S

Solubility CO2 in geological storage sites and remediation of contami-

nated sites. The concept is also needed in experimental sim-
Jean-François Boily ulations of geochemical reactions, such as in the study of
Department of Chemistry, Umeå University, Umeå, Sweden mineral growth, nucleation and dissolution, as well as in
laboratory (e.g., preparation, purification) procedures and
analytical determination of natural samples. This concept is
Definition even needed in molecular simulations of geological fluids for
determining the maximal number of ions that can be realisti-
Solubility is the maximal quantity that a substance can dis- cally inserted into a box of solvent molecules.
solve as a solute in a host solvent, forming a saturated solu- Solubility is expressed as a concentration for a given
tion. Any additional quantity occurs as a precipitate and temperature and pressure. The mass fraction (wi) of solute
results in a suspension of particles. Solubility is expressed in i in solvent solv is given by:
terms of concentration units for a given temperature and
pressure. si
wi ¼ (1)
si þ ssol

Introduction where si and ssolv are respectively the masses of the solute
and of the solvent. Other concentration units relate to wi
Solubility is an essential concept needed in many areas of through:
geochemistry to describe the ability of gases (g), liquids (l), or
wi
solids (s) to dissolve in substances. The resulting solutions are ½i ¼ 1000 rsolv for Molarity (2)
Mi
homogeneous mixtures of two or more substances, where the
solvent is typically, although not necessarily, one of high mole wi
fraction and the solute of lower mole fractions. Examples mi ¼ 1000 for Molality (3)
Mi ð1  wi Þ
include the dissolution of sodium chloride (halite) in water,
or of water in magmas. In contrast to miscibility where sub-

wi wi 1  wi 1
stances can mix in all proportions (e.g., water and ethanol), x2 ¼ þ for mole fraction (4)
Mi Mi Msolv
solubility pertains to the maximal quantity that can be
accomodated by the host solvent. This quantity is predomi- wi
nantly one determined by laboratory experiments, explained r i ¼ 100 for mass of solute per 100 g water (5)
1  wi
by molecular details of solute–solvent interactions, and pre-
dicted by thermodynamics. and where rsolv is the density of the solvent and M is
The concept of solubility is essential for understanding and molecular weight.
predicting numerous forms of mass transport processes in A conventional rule of thumb for rationalizing solubility
fluids of Earth’s surface and subsurface. Examples include involves the concept of polarity. A polar solute is soluble in a
fluid flow in geothermal systems and ore deposits, diagenesis, polar solvent, like water, but insoluble in a nonpolar solvent,
secondary porosity generation in sedimentary basins, contam- like benzene. Likewise, the solubility of an organic molecule
inant transport in aquifers, as well as injection of supercritical

# Springer International Publishing AG 2017

W.M. White (ed.), Encyclopedia of Geochemistry,
DOI 10.1007/978-3-319-39193-9_72-1
2 Solubility

in water can be largely controlled by the population and entropy during as substances randomly mix in a solution.
distribution of polar groups (e.g., carboxyls) on a molecule. Entropy is the driving force for dissolution when no heat is
On the molecular front, solubilization can be explained by the exchanged, such as in the mixing of organic liquids. However,
differences between solute/solvent interaction energies with when heat is exchanged, dissolution will be favored by reac-
those of solute/solute and solvent/solvent interactions. tions that are strongly exothermic to mildly
Intermolecular, ion-dipole, ion- and dipole-induced dipole endothermic. These reactions proceed until saturation is
as well as (London) dispersion forces govern the ability of a achieved, namely, at solubility, the point beyond which no
solvent to host solutes. Hydrogen bonding in water is notably additional solute can be accomodated by the solvent.
a well-known form of interaction solubilizing hydrophilic A thermodynamic description of solubility in non-ideal
substances. Solubility can be strongly affected by temperature (real) solutions begins with the chemical potential
 
due to changes in the hydrogen-bonding capability of water, mi ¼ ðdG=dN i ÞT , P, Nj of solute species i at temperature
explained through changes in its dipole moment (dielectric
constant) and self hydrogen bonding populations and T and pressure P:
structure.
In this article, thermodynamic expressions describing sol- mi ¼ mi ðT, PÞ þ RTgi xi ¼ mi ðT, PÞ þ RTai (8)
ubility are presented alongside several key examples also
detailing the molecular-level details of solute–solvent where mi* is the chemical potential of the ideal species, gi is
interactions. the activity coefficient, xi the mole fraction, and the activity is
given by ai = gi mi. When, say, a gas (g; myi ) is in equilibrium
with a liquid solution (l), the chemical potential of species
Thermodynamics of Solubility i will be identical in both phases:

The Gibbs free energy of formation of solutes generated by mi ðgÞ ¼ mi ðlÞ ¼ mi ðT, PÞ þ RT ln gi xi
the dissolution of a substance into a solvent at a given tem- ¼ myi ðT Þ þ RT ln Pi (9)
perature (T) relates to, just as all in chemical processes, the
enthalpy (H) and entropy (S) of the reaction, DG = DH  TDS. From this equation:
The enthalpy change resulting from this reaction is accounted
for by Hess’ law: Pi ¼ g i x i k i (10)

DHsoln ¼ DHsolute þ DHsolvent þ DHsolutesolvent (6) with

where DHsoln is the total heat of solution, DHsolute is the
 

mi ðT, PÞ  myi ðT Þ
(endothermic) heat involved in separating the solute particles ki ¼ exp (11)
RT
against intermolecular forces of the original substance (e.g.,
lattice energy of solids), DHsolvent is the (endothermic) heat Thus, if the mole fraction xi is 1, we have Raoult’s law
involved in separating the solvent molecules to accomodate (Convention I) where ki relates to the vapor pressure of the
the solute, and DHsolute–solvent is the (exothermic) heat solvent of dilute solutions or mixtures of similar types of
resulting from solute–solvent interactions. For the specific liquids.
case of the dissolution of a solid in water, hydration energy As many solutions cannot form high mole fractions of each
(DHhydr) is the sum of the DHsolvent and DHsolute–solvent terms substance, the main species present at high mole fractions is
and is exothermic when ion–water dipole interactions are referred as the solvent and the minor species the solute. As
stronger than the sum of water–water hydrogen bonds broken stated above, solubility is the composition of a solution
to accomodate the ion. containing the maximal quantity of a solute that can be incor-
The entropic contribution to these reactions pertains to the porated by a solvent. In this case, Convention II for activity
dispersion of solutes from a solid to the solvent. In a system of coefficients (g) stipulates distinct thermodynamic treatments
r components, the entropy of mixing is: for the solvent and the solutes. Thus, as in Raoult’s law, the
Xr chemical potential of the solvent is expressed through:
DSmix ¼ NkB x
i¼1 i
ln xi (7)
msolv ¼ msolv ðT, PÞ þ RT ln ðgsolv  xsolv Þ (12)
where N is the number of molecules, kB Boltzmann’s
constant, xi is a mole fraction (also, probability of finding a with gsolv!1 when the mole fraction xsolv!1. For the
component in a given site). As such, there is always a gain in solute, the chemical potential also becomes:
Solubility 3

msolute ¼ msolute ðT, PÞ þ RT lnð gsolute  xsolute Þ (13) K ¼ K sp ¼ aA x aC y (17)

except that gsolute ! 1 when xsolute ! 0. The chemical where the activities of Az- and Cz+ at saturation are used to
potential of the solute (msolute ) is that of the ideal (gsolute = 1) obtain Ksp. As such, when the ion activity product (IAP)
pure solute at xsolute = 1 for a given T and P. Under dilute product exceeds the solubility product (IAP > Ksp) precipi-
conditions are solute–solute interactions negligible, and tation occurs, but when it is below this value (IAP < Ksp),
msolute is dominated by solute–solvent interactions, and dissolution occurs. Concrete examples include the solubility
Equation 10 is then Henry’s law with the constant kHi = ki. of silica:
In this case, the chemical potential of the solute is one behav-
ing as an ideal liquid under infinite dilution (gsolute ! 1). SiO2 ðsÞ þ 2H2 O Ð H4 SiO4 (18)
Stated otherwise, at equilibrium, the mass of gas dissolved
in the solvent is proportional to its partial pressure in the with K sp ¼ aH4 SiO4 because aSiO2 ¼ 1 and aH2 O ¼ 1 and that
adjacent gas phase. The temperature dependence of kHi can of calcite with
be expressed via the van’t Hoff equation (e.g.,
h  i
kHi ¼ kHi ref exp C T1  T1ref , where ref denotes a reference CaCO3 ðsÞ þ Hþ Ð Ca2þ þ HCO3  (19)
(e.g., standard state) value and C a constant in Kelvin related
with K sp ¼ aCa2þ aHCO3 a1
Hþ
because aCaCO3 ¼ 1.
to the enthalpy of the solution.
As solubility is affected by the ability of the solvent to
When working within solutions only and the mole fraction
overcome the attractive forces of the solute molecules of the
of the solute i is sufficiently low, the chemical potential (mm, i)
incipient substance, temperature increases the kinetic energy
of species i is more conveniently expressed on a molality (mi;
of the solvent molecules and therefore generally favors solu-
moles per kg solvent) scale, such that
bility. This can be generally observed in inorganic salts
  (Figure 1). Still, retrograde (or reverse) solubility occurs for
mm, i ¼ mm, i ðT, PÞ þ RT ln gm, i mi (14) solids whose dissolution is exothermic, such as cerium sul-
fate, calcium carbonate, calcium sulfate, calcium phosphate,
and where the activity of i is am , i = gm , imi. In this case, and magnesium silicate.
the standard state is a fictious 1 molal (or molar) solution The pressure (P) dependence, in turn, is given by the
where gm, i = 1. This is a widely used convention in the differences in the partial molal volumes (DV) of the solids
realms of low temperature aqueous and hydrothermal geo- involved
chemistry. Activity coefficients can be neglected for solutions
containing a constant ionic strength or predicted using

K sp, P DV ðP  Pref Þ
extended Debye-Hückel (I < 0.1 m), Güntelberg log ¼ (20)
K sp, Pref 2:303 RT
(I < 0.1 m), or Davies (I < 0.5 m) theories. The
(semiempirical) Pitzer ion interaction model can alternatively While this dependence is typically negligible for Earth
be used for a wider range of ionic strengths, as, for instance, surface processes, it becomes essential for predictions of
simultaneously applied for various minerals of eight compo-
nent Na-K-Mg-Ca-H-Cl-SO4-OH-HCO3-CO3-CO2-H2O sys-
tem (Harvie et al. 1984).
Building upon this framework, the dissolution of, say, the
binary AxCy solid, with the soluble anion Az- and cation Cz+

Ax Cy ðsÞ Ð xAz ðaqÞ þ yCzþ ðaqÞ (15)

is predicted with the equilibrium constant

aA x aC y
K¼ ¼ gA x gC y  m A x m C y (16)
aAC

Taking the activity of AxCy to 1 (aAC = 1), the solubility

product is defined as

Solubility, Figure 1 Solubility of common inorganic salts in water.

4 Solubility

deeper Earth and deep sea processes. Examples include pres-

sure dissolution phenomena within solutions at grain bound-
aries under where pressure (cf. stylolites), or the enhanced
solubility of calcite in oceanic bottoms (cf. carbonate com-
pensation depth).
Prediction of solubility can be made using thermodynamic
parameters derived by experimentation and with values tabu-
lated in comprehensive reviews on the subject (e.g., IUPAC-
NIST Solubility Database). As a note of caution, databases of
solubility constants can contain large variations for the same
substances, such as minerals. These disprepancies can arise
from numerous factors including (1) variations in crystallin-
ity, (2) impurities or trace elements hindering congruent dis-
solution, (3) unaccounted complexing agents (e.g., pH buffers
such as potassium hydrogen phthalate), (4) differences in
equilibration criteria, (5) variations in ionic strength, and Solubility, Figure 2 Impact of salinity (NaCl) on solubility of oxygen
(6) heterogeneous particle size distributions. Ideally, solubil- in water.
ity constants of solids ought to pertain to highly crystalline,
pure, large monocrystalline particles of low specific surface released via organic solvent–solute interactions. Thus, fol-
area. Still, even for pure samples, particles of high specific lowing Le Ch^atelier’s principle, the solubility of gases in
surface area (cf. submicrometer-sized particles) may have water is generally smaller at larger temperatures where heat
shifted reaction energies. This shift can be expressed through is added to the system. Additionally, as dissolved salts effec-
and apparent (app) solubility constant with: tively compete for hydration waters, gas solubility is gener-
ally lowered by salinity, as, for example, shown in the case of
gA O2 solubility in aqueous solutions of NaCl (Figure 2) (Benson
log K s, app ¼ log K s þ (21)
3:454 RT and Krause 1984).
Gases undergoing reactions with water, such as CO2 and
where g is surface tension, A is the molar surface area of
NH3 are, on the other hand, more soluble. The case of CO2
the solute. Tightly bound water molecules on small (e.g.,
solubility in water
nanometer-sized) particles can impact surface tension and
therefore reaction energies. The impact of particle size on
CO2 ðgÞ Ð CO2 ðaqÞ (22)
the thermodynamics of deydroxylation reactions of iron
oxyhydroxides to iron oxides was notably demonstrated by is a pivotal process to many geochemical processes and
(Navrotsky et al. 2008). can be predicted through Henry’s constant (KH) with:

½CO2 ðaqÞ ¼ K H PCO2 (23)

Examples of Solubility
where KH relates to Dalton’s law of partial pressure
Gases in Water (KD = [CO2(aq)])/[CO2(g)] with ½CO2 ðgÞ ¼ PCO2 =RT and
Dissolution of gases in liquids require the accommodation of (KH = KD/RT). CO2 solubility in water over a wide range of
the host solvent for the gaseous molecules. Those reacting pressures and temperatures can be accurately predicted using
with the solvent will have greater solubilities than those that equations of states, such as the virial-like formulation of
are relatively inert. For example, the solubility of common Henry’s law by (Diamond and Akinfiev 2003) (Figure 3):
gases in geochemistry is greater in the order
He < N2 < O2 < CH4 < CO2 < NH3. Generally, the solubi-

RT 0
lization of a nonpolar gas in a solvent first involves an endo- ln ðkH Þ ¼ ð1  xÞln f 0H2 O
þ x ln r
M W H2 O
thermic component to make space for the molecule. An "
 #
exothermic component then results from the formation of a 0 1000 0:5
þ 2rH2 O a þ b (24)
hydrophobic hydration shell via van der Waals dispersion T
interactions (e.g., H2O gas), forming a clathrate-like environ-
ment. Reactions are generally more endothermic in organic
solvents than in water because the energy required to
accomodate space for the solute is larger, and less energy is
Solubility 5

Solubility, Figure 3 P-T

diagram of solubility isopleths of
CO2. Reproduced with permission
from Elsevier. Diamond LW,
Akinfiev NN (2003) Solubility of
CO2 in water from -1.5 to
100 degrees C and from 0.1 to
100 MPa: evaluation of literature
data and thermodynamic
modelling. Fluid Phase Equilib
208 (1-2):265-290. doi:10.1016/
s0378-3812(03)00041-4.

where x is a scaling factor (0.088) for predicting standard solubility increases strongly at pH values exceeding the first
partial molar volume of CO2, f 0H2 O is the fugacity (MPa), r0H2 O deprotonation step from carbonic acid to bicarbonate.
the density (gcm3) of pure water, and a and b are empirial fit
parameters predicting solubility over 1.5–100  C and Gases in Melts
0.1–100 MPa. The solubility of volatiles in melts can, in principle, be calcu-
Soluble CO2 (aq) can be regarded as a hydrophobe (e.g., lated via thermodynamic properties of a given decomposition
0.56 hydrogen bond per molecule in water (Lam et al. 2015; reaction (e.g., biotite ! feldspar + magnetite + water) involv-
Kumar et al. 2007)) embedded in a cylindrical void forcing a ing a rock of a given mineralogical composition at the time of
local ordering of water. Carbonic acid (H2CO3), which results crystallization. The study of fluid inclusions also supplies
from the association of CO2(aq) and H2O, represents however estimates for the pressure during fluid entrapment. At the
about 1% of the total soluble CO2 and has a short lifetime of same time, experimental studies provided direct means at
~300 ns. This lifetime was recently suggested to arise from assessing the ability of melts at solubilizing gases. For
deprotonation to bicarbonate (Adamczyk et al. 2009) and thus instance, water solubilizes in silicate melts as OH- and molec-
not to the backdissociation as often stated in textbooks. The ular (H2Om) species. A representative H2O dissolution reac-
total analytical concentration of dissolved CO2 is expressed tion for melts can be written as (Zhang et al. 2007):
as [H2CO3] = [CO2(aq)] + [H2CO3]. When accounting for
the deprotonation to bicarbonate and carbonate, the total H2 OðgÞ Ð H2 Om ðmeltÞ K1 (26)
concentration ([CO2]T)of CO2 in a solution is given by
[CO2]T = [H2CO3] + [HCO3] + [CO32]. Moreover, the H2 OðmeltÞ þ OðmeltÞ Ð 2OHðmeltÞK2 (27)
pH-dependent total concentration of CO2 in a solution
exposed to an open atmosphere relates to Henry’s law via: The pressure dependence of the total solubility H2O
(H2OT) does not follow Henry’s law as it follows the expres-
!
KH K1 KH K1 K2 sion of the type:
½CO2 T ¼ KH þ þ pCO2 (25)
½H þ  ½H þ 
2

where K1 and K2 are the first and second deprotonation

constants of carbonic acid, respectively. As a result, CO2
6 Solubility

1 1=2 a  AlðOHÞ3 þ 3Hþ Ð Al3þ þ H2 O (32)

½H2 OT  ¼ ½OH  þ ½H2 Om  ¼ k1 pH2 O þ k2 pH2 O (28)
2
involves the pH-dependent hydrolysis of this ion via:
The temperature-dependent equilibrium constant of reac-
tion n is represented by
qAl3þ þ pH2 O þ rLz Ð Alq ðOHÞp Lr 3qprz


XH2 Om DH 0n þ pHþ (33)
Kn ¼ 0 ¼ An exp  (29)
f H2 O RT
In the absence of a foreign ligand (Lz-), common species
where X is the mole fraction of H2O in the melt, fH2O is the AlOH2+ (p,q = 1,1), Al(OH)2+ (p,q = 2,1), Al(OH)3+
fugacity of water vapor, A is a preexponential factor, and DH0 (p,q = 3,1), Al3(OH)45+ (p,q = 4,3), and
is the standard state reaction enthalpy. Improved, yet semi- Al13O4(OH)24 (p,q = 32,13) control gibbsite solubility.
7+

empirical, expressions have been formulated to better account The pH-dependent concentration of aluminum is then given
for composition of Fe-free silicate melts such as (Zhang et al. by the sum of the concentrations of its aqueous species
2007): (q = 0):

 Xp Xq Xr h i
651:1 pffiffiffi ½Altot ¼ q Al q ð OH Þ p Lr
3qprz
(34)
Cw ¼ 0:231 þ P 0 0 0
T


32:57 and forms the characteristic U-shaped solubility curve of
þ 0:03424  þ 0:02447  AI (30) many metal (oxy)(hydr)oxide minerals in water and with the
T
solubility minimum controlled by the crystallinity of the
where Cw is the total solubility of water (weight %), P is minerals involved (Figure 4). The slopes (s) of the solubility
pressure (MPa) and AI is a compositional parameter given by curves on a log (base 10) [Al3+] vs. pH plot are the result of
AI = Na + K  Al given in mole fractions. Predictions of the pH-dependent speciation of the aqueous species, with
mixed volatile solubility (e.g., H2O-CO2) can be more readily s =  (3q-p)/q when a single species dominates the specia-
achieved via computer codes (Newman and Lowenstern tion (e.g., s = 3 for Al+ but 0 for Al(OH)30). Increased salt
2002). (e.g., NaCl) can favor more highly charged ions and therefore
enhance solubility. Addition of metal-complexing ligands can
Solids in Liquids considerably enhance solubility, such as in the case of fluoride
Dissolution of an ionic solid compound, such as NaCl, pro- forming various mixed aluminum-hydroxo-fluoro complexes
duced hydrated Na+ and Cl species each surrounded by a (Figure 4). Formation of (nano- to micro-meter thick) surface
number of hydration shells. The molar enthalpy of the coatings can, in turn, block dissolution reactions and therefore
resulting solution corresponds to the dissolution of one mole suppress solubility. One example includes aluminum hydroxo
of a molecule, such that fluoride precipitates when gibbsite is exposed to concentrated

NaClðsÞ ¼ Naþ ðaqÞ þ Cl ðaqÞ DH ¼ 3:9kJ=mol (31)

for pure NaCl(s), with a solubility of 35.9 g per 100 g of

water at 298 K. A saturated aqueous solution of NaCl thus
contains an average of ~9 H2O per NaCl molecule. It there-
fore not only consists of hydrated Na+ and Cl species but
also of solvent-shared and contact ion pairs (e.g., [Na (H2O)n
Cl]0). Molecular simulations of NaCl aqueous solutions at
elevated temperatures also suggest the possible presence of
polyatomic complexes of the type NaxClyx-y, whose formation
could be induced by the decrease in the dipole moment
(cf. dielectric constant) of water (Driesner et al. 1998).
Dissolution of solid compounds into water and whose
components undergo aqueous speciation reactions can dis-
play strongly pH- and complexation-dependent solubilities.
For instance, the dissolution of gibbsite (a-Al(OH)3) to Al3+: Solubility, Figure 4 Solubility of gibbsite (a-Al(OH)3) in water at
25  C with and without NaF.
Solubility 7

fluoride solutions. The fluoridation of apatite surfaces, Franck 1981). Accounting for variations in the solvent den-
forming a non-stoichiometric coating of low solubility, is sity (Dolejs and Manning 2010) provides advantages for
another example of how passivation of mineral surfaces rationalizing changes in solubility with temperature and pres-
leads to important changes in solubility. sure. This view accounts for the energetic contributions of the
Temperature and pressure effects on solubility are strongly compression of the hydration shell (electrostriction) on the
affected by changes in aqueous speciation. This directly arises solute – which are smaller in neutrally charged and larger in
from changes in the ability of water to hydrate ions. For highly charged aqueous species. As such, the solubility of
example, changes seen in metal (oxy)(hydr)oxides include minerals generating highly-charged species (e.g., Ca-bearing
shifts in the solubility minima to more acidic conditions calcite, apatite, fluorite, and portlandite of Dolejs and Man-
with temperature and to the increase in the concentration of ning (2010)) undergo important hikes in solubility in solvents
species of lower charge due to the decreased dielectric con- of greater density (at constant temperature). This dependence
stant (static permittivity) of water. Born theory can be used to is weaker in minerals generating neutrally charged species
predict the thermodynamics of solvation of ion i: (e.g., Si(OH)40, Al(OH)30 at circumneutral pH). Likewise, the
retrograde solubility of minerals generating charged species


1 tends to occur under conditions of lower solvent density (e.g.,
DGsolv ¼ oi 1 (35)
ϵr high temperature, low pressure).

where o is a species-dependent Born interaction parameter Solids in Gases

and ϵ r is the relative permittivity of the solvent. Knowledge of Although many solid phases of interest for geochemistry are
the latter value for water (Pan et al. 2013; Sverjensky et al. of low volatility, gaseous reactants can act as transport agent.
2014) enables use of Equations of State (e.g., Helgeson- For example, boiling of hydrothermal fluids can produce
Kirkham-Flowers model; (Helgeson et al. 1981) to predict species CO2, H2S, HCl, SiO2 as well as low molecular weight
thermodynamic properties of aqueous species at elevated metals complexes (e.g., AgCl(H3O)3(g); (Migdisov et al.
temperatures and pressures. For example, the solubility of 1999)) in high temperature steam. Transport of such species
quartz to several kb and hundreds of  C (Sverjensky et al. along a temperature gradient can drive crystallization reac-
2014) can be effectively described in this fashion (Figure 5). tions which can, for example, account for epithermal transport
An alternative approach stems upon the log-log relationship and deposition of ore deposits, such as those of gold (Hurtig
of equilibriums constants with the density of water and with and Williams-Jones 2014). Chemical vapor deposition
relative permittivity (Dolejs and Manning 2010; Marshall and

a −1 b 0.5
SOLUBILITY OF α-QUARTZ SOLUBILITY OF α-QUARTZ 20.0 kb
10.0 kb
−1.5 0
1.0 kb 5.0 kb

−2 −0.5
log molality

0.5 kb 2.0 kb
log molality

Psat.
−2.5 −1
Hemley et al. (1980)
1.0 kb
Morey & Hesselgesser (1951)

−3 Rimstidt (1997) −1.5

Hemley et al. (1980)
Crerar & Anderson (1971) Morey & Hesselgesser (1951)

Kenedy (1950) Kenedy et al. (1950)

−3.5 −2
Manning (1994)
Walther & Orville (1983)
0.5 kb Weill & Fyfe (1964)
Weill & Fyfe (1964) Walther & Orville (1983)
−4 −2.5
0 100 200 300 400 500 600 700 200 300 400 500 600 700 800 900 1000
TEMPERATURE (°C) TEMPERATURE (°C)

Solubility, Figure 5 Solubility of quartz (a-SiO2) as a function of temperature and pressure, and thermodynamic predictions. Reproduced with
permission from Elsevier. Sverjensky DA, Harrison B, Azzolini D (2014) Water in the deep Earth: The dielectric constant and the solubilities of
quartz and corundum to 60 kb and 1200 degrees C. Geochim Cosmochim Acta 129:125-145. doi:10.1016/j.gca.2013.12.019.
8 Solubility

methods are also used in the production of solid materials for Cross-References
technological uses.
R. Bunsen (Biot et al. 1852) first invoked vapor-driven ▶ Activity and Activity Coefficients
transport to describe the formation of crystalline Fe2O3 by ▶ Aqueous Solutions
HCl-bearing volcanic gases. In the absence of a transport ▶ Debye-Hückel Equation
agent, solid/gas transfers include direct sublimation ▶ Enthalpy
(e.g., AlCl6(s)ÐAl2Cl6(g)) and decomposition sublimation ▶ Entropy
(e.g., CuCl2(s)ÐCuCl(s) + ½ Cl2(g), followed by ▶ Equilibrium
CuCl(s)ÐCuCl(g)) reactions. In contrast, a transport agent, ▶ Equilibrium Constant
such as HCl(g), facilitates solid/gas conversion via a reaction ▶ Fluid–Rock Interaction
of the type: ▶ Fugacity
▶ Geochemical Thermodynamics
Fe2 O3 ðsÞ þ 6HClðgÞ Ð Fe2 Cl6 ðgÞ þ 3H2 OðgÞ (36) ▶ Gibbs–Duhem Equation
▶ Henry’s Law
The transport direction can be understood by the sign of ▶ Hydrothermal Solutions
the reaction enthalpy, in the sense that Le Ch^atelier’s principle ▶ Low-Temperature Geochemistry
will stipulate that endothermic transport proceeds toward ▶ Silicate Melts
cooler while exothermic reactions to warmer reaction. As ▶ Water
the reaction of Equation 36 is endothermic, transport of
Fe2Cl6(g) along a thermal gradient favors deposition of hema-
tites in colder regions, such as vents of volcanoes. References
In the presence of water vapor, species can be stabilized via
solubilization in small molecular clusters of water (e.g., Adamczyk K, Premont-Schwarz M, Pines D, Pines E, Nibbering ETJ
(2009) Real-time observation of carbonic acid formation in aqueous
H3O+. . . (H2O)m, with m = 1  5) (e.g. Lemke and Seward
solution. Science 326(5960):1690–1694. doi:10.1126/
2008). Gas-phase solubility metals (or metal fugacity) and science.1180060
speciation can therefore be strongly affected by steam com- Benson BB, Krause J Jr (1984) The concentration and isotopic fraction-
position and properties. For example, Hurtig and Williams- ation of oxygen dissolved in freshwater and seawater in equilibrium
with the atmosphere. Limnol Oceanogr 29:620–632
Jones (2014) show that gold solubility in steam increases with
Biot BC, Henry O, Barreswil R, Pohl JJ (1852) Notizen. J Prakt Chem
water density, such as along the water vapor saturation curve. 56(1):53–64. doi:10.1002/prac.18520560106
Diamond LW, Akinfiev NN (2003) Solubility of CO2 in water from 1.5
to 100 degrees C and from 0.1 to 100 MPa: evaluation of literature
data and thermodynamic modelling. Fluid Phase Equilib
Summary
208(1–2):265–290. doi:10.1016/s0378-3812(03)00041-4
Dolejs D, Manning CE (2010) Thermodynamic model for mineral sol-
The concept of solubility is essential in geochemistry for ubility in aqueous fluids: theory, calibration and application to model
describing the maximal quantity of a substance that can be fluid-flow systems. Geofluids 10(1–2):20–40. doi:10.1111/j.1468-
8123.2010.00282.x
dissolved in solutions. It often becomes a starting talking
Driesner T, Seward TM, Tironi IG (1998) Molecular dynamics simula-
point for understanding whether solutions are homogeneous tion study of ionic hydration and ion association in dilute and 1 molal
or prone to bear particular matter. Preditions of solubility aqueous sodium chloride solutions from ambient to supercritical
often require a distinct thermodynamic treatment of the solute conditions. Geochim Cosmochim Acta 62(18):3095–3107.
doi:10.1016/s0016-7037(98)00207-5
and the solvent that can be built in different modeling frame-
Harvie CE, Moller N, Weare JH (1984) The prediction of mineral
works. Equations of states or more empirical sets of equations solubilities in natural-waters – the NA-K-MG-CA-H-CL-SO4-O-
can be used to effectively predict solubility over a range of HCO3-CO3-CO2-H2O system to high ionic strengths at 25-degrees-
conditions relevant to the solvent. Examples include salinity C. Geochim Cosmochim Acta 48(4):723–751. doi:10.1016/0016-
7037(84)90098-x
and pH of water, partial pressures of gases, as well as pressure
Helgeson HC, Kirkham DH, Flowers GC (1981) Theoretical prediction
and temperature of water and melts. The concept of solubility of the thermodynamic behavior of aqueous-electrolytes at high-
thus applies to diverse aspects of Earth Sciences and enables a pressures and temperatures.4. Calculation of activity-coefficients,
means for formulating predictive models on mass transport in osmotic coefficients, and apparent molal and standard and relative
partial molal properties to 600-degrees-C and 5 KB. Am J Sci
nature.
281(10):1249–1516
Solubility 9

Hurtig NC, Williams-Jones AE (2014) An experimental study of the pressures. Geochim Cosmochim Acta 63(22):3817–3827.
transport of gold through hydration of AuCl in aqueous vapour and doi:10.1016/S0016-7037(99)00213-6
vapour-like fluids. Geochim Cosmochim Acta 127:305–325. Navrotsky A, Mazeina L, Majzlan J (2008) Size-driven structural and
doi:10.1016/j.gca.2013.11.029 thermodynamic complexity in iron oxides. Science
Kumar PP, Kalinichev AG, Kirkpatrick RJ (2007) Dissociation of car- 319(5870):1635–1638. doi:10.1126/science.1148614
bonic acid: gas phase energetics and mechanism from ab initio Newman S, Lowenstern JB (2002) VolatileCalc: a silicate
metadynamics simulations. J Chem Phys 126(20). doi:10.1063/ melt–H2O–CO2 solution model written in visual basic for excel.
1.2741552 Comput Geosci 28(5):597–604. doi:10.1016/S0098-3004(01)
Lam RK, England AH, Smith JW, Rizzuto AM, Shih O, Prendergast D, 00081-4
Saykally RJ (2015) The hydration structure of dissolved carbon Pan D, Spanu L, Harrison B, Sverjensky DA, Galli G (2013) Dielectric
dioxide from X-ray absorption spectroscopy. Chem Phys Lett properties of water under extreme conditions and transport of car-
633:214–217. doi:10.1016/j.cplett.2015.05.039 bonates in the deep Earth. Proc Natl Acad Sci U S A
Lemke KH, Seward TM (2008) Solvation processes in steam: Ab initio 110(17):6646–6650. doi:10.1073/pnas.1221581110
calculation of ion-solvent structures and clustering equilibria. Sverjensky DA, Harrison B, Azzolini D (2014) Water in the deep Earth:
Geochim Cosmochim Acta 72:3293–3310. doi:10.1016/j. the dielectric constant and the solubilities of quartz and corundum to
gca.2008.02.024 60 kb and 1200 degrees C. Geochim Cosmochim Acta 129:125–145.
Marshall WL, Franck EU (1981) Ion product of water substance, doi:10.1016/j.gca.2013.12.019
O-degrees-C-1000-degrees-C, 1-10,000 bars – new International Zhang Y, Xu Z, Zhu M, Wang H (2007) Silicate melt properties and
formulation and its background. J Phys Chem Ref Data volcanic eruptions. Rev Geophys 45(4). doi:10.1029/
10(2):295–304 2006RG000216
Migdisov AA, Williams-Jones AE, Suleimenov OM (1999) Solubility of
chlorargyrite (AgCl) in water vapor at elevated temperatures and