C. Remigius Fresenius.

Quantitative chemical analysis online

. (page 10 of 69)
Online LibraryC. Remigius FreseniusQuantitative chemical analysis → online text (page 10 of 69)
Font size
QR-code for this ebook

paratively short time quite large vessels to a point where the
pressure is exactly equal to the tension of aqueous vapor at the
temperature of the water, i.e., 7 to 11 mm. of mercury.

After connecting i with the pump at ~k by means of a rubber
tube, the filtration is effected by first pouring the supernatant
liquid into the filter, and afterwards the precipitate. The liquid
at first runs off in a rapid stream, then in rapidly flowing drops.
It is advisable to keep the filter filled to the edge. The precipi-
tate will be compressed by the pressure to a thin layer, permeated
by channels. As soon as the fluid has ceased to drop, and the first
channels are visible, the precipitate will adhere so firmly to the
paper that it will not be disturbed on adding water carefully.
The washing is completed by filling the funnel 1 cm. above the
edge of the filter, not using a wash-bottle, however, but pouring
the water carefully down the side from an open vessel. When the
filtration has been completed, which is usually the case in from one
to four fillings of the funnel, the filter is allowed to drain com-
pletely, when it will be obtained half-dry so that it may be fre-
quently transferred without further manipulation, together with the
precipitate, to the crucible, and ignited (see 52). It will be
readily seen that, by this method of filtering, the operation is
greatly shortened. In this process, the filters employed may be
smaller than usual, because the precipitates obtained under pres-
sure occupy less space than is otherwise the case ; besides, precipi-
tates which are ordinarily very difficult to wash can be completely
washed with a relatively small quantity of water. Finally, the
half-dried condition in which precipitates are obtained after com-
plete withdrawal of the water from them, permits them to be easily
and completely separated from the filter, and free from fibres.

"We have described above the hydraulic air-pump in its most
perfect form, in which it is applicable for use not only in filter-
ing, but also for all other purposes (exhausting desiccator jars,




stc.) ; for pumps, however, which are intended only for filtering
and drawing air through drying-closets (28), a fall of 10 to 15
feet of water will suffice. Such a fall may be readily obtained

Fig. 65.

even on the ground floor and without a deep drain, by affixing the
pump to the upper part of the wall. The cock on the water sup-
ply pipe must then, of course, be placed where it can be manipu-
lated from the floor ; and the air exhaust pipe should be connected
with a thin lead pipe which is conducted down to the work-table,
and fastened as shown in Fig. 65. In this a is the lead pipe lead-
ing to the air-pipe ; b is the rubber tube connecting the flask a by
means of the glass tube c ; d is a glass tube of suitable height, dip-
ping into mercury at 0, and serving as a manometer ; f is a glass
tube to which is connected the rubber tube g closed by the pinch-
cock h. On connecting this tube with the. suction- tube of the
filtering-flask, starting the pump, and opening A, filtration begins.
Slighter but still quite effective suction for filtering purposes is
also obtainable without the aid of air-pumps or aspirators. Such
apparatus is shown in Figs. 66 and 67. The apparatus illustrated
in Fig. 66 is recommended by WEIL.* On applying suction at 0,
the liquid is raised in A to any desired height, thereby effecting

* Zeitschr.f. analyt. Clem., H, 359,




filtration through the suction caused by the raised column of liquid.
The filter may be strengthened by first inserting a small filter, a,

Fig. 67.

Fig. 66.

into the funnel, and then the filter-paper, J. The filters should lie
close to the funnel, and be free from folds.

The apparatus shown in Fig. 67 was devised by PICCARD.*
If the column of water is not given a greater length than 30 cm.
the filter need not be reinforced. However, it is always advisable
even in this case to use the small extra filter. If the filter lies
close to the funnel and is free from folds, filtration is much more
rapid (10 to 12 times, according to PICCARD) than when the bent
tube is used.

The use of exhaust apparatus renders it possible to dispense
entirely with paper filters, and to effect filtration through asbestos
or glass powder ; in cases where the precipitates must be dried at a
certain temperature and then weighed, such, arrangements for filter-
ing are frequently very useful.

* Zeitschr. f. anatyt. Chem. , iv, 47.

108 OPEKATIOtfS. [ 48.

Fig. 68 is a filter-tube which is recommended for the
weighing of small quantities of anti-
mony sulphide.* The tuhe is charged
at a with long-fibred asbestos which
is then washed with water to free it
from all the finer particles, next fixed per-
JL|| pendicularly in a support, suction applied
at J, and finally heated by suitable means
until the asbestos has been completely
dried. It is then weighed, after which it
is fixed at 5 in the perforated stopper of
the filtering-flask, when a small funnel is
inserted at *, and slight suction applied,
the fluid to be filtered being carefully
poured in.

Fig. 69 is an apparatus recommended
by "W. GIBBS and TAYLOR f for a
similar purpose. The tube contains at a
first fragment of glass, then glass or
sand in coarse powder and finally in fine
Fig. 68. Fig" 69. powder.




In the case of precipitates which, from their gelatinous nature,
or from the firm adhesion of certain coprecipitated salts, oppose
insuperable, or, at all events, considerable obstacles to perfect wash-
ing on the filter, the following method is resorted to : Let the
precipitate subside as far as practicable, pour the nearly clear super-
natant liquid on the filter, stir the precipitate up with the washing
fluid (in certain cases, where such a course is indicated, heat to
boiling), let it subside again, and repeat this operation until the
precipitate is almost thoroughly washed. Transfer it now to the
filter, and complete the operation with the washing-bottle (see
46). This method is highly to be recommended ; there are
many precipitates that can be thoroughly wa>liel only by its

* Zeitschr. f. analyt. Chem., vm, 154.
f Silliman's Amer. Jour., (n) XLIV. 215.

49.] FILTKATION. 109

in cases where it is not intended to weigh the precipitate
washed by decantation, but to dissolve it again, the operation of
washing is entirely completed by decantation, and the precipitate
not even transferred to the filter. The re-solution of the bulk of
the precipitate being effected in the vessel containing it, the filter
is placed over the latter, and the solvent passed through it.
Although the termination of the operation of washing may be
usually ascertained by testing a sample of the washings for one of
the substances originally present in the solution which has to be
removed (for hydrochloric acid, for instance, with silver
nitrate), still there are cases in which this mode of proceeding is
inapplicable. In such cases, and indeed in processes of washing by
decantation generally, BTTNSEN'S method will be found convenient
viz., to continue the process of washing until the fluid which
had remained in the beaker, after the first decantation, has under-
gone a ten thousand- fold dilution. To effect this, measure with a
slip of paper the height from the bottom of this beaker to the
surface of the fluid remaining in it, together with the precipitate,
after the first decantation ; then fill the beaker with water, if
possible, boiling, and measure the entire height of the fluid ;
divide the length of the second column by that of the first. Go
through the same process each time you add fresh water, and
always multiply the quotient found with the number obtained in
the preceding calculation, until you reach 10000.


Before proceeding to weigh a precipitate, it still remains to
convert it into a form of accurately known composition. This is
done either by igniting or by drying. The latter proceeding is
more protracted and tedious than the former, and is, moreover, apt
to give less accurate results. The process of drying is, therefore,
as a general rule, applied only to precipitates which cannot bear
exposure to a red heat without undergoing total or partial volatili-
zation ; or which leave upon ignition residues having MO constant
composition ; thus, for instance, drying is resorted to in the case
of mercuric sulphide, arsenous sulphide, and other metallic sul-
phides ; and also in the case of silver cyanide, potassium-platinio
chloride, etc,


But whenever the nature of the precipitate (e.g., barium sul-
phate, lead sulphate, and many other compounds) leaves the
operator at liberty to choose between drying and heating to red-
ness, ignition is almost invariably preferred.

aa. Drying of Precipitates.

"When a precipitate lias been collected, washed, and dried on a
filter, minute particles of it adhere so firmly to the paper that it is
found impossible to remove them. The weighing of dried precipi-
tates involves, therefore, in all accurate analyses, the drying and
weighing of the filter also. Formerly the collection of precipi-
tates to be dried was frequently done in two filters of equal size,
one within the other, the outer one being removed and used as a
counterweight to the filter containing the precipitate. It was
assumed that filters of equal size were also of equal weight.
This assumption is, however, inadmissible in accurate analyses,
as every experiment proves that two even very small filters of
equal size differ in weight by as much as 20, 30, or even
more milligrammes. To obtain accurate results, it is neces-
sary to dry and weigh the filter before using it ; the temperature at
which the filter is dried must be the same as that to which it is
intended subsequently to expose the precipitate. Another condi-
tion is that the filtering-paper must not contain any substance
liable to be dissolved by the fluid passing through it.

The drying is conducted either in the water-, air-, or oil-bath,
according to the degree of heat required. The weighing is per-
formed in a closed vessel, mostly between two clasped watch-
glasses (Fig. TO), in a platinum crucible, or in two glass tubes,
each sealed at one end, and placed one within the other, as
shown in Fig. 71. When the filter appears dry, it is placed
between the warm watch-glasses, or in a warm crucible, allowed
to cool under a bell-glass over sulphuric acid, and weighed.
The reopened crucible or watch-glasses, together with the filter, are
then :i-:iin exposed for some time to the required degree of heat,
and, after cooling, weighed once more. If the weight does not
differ from that found at first, the filter may be considered dry,
and we have simply to note the collective weight of the watch-
glasses, clasp, and filter, or of the crucible and filter.




After the washing of the precipitate has been concluded and
the water: allowed to run off so far as possible, the filter with the

Fig. 70.

Fig. 71.

precipitate is taken off the funnel, folded up, and placed upon
blotting-paper, which is then kept for some time in a moderately
warm place protected from dust ; it is finally put into one of
the watch-glasses, or into the uncovered platinum crucible, with
which it was first weighed, and exposed to the appropriate degree
of heat, either in the water-, air-, or oil-bath. "When it is judged
that the precipitate is dry, the second watch-glass, or the lid of the
crucible is put on (with the clasp pushed over the two in the former
case), and the whole, after cooling in the desiccator, is weighed.
The filter and the precipitate are then again exposed, in the same
way, to the proper drying temperature, allowed to cool, and
weighed again, the same process being repeated until the weight
remains constant or varies only to the extent of a few deci-milli-
grammes. By subtracting from the weight
found the tare of the crucible or watch-glasses
and filter, we obtain the weight of the dry
precipitate. [The filter must not be dried too
long, as it slowly loses weight, and even be-
comes brown from decomposition when heated
to 100 for days together.]

It happens sometimes that the precipitate
nearly fills the filter, or retains a considerable
amount of water; or sometimes the paper is so thin that its re-

Fis:. 72.

112 OPERATIONS. [ 61.

moval from the funnel cannot well be effected without tearing.
In all such cases, the best way is to let the
filter and precipitate get nearly dry in the
funnel, which may be effected readily by
covering the latter with a piece of blotting
paper* to keep out the dust, and placing
it, supported on a broken beaker (fig. 72),
or some other vessel of the kind, on the
steam-apparatus or sand-bath, or stove, or
on a heated iron plate. For support to a

funnel while drying a hollow frustum of a cone open both ends,
made of stoneware or tinned iron (Fig. 73), is convenient. Two
sizes may be used, 10 cm.^and 12 cm. high respectively. The
lower diameter should be from 7 to 8, the upper from 4 to 6 cm.

II. Ignition of Precipitates.

It was formerly customary to first dry the precipitate
together with the filter, then to scrape it clean from the latter
into the crucible, and ignite it. In spite of the most careful
scraping, however, some of the precipitate was inevitably lost by
adherence to the filter. Experience has shown that more accurate
results are obtained by igniting the filter with the precipitate and
deducting the weight of the filter-ash from the weight found.

If care be taken to make the filters always of the same paper,
and to cut every size by a pattern, as advised in 45, an, the
quantity of ash which each size yields upon incineration may be
readily determined. It is necessary, however, to determine sep-
arately the quantity of ash left by ordinary filters, and that left by
filters which have been washed with hydrochloric acid and water ;
on an average the latter leave about half as much ash as the former.
To determine the filter ash take ten filters (or an equal weight of
cuttings from the same paper), burn them in an obliquely placed
platinum crucible, and ignite until every trace of carbon is con-

* Turned down over the rim. Or more neatly as follows : Wet a com-
mon cut filter, stretch it over the ground top of the funnel, nnd then gently
tear off the superfluous paper. The cover thus formed continues to adhera
after drying with some force.


sumed ; then weigh the ash, and divide the amount found by ten ;
the quotient expresses, with sufficient precision, the average quan-
tity of ash which every individual filter leaves upon incineration.
In the ignition of precipitates, the following four points have
to be more particularly regarded:

1. No loss of substance must be incurred ;

2. The ignited precipitates must really be the bodies they are
represented to be in the calculation of the results ;

3. The incineration of the filters must be complete ;

4. The crucibles must not be attacked.

The following two methods seem to me the simplest and most
appropriate of all that have as yet been proposed. The selection
of either depends upon certain circumstances, which I shall imme-
diately have occasion to point out. But no matter which method
is resorted to, the precipitate must always be thoroughly dried,
before it can properly be exposed to a red heat. The application
of a red heat to moist precipitates, more particularly to such as are
very light and loose in the dry state (silicic acid, for instance),
involves always a risk of loss from the impetuously escaping
aqueous vapors carrying away with them minute particles of the
substance. Some other substances, as aluminium hydroxide or
ferric hydroxide, for instance, form small hard lumps ; if such
lumps are ignited while still moist within they are liable to fly
about with great violence. The best method of drying precipitates
as a preliminary to ignition is as described in 50, the last
paragraph. These methods may also, according to circumstances,
be modified ; for instance, BUNSEN * has pointed out that when
a precipitate has been sufficiently washed and freed from its
excess of water by means of the hydraulic air-pump, it may fre-
quently be at once ignited together with the filter, and without
further drying. This is more particularly described in 52.

Respecting the ignition, the degree of heat to be applied and
the duration of the process must, of course, depend upon the
nature of the precipitate and upon its deportment at a red heat.
As a general rule, a moderate red heat, applied for about five
minutes, is found sufficient to effect the purpose ; there are, how-

* Annal. d. Chem. u. Phar., CXLVIII, 285; ZeitscJir, f. anatyt, Chem. YIII,

114 OPERATIONS. [ 51.

ever, many exceptions to this rule which will he indicated where-
ever they occur.

"Whenever the choice is permitted between porcelain and
platinum crucibles, the latter are always preferred, on account of
their comparative lightness and infrangibility, and because they
are more readily heated to redness. The crucible selected should
always be of sufficient capacity, as the use of crucibles deficient in
size involves the risk of loss of substance. The proper w/e, in
most cases, is 4 cm. in height, and 3*5 cm. in diameter. That the
crucible must be perfectly clean, both inside and outside, need
hardly be mentioned. The analyst should acquire the habit of
cleaning and polishing the platinum crucible always after using it.
This should be done, as recommended by BUNSEN, and more
lately by ERDMANN, by friction with moist sea-sand, the grains of
which are all round and do not scratch. The sand is rubbed
on with the finger, and the desired effect is produced in
a few minutes. The adoption of this habit is attended with
the pleasure of always working with a bright crucible and
the profit of prolonging its existence. This mode of cleaning
is all the more necessary, when one ignites over gas-lamps, since
at this high temperature crucibles soon acquire a gray coat-
ing, which arises from a superficial loosening of the platinum.
A little burnishing with sea-sand readily removes the appear-
ance in question, without causing any notable diminution of the
weight of the crucible (ERDMANN).* The foregoing remarks
on platinum crucibles refer equally to those of indium-platinum
which, by the by, are now much used, and very highly to be recom-
mended only the restoration of the polish is somewhat more diffi-
cult with the latter, on account of the greater hardness of the alloy.
If there are spots on the platinum or indium-platinum crucibles,
which cannot be removed by the sand without wearing away too
much of the metal, a little potassium disulphate is fused in the
crucible, the fluid mass shaken about inside, allowed to cool, and
the crucible finally boiled with water. There are two ways of
cleaning crucibles soiled outside ; cither the crucible is placed in ;i
larger one, and the interspace filled with pntaium disulphate,
which is then heated to fusion; or the crucible is placed on :i
platinum-wire triangle, heated to redness, and then sprinkled over

* Jour.f. prakt. Chem., LXXIX. 117.


with 'powdered potassium disulphate. Instead of the disulphate
yon may, use borax. Never forget at last to polish the crucible
with sea-sand again.

"When the crucible is clean, it is placed upon a clean platinum-
wire triangle (Fig. 74), ignited, allowed to cool in the desiccator,
and weighed. This operation, though not indispensable, is still
always advisable, so that the weighing of the empty and filled cru-
cible may be performed under as nearly as possible the same cir-
cumstances. The empty crucible may of course be weighed after
the ignition of the precipitate ; however, it is
preferable in most cases to weigh it 'before.
The weighing of the crucible after ignition
of the precipitate is only then necessary
w r hen this requires to be subjected to the
action of the gas blowpipe for some time,
since experience has shown that in this case
the platinum crucible often loses weight.* &'

If a platinum triangle is wanting, one of iron wire may be used,
only that part coming in contact with the platinum crucible
should be covered with platinum foil or wound with platinum
wire, or pieces of clay tobacco pipes may be slipped over the
iron wire.

The ignition is effected with a BERZELIUS spirit-lamp or a gas-
lamp, or else in a muffle. In igniting reducible substances over
lamps, the analyst must always be on his guard against the con-
tact of unconsumed hydrocarbons even in covered crucibles.
"When gas-lamps are used there is especial need of caution in this
respect. Reduction will be avoided if the flame is made no larger
than necessary, if the crucible is supported in the upper part of
the flame, and if, when the crucible is in a slanting position, it is
heated from behind.

"We pass on now to the description of the special methods.

* According to WITTSTEIN platinum crucibles lose weight on i.nition only
when they still contain small quantities of osmium (Zeilschr.f anal. Chem.,
v, 98). STOLBA ascribes the loss of weight to formation ot platinum carbide,
and states that the rougher the surface, the greater will be the loss (Poly I. Jour. t
cxcvm, 177).




FIRST METHOD. (Ignition of the Precipitate with the Filter?)

This method is resorted to in cases where there is no danger of
a reduction of the precipitate by the action of the carbon of the
filter. The mode of proceeding is as follows :

The perfectly dry filter, with the precipitate, is removed from
the funnel, and its sides are gathered together at the top, so that
the precipitate lies enclosed as in a small bag. The filter is now
put into the crucible, which is then covered and heated over a
spirit-lamp with double draught, or over gas very gently, to effect
the slow charring of the filter; the cover is now removed, the
crucible placed obliquely, and a stronger degree of heat applied,
until complete incineration of the filter is effected ; the lid, which
had in the meantime best be kept on a porcelain plate, or in a por-
celain crucible, is put on again, and a red heat applied for some
time longer, if needed ; the crucible is now allowed to cool a little,
and is then, while still hot, though no longer red hot,* taken off

Fig. 75.

Fig. 76.

Fig. 77.

with a pair of tongs of brass or polished iron (Figs. 75 and 76),
and put in the desiccator, where it is left to cool ; it is finally

* Taking hold of a red-7iot crucible with brass tongs might cause the forma-
tion of black rings round it.


The combustion of the carbon of the filter may be promoted,
in caseswhere it proceeds too slowly, by pushing the non-consumed
particles, with a smooth and rather stout platinum wire, within the
focus of the strongest action of the heat and air. And the oper-
ator may also increase the draught of air by leaning the lid of the
crucible against the latter in the manner illustrated in Fig. 77.

It will occasionally happen that particles of the carbon of the
filter obstinately resist incineration. In such cases the operation
may be promoted by putting a small lump of fused, dry ammonium
nitrate into the crucible, placing on the lid and applying a gentle
heat at first, which is gradually increased. However, as this way
of proceeding is apt to involve some loss of substance, its applica-
tion should not be made a general rule.

in cases where the bulk of the precipitate is easily detached
from the filter, the preceding method is occasionally modified in
this, that the precipitate is put into the crucible, and the filter,
with the still adhering particles, folded loosely together, and laid
over the precipitate. In other respects, the operation is conducted
in the manner above described.

As above stated, precipitates quite thoroughly freed from
water by aspiration may, according to BTJNSEN,* be at once ignited
without further drying, but, of course, only when the precipitates
are not reducible by the filter-carbon. The process is carried
out as follows : The portion of filter-paper free from precip-
itate is tightly wrapped round the remainder of the filter in such

Online LibraryC. Remigius FreseniusQuantitative chemical analysis → online text (page 10 of 69)