Water site headerMasthead Island, Great Barrier Reef Print-me keygo to Water Visitor Book contributions
 Go to my page Water Structure and Science


Glacier iceberg: pure water


Glacier ice floating

Pure water and solubility

Water is difficult to purify as it dissolves many materials.

< Ion-pairs

< Moles, molarity, and molality

V Pure water

V Hydrophobicity

V The solubility of gases

V Solubility; organic; inorganic

V Conductivity

V Henry's constant Henry's volatility constant


Water is often called the 'Universal solvent'     

Pure water

Water may be purified by distillation, followed by deionization, reverse osmosis, and activated carbon filtering (a mainly hydrophobics absorbent with large surface area; > 3000 m2 ˣ g−1) [3067]. It is challenging to obtain pure water (for example, < 5 ng ˣ g−1 solutes; < 5 ppb). Distillation is expensive but can remove a wide range of contaminants, but not all. However, unexpected molecules, including ions, can carry over, and both further contamination and microbial/algal growth can occur on subsequent storage. Deionization is necessary to reduce the conductivity to minimum values. However, deionization has poor effectiveness in removing many organic compounds, and the ion exchange columns can harbor microbial growth and release fine particulates. Usually, combinations of purification steps are made to produce 'pure water'. Also, even 'pure' water contains nm-sized dust particles, colony-forming units (CFU), and nanobubbles. Nanobubbles may be removed by extensive degassing by freeze-thaw cycling under reduced pressure or (preferably) by helium washing [1825]. Removal of oxygen and carbon dioxide may also be achieved by saturation with oxygen-free nitrogen gas. This nitrogen may increase the aqueous clathrate formation in combination with low-pressure glow plasma [4096]. Fine dust (≈ ag) and nanoparticles are difficult to remove, but it is possible to use several distillations in wax-coated vessels (not naked glass or silica as they release fine particles). After production, pure water should be stored in appropriate vessels to avoid atmospheric gases and impurities from the atmosphere and the vessel. Autoclaving at 121 °C, 103 kPa, for 20 minutes may be used to obtain sterile water. The purest water may still contain significant amounts of impurities according to its use and storage conditions.


There are standard specifications for the purest water normally obtainable (ASTM D 1193). Ultra-pure water has >18.0 MΩ ˣ cm resistivity at 25 °C (5.56 µS ˣ m−1 conductivity), e < 50 µg ˣ L−1 total organic carbon (TOC), < 1 µg ˣ L−1 Na+, < 1 µg ˣ L−1 Cl, < 3 µg ˣ L−1 silica, < 10 CFU ˣ L−1 ; the pH is unimportant at this level of purity. The purest water, made by a combination of several processes, is used in the manufacture of semiconductors (ASTM D 5127) with Type E-1.2 (also called Type I+) , intended for the most critical uses, limited by < 1 µg ˣ L−1 total organic carbon (TOC), < 0.005 µg ˣ L−1 Na+, < 0.02 µg ˣ L−1 Cl, dissolved silica 0.5 µg ˣ L−1, <10 CFU ˣ L−1, and 18.2 MΩ-cm resistivity. Such water should be stored under high-purity nitrogen and may be used for processes such as polymerase chain reaction, DNA electrophoresis, HPLC, capillary electrophoresis, in vitro fertilization, and cell culture. Typical general laboratory 'pure' water has >10.0 MΩ-cm resistivity (0.1 - 2 mS ˣ m−1 conductivity), < 50 ng g−1 total organic carbon (TOC), < 50 CFU ˣ ml−1. Ultra-pure water may contain 12C, 14N, 16O, 40Ar, 12CmHn, and 14N mH n from atmospheric gases plus Na, Mg, Cl, Ca, Cr, Fe, Ge, Br, Sb, Ag, and I as detected by ICP-MS [3068]. Pure water should be protected from atmospheric contamination, particularly reactive O2 and the highly soluble and acidic CO2 that seriously affects pure water's electrical conductivity (ultra-pure water saturated with CO2 has conductivity of 0.110 mS ˣ m−1 at 25 °C). Care should be taken on storage as ions may leach from glass, and organic materials may leach from plastics. Characterization of a 'pure' water sample should include electrical conductivity, the concentration of silica and common ions, the total organic carbon concentration, the residue after evaporation, the pH, and the optical absorbance at 254 nm. The NIH has published an excellent guide to laboratory water.


Note that (hexagonal) ice, in contrast to liquid water, is a very poor solvent and this may be made use of when purifying water (for example, degassing) using successive freeze-thaw cycles.



The H2O(H2O)20 cluster; the central O-atom is shown centrally


The H2O(H2O)20 cluster; the central O-atom is shown centrally


There is a question about how many solvent water molecules are required to solvate one water molecule [4254]. This question has been answered by Using semi-classical spectroscopy to determine the minimal network of surrounding water molecules on quantum dynamical grounds to make a central water display the same vibrational features of liquid water. It was found that twenty water molecules were sufficient, arranged rather like the twenty molecules surrounding an H3O+ in the magic number ion [3998].

The H3O+(H2O)20 magic number cluster


H3O+(H2O)20 cluster

The central water molecule of the solvated water is tetrahedrally hydrogen-bonded to four water molecules with each of these further hydrogen-bonded to three water molecules, with a further four water molecules connecting these; the inner five water molecules have four hydrogen bonds whereas all sixteen outer water molecules possess three hydrogen bonds.

[Back to Top to top of page]


An important factor in the solubility of organic molecules is their hydrophobicity. Compounds with lower polarity (i.e., greater hydrophobicity) are less able to disrupt the structure of the water molecules. The best measure of polarity is the logarithm of the partition coefficient (LogP) of the organic molecule between n-octanol and water [3073]; the higher the LogP, the more hydrophobic (nonpolar) is the compound; for example, a LogP = 1 means that there is a 10:1 partitioning between the organic and aqueous phases.


 LogP = Log10(concentration in octanol/concentration in water)


LogP values of some organic compounds






< 0.3



Ethyl acetate


Carbon tetrachloride




Dibutyl ether


Diethyl ether




Methylene chloride




Butyl acetate


Petroleum ether (60-80)


Di-isopropyl ether


Petroleum ether (80-100)




Dipentyl ether








Petroleum ether (100-120)







LogP values increase by about 0.52 for every methylene group (-CH2-) added in a homologous series. Thus, the LogP of hexanol is that of butanol (0.8) plus 2 ˣ 0.52 (i.e., approximately 1.8). In order to gather a more efficient use of lipophilicity in the intramolecular hydrogen-bonding of potential drugs, a more apolar organic partition system utilizing toluene rather than octanol may be used [3425].


Adding cosolvents, hydrotropes, and surfactants can dramatically increase the solubility of hydrophobic molecules by the incorporation of solutes into micelles, the enlargement of micelles, and the reduction of critical micelle concentrations [4297]


[Back to Top to top of page]

The solubility of gases

the amount of dissolved gas is proportional to its partial pressure in the gas phase

  William Henry, 1803  c 


Equilibrium solubility of oxygen under pressure

from [ IAPWS ]

Solubility of oxygen under pressure

Dissolved gases are often mistakenly ignored in aqueous solutions, although they may impart properties different from pure water [4347]. At equilibrium, molecules of the solute in the gas phase enter the liquid phase at the same rate as molecules of the solute in the liquid phase escape to the gas phase. nonpolar gases are poorly soluble in water, with their equilibrium solubility proportional to their partial pressure (p, fugacity), see right. The partial pressure is the pressure that that gas would exert in a mixture of gases if it occupied the same volume on its own.


Solubilities are measured in terms of their mole fraction (x2);


x2 = moles solute/(moles water + moles solute)


Equilibrium solubilities are often described by the use of the Henry's Law constant at a specified temperature (Henry's volatility constant, high Henry's constant = high volatility= low solubility; often a cause of confusion as other definitions occur in the literature. It has been recommended that the Henry's constant used here, and in much of the historical literature, should be alternatively be called Henry's volatility) where Henry's Law states that


Henrys constant= partial pressure/mole fraction


More precisely Henry's volatility constant is defined as

KH= the limit as x2 tends to zero of fugacity/mole fraction


where f2 and x2 are the fugacity (effective partial pressure) and mole fraction of the solute, respectively (IAPWS). In this case, the units used for Henry's constant are Pa. f Conversion factors for other (equally valid) Henry's constant definitions have been tabulated, together with a large number of Henry's constants for various gases [3406]. Whatever constants are used, the solubility of the gas can be calculated given its partial pressure. Henry's Law, although not an exact law, can be used for converting solubility data from the experimental pressure to unit atmosphere partial gas pressure, provided the mole fraction of the gas in the liquid is small and that the difference in pressures is small. As an example, the solubility of oxygen at 101.325 Pa oxygen partial pressure and 25 °C is a mole fraction of 2.301 ˣ 10−5 (Henry's volatility constant = 4.4038 ˣ 109 Pa) [4120].


Equilibrium solubility of gases with temperature

from [ IAPWS ]


Solubility of gases with temperature

Most solid solutes dissolve more in water as the temperature is raised. However, while most gaseous solutes also dissolve more in most solvents as the temperature is raised, nonpolar gases are much more soluble in water at lower temperatures than would be expected from their solubility behavior at high temperatures (see right and anomaly M7).


It may also be seen from the solubility profiles (right) that the gases are relatively somewhat soluble even up to 100 °C, in contrast to the common mistaken belief that aqueous solutions are efficiently degassed at high temperatures.


Somewhat surprisingly, no inflections have been found in the solubility data around the density maximum at ≈ 4 °C [3403].


The solubilities for the noble gases, from [IAPWS, 1166]

Solubilities for the noble gases in liquid water, http://www.iapws.org/relguide/HenGuide.pdf, Radon data from refs. 1166, 678


The solubilities of the noble gases are shown opposite [IAPWS, 1166]. Their hydration may be considered as the sum of two processes: (A) the endothermic opening of a clathrate pocket in the water, and (B) the exothermic placement of a molecule in that pocket due to the multiple van der Waals dispersion interactions (for example, krypton dissolved in water is surrounded by a clathrate cage with 20 Kr···OH2 such interactions [1357]). In water at low temperatures, the energy required by the process (A) is minimal as such pockets may be easily formed within the water clustering (by CS -> ES).


Using the noble gases to investigate the solvation of nonpolar gases is useful as they are spherically symmetrical and have low polarizability, whereas shape and polarizability may confuse the hydration of other gases. The solubility of the noble gases increases considerably as the temperature is lowered. Their enthalpy and entropy of hydration become more negative as their fit into the water dodecahedral clathrate improves.


Equilibrium solubility of gases from the air, ml, STP

from [ IAPWS ]

Equilibrium solubility of gases from the air, ml, STP


Under pressure of 101,325 Pa of each gas, the solubilities of the following atmospheric gases i at 0 °C are: nitrogen 1.11 mM, oxygen 2.31 mM, carbon dioxide 77.6 mM, argon 2.51 mM, neon 0.603 mM, helium 0,457 mM, methane 2.61 mM, krypton 5.05 mM, hydrogen 1.07 mM, carbon monoxide 1.71 mM, xenon 10.32 mM. At equilibrium with air at 25 °C and under the atmospheric pressure of 101.325 kPa, the following concentrations of the atmospheric gases are present in water: nitrogen 0.549 mM, oxygen 0.288 mM, carbon dioxide 14.3 µM, argon 14.1 µM, neon 9.05 nM, helium 2.25 nM, methane 2.71 nM, krypton 3.10 nM, hydrogen 0.454 pM, carbon monoxide 0.158 pM, xenon 4.07 pM (see right, IAPWS ). The solubilities of the inert gases are given in more detail elsewhere as are those of carbon dioxide and carbon monoxide.

The solubility of methane under high pressure,

from [3407]

Solubility of methane under high pressure, from [3407]



It should be noted that Henry's law only applies under equilibrium conditions in dilute solutions (<~1% w/w). Also, the data has usually been obtained in pure water, and other solutes may interfere if they interact with the solute at issue. The solubility of gases diverges from Henry's Law above about one MPa. At very high pressures (see that for methane left), there may be a transformation into clathrate hydrates of filled ices [3407]. It is expected that other gases such as O2 and N2 behave similarly.


The effect of salts on the solubility of gases has been investigated [4062]. Mostly the salts 'salted-out' the gases. The general relationship was,


log(CG,0/CG) = KS ˣ CS



where CG is the solubility of a sparingly soluble gas (e.g., O2, C2H4, CO2 ) compared with that in pure water, CG,0 in the presence of salt of concentration concentration CS. KS is the 'Sechenov constant', specific to the gas and the salt and with a weak dependence on the temperature.

[Back to Top to top of page]

The solubility of solids

Solubility is the capacity of a solute to dissolve in a solvent. Traditionally water arranged around a solute 'belongs' to the bulk water. However, it is now realized that some of this water may be 'bound' to the solute and should not be treated as bulk water but as part of the solute. Generally, this 'bound' water may be defined as water bound to solutes with energies greater than about 56 kJ ˣ mol−1 at 25 °C, which is the average binding energy of water molecules binding to other waters; ≡ two hydrogen bonds [4100, 4102]. For a review of aqueous solubility prediction, see [744]. The equilibrium solubility depends in a complex manner on the molecular properties of the solute, any cosolute(s) including their concentration(s), solvent and any cosolute(s) including their concentration(s), as well as the temperature, pressure, pH, ionic strength, and sometimes on other factors. Solubilities are challenging to predict quantitatively. The rules for expressing solubility vary as; soluble, greater than 33 g dissolves in a liter; slightly soluble, from 1 g to 33 g dissolves in a liter, and practically insoluble, less than 1 g dissolves in 10 liters.


The excellent solvent properties of water, together with its non-toxic nature, make water a preferred solvent for many chemical reactions [1566]. It has been shown that aqueous solvation takes place in two stages; a rapid partial water rearrangement (< ps) counteracting the polarity of the solute but building up a strain within the water's hydrogen-bonding, followed by a slower relaxation (> ns) of this hydrogen-bonding involving reorientation of some of the water molecules [3389].


Several factors determine aqueous solubility: (1) the crystallinity of the solute, (2) the interactions between the solute and water, (3) any ionization, dissociation, and stability issues, (4) the temperature and (5) the ionic strength and cosolutes. Hydrophilic organic compounds containing several oxygen atoms or nitrogen atoms are generally soluble in water, particularly if they possess positive and negative charges. This is due to their strong interactions with water molecules. At high concentrations, the solute may form soluble aggregates to various extents that can be compared by the use of molecular dynamics and graph theory [3371].


Organic compounds


The hydration free energies of neutral molecules may be estimated [3698] , with solubility of organic molecules estimated from the general solubility equation [3078]


logS = 0.5 - 0.01 ˣ (MPt - 25) - LogP


where S is the molar aqueous solubility, MPt is the melting point in °C, and LogP is derived from the octanol-water partition coefficient of the solute [3077]. If MPt < 25 °C, the term (MPt - 25) is set to zero. This equation can also be used for weak acids (so long as pKa + logS ≤ 0) and bases (so long as pKa - logS ≤ 14). This covers most weak electrolytes [3079].


Often, changes to the solubility of pharmaceuticals are of real benefit; particularly to allow the dissolution of the targeted dose. The general solubility equation can be used together with the known changes in LogP and MPt with minor structural changes to predict the most valuable changes in the structure required [3074].


Inorganic salts and common ion effects


The small molar volume and high permittivity of water contribute to water's high dissolving power for salts as they reduce the attractive Coulombic forces between oppositely charged ions and allow multiple stabilizing interactions between the dissolved ions and the water molecules. Inorganic salts are often classified as soluble, sparingly soluble, or insoluble, although all are soluble in water to some extent, even if they may be minimal. There are no strict limits for the solubility nomenclature but, generally, soluble salts have solubilities above about 0.1 M, whereas insoluble salts have solubilities below about one mM.


The solubility product (Ksp) is the product of the molar concentrations of the ions in a saturated solution. It depends on the crystal structure of the undissolved salt. Due to the common ion effect, increasing the anion concentration will reduce the cation concentration and vice versa. As an example, the Ksp of iron(III) hydroxide Fe(OH)3 is 2.79 ˣ 10−39 M4 = [Fe3+] ˣ [OH] ˣ [OH] ˣ [OH] and that of iron(II) hydroxide Fe(OH)2 is 1.4 ˣ 10−15 M3 = [Fe2+] ˣ [OH] ˣ [OH]. Thus, the solubility depends on the pH. The cations ion-pair to hydroxide, however, with both FeOH2+(aq) 6% and Fe(OH)2+(aq) being formed. Without considering this ion-pairing, at pH = 7 there would be 2.79 ˣ 10−18 M iron(III) or 0.14 M iron(II) in saturated solutions with the ferrous ions appearing far more soluble than the ferric ions. However, due to ion-pair formation, hydrolysis, complex ion formation, and ion-water complexes, there are only a few cases in which solubility and Ksp are related in such a simple way [3076].


With metal hydroxides having widely different solubilities, they can often be separated from each other by changing the pH, with one cation precipitating at a particular pH while the other remains in solution. Many metal hydroxides are amphoteric, with the precipitated solid hydroxides redissolving in excess hydroxide ion,

Al(OH)3 (s, ppt) + OH(aq, pH >12) → Al(OH)4(aq)                b


Some rules concerning inorganic salt solubilities are



The solubilities of the alkali metal halides (and ammonium halides) d

Solubilities g ˣ L−1 (20 °C) F Cl Br I
  ionic radius, pm


167 182 206
NH4+ 152 850 (25 °C) 370 760 1630


3 840 1600 1650



40 360 910 1780


950 340 650 1440


1310 910 1080 1440


3220 1870 1060 770


[Back to Top to top of page]


a These Ksp values are from Stefan Franzen; all are at 25 °C. Generally, Ksp values may depend on how they are determined, and there are discrepancies between different sources [3076]. These values ignore any possible ion-pair formation; for example, 0.1 M FeCl3 contains only 10% Fe3+ along with a mixed solution of 42% FeCl2+, 40% FeCl2+, 6% FeOH2+, and 2% Fe(OH)2+ [3075]. See also the calcium carbonate equilibria. Ion-pair formation is particularly relevant at high concentrations. [Back]


b s = solid; aq = aqueous solution; ppt = precipitate. [Back]


c W. Henry, Experiments on the quantity of gases absorbed by water, at different temperatures, and under different pressures, Philosophical Transactions of the Royal Society of London, 93 (1803) 29-42,274-276. [Back]


d From the Wikipedia entry. [Back]


e Conductivity. Electrolytic (electrical) conductivity of a solution is the reciprocal of its alternating current (AC, ~2 kHz) resistance in ohms, as measured between pairs of parallel electrodes at a specified temperature. In poorly conducting solutions, there should be an allowance for sources of error and a temperature correction (D 1125 standard). The electrical conductivity is widely used to characterize the purity of water [4097], due to its high sensitivity to ionic contaminants (1 ppb NaCl increases the conductivity by 4%). A usual reference standard is 0.01 mol(KCl) ˣ kg−1(aq) with 140.823 mS ˣ m−1 at 25 °C (see K. Pratt, W. Koch, Y. Wu, and P. Berezansky, “Molality-based primary standards of electrolytic conductivity (IUPAC Technical Report),” Pure and Applied Chemistry, 73 ( 2001) 1783-1793). [Back]


f The solubilities of gases in older literature are given in terms of the Bunsen coefficient (α); defined as the volume of gas reduced to 273.15 K and 101.325 kPa pressure which is absorbed by unit volume of solvent (at the temperature of measurement) under a partial pressure of 101.325 kPa; it is dimensionless. If ideal gas behavior and Henry's law are assumed to be obeyed,


α = {V(g)/V(l)} ˣ (273.15/T)


where V(g) is the volume of gas absorbed, V(l) is the original (starting) volume of absorbing solvent, and T is the temperature (K). [Back]


h Henry's constant (Henry's law volatility constant [3406], Henry's volatility constant, its SI unit is Pa) = partial pressure/mole fraction (Henry's volatility constant, KH below) may be described exactly by the following equation,

KH= limit as solute tends to zero()fugacity of solute/mole fraction of solute

where f2 and x2 are the liquid-phase fugacity (effective partial pressure) and mole fraction of the solute (2). This is generally simplified to,


Henrys constant= partial pressure/mole fraction =(RT/molar volume of water)exp(excess chemical potential of hydration/RT)

where p is the partial pressure of the solute in the gas, X is the solute mole fraction that should be less than 0.01, R is the gas constant, T is the absolute temperature, VH2O is the molar volume of water and μ is the temperature-dependent excess chemical potential of hydration for the solute [1276]. There are other variants for the definition of the Henry's constant including the inverse of the one used at this site. Henry's volatility constant in atm = 1.80695 ˣ 10−5 Henry's volatility in concentration in m3 ˣ atm ˣ mol−1, where Henry's volatility in concentration is the pressure ˣ concentration−1 Henry's volatility [3406]. [Back]


i. Chemical composition of dry air with a fixed CO2 level is


Chemical composition of dry air, from Thermodynamic Equation of State- 2010 (TEOS-10)

Gas Mole fraction Mass fraction Gas Mole fraction Mass fraction
N2 0.780 847 9 0.755 184 73 CH4 0.000 001 5 0.000 000 83
O2 0.209 390 0 0.231 318 60 Kr 0.000 001 1 0.000 003 18
Ar 0.009 332 0 0.012 870 36 H2 0.000 000 5 0.000 000 03
CO2 0.000 400 0 0.000 607 75 N2O 0.000 000 3 0.000 000 46
Ne 0.000 018 2 0.000 012 6 CO 0.000 000 2 0.000 000 19
He 0.000 005 2 0.000 000 72 Xe 0.000 000 1 0.000 000 45




j The Law of Mass Action indicates that if a reaction is at equilibrium and an additional reactant is added, the equilibrium is shifted away from the reactant. [Back]




Home | Site Index | Moles, molarity and molality | Water: Introduction | Water activity | LSBU | Top


This page was established in 2017 and last updated by Martin Chaplin on 27 October, 2021

Creative Commons License
This work is licensed under a Creative Commons Attribution
-Noncommercial-No Derivative Works 2.0 UK: England & Wales License