution having
a density of 1.18) . After syneresis was complete, the gels were washed
in running cold distilled water until the washings were only slightly
Opalescent With silver chloride. After drying the gels were further ac-
tivated as previously described at 200° 0. and compared as to their ad-

sorptive capacity for benzene vapors.

-m-

W-
armor OF meson or DRYING SILICA GELS UPON mm:
mm; ansoapmvn caracxn

Air Stream (520 c.c. per minute) saturated with benzene at 12.50 0.

Temperature of adsorption bath 25° C.

 

Method of Time of Grams of 06H5
Adsorbed per

mm Drying gram of Gel

18°— 25° 0. in still air 72 hours 0.330

40°— 55° 0. .. " " 24 " 0.250

7 inch fan 18° 0. 35 n 0.460

13 n .. n .. 12 .. 0.260

The ill effects of rapid drying are not confined to unwashed
gels but have been observed in Cl" free gels in which syneresis has been
completed. There is apparently a mechanical inJury to the gel structure
due to the rapid removal of the water. A moderate blast of air at roan-
temperature however, if it is so controlled that there is a uniform shrink-
ing of the gel pieces unaccompanied by any checking of the surface, re-
sults in a very active gel. 1

A seven inch electric fan was used for the purpose of drying
the gels, it being arranged so that the blast was directed downwards.
Different methods of placing the gels in this blast were tried, using
cheese cloth racks, shallow evaporating dishes, and glass jars six inches
deep. It was found that if the gel pieces were spread out on the cheese
cloth racks, that drying took place much too rapidly. The best results

were obtained by drying the gel in the glass Jars. These glass jars were

-25-

later replaced by a large rectangular glass vessel having five inch
walls. The gel lumps were spread out on the bottan of this dish in a
layer approximately 3 cm. thick. Under these conditions it required
nearly thirty-six hours to dry 1800 grams of fresh gel to a point where
no further change in volume occurred.

One of the factors in drying a gel, Which is usually beyond
the control of the average laboratory, is the relative humidity of the
air. It has been found that much more active gels have been produced
during rainy periods than otherwise. As a means of correcting this
difficulty, it was found that if a ten inch evaporating dish filled

with water was placed near the fan, better results were secured.

W

The process of activating silica gel consists essentially of
removing the excess of water held in the capillaries. It has been found

(51)

by Patrick that for the adsorption of sulphur dioxide there was a

minimum limit for the amount of water held by the gel beyond which there

(52)
has shown that as the tem-

was a loss in activity. J. van Bemmelen
perature of activication is raised, there is a decided loss of adsorp-
tive activity as soon as the gel particles show signs of sintering.
Holmes“) in a recent contribution stated that the amount of water lost
at 300° 0. was almost the same as at 150° C. Ruthsso’ working in this
laboratory, has danonstrated that there was a progressive loss in adsorp-
tive capacity for dry ammonia gas as the temperature of activication was

raised above 250° 0.

-25-

The effect of the tauperature of activication upon the adsorp-
tive capacity of a very active gel for benzene was investigated. {the

results are given in Table VIII.

TABLE EQLI
EFFECT Ob‘ THAI WWW 0F ACTIVICATION UPON THE ADSORPTIVE
CAPACITY OF SILICA GEL FOR BENZEME VAPORS
Air stream (320 c.c. per minute) saturated with benzene vapors

at 12.5° 0.

Temperature of adsorption bath 25° C.

 

Temperature of - Gram of Water Grams of 0636
Adsorbed Adsorbed
Aotivication Per Gram of Gel per Gram $2.631...
25° 0. 0.150 0.320
110° n 0.080 0.380
150° " 0.072 0.482
250° n 0.070 0.460
800° n 0.068 0.450

The same precautions in activating a fresh sample of silica
gel must be observed that have been established in drying the recently
prepared soft gel. The preliminary tauperatures mist be low and a too

rapid dehydration must be avoided.

W

The factors influencing the physical structure of silica gel

have been found to depend largely upon the mode and extent of syneresis

-27...

and also upon the method of drying.

Washingza crude gel free of its acid and salt after syneresis
had reached completion produced a very active gel. This gel had an ad-
sorptive capacity fer benzene vapors of 0.462 grams of benzene per gram
of gel, while one dried to a rigid solid in the Open at room.temperatures
before washing had an adsorptive capacity of 0.330 grams of benzene per
gram of gel.

It has been found that an active gel can‘be produced by drying
at room.temparatures by'means of the air blast from an electric fan. The
force of the air blast should be so regulated that there is a gradual
shrinkage of the gel particles and that no checking of the gel surface
occurs.

For the adsorption of benzene vapors, there is no gain obtained
by activating the gel at temperatures above 200° 0.

The concentration of the acid and water glass solutions used in
preparing the gel should be such that it requdres an appreciable time
(at least an.hour) for the gel to set. This effect is probably due to
the fact that the colloidal aggregates can better orientate themselves
into a homogeneous structure.

Prolonged washing with hot water in order to remove all traces

of chlorides tends to lower the adsorptive capacity of a gel.

mm M 0 egg 0F :31me AClD

 

In the first pert of this discussion of the factors influencing
the activity of silica gel, only those have been considered which affected

its physical structure per as.

-m-

The two factors influencing adsorptive activity are, of course,
the extent and the nature of the surface. Various methods have been used
to change the nature of the surface. Patricia” has develOped methods of
depositing metallic colloids on the surface of the gel and in this way
radically change the prOperties of the gel. Bohringer and Schwarz‘“’
have produced a highly dispersed colloidal sol of silica by dissolving
silica in aqueous ammonia.

Ammonium silicate has not been isolated as a solid although

(85)

Struclnnan has shown that gelatinous silica is soluble to the extent

of 0.02 grams per 100 grams of a 5% solution of amonium carbonate.

86)

Schwarz( has found that the solubility of silica in aqueous amonia

depended upon the concentration and temperature of the anmonia and also
upon the physical conditions of the silica. Schwarz ani Liedese’n using
quartz vessels to avoid contamination by the alkalies from glass, found
that in a concentrated solution of ammonia equilibrium was established in
seventy-two hours, that 64% of the silica was in actual solution, 8% in a
colloidal dispersion, and 28% undissolved.

(65) by adding

The gelatinous precipitate produced by Struckman
an ammonium carbonate solution to one of sodium silicate, When thoromly
washed free of carbonate ions, retained 0.78% of ammonia after being thor-

(68) claimed however that all of the ammonia

ougily air dried. Liebig
could be washed from the gel. Schwarz(66) has stated that it is difficult
to determine Whether the ammonia neutralizes the silicic acid or forms an
adsorption complex with it.

It has been found in this laboratory that a gel possessing un-

usual qualities can be produced by treating a crude gel after syneresis

-29..

was complete with concentrated aqueous ammonia, and after permeation,

which should not exceed two hours, washing the gel free of chlorides.

The following method of preparing this ammo gelhas been found to give
the most satisfactory results.

To a 3.8K solution of hydrochloric acid an equal volume of a
water glass solution having a specific gravity of 1.18 was added with
vigorous agitation by means of compressed air. The water glass used in
this work was a water-white product having an original specific gravity

of 1.402 and an Na 0 3 SiO ratio of 1.3.46.

2 2
The hydrosol of silicic acid formed in this way was poured into

shallow crystallffng dishes so that the layer of gel formed was not more
than three centimeters thick. The dishes were then covered with large
watch glasses to prevent excessive evaporation and allowed to stand until
syneresis was complete. This point was arbitrarily selected as that at
which no more liquid was squeezed from a covered gel during a twenty-four
hour period. The concentration of water glass and hydrochloric acid used
in this work required from three to four hours to set and syneresis was
complete within seven days.

When syneresis was commete, the gel was broken into lumps with
edges approximately three centimeters long, placed in a wide mouthed bottle,
and covered with a concentrated solution of amnonium hydroxide having a
specific gravity of 0.90, containing 255 grams of NHS per liter. “t the
end of the first half hour the ammonimn hydroxide solution was decanted
ani replaced with fresh.

The anmonium hydroxide penetrated the gel pieces uniformly as

was evidenced by the changing color. The yellow green color of the orig-

«'50-

inal. clear gel was replaced by the Opalescent appearance of the ammo-gel.
After a period of not more than two hours, at which time even the largest
gel pieces had been invaded by the ammonia, the solution was poured off
and the gel placed in a dry rack.

The terminology contributed by Grahamus)

to the science of
colloid chemistry has been preserved largely in its original form. It
has been thought best to follow his lead in taming the gel produced by
treating the unwashed hydrochloric acid gel of silicic acid with concen-
trated ammonia an ammo gel of silicic acid.

The activity of this gel depends to a great extent upon the
method of drying. To prevent too rapid drying the gel pieces should be
spread upon the bottom of a container having sides of at least five inches
in a layer not less than five centimeters thick. If the gel is preperly
dried, the pieces will show at; uniform shrinkage with no checking of the
surface. During the process of drying there is a definite cycle of changes
in the physical appearance of the gel. The bluish 0palescence of the fresh
ammo gel disappears, followed in turn by a condition in which the surface
of a fresh fracture is dull; this is again followed by a stage in Which
the fresh fracture has a bright glassy appearance. If the gel is dried
much beyond this stage, the concentration of the salt solution held in
the capillary spaces increased to such a point that crystallization takes
place in sig, shattering the gel pieces and inJuring its activity.

Washing is started when the majority of the gel pieces show
this glassy fracture. Due to the fact that a silica gel so treated with
amnonium hydroxide is soluble in hot water, all of the washings should be
done with cold water. As the chloride content of the gel falls to a cer-

tain level, which varies with the mode of drying, the solubility of the

-31-

amnonium silicate increases and the point at which the preliminary wash-
ing should step is that at which there is only a slight cloudiness of
silver chloride after the amenium silicate has dissolved with an excess
of nitric acid.

The citron yellow precipitate formed when a solution of silver
nitrate is added to a solution of a soluble silicate is, according to
Havikinssw’ silver meta silicate, AgZSiOS. As this silver silicate is
decomposed by nitric acid, the procedure in testing for chlorides to
follow in washing these gels is to add nitric acid to the test solution
after the addition of the silver nitrate. The amount of ammoniun 1w-
droxide in the wash water is relatively very small, and consequently
the amount of silver chloride which it would dissolve is below that of
the minimum concentration of chloride necessary to inhibit the solubil-
ity of the ammonium silicate.

Washing is stepped at the point where a decided yellow preci-
pitate is forned, and the gel further dried until it reaches its final
degree of hardness. During this final drying, the gel passes fran a
clear water-white appearance to an Opaque condition which finally changes
to a clear bluish epalescence.

If the maximum amount of chlorides consistent with the solu-
bility of the ammonium silicate has not been removed, sane of the larger
gel pieces will not pass into the final clear stage, but will remain

Opaque. A final washing with iced, distilled water will correct this.

-32-

PHYSICAL Psomagmg or rm Aime-Gap OF sxycuc sow

 

The gel, prepared by treating a crude silica gel after syner-
esis is complete with concentrated aqueous amnonia, is a clear opales-
cent substance somewhat softer than the ordinary silica gel. When ground
to a fine powder, there is a decided increase in the apparent volume due
either to the adsorption of air or to some influence of the mania.

This can be danonstrated by heating some of the powder in vacuo to s
tennperature of 150° 0., at which tenperature it loses its fluffiness

and occupies relatively the same apparent volume as that of the untreated
gel.

The action of the dehydrated gelin water differs from the
Patrick gel in that, instead of an obvious displacement of air by the
water accelnpanied by a decided splitting of the gel particles, there is
scarcely any displacement and the water permeates the gel structure uni-
formly without splitting. The fact that water does not displace air fran
the dehydrated gel strengthens the contention that the amonia has entered
into the structure of the gel.

Wham strongly heated over a blast flame, the ordinary silica gel
splits and swells into a bright "pep-corn" appearance, apparently due to
the sintering of the silica. The anmo-gel, however, when heated does
not swell but, instead, loses volume and assunes a dull, lusterless form.
me ash of silica gel is slightly acidic to methyl orange, while that of

sumo-gel is alkaline to phenolphthalein.

-33-

Iflflé'fiTIGATIQfl OF am: Q§ORPTIVE PEDPE-RTIBfi 0F M04513;

The outstanding property of this ammo-gel is its ability to
decolorize aqueous solutions of basic dyes, especially methylene blue.
The methylene blue solution used in this work was of a 400th molar con-
centration, prepared from a high-grade zinc-free sample.

the gel, activated at 150° 0., was so pulverized that it
would pass through a 200 mesh sieve and kept in ground glass stoppered
bottles until used. A half gram saniple was weighed in an 100 c.c.
riherlemeyer flask and the methylene blue solution added to it from s.
burrette. The flask was then stappered with a cork which had been pre-
viously impregnated with paraffin. After vigorous shaking for two min-
utes, the solution was poured onto a #3 3711an filter paper and the fil-
trate caught in a clean test tube the dimensions of which were 2 x 17 cm.
Complete decolorization was defined as that of a "solution in which no
color could be detected looking through a column eight centimeteres long
in a standard dimension test tube towards a "day-light" screen.

Table I1 shows a comparison of the decolorizing preperties of

Borite, ammo-gel, and silica gel (Patrick).

 

TAQ‘ E
Cubic Centimeters of Degree of
Adsorbent used Amount 400 th Molar
Methylem Blue 401. Decoloriz.
Silica Gel 10 grams 5 Partial
Norite 0.5 " 55 complete
Anna-gel 0.5 " 45 Complete

-34- -

Because of the retention of an appreciable amount of free am-
monia by the activated gel, it was thought that the decolorization might
be due to this factor. The addition of a concentrated aqueous solution
of ammonia to the methylene blue solution only intensified the purple

tint of the dye.

If washing is continued to any extent after the chlorides
have been removed, the gel particles become more Opalescent, finally
reaching a condition in which they are no longer translucent. With this
increase in the opalescence of the gel, there is a progressive falling
off of its adsorptive capacity for benzene vapors,.but this is not ac-
companied by a marked decrease in its decolorizing capacity.

Another factor that effects the benzene adsorption but has no
effect upon the decolorization capacity of the gel is age. If a tightly
stOppsred bottle containing activated gel is allowed to stand for several
weeks, it has been found that at the! end of this period the benzene
adsorptive capacity has fallen off by at least 50% While there has been
only a loss of 4.57. in the decolorization capacity.

Table X gives the results of the effect of aging of an ammo-

gel upon the decolorizing and benzene adsorptive capacity.

-35..

 

 

c.c. of Methylene Blue Grams of c H adsorbed per
400th Molar Solution De- Gram of Ge? from an Air Stream

Age colorized by mg Grams 320 c.c. peg minute, satura-

of Gel at 25 C. ted at 12.5 8., adsorbed at
25 Q.

mall 45 0.0. 0.36

1 Week 45 " 0032

2 Weeks 40 " 0.28

4 Weeks 35 " 0.26

8 Weeks 35 n 0018

The effects of treating a freshly prepared silica gel with con-
centrated aqueous ammonia are very marked, simulating those of syneresis
in many Ways. There is a decided shrinkage in volume achpanied by an
increase in hardness, but the activity of a gel so treated does not com-
pare in any way with one in which syneresis has been allowed to proceed in
a normal manner.

Table XI sumarizes the effects of different treatments upon the
adsorptive capacity of ammo-gel for benzene and water vapor together with
its decolorizing action upon a 400th molar metlwlene blue solution. In
the case of water vapor, air at the rate of 320 c. 0. per minute was satur-

ated with water at 18° c. and passed through the gel maintained at a tem-

perature of 25° C.

TABLE 3; ~
ADSORPTION 0F VAPORS AT 25° 0. BY into-ems Pmpazdm BY

DIFFERfiNT METHODS.

C.C.400th.Molar Grams of water Grams of Ben-

Methylene Blue Adsorbed per zene Adsorbed

Solution De- Gram of Gel from per Gram of Gel
Juethod colorized Air Saturated at from Air Satur-

180 CL, ages at 12.50 g.

 

1.

2.

3.

4.

Tree ted with NH
as soon as Gate 22 0.18 0012

Treated with N3
after completio

of syneresis. 45 0.42 0.38
Treated with 6H HNO

after dehydration. 3 0.68 0.24
Patrick 691. 2 0038 0028

It was thought that the mechanical advantages gained by the

peptizing action of the ammonia upon the fresh gel might be retained and

the further action of the ammonia checked by treating the ammonified gel

with an acid. Investigations of this treatment were carried out and it

was found that while the adsorptive capacity of the gel for benzene and

water vapors were enhanced by neutralizing the monia, the decoloriza-

tion capacity was eliminated.

gunman}: ,

The effect of a concentrated solution of aqueous amonia upon

a crude silica gel has been investigated. This has shown that while the

adsorptive capacity for benzene vapors of the dehydrated gel have not

been materially changed, there has been a decided change in the decolor-

izing capacity for solutions of basic dyes.

.37-

The effect of acid upon this ammonified gel has been to des-
troy the decolorizing preperties but to enhance its adsorptive capacity
for water vapors. There has been shown that there is a progressive de-
crease in the adsorptive capacity of the sumo-gel for benzene vapors
with age.

The decolorizing action of these sumo-gels is directly cor-
related with the presence of ammonia in the gel structure. The apparent
relation between pore size and decolorizing action is not clear since a
gel such as No. 1, Table XI, with a poor adsorptive capacity for benzene

vapors still shows a relatively high decolorizing activity when cmnpared

with the ordinary gel.

438-

1.

2.
3.

4.

5.
6.
7.
8.
9.

10.

11.

13.

14.

15.

16.

17.

18.

19.

BI BLIOGRAPHY

W. A. Patrick, Inaugural Dissertation, Gottingen (1914); U. 8.
Patent 1,297,724 (March 18, 1919); 1,335,348 (March 30, 1920).
H. Briggs, Proc. Roy. Soc., _1_Q_Q_A_, 88-92 (1921).

H. A. Fells and J. B. Firth, J. Phys. Chem., 29., 241,249 (1925).
H. N. Holmes, J. Ind. Eng. Charm, ll, 280-282 (1925); _];_8_, 386-
388 (1926).

R. Fruhling, Chem. 2., .12. 117—121 (1895).

P. Walden, Z. Koll., 6, 2331-235 (1910).

A. Baume, Ghymie experimental et raisonne, (Paris) 1., 329 (1773).
II. Watts, Dictionary of Ghenistry, 3rd Edition, page 240 (1883).
A. Noble, Proc. Norwich Soc., 1867.

J. J. Berzelius, Lehrbuch der Chemie (Dresden) _2_, 122-126 (1833);
Ann. Chim. Phys. (2) _1_g_, 366-370 (1830).

J. J. “belman, Ann. Chim. Phys. (3) 1.6, 129-131 (1846).

A. 0. Becqueral, Compt. Rend., 53., 1196-1202 (1861).

M. Kroger, K011. 3., pg, 16-18 (1922).

J. F. Spencer and K. Proud, K011. 33., 33;, 36-40 (1922).

T. Graham, Phil. Trans., (London), _l__5_l_, 181-224 (1861).

T. Graham, J. Chem. Soc., I_l_5_, 246-254 (1862); J. Chem. Soc., 11,
618-626 (1864).

J. Tyndall, Proc. Roy. Soc., ll, 223-226 (1869); Phil. Mag” (4)
31, 384-387 (1869); Proc. Roy. Inst., 6, 365,367 (1871).

E. M. Plessey, Compt. Rend., _4_1_, 599-603 (1855).

B. Monier, Compt. Rend., §_5_, 1053-1054 (1877).

459-

20.
21.
22.

23.

25.
26.

27.

29.
30.

31.

33.

34.

35.

36.

37.

S. meunier, Compt. Rend., 1125 953-954 (1891).

H. N. H01mes, J. Phys. Chem.,.§2, 510-519 (1918).

H. N. Holmes, J. Am. Chem. Soc., 3;, 1329-1333 (1922).

Gmelin-Kraut, Handbuch der anorganische Chemie (Heidelburg), page
143 (1912).

w. Fleming, 2. Phys. 0118111., 5;, 427-457 (1902).

IE. Jordis, z. anorg. Chem., 44, 200-207 (1905).

R. Zsigmondy and W. Heyer, Z. anorg. Chem., 68, 169-187 (1910).

M. Wunder and A. Suleimann, Am. Chem. Ana1., _1_9_, 45-49 (1914).

J. 11. von Bernmelen, J. Prak. Chem, 23, 324-326 (1881); Ber., 13,
1466-1469 (1880).

IE. Jordis, Z. Eflectrochem.,.;;, 835-841 (1905).

in. Jordis andlfi. H. Kanter, z. anorg. Chem.,‘§§, 16-22 (1903).

F. Mylius and 1d. Groschuff, Ber., 352., 116-125 (1906).

V. Lenher and H. B. Merrill, J. Am. Chem. Soc., 32, 2630-2635 (1917);
v. Lenher, J. Am. Chem. Soc., 513, 391.393 (1921).

J. WtJMellor, A Treatise on Quantitative Inorganic Analysis, (London)
page 175 (1913).

n. Frany, Ann. Chim. Phys., (3) 38-42; 328-331 (1853).

L. Doveri, Ann. thn. Phys., (3), 21, 40-43 (1847); Liebig's Ann.,
pi, 256-259 (1847).

J. N. van Fuchs, Liebigs Ann..,§§, 119-123 (1852).

0. Roberts, J. Chem. Soc., 21, 274-277 (1868).

J. M. von Bemnelen, J. Prak. Chem" (2) _2_3_, 324-326 (1881); Z. anorg.
Chem., _1_§_, 233-356 (1898); 2. anorg. Chem” _1_§,, 14-36 (1898); z.

anorg. (menu, _1_8_, 98-146 (1898); Z. anorg. Chem. 32, 265-279 (1902);

39.

41.

42.

45.
46.

47.

49.
50.
51.

52.

53.

55.

Z. anorg. Chem. 39,, 380-402 (1903); Z. anorg. Chem. _5_9_, 325-347
(1908); Z. anorg. Chem. 62,, 1-23 (1909); Chem. Week-b., 9, 63-82
(1909).

a. Jordis, z. angew. Chem., 1.9.. 1697-1702 (1906).

R. Zsigmondy, Z. anorg. Chem., 2;, 356-377 (1911).

T. Garnelley and J. Walker, J. Chem. Soc., 53., 59-101 (1888).

J. W. Mellor, A Comprehensive Treatise on Inorganic and Theoretical
Chemistry (London), _6, page 293 (1925).

E. M. Faber and H. G. Olson, Chem. Met. ling" _2_8_, 805-806 (1923).
0. Tschermak, 2. Phys. Chem., _5_§, 349-367 (1905); 2. anorg. them,
55, 101 (1907); Zentralblatt Min. 0601., 225-230 (1908); Z. anorg.
Chem., §_3_, 230-274 (1910).

o. Mugge, Centr. Min., 129-134 (1908).

0. Suida, Setzber Akad. Wien., 113., 725 (1904).

A. Sabeneff, J. Russ. Phys. Chem. Soc., 2;, 515-525 (1890), 23,
80-91 (1891). 1

J. 8. Anderson, 2. Phys. Chem.,h_8_8_, 191-228 (1914).

hi. Earl and W. Urban, Z. angew. Chem, 36., 57-62 (1923).

3. F. Ruth, M. Sc. Thesis, Mich. State College (1925).

0. Butschli, Verb-Rat. Med. Ver., (Heidelburg) (2) 6, 287-301 (1901).
W. Biltz and W. Geibel, Rachr. K. Ges. Wiss., (Gottingen), 141-156
(1906).

R. Zsignondy, 23. anorg. Chem., 1;, 356-377 (1911).

W. Bachmann, 35. anorg. Chem, 13, 125-172 (1911).

P. P. van Weimarn, Grundzuge der Dispersoidchemie (Dresden) page 30

(1911) Rev. Gen. 0011., _2_, 193-195 (1924).

-41-

O
1 . c
p _n. 0 O
O
P. O O
. (. .
I
7 '
-—- ‘
7° O
'4 . I
v o. O
I
\e.
f. O O O
.F- e
0

53340404

Mill“

28 i;

I

56.
57.
58.

59.

63.

64.

R.

H.

J.

J.

Pappada and C. Sedowsky, Z. K011., _6_, 292-297 (1910).
c. Ghatelier, 001npt. 38:18., 147, 660-662 (1908).
M. von Bemmelen, Z. anorg. Chem., 1__8_, 98-146 (1898).

W. Mellor, A Comprehensive Treatise on Ino rganic and Theoretical

Chemistry (London), _6_ pages 304-307 (1925).

ii.

VI C

Schwarz and F. Stowener, K011. Beiheft., _12, 171 (1924).

A. Patrick, J. Am. Chem. Soc., 43., 946-978 (1920).

J. M. von Demelen, Arch. Near. Sci. exact list., (2) _6_, 607-624

(1901) .

7’?
"ll 0

C.

C.

R.

R.

J.

J.

A. Patrick, British Patent, 212,034 (1923); 212,035 (1923).

F. Boehringer and R. Schwarz, German Patent D.R.P. 373,110 (1920).
Struclnnann, Liebig's Ann., 24, 341 (1855).

Schwarz, Ber., _42, 2359 (16) 531;, 601 (1918).

Schwarz and O. Liede, Ber., 53;, 1509-1513 (1920).

von Liebig, Liebig's Ann., 24;, 375 (1855).

D. Hawkins, Am. J. Science, (3) 32, 311-312 (1890).

-42-

)Il)‘|))|“|))|)|l 111111)) )

 

31293 02244 8074