(G S Manku, email: manku.gs@gmail.com)
Assuming ions to be spherically charged spheres, solubility of ionic electrolyte in a polar solvent of dielectric constant ε can be calculated by equating electrostatic work done for charging a sphere in a dielectric medium with the free energy and solute concentration (Born’s equation). The work done W in charging a sphere of radius r to charge z is given as
W = z2e2 / 8πεr
Here e is the electronic charge. Assuming similar expressions to hold for cation of radius r+ , charge ν+ and anion of radius r– , charge ν – charge ν –, total work done for N molecules ionizing to will be
W = Nν+z+2e2 / 8πεr+ + Nν–z–2e2 / 8πεr–
= (Ne2 / 8πεr) (ν+z+2 + ν–z–2 )
Here r is the average ionic radius of the two types of ions. Work done in charging the ions of a given electrolyte in solvent 1 with respect to work necessary to charge in another solvent 2 will be (if similar equations hold for both the solvents)
W = {(Ne2 / 8πε1r) (ν+z+2 + ν–z–2 )}1
– {(Ne2 / 8πε2 r) (ν+z+2 + ν–z–2 )}2
= (Ne2 / 8πr) (ν+z+2 + ν–z–2 ) (1/ ε1 -1/ ε2 )
Here, r is now the average ionic radius of the electrolyte ions in the two solvents.
The ideal relation between the free energy G and the solute concentration c is:
ΔG = ΔG0 + RT ln c
Equating this with energy required for transferring ions from solvent 1 to solvent 2, (ionization of electrolyte giving ν+ cations and ν– anions)
ΔG1 – ΔG2 = RT ln c1 – RT ln c2
= RT ln {( ν+c1) ν1 – (ν–c 1 ) ν2) / ( ν+c2) ν1 – (ν–c 2 ) ν2 ) }
From these relations,
(Ne2 / 8πr) (ν+z+2 + ν–z–2 ) (1/ ε1 -1/ ε2 )
= RT ln {( ν+c1) ν1 – (ν–c 1 ) ν2) / ( ν+c2) ν1 – (ν–c 2 ) ν2 ) }
Because of charge symmetry of z with respect to ν,
ν+z+2 + ν–z–2 = z+z– (ν+ + ν– )
And the above equation becomes
(Ne2 / 8πr) z+z– (ν+ + ν– ) (1/ ε1 -1/ ε2 ) = RT (In c1 – ln c2) (ν+ + ν– )
Or, ln RT ln c1 = ln c2 + (Ne2 z+z– / 8πrRT) (1/ ε1 -1/ ε2 )
This is Born equation and can be used to find the solubility of a solute in any solvent 1 by knowing
The Born equation is too ideal and has many obvious limitations:
The agreement between solubility of ionic solutes and Born equation is rather poor, even after considering the activity coefficients.
Ricci and Davis proposed an empirical equation for solubility of ionic solutes in nonaqueous solvents. They assumed that the activity coefficients of the ions (γ) in saturated solutions of sparingly soluble salts is almost constant regardless of the solvent (for silver acetate, γ = 0.79 ± 0.01 in 10 – 30 % aqueous acetone and dioxane), and applied Debye – Hückel theory to saturated but dilute solutions to get
A ε1– 3/2c11/2 = A ε2– 3/2c21/2
Or, log c1 = log c2 + 3(log ε1 – log ε2)
This equation is better for solubility of sparingly soluble salts.
[10.2],2 ION ASSOCIATIONS
In concentrated solutions, electrostatic interactions between ions become significant, If ions come close enough, electrostatic interactions become more than the thermal agitations, which leads to disorder the ions. If two oppositely charged ions come close together, they may form an ion pair due purely electrostatic attractions. This is different from the covalent bond formation and is shown as A+B–. In solvents of low dielectric medium, ion pair formation is, however, rather extensive.
Bjerrum’s model for ion pair formation assumes that (1) ions are hard non-polarizable spheres, (2) only coulombic forces exist between the ions, and (3) dielectric constant of the medium between ion pair is same as that in the bulk of the solvent. Number of ions (dn) of the type i in spherical shell of thickness dr at a distance r from the selected j–ion by given by Beckmann’s distribution law as
dn = ni exp (U/kT)(4πr2dr)
Where k = R/N is the Boltzman’s constant, n1 = number of i-ions per unit volume (cm3 or ml) solution, U = work required to separate the i-ions at a distance r to infinity from a particular j-ion, 4πr2+dr is the volume of the spherical shell of thicken dr at a distance r from the center. As U is the coulombic potential energy of the charge in the presence of dielectric constant ε,
U = – z1z2e2/4πεr.
Hence, Beckman’s distribution equation becomes
dni = 4πr2ni exp (– z1z2e2/4πεkTr) dr
The probability of finding i-ion in the shell of thickness dr (may be as small as 1 pm) from a selected j-ion as a function of r shows a minimum at r = r0. The value of r0 can easily be calculated by differentiating the above equation and putting d/dr (dni/dr) = 0:
d/dr (dni/dr) = 2r.4πrni exp (– z1z2e2/4πεkTr) 4πr2+dr –
(z1z2e2/4πεkTr2 ) 4πr2ni exp (– z1z2e2/4πεkTr)
= [ 8πr – z1z2e2/εkT ] ni exp (– z1z2e2/4πεkTr)
Hence d/dr (dni/dr) = 0 at r = r0 gives
r0 = z1z2e2/4πεkT
The Bjerrum’s equation applies only to electrostatic interactions and breaks down in case of covalent bonding, e.g. for weak electrolytes.
The distance r0 for ion pair formation cannot be determined directly, but can be evaluated from the equilibrium between the ion pair formation and dissociation of ions. Degree of association α will be proportional to the number of i-ions that are at a distance ≤ r0 and can be calculated by integrating the expression for dni/dr from a distance a (point of contact of the two ions) to the distance r0..
The different solvents can be classified in many ways depending upon their physical and chemical properties.
Protonic and Aprotic solvents: Protonic solvents autoionize to form a solvated hydron (proton), e.g. water, ammonia, hydrogen fluoride, sulphuric acid, acetic acid, etc. Aprotic solvent do not ionize to give a hydronium ion, e.g. SO2, BrF3, CS2, CHCl3, etc.
The protonic solvents can be further classified as
However, this classification over-emphasizes the acid – base behavior as all the protonic solvents can accept as well donate a hydronium ion depending upon the nature of the solute.
The aprotic solvents can be further classified as
However, such classifications are of no use as they are based on the studies of only a limited number of solvents and do not give any basic information of the solvent properties
Solubility of ionic compounds, especially in aprotic solvents, involves electrostatic as well as specific solute–solvent interactions Gutman has proposed a donor number (DN) as a measure of basicity of a solvent. It is defined as the negative enthalpy of the reaction of aa solvent with antimony trichloride:
Solvent + SbCl3 → [(Solvent)nSbCl3] ΔH = – DN
However, there seems to no relation between DN and the dielectric constant (ε) or polarity of solvents. Solvents with high ε may be poor donors (nitrobenzene, ε = 35.9, DN = 2.7), diethyl carbonate (ε = 89.1, DN = 16.4) and best donor solvents may have a low ε (pyridine, ε = 12.3, DN = 33 or tributyl phosphate ε = 12.3, DN = 33).
Water has long been recognized as a good solvent because of its ease of availability and its purification. No other solvent is so versatile to handle. High availability, purity, wide liquid range, dissolution tendencies for both covalent and ionic compounds, non-corrosive nature, high dielectric constant that reduces the interionic interactions, dipolar nature and coordination tendencies that stabilizes many ions and molecules giving high hydration energies, amphoteric behavior and other physico-chemical properties make water almost a unique solvent. However, many other liquids possess these properties to different extents, but have similar behavior.
The most important properties of the ionizing solvents are given below.
NH3 + NH3 ⇌ NH4+ + NH2–
H2O + H2O ⇌ H3O+ + OH–
HF + HF ⇌ H2F+ + F–
SO2 + SO2 ⇌ SO2+ + SO32 –
An examination of commonly used solvents (Table [10.2].1) shows that these properties may not be available in many solvents. HF and HCN are dangerous to health; NH3, SO2 and H2S have low boiling points;acetic acid and dioxane have low dipole moments. Many organic solvents (alcohols, benzene, chloroform, carbon tetrachloride and dichloromethane have favorable handling properties but are not good solvents for ionic solutes.
Table [10.2].1 Physical Characteristics of Some Solvents
olvent M.P. B.P. Dipole Moment Dielectric constant
Water 273 K 373 K 1.84 D 81.3 at 288 K
Ammonia 195.5 K 240 K 1.46 D 22 at 240 K
Hydrogen fluoride 190 K 293.1 K 1.90 D 83.6 at 273 K
Sulphur dioxide 198 K 290 K 1.61 D 17.3 at 298 K
Hydrogen cyanide 299 K 2.93 D 106.8 at 298 K
Hydrogen sulphide 187.5 212.9 K 1.10 D 10.2 at 213 K
Acetic acid 289.8 K 391 K 0 7.1 at 298 K
Benzene 278.8 K 353 K 0 13.1
Sulphuric Acid 283.7 K 611 K (d) 1.83 D 110
Ethanol 161 K 353.5 K 1.7 D 25
Chloroform 210 K 334 K 1.1 D 5.1
Carbon tetrachloride 251 K 353 K 0 13.1
Carbon disulphide 160 K 319 K 1.1 D 2.7
Diethyl ether 155 K 308 K 1.1 D 4
Nitrobenzene 279 K 484 K 4.18 D 35.5
———————————————————————————
d = decomposes
Though in solvents, specific reactions may be different, same types of reactions are defined for solvents as for water. Activity coefficients of solutes can be determined from measured half-cell potentials using using Nernst equation and Debye-Hückel theory in aqueous as well as nonaqueous solutions. Some typical reactions are given below.
By the Cady–Elsey solvent system definition of acids and bases (Article 1.3, Lewis and Other Definitions), acids are those substances that by direct interaction with solvent, increase the concentration of its characteristic cation formed by autoionization, whereas bases are those substances that by direct interaction with solvent, increase the concentration of its characteristic anion. The acid–base behavior depends upon (i) inherent acid–base nature of the solvent, (ii) inherent character of the solute and (iii) solute–solvent interactions. The solvent with high dielectric constant ε only can support ionization of solutes.
In a protonic solvent, the cation is just a solvated hydronium ion (proton), so that the Brönsted and the solvent system definitions are the same. Thus, in ammonia as solvent, NH4+ is acid whereas NH2–, NH2– and N3– ions (amide, imide and nitride respectively) are bases. Typical neutralization reactions in ammonia (solvent) are:
NH4Cl + NaNH2 → NaCl + 2NH3
2NH4Cl + PbNH → PbCl2 + 3NH3
3NH4Cl + Na3N → NaCl + 4NH3
NH4Cl + NaOH → NaCl + NH3 + H2O.
In non-protonic solvents (benzene, carbon tetrachloride), Lewis type of neutralization reactions can take place. The solvent system definition of acid and base is useless here. However, for self-ionizing solvents (SO2, BrF3, POCl3), the acid–base reaction will be
2SO2 ⇌ SO2+ + SO32–
SOCl2 + Na2SO3 ⇌ 2NaCl + 2SO2
A solute may act acidic or basic depending on the solvent. For example, acetic acid is a strong acid in ammonia, a weak acid in water and a base in sulphuric acid solutions:
CH3COOH + NH3 ⇌ CH3COO– + NH4+ (strong acid)
CH3COOH + H2O ⇌ CH3COO– + H3O+ (weak acid)
CH3COOH + H2SO4 ⇌ CH3COOH2+ + HSO4 – (weak base)
The differentiating range of a solvent is controlled by the autoionization constant for the solvent. It is 0–14 for aqueous solutions, 0–27 for ammonia and 0–3 for sulphuric acid. All acids stronger than the characteristic cation of the solvent behave as strong acids, and all bases stronger than the characteristic anion of the solvent behave as strong bases. These get leveled off to the solvent cation or the anion.
A metathetical reaction can be predicted only we know the solubility product of the compound in the solvent. The solubility product of a salt changes so much from solvent to solvent, that it can be reversed by a change of solvent. For example, in aqueous solutions, addition of KI to AgNO3 solution forms a precipitate of AgI, but in ammonia solutions, addition of KNO3 solution to a solution of AgI precipitates KI:
AgI + KNO3 → KI ↓ + AgNO3 (in ammonia solutions)
Metathetical reactions cannot be generalized due to specific solute–solvent interactions.
In a solvolytic reaction, the solvent reacts with the solute changing the normal concentration of the solvent cations or anions, and the phenomenon is called solvolysis. As an example, solvolysis of POCl3 in some solvents is as:
POCl3 + 3HOH → OP(OH)3 + 3HCl
POCl3 + 3ROH → OP(OR)3 + 3HCl
POCl3 + 6HNH2 → OP(NH2)3 + 3NH4Cl
In nonprotonic solvent, an example is
SnCl2 + 2SOCl2 → 2SOCl+ + SnCl3 –
If the solvent does not undergo redox reaction, the redox reactions in aqueous and in nonaqueous solvents are similar. In ammonia solutions, the reference hydrogen electrode is the half-cell for the reaction
(1/2)H2(g) + NH3 ⇌ NH4+ + e – E0 = 0.000 V
However, the E0 values for half reactions vary considerably from solvent to solvent due to the solvation energy and the chemical reactivity of solvents. In many cases, the order of reactivity of the substances may get altered in nonaqueous solutions. In many cases, the reactions appear very unique. For example, sodium can be electrolyzed from aqueous solutions containing pyridine using mercury cathode. Sodium and beryllium can be electro-deposited from ammonia solutions. The dimercury(2+) cation (Hg22+) ion disproportionates to Hg + Hg2+ in ammonia solutions.
The order of – E0 (in volts) for alkali and alkaline earth metals in aqueous solutions is
Li (3.05) > Cs (3.04) > K (2.93) ) Rb (2.92) > Ba (2.90) >
Sr (2.89) > Ca (2.87) > Na (2.71) > Mg (2.37)
In ammonia solutions, it changes to
Li (2.34) > Sr (2.30) > Ba (2.20) > Ca (2.17) > Cs (2.08) >
Rb (2.06) > K (2.04) > Na (1.85) > Mg (1.74)