hydrolysis of a great number of salts formed of strong acids and
weak bases; the acids being crystalloids and the bases colloids, the
first alone permeating the membrane of the dialyzer.
Finally, we can, in certain cases, bring into evidence the hydroly-
sis of salts of weak acids or bases, by the following process, based
upon the phenomena of absorption (described in Chapter I, i).
We have seen that filter paper has a greater absorbing power
for bodies with basic properties than for those of an acidic nature.
If, then, a few drops of a dilute solution of a strongly hydrolyzed
salt be allowed to spread upon a sheet of filter paper, the hydrate
will be fixed upon the paper by absorption at the center of the
disc, formed by the spreading of the liquid; while the water, con-
taining no longer anything but acid, will withdraw toward the
periphery, and, with a proper reagent, one can determine the zone
beyond which the hydroxide has not passed. I have thus ascer-
tained that a concentrated solution (10 per cent) of lead acetate
spreads out uniformly upon the filter paper. If we mark the limit
obtained and place the paper in an atmosphere of hydrogen sulphide,
the disc turns uniformly black ; but, with a 0.5 per cent solution, the
result is entirely different, the hydrogen sulphide blackens nothing
more than the center of the disc upon which the hydroxide of lead
is precipitated by absorption, and the water, which is absolutely free
from lead acetate, has diffused several millimeters beyond the
clearly defined zone.*
All the cases of hydrolysis enumerated above, which are placed
beyond a doubt by incontestable chemical reactions, are accompanied
by heat effects (positive or negative, according to the case) in a
ratio with the proportion of the salt decomposed by water. It is
seen that between an hydrolysis like that of bismuth chloride, be-
longing in the category of very active chemical reactions with a
large evolution of heat, up to the hydrolysis of ammonium salts of
* This experiment, which is easy to repeat with very dilute solution of
numerous salts of heavy metals, explains why washing alone is incapable of
removing the oxide thus precipitated, which has formed a union with the
paper; this, in many cases, necessitates washing with acidulated water, in
order to eliminate completely from the filter and from the precipitate the
oxide of the metal contained in the filtrate (washing of barium sulphate pre-
cipitated in the presence of ferric solution, according to the Arnold method
for the determination of sulphur in iron, cast iron and steel, of lead sulphate
precipitated in the presence of copper, etc.)
BASED UPON CHEMICAL REACTIONS 61
the strong acids, almost imperceptible in the calorimeter, all inter-
mediate cases are encountered and are demonstrated by irrefutable
phenomena.
In a general way, experiments show that hydrolysis increases
with the dilution, whatever may be the sign of thermal change. It
generally increases with the temperature, because these reactions
almost always absorb heat which is the habitual case of simple dis-
sociations into free acid and base. When heat is liberated (the cases
of hydrolysis of the chlorides of antimony and of bismuth giving
insoluble oxychlorides) the hydrolysis diminishes with the tempera-
ture conformably to Le Chatelier's principle of the opposition of
the reaction to further change, which has been formulated in all its
generality by its author, precisely apropos of the anomaly which
the hydrolysis of antimony chloride appeared to present in compari-
son to other known hydrolyses like that of mercuric sulphate.*
The speed with which hydrolysis occurs is extremely variable.
The limit is obtained almost instantaneously within the calorimeter,
for all the decompositions in which the water is not itself decom-
posed and forms only soluble hydroxides with the two parts into
which the molecule of salt is subdivided. This is the case of all the
hydrolyses of inorganic salts that do not precipitate upon solution,
as ammonium salts, sodium borate, etc., where the heat effect is
produced in a few seconds, as in the neutralization of acids by bases.
It is not the same in the case in which the molecule of water is to
divide in order to give with its elements, new groupings with those
of the hyclrolyzed bodies, as in the hydrolysis of ferric chloride, the
saponification of esters by water, etc. :
Fe 2 Cl e +6H,0 = Fe,(OH) 6 -f 6HC1
CH 8 COOCH 8 +H 2 .= CH 8 COOH+CH 8 OH.
Reaction is then decreased in general by the preliminary work of
decomposition of the intervening molecules of water and the limit
is often long in being attained, notably in the hydrolysis of esters.
However, in the analogous case of salts of bismuth and of anti-
mony, the limit is reached in the cold in a few minutes, to such a
degree that one can determine quantitatively the bismuth in the
form of the oxychloride by precipitating with an excess of water.
The same is true for ferric acetate, when precipitated hot in the
*H. Le Chatelier, C. R., c, 737 (1885).
62 METHODS OF ANALYTICAL CHEMISTRY
form of the basic salt. On the other hand, the decomposition of
stannic and titanium oxysalts is accomplished by boiling for a very
long time. We will see later some very important applications in
analytical chemistry of these unequal speeds of hydrolysis.
Hydrolytic dissociations are, moreover, phenomena of equilib-
rium, in which the direct reaction is limited by the reverse reaction,
and are easy to realize by putting- the products of the direct reaction
in contact in water. The verification of the reversible character as
related to the concentration factor is very simple for salts which
hydrolyze instantaneously, like the alkali borates. For example, the
absorption of heat, when its volume of water is added to a solution
of ammonium borate, containing one equivalent in four liters, is
i.o Calorie and, if the ammonium borate is produced directly by
the saturation of the acid and base, the following is obtained :
H 3 BO 3 (i eq. = 2l)+NH 8 (i eq. = 2!) liberates 9.44 Calories
H 3 BO 3 (i eq. =4l)+NH 8 (i eq. = 4!) liberates 8.44 Calories
Difference i.oo Calorie
The limit of combination is then rigorously the same as the limit
of decomposition for the same dilution. With hydrolyses causing
insoluble precipitates, the same verification is rarely possible, on
account of the physical transformation of the precipitate (poly-
merization, without doubt), in the cases, and these are most fre-
quent, where the precipitate is of amorphous (colloidal) nature, and
becomes rapidly insoluble in free acid, having lost the function of a
normal hydroxide. The studies of Le Chatelier, upon the decompo-
sition of mercuric sulphate by water, in which the precipitate is of
crystalline nature and does not undergo physical modifications,
leave in all cases no doubt of the influence of the concentration upon
the character of reversibility.
The same observations also show the dependence of these re-
versibilities upon the temperature. Salts of simple hydrolysis, giv-
ing only soluble products, return to the initial state after heating, as
Berthelot has shown in a number of cases by the determination of
the heat liberated by the addition of one equivalent of sodium
hydroxide to one equivalent of the salt before heating and after
cooling. The heat liberated is the same. But, if there is produced
by hydrolysis, an amorphous precipitate which is modified by heat-
ing, or by standing for a considerable time, as in the case of ferric
salts, necessarily the system no longer returns to the initial state
BASED UPON CHEMICAL REACTIONS 63
after long heating. For example, ferric acetate, which is very
readily hydrolyzed when heated for some minutes at 100, then
cooled to the ordinary temperature, is entirely decomposed, for
potassium hydroxide gives an evolution of +12.72 Calories and it
is only at the end of a rather long time that the acetic acid has
redissolved a very small part of the basic acetate precipitate. On
the other hand, heating for a short time does not keep the sulphate
and even the ferric chloride from returning to the initial state.*
Measurements of the conductivity made by M. Foussereau, of solu-
tions of ferric chloride at different temperatures lead to the same
conclusions.f
Summing up, we can deduce from the numerous calorimetric
measurements of salts in aqueous solutions that, in general, all salts
are more or less hydrolyzed into free acids and bases (or, occasion-
ally, into complex compounds such as salts called basic). Salts
formed by a strong acid and a strong base are but very slightly
decomposed (sulphates, nitrates, and chlorides of potassium and
sodium for example). Salts formed by a strong acid and a weak
base or reciprocally (ferric chloride, sodium borate), are notably
decomposed by water. Finally, salts formed by a weak acid and a
weak base (ferric acetate) are very extensively decomposed by
water, and, at times, practically completely.
If we designate Q the heat of formation of a dissolved salt,
starting from the acid and base dissolved in a little water, q the
heat liberated for a definite dilution, then the ratio -. will represent,
as a general rule, the fraction of the salt hydrolyzed. As it is
necessary to consider the heats of solution as well as the heats of
dilution of the free acid and base, it is often quite difficult to
determine the correction to give to the values determined by the
calorimeter, in order to obtain the exact value oi-~. As a result
the calorimeter rarely gives the exact limit of hydrolysis and, in
fact, it is by other methods applied only to hydrolysis of very slow
speed that the study of these limits has been possible for a number
of reactions otherwise restricted ; but the general significance of the
* M. Berthelot, loc. cit., II, 284 and following.
t Foussereau, Ann. Chim. Phys., (6) xi, 383 (1887) ; (6) xii, 393, 553
(1887).
64 METHODS OF ANALYTICAL CHEMISTRY
phenomenon is no less firmly established by them, and that suffices
in order to understand the processes of double decomposition.
Mutual Action of Two Salts. With this notion of hydrolysis of
salts, we ought to consider the aqueous solutions of an inorganic
salt as containing the non-hydrolyzed salt in equilibrium with the
greater or less fraction of free acid and base. If, then, we intro-
duce another acid into the solution, it will combine with the free
base and the equilibrium will be destroyed; a new quantity of the
salt in solution will be decomposed, and so on until a state of
equilibrium is established. At this moment there will be in solution
two salts, non-hydrolyzed and a certain portion of the two free
acids and free base.
Likewise, if we mix the solutions of two salts differing in acid
and base, there will be a mutual combination of the free acids and
bases, causing a new decomposition of non-hydrolyzed salts, and
finally, there will be established a new equilibrium among the four
non-hydrolyzed salts in presence of a certain proportion of the two
free acids and free bases, an extremely small proportion if it is a
question of salts of strong acids and bases, and a notable decomposi-
tion, on the contrary, with salts of weak bases and acids. The
reaction can be represented by the equation:
i <->
A and A t being the acid radicals, B and B x the basic radicals,
and it being understood that the salts are partially hydrolyzed into
free acids and bases.
As previously indicated, if we call Q the quantity of heat which
would correspond to the total transformation of the first system into
the second and q the quantity of heat liberated by mixing the two
salts, the expression -~- gives, with the restrictions already seen,
apropos of hydrolysis, the relative proportions of the bodies of the
first system, which were transformed into bodies of the second
system.
The neutralization of strong acids by strong bases, liberating
almost equal quantities of heat, was first shown by Hess. From this
it follows that double decompositions of neutral salts of strong acids
and bases, stable in the presence of water, do not occasion an
appreciable heat effect, even when the transformation is practically
BASED UPON CHEMICAL REACTIONS 65
complete. The thermal effects are more important in proportion as
the stability of the two salts in the presence of water is more differ-
ent from what we have seen previously concerning the classification
of salts in regard to this stability.
As it is a question here of essentially reversible reactions, in
which there is consequently the production of no non-compensated
work, the principle of maximum work necessarily does not find any
application in double decomposition of salts, and the transformation
of one system of two salts into the opposite system is done either
with evolution or with absorption of heat. What controls the direc-
tion of the transformation is the tendency toward the formation of
the most stable salt in the presence of water, and also, as an inevit-
able consequence, to the correlative formation of the least stable
salt, more or less hydrolyzed. As the numerous experiments of
Berthelot prove, "the strong acids unite by preference with strong
bases, leaving the weak bases to the weak acids."* It is, moreover,
to be noted that the more stable salt is the one which corresponds
to the maximum liberation of heat, but it does not follow that the
sign of the heat of the reaction ought necessarily to be positive. One
of the most striking examples in this respect is the action of potas-
sium carbonate on ammonium sulphate, whose equimolecular quan-
tities absorb 6.36 Calories. The total transformation into
potassium sulphate and ammonium carbonate would correspond to
6.81 Calories. There is, then, more than nine tenths of the sys-
tem transformed, although the reaction is strongly endothermic.
Experiments show that the transformation is often much more
complete when one of the bodies can be eliminated in the form of an
insoluble precipitate or a volatile compound, without, however, this
physical property being the determining cause of the transformation,
for, in a great many cases, it takes place in the inverse direction.
For example, in the reaction :
PbSO 4 +2NaC 2 H 3 O 2 = Na 2 SO 4 +Pb(C 2 H 3 O 2 ) 2 ,
it is the insoluble lead sulphate that is dissolved by the sodium ace-
tate, and, in the still more numerous cases of insoluble bodies, car-
bonates, phosphates, etc., being acted upon by strong acids. These
reactions are used frequently in analytical chemistry. However, we
will see further, as an application of the numerical law of equilib-
* Loc. cit., II, p. 712.
5
66 METHODS OF ANALYTICAL CHEMISTRY
rium that every time it is possible, insolubility or volatility plays a
very important part in accentuating the direction of the transform-
ation.
Equilibrium in double decomposition of salts, is attained in
general with great rapidity, as is shown by the reactions effected in
the calorimeter where the maximum or minimum temperature is
attained in a few seconds. This is true for reactions in which all the
bodies remain soluble, and, for a great number of double decompo-
sitions giving crystalline precipitates. For example, in the action of
sulphuric acid on barium chloride, of ammonium oxalate on soluble
salts of calcium, etc., and, if, in other cases, we allow the precipitate
to digest in the hot mother liquor, it is less to complete the precipi-
tate than to increase the size of the grains and to facilitate the filtra-
tion (see Chapters I and II). If, in certain cases of simple double
decompositions (consisting in the pure and simple exchange of
bases and acids), we are obliged to wait some time for the
appearance of the expected precipitate or for the precipitation to
become complete, that depends upon the fact that the insoluble
precipitate begins by being more soluble, whether on account of its
minute state of division, and that its supersaturation must be com-
pletely destroyed in order that the equilibrium be established, or
because it forms first in the colloidal state, giving a pseudo-solution
which requires a certain time of contact with the salts or acids in
solution in order to be precipitated. It is especially true that retard-
ation of the formation of the precipitates occurs in precipitations
taking place in extremely dilute solutions (precipitation of traces of
sulphuric acid by BaCl 2 , of traces of calcium by ammonium oxalate,
of metals in extremely dilute solutions by hydrogen sulphide which
gives colloidal sulphur).
When the double decomposition is not simple and produces salts
formed of acid radicals or complex bases, necessitating, as a conse-
quence, an intermediate reaction of decomposition and, of recom-
bination of the elements of the bodies of the first system, the limit is
often much longer in being attained. Such is the case in the forma-
tion of magnesium ammonium phosphate from ammonium phos-
phate and magnesium chloride, of potassium chlorplatinate by the
action of potassium chloride on platinic chloride, of potassium
cobaltonitrite by the action of potassium nitrite on cobalt salts, etc.
But, taking these particular facts into account, and with the
same restrictions as for the phenomena of hydrolysis, in that which
BASED UPON CHEMICAL REACTIONS 67
concerns the possible modifications in the chemical function of pre-
cipitates, the reversibility of the phenomena of double decomposi-
tion in proportion to concentration and temperature, is placed be-
yond doubt, as a general rule, by the multitude of experiments and
numerous examples which will be given later.
2. Experimental Research upon the Numerical Law of
Equilibrium in Double Decompositions
We have just seen how, by the conception of the hydrolysis of
salts in aqueous solutions, one may conceive the exchange of acids
and of bases in the mixture of two salts. The calorimeter taught us
that these reactions are not generally complete and that they cause
equilibrium between bodies of two opposed systems; even, in cer-
tain cases, it permits the determination of the state of equilibrium
of the system when it is a question of almost instantaneous reac-
tions. But, in order that the data furnished by the calorimeter may
be utilized in ascertaining the conditions to be fulfilled in view of
the practically total precipitation of insoluble bodies, it is necessary
that we have precise conceptions concerning the influence of the fac-
tors of equilibrium, namely; the temperature t and the concentra-
tion c, of the reacting bodies. The other factors such as the pres-
sure, electric state, etc., remain generally constant and are not able,
consequently, to vary the state of equilibrium of the system.
The mathematical law f (c, c', c" > which gov-
erns the equilibrium in double decomposition, has engrossed the
minds of chemists for more than a century. It is by a series of
successive trials and improvements that it has been established, at
first experimentally, then in a theoretical manner, by basing it upon
the principles of thermodynamics, and thus bringing it in a form
which is probably only one stage toward the exact solution, but
which appears already sufficiently near, as we shall see, for opera-
tions of a practical character like those of analytical chemistry.
Historical. Bergmann is the first chemist who occupied himself
with explaining the mechanism of double decomposition in building
up his general theory of affinities, founded upon the mutual "attrac-
tion" of bodies. His theory may be summarized as follows: All
bodies exercise an attraction upon each other, the magnitude of this
attraction can be expressed by a definite number. Different bodies
have different attractions for the same body. If the body A has a
greater attraction for the body B than for the body C, the body B
will drive the body C from its combination with A or,
68 METHODS OF ANALYTICAL CHEMISTRY
AC+B = AB+C.
It is upon this hypothesis as a foundation that Bergmann estab-
lished in 1775 his celebrated Table of Affinity. In Bergmann's
theory, insolubility, volatility, and the relative concentration of the
reacting bodies have no influence upon the chemical reaction.
This theory contained manifest errors, and, in 1801 to 1803,
Berthollet developed a theory of affinity, in a contrary direction to
that of Bergmann, in the Annales de Chimie and in I'Essai de statique
chimique. This theory rests upon the following principles : Accord-
ing to Berthollet, all bodies have an affinity for each other, but this
affinity varies, according to the bodies under consideration, and de-
pends essentially upon the physical properties of the combination
which they can produce; cohesion (insolubility), and volatility
(elastic force of the gases resulting from the reaction). Bodies can
react upon each other only if their smallest particles are in close
contact, for example, in solution. The chemical action of bodies
depends upon the affinities and quantities of the two bodies ; a body
is withdrawn from chemical action each time that it is precipitated
in an insoluble form, or when it takes the gaseous form.
Berthollet did not fall into the error, with which he is wrong-
fully blamed, of believing that insolubility suffices to determine the
direction of the reaction, for he has called attention to the fact that
it can be produced in inverse order. Take, for example, the insol-
uble salts of calcium; "oxalic acid," said Berthollet,* "precipitates
as calcium oxalate only a part of the calcium which forms a neutral
combination with another acid. As soon as the acid of the combina-
tion has acquired a certain energy by the diminution of the base, it
counterbalances the effort of insolubility, and the calcium oxalate
ceases to separate. Again, the insolubility of the phosphate or the
sulphate of calcium is overcome much more readily ; a weak acidity
suffices to make the effect disappear."
One cannot define more clearly a state of equilibrium between
two reversible reactions, dependent upon the concentration of the
bodies in solution. But Berthollet's attention being doubtless
turned toward practically complete reactions, with insoluble precipi-
tates or with volatile products, he has not made any direct measure-
ment of the partition of the base between two acids in solution. He
contented himself with admitting without proof that this partition
takes place in all cases, proportionally to the reacting masses of the
* Berthollet, Essai de statique chimique, I, p. 78.
f'\ i>
OF THE
UW
BASED UPON CHEMICAL REACTIONS
69
acids (when, in fact it is produced according to an entirely different
law, as we shall see), and this error cast for a long time unmerited
discredit upon his very correct conception of the equilibrium be-
tween the two opposed states of a double decomposition.
It is barely half a century after the work of Berthollet that the
study of the law of this equilibrium was resumed. This study at the
very first bore upon the verification of the existence of a common
limit between the two opposed states of the system and upon the de-
termination of this limit. Malaguti* investigated a great number of
double decompositions by pouring the mixture of the solutions of two
salts into a large excess of alcohol. In all cases where two of the
four salts produced by the mutual reaction are insoluble in alcohol,
the analysis of the precipitate permits the easy determination of the
proportion of the salts of the first system changed into those of the
opposed system, if we admit that the addition of alcohol does not
modify the repartition of the acids among the bases and that the
precipitate well represents the state of the combinations in the
aqueous solution. This last point is rather questionable, a priori,
however the experiments of Malaguti seem, indeed, to justify this
hypothesis, the coefficient of partition having been found to be the
same when starting from the two opposed systems, as is shown by
the following results obtained by this author:
REACTING SUBSTANCES
MIXED (ONE EQUIVA-
LENT OF EACH SALT)
( Lead Nitrate
( Potassium Acetate. . .
(Lead Acetate
( Potassium Nitrate. . .
( Zinc Sulphate
(Potassium Chloride..
( Potassium Sulphate.
(Zinc Chloride
COMPOSITION AFTER PRECIPITA-
TION BY ALCOHOL
0.92 Potassium Nitrate and
Lead Acetate
0.08 Lead Nitrate and Po-
tassium Acetate
0.91 Potassium Nitrate and
Lead Acetate
0.09 Lead Nitrate and Potas-
sium Acetate
0.84 Potassium Sulphate and