D. OTHER NUCLEOPHILIC SUBSTITUTION REACTIONS
1. Reactions with Naturally Occurring Nucleophiles
Thus far, discussion of nucleophilic substitution processes has been limited to reactions with the oxygenated nucleophiles, OH~ and H2O. Recently, however, there has been a great deal of interest in determining the contribution of other naturally occurring nucleophiles to the abiotic transformation of organic pollutants; namely, the halogenated aliphatics. Much of this interest comes from field studies that have detected organic sulfur compounds in groundwaters that presumably were formed from nucleophilic substitution reactions of bisulfide anion, HS“, with haloaliphatic compounds (Schwarzenbach et al., 1985; Jackson et al., 1985; Weintraub and Moye, 1987; Barbash and Reinhard, 1989; Roberts et al., 1992). Laboratory studies have demonstrated the reactivity of HS", as well as other sulfur nucleophiles, for halogenated aliphatics (Schwarzenbach et al., 1985; Barbash and Reinhard, 1989a and 1989b; Haag and Mill., 1988b). In addition to having the capability to correctly predict the concentrations of halogenated compounds in aquatic ecosystems, there is concern over the formation of reaction products (e.g., thiols) that may be more persistent and hazardous than the parent compounds or their hydrolysis products.
HYDROLYSIS 137
0.1 IQ) ' ¥
10 30
pH
Figure 2.11. Chlorimuron-ethyl hydrolysis rate constant versus pH at 45°C in buffered aqueous solution. From Hay (1990). (Reprinted by permission of the Society of Chemical Industry).
a. Nucleophilic Reactivity
To assess the importance of nucleophilic substitution reactions of naturally occur
ring nucleophiles it is necessary to have some measure of their reactivity, relative to OH" and H2O. A number of properties of nucleophiles, all of which are some measure of the nucleophile’s ability to donate electrons to an electrophile, have been used to correlate nucleophilic reactivity. These closely related properties include basicity, oxidation potential, polarizability, ionization potential, electronegativity, energy of the highest filled molecular orbital, covalent bond strength, and size (Jencks, 1987).
Hard-soft acid base model, Pearson (1963) has developed an empirical model for nucleophilic reactivity that has the ability to predict reactivity trends among differ
ent nucleophiles and electrophilic substrates. The model, known as Pearson’s princi
ple of hard and soft acids and bases (commonly abbreviated HSAB principle) classi
fies Lewis bases (nucleophiles) and Lewis acids (electrophiles) as either “hard” or
“soft.” Hard bases and acids are relatively small, of high electronegativity, and of low polarizability. Soft acids and bases are relatively large, of low electronegativity, and of high polarizability. Hardness is associated with a relatively large amount of ionic character and softness with a large amount of covalent character in a bond or transition state. On the whole it is found that hard acids form stronger bonds and react faster with hard bases, and soft acids form stronger bonds and react faster with soft bases. The HSAB principle summarizes this observation: Hard acids prefer to bind to hard bases and soft acids tend to bind to soft bases. Because nucleophilic- electrophilic reactions are an example of a Lewis base-Lewis acid reaction, the HSAB principle applies equally to the reaction of nucleophiles with electrophiles.
Environmentally significant nucleophiles are classified, according to Pearson’s HSAB principle, as either hard, soft, or borderline (possessing intermediate hard- soft character) as follows:
Hard
OH-, H2PO4- NO3-, SO42-, ci- NH„ CH,COO-
HCO3- F-
Borderline H2O, NO2', S03^
B r, C6H5NH2
Soft
HS-, RS-, PhS- S2O3-, 1-, CN-
The HSAB principle has been criticized because of the difficulty in defining the terms “hard” and “soft” quantitatively. More recently, Pearson (1983) has attempted to quantify the term hardness by defining hardness operationally as half the differ
ence between ionization potential and electron affinity.
Because of the qualitative nature of the HSAB model, it tells us nothing about the absolute value of kinetic rate constants. The usefulness of the HSAB model arises from its ability to predict the relative reactivities of nucleophiles with various sub
strates. For example, C-O cleavage of dimethyl phosphates results from nucleo
philic attack at carbon, a soft electrophilic site (carbonium ions are soft elec
trophiles), whereas cleavage of the P-O bond results from nucleophilic attack at the phosphorus atom (P-O), a hard electrophilic site. Accordingly, reaction of dimethyl phosphate with a soft nucleophile, such as sulfide, would expect to occur mainly at the soft carbon center, resulting in C-O cleavage (2.79). On the other hand, nucleo
philic attack by a hard nucleophile, such as hydroxide ion, is expected to occur predominantly at the hard phosphorus center (2.80).
OII
CH3O—P—XR
¿QCH3
HS— ^
"S>j2" attack at carbon
OII
CH3O—P—XR + HSCH3
O - (2.79)
CH3
"Sn2" attack at phosphorus
(2.80)
These predictions are consistent with experimental data on the competition between dealkylation and hydrolysis for the reactions of HS' and OH" with dimethyl phos
phate (Schmidt, 1975).
The HSAB model also provides a rationale for the relative reactivities of HS", H2O and OH" toward halogenated aliphatics. Laboratory studies have demonstrated that HS" reacts with 1,2-dibromoethane predominantly through nucleophilic substi
tution (2.81) (Haag and Mill, 1988b; Barbash and Reinhard, 1989a). By contrast, reaction of OH" with 1,2-dibromoethane occurs predominantly through elimination (2.82) (Hine and Langford, 1956; Vogel and Reinhard, 1986; Weintraub, et al., 1986; Haag and Mill, 1988b; Weintraub and Moye, 1987). Reaction of water with 1,2-bromoethane favors nucleophilic substitution; however, a significant amount of elimination is also observed (2.83) (Junglaus and Cohen, 1986; Pignatello, 1986;
Vogel and Reinhard, 1986; Weintraub et al., 1986; and Haag and Mill, 1988b).
HYDROLYSIS 139
BrI CH2-
BrI -GH2 HS"
"Sn2"
SHI CH2-
SHI
-CH2 (2.81)
Br Br I I
CH2—CH2 OH"
'E2"
.Br H2C =C^
H
(2.82)
Bri CH2-
BrI
-CH2 H2O ,
"Sn2"
’E2"+
? “ .Br
CH2— CH2 + H2C=C^
H (2.83)
These results are consistent with the HSAB model if we consider that bimolecular elimination is a hard-hard interaction resulting from attack of the hard nucleophile, OH", at the proton, a hard electrophilic center. Whereas, nucleophilic substitution occurs by attack of the soft nucleophile, HS", at saturated carbon, a soft electro
philic center. The softness of H2O is intermediate between that of OH" and HS", resulting in a mixture of substitution and elimination products.
Correlation equations. Duboc (1978) has critically reviewed the equations used for correlating nucleophilic reactivity. The correlation equation that has been used to the greatest success is a two-term equation (Edwards, 1954 and 1956):
log k/ko = aEn + jSH (2.84)
where En is a measure of the nucleophile’s polarizability, H is a measure of its basicity, and a and /3 are the measures of susceptibility of the substrate to E„ and H, respectively. E„ has been defined in terms of the molar refractivity of the nu
cleophile. The limitations of this correlation result from the fact that it requires information that may not be generally available.
A simpler correlation that allows for direct comparison of reactivity for a series of nucleophiles for a particular substrate (RX) is that of Swain and Scott (1953), which is given by:
log (k„„,/kHjo) = ns (2.85)
where is the second-order rate constant for the reaction of the nucleophile of interest with RX, ICH20 is the second-order rate constant for the reaction of RX with water, n is the nucleophilicity constant, and s is a parameter for the sensitivity of the substrate to nucleophilic reactions. Nucleophiles that are stronger than H2O will have values for n that are positive (the larger the value of n, the stronger the nucleophile). If the substrate constant, s, is known for a particular substrate, the rate constant for the reaction of any nucleophile with the same substrate can be predicted using n values from a reference reaction. The Swain-Scott equation has been used successfully to correlate rates of reactions with similar transition states.
The correlation is found to fail if reactions with fundamentally different transitions states (e.g., nucleophilic reactions at saturated and unsaturated carbons) are com
pared.
Values of n are determined by setting n = 1 for H2O and s = 1 for a reference reaction, typically the 8^2 reaction of CH3Br.
CH.Br -E H.O CH3OH + HBr (2.86)
CH.Br -E Y: CH.Y + Br" (2.87)
Table 2.7 lists nucleophilicity constants in water for a number of naturally occur
ring nucleophiles, values of rate constants for nucleophilic substitution (ky) relative to H2O, and the calculated concentrations, [Nuc]5oo7o, at which the rate constant for the Sn2 displacement by the nucleophile of interest is equal to the rate of 8^2 displacement by H2O (kNuc[Nuc]5o% = kn2o[H20]). It is readily apparent that as the magnitude for the value of n increases, the concentration of nucleophile needed to compete with hydrolysis decreases.
b. Reactions o f Sulfur-Based Nucleophiles with Halogenated Aliphatics
The data in Table 2.8 demonstrate that the sulfur-based nucleophiles will have the greatest reactivity toward halogenated substrates (largest n values). In general, the concentrations of the naturally occurring nucleophiles typically found in uncontami
nated, freshwater ecosystems are several orders of magnitude below their concentra
tions. At the concentrations of hydrogen sulfide, polysulfides, sulfite, and thiosulfite found in saltmarshes, anoxic surface waters and contaminated ground- waters, however, nucleophilic reactions with these nucleophiles can compete with hydrolysis (Haag and Mill, 1988b; Barbash and Reinhard, 1989b; Roberts et al..
HYDROLYSIS 141
Table 2.7. Relative Reactivities of Naturally Occurring Nucleophiles
Nucleophile n ^rel [NUC]50%
H2O 0.0 1.0 0.0
NO3- 1.0 ~6
s o / - 2.5 3x10^ 2x10-^
C|- 3.0 1 x10^ 6x10-2
HC03- 3.8 6x10^ 9x10-2
HP042- 3.8 6x10^ 9x10-2
B r 3.9 8x10^ 7x10-2
OH- 4.2 2x10^ 3x10-2
|- 5.0 1 x10® 6x10-“*
HS- 5.1 1 xIO® 6x10-“*
SO32- 5.1 1 x10® 6x10-4
S4"- 7.2
Ss^- 7.2
1992). Table 2.8 compares hydrolysis half-lives with half-lives for reaction with sulfur-based nucleophiles for several halogenated aliphatics. These data show that the environmental half-lives for substrates such as 1-bromohexane and 1,2- dibromoethane can be substantially reduced in the presence of HS" and polysulfides.
Enhanced degradation of 2-bromopropane and 1,1,1-trichloroethane, as well as chloroform and carbon tetrachloride (results not shown) was not observed, suggest
ing that steric hindrance significantly impedes reaction with the sulfur based nu
cleophiles (Haag and Mill, 1988a).
Table 2.8. Comparison of Hydrolysis Half-Lives (days) with Half-Lives for the Sn2 Reaction of Sulfur-Based Nucleophiles with Halogenated Aliphatics at 25®C®
Nucleophile‘s hexBr^ DBE<= 2BP= TrCE®
H2O 2 0 1 , 0 0 0 2 350
5.0 mM HS- 1 . 0 3.9 9.4 >3,200
0.07 mM S /- 1 . 1
0.5 mM 8 2 0 3^“ 17 170 240 >160,000
0.2 mM SO3 2- 2 2 170
^Data taken from Haag and Mill (1988b).
•^Assuming complete ionization of sulfur-based nucleophiles.
^hexBr = 1-bromohexane; DBE = 1,2-dibromoethane; 2BP = 2-bromopropane; TrCE =
1,1 , 1 =trichloroethane.
In addition to accelerating degradation rates for halogenated aliphatics, reaction with sulfur nucleophiles will have significant consequences with respect to reaction product distributions. Schwarzenbach et al. (1985) have observed the formation of thiols and dialkyl sulfides from the Sn2 reactions of primary alkyl bromides with HS- (Figure 2.12).
RCH2OH + Br
RCH.SH + Br"