Cent. Eur. J. Chem. • 10(5) • 2012 • 1600-1608
DOI: 10.2478/s11532-012-0080-8
Central European Journal of Chemistry
Role of water in determining organic
reactivity in aqueous binary solvents
Research Article
Siim Salmar*, Jaak Järv,
Tiina Tenno, Ants Tuulmets
Institute of Chemistry, University of Tartu, 50411 Tartu, Estonia
Received 29 February 2012; Accepted 18 May 2012
Abstract: Kinetic data for organic reactions in various binary water-organic solvent mixtures were collected and quantitatively analysed in terms
of linear-free-energy relationships by using tert-butyl chloride (2-chloro-2-methylpropane) solvolysis as the reference system. Linear
similarity plots for these kinetic data were determined for solvent systems ranging from pure water mixtures up to considerable
amount of cosolvent, and 161 similarity coefficients were calculated from slopes of these plots. The existence of these linear plots
demonstrated that the solvent effects are of some common nature in all analysed reaction mixtures independent of the reaction type
and the cosolvent used. Therefore it was concluded that the observed effects could be connected to the specific solvating properties
of water, which govern reactivity even in significant dilution of water by an organic cosolvent. This conclusion was supported by the
linear interrelationship between the slopes of similarity plots of different reactions, and hydrophobicity parameters log P of the reacting
compounds. The relative solvent effects observed in binary water-organic solvent mixtures were for the first time directly related to the
structure of reacting compounds.
Keywords: Solvent effects • LFE relationships • Organic-aqueous solutions • Similarity coefficients • Hydrophobic interactions
© Versita Sp. z o.o.
1. Introduction
Water is a unique solvent and its crucial role is generally
recognised in almost all chemical processes occurring
in living systems [1-3]. Specificity of water stems from
the complexity of its dynamic bulk structure. While the
current understanding of the properties of pure water
implies not so much ambiguity, the interactions between
bulk water and solutes, particularly in binary aqueous
solvent mixtures, are not yet fully understood. In highly
structured liquids, which can form extensive networks
of hydrogen bonds, solvation effects also depend on
the structure of the reaction medium. This structure
is extensively micro-heterogeneous in binary waterorganic solvent mixtures, where the domains surrounded
by water exist with domains of water surrounded by the
organic solvent [3-8].
The involvement of water in solvent mixtures brings
forth a characteristic phenomenon known as hydrophobic
effect, defined as a tendency of apolar species to
aggregate in aqueous solutions and thereby reduce
their contact surface with water [7-13]. This property
of apolar molecules or apolar fragments of otherwise
polar molecules is an important driving force for interand intramolecular binding and assembling processes in
water, and plays a significant role in the functioning of living
cells [1,10-12]. Today it is well recognised that the same
factor can also strongly influence chemical equilibria and
reaction rates in aqueous media [7-15]. More generally,
the possibility that lipophilicity of substrates may even
play an important role in pure solvents has recently been
realised [16].
Different aspects of reactivity in binary watercosolvent mixtures have attracted interest of researchers
for a long time [6-9,17]. Since the pioneering work of the
Engberts and Blandamer group [18], a series of papers
about the quantitative interpretation of cosolvent induced
rate effects have been published (for reviews see, e.g.
[10] and [11]), covering data for water-rich media, i.e.,
systems with concentrations of cosolvents of about a
few mole percentage points. Recent works from this
group [19,20] contain important findings related to kinetic
effects observed in aqueous binary mixtures over a
large concentration range and about the interactions
* E-mail: [email protected]
1600
Unauthenticated
Download Date | 6/15/17 6:39 PM
S. Salmar et al.
responsible for these effects. Binary mixtures of lower
water content have attracted the attention of many
researchers because of their variable polarity and some
other properties and therefore important mechanistic
studies have been carried out within these concentration
ranges by the groups of Bentley [21,22], Kevill [23-25],
and others. However, unlike these studies, our work was
focused mainly on water-rich mixtures, because these
are the media where hydrophobic interactions can play
an important role in determining reactivity.
Hydrophobic interactions have been studied by a
large variety of experimental techniques [7,10,11,26].
Our contribution to studying hydrophobic interactions
in different solvent systems consists of applying
power ultrasound to kinetic investigations of polar
homogeneous reactions in solutions [27-30]. The
progress in this approach was attained through a
combination of sonication data [29,30] with application
of linear-free-energy relationships. In our subsequent
work in this field [31], rate constants for some organic
reactions were compared with those for the solvolysis
of tert-butyl chloride (2-chloro-2-methylpropane) in the
same organic-aqueous solutions and linear similarity
plots [32] were observed in a wide range of the
cosolvent content. These results, indicating that the
same intermolecular interactions should govern solvent
effects in different reactions, have inspired us to extend
this extra-thermodynamic approach for the further study
of solvent effects in binary water-organic systems.
In this report we use the database covering all
published kinetic data related to solvent effects in binary
water-organic media. It appeared that a similarity to the
tert-butyl chloride solvolysis holds for an unexpectedly
wide range of reactions, which represent largely
different mechanisms and proceed in a great number
of solvent mixtures. Unlike in conventional approaches,
where attempts have been made to describe solvent
effects stemming from the properties of the solvents,
this investigation is focused on the structure of the
reagents and their interaction with reaction media.
The practical importance of organic reactions in watersolvent mixtures, which are generally recognised as the
“greenest” reaction media for chemical technology [1,2],
seemed to be an additional important reason for a wider
analysis of this phenomenon.
2. Similarity relationships in aqueous
binary mixtures
Most generally, a free energy relationship can be defined
by Eq. 1 where parameter a is the similarity coefficient
[32].
ΔG = a ΔGst + b
(1)
The term ΔG is the free energy (the Gibbs energy) of a
process such as a rate or equilibrium and ΔGst is the free
energy of a standard process.
For practical data processing of free energy-related
parameters, logarithms of rate or equilibrium constants
of chemical reactions can be used, so Eq. 2, equivalent
to Eq. 1, can be applied to kinetic data.
log k = Csim log kst + b
(2)
In Eq. 2 Csim is the similarity coefficient and kst is the
rate constant of the standard reaction selected on the
basis of certain theoretical or practical considerations.
In our approach log k and log kst values characterise two
different reactions, studied at identical compositions of
the reaction medium.
By analogy with numerous other approaches,
the solvolysis of tert-butyl chloride (2-chloro-2methylpropane) in aqueous organic binary solvents
at 25oC was selected by us as the standard process
because of the most extensive set of available data. The
appropriate kinetic data were taken from [35-37]. Rate
constants for other reactions in binary aqueous organic
solvents were collected from literature and these data
were compared with those for the standard reaction.
This way we were able to avoid the commitment to any
previously determined solvent parameters and proposed
intermolecular interactions and we were able to rely on
the most fundamental principles of the LFE analysis, i.e.,
if concerning our subject, the solute-solvent interactions
in the model process and in the process under
investigation must be closely related [4]. In contrast, if
Eq. 1 holds, the similarity between the solute-solvent
interactions in processes under investigation is reliable.
For practical reasons we define log kst as in Eq. 3
(3)
In this case intercept b in Eq. 2 has a reasonable physical
meaning as the rate constant for pure water.
It has to be mentioned that the procedure used in
this work has certain conformity to correlations with the
Grunwald-Winstein equation [4,33] (Eq. 4)
,
(4)
where Y is defined as
,
and m is the susceptibility of the reaction to solvent
ionising power, while log k0 is measured in 80 vol% EtOH1601
Unauthenticated
Download Date | 6/15/17 6:39 PM
Role of water in determining organic
reactivity in aqueous binary solvents
Figure 1.
Plots of log k for reactions vs log kt-BuCl for the solvolysis of tert-butyl chloride (2-chloro-2-methylpropane) in binary water-ethanol mixtures
at 25oC [35]. ○-alkaline hydrolysis of ethyl acetate [40], ●-solvolysis of 2-bromo-2-methylpropane [41], ∗-solvolysis of benzoyl fluoride
[42], □-reaction between 1-chloro-2,4-dinitrobenzene and aniline [43], ■-decomposition of trimethylsulfonium ion [44].
H2O. Importantly, in some cases the Csim parameter has
the same numerical value as the susceptibility factor
in Eq. 4; however, the physical meaning of Csim in our
approach is not identical to that of the parameter m.
The same holds for the extended Grunwald-Winstein
equations [38,39].
Transformation of Eq. 4 to the form in Eq. 5,
,
(5)
where m and c are to be fitted, clearly demonstrates
the principal difference between Eq. 2 and Eq. 4. While
Grunwald-Winstein equations have been extensively
exploited in mechanistic investigations of solvolytic
reactions [21-23], our approach was focused on
solvation issues relative to the solvolysis rate of tertbutyl chloride (2-chloro-2-methylpropane) and involved
a great variety of reactions in binary water-organic
solvent mixtures, including solvolyses, hydrolyses, and
many other processes.
To illustrate this principle, the kinetic data for five
reactions in aqueous ethanol mixtures were compared
with the data for the solvolysis of tert-butyl chloride in
the same media as shown in Fig. 1. It is evident that
all these reactions, occurring by principally different
mechanisms, are strongly influenced by the addition of
alcohol into the reaction medium. Log k plots for these
processes have extensive linear regions, in some cases
up to 100 mol-% of ethanol. In this solvent system log
k for the base-catalysed hydrolysis of ethyl acetate are
linear to log k for tert-butyl chloride solvolysis in the
range up to 55 mol% of the alcohol. It can be seen that
the slopes of these linear plots significantly depend on
the nature of the reagents and change from positive to
negative values.
Similar linear plots were obtained for 70 different
reactions in water-methanol, water-ethanol, wateracetone, water-1,4-dioxane, water-acetonitrile, and
water-DMSO mixtures (Table 1). All these correlations are
listed in the Supplemental Materials, where the values of
the similarity coefficients Csim and all the statistical data
are presented. No compilation of data from different
literature sources was performed; however, if needed,
interpolation of data for the tert-butyl chloride solvolysis
was used to find points for the corresponding solvent
compositions. Exclusively, data comprising at least
four points were subjected to statistical analysis. As a
rule, extrapolated data were not used in this analysis
and coincidence of intercepts of Eq. 2 for cosolvents
was interpreted as an indication of reliability of the
correlations. In addition, data determined at 25oC or at
temperatures close to 25oC were preferred. In a number
of cases the studied range of the solvent composition was
determined by the availability of data, since the reported
measurements do not always cover the 0 to 100% range
of the molar content of the cosolvent in the mixture.
For a few reactions the bell-shaped dependences on the
log ktBuCl were obtained. If linearity in such plots could
be found, the water-rich region was preferred. For all
161 linear plots listed in the Supplemental Materials the
correlation coefficient is large and remains greater than
0.996 in a half of the cases. These results demonstrate
significant generality of the LFE relationship, Eq. 2, in
different aqueous solvents.
A certain similarity was found in the large ranges
of solvent composition, up to 100% of cosolvent
if experimentally available, not only for solvolysis
1602
Unauthenticated
Download Date | 6/15/17 6:39 PM
S. Salmar et al.
Table 1. Survey of the similarity analysis of kinetic dataa.
Reaction type
Number of
reactions
Number of cosolvents
in entries
Similarity
coefficientsb
solvolysis of haloalkanes
22
1-5
0.84 – 1.37
solvolysis of carbonyl halides
1
3
0.44 – 0.60
solvolysis of chloroformates
3
1-3
0.212 – 0.356
solvolysis of carbamoyl halides
4
3
0.59 – 0.89
solvolysis of sulphonyl halides
6
2-3
0.28 – 0.64
solvolysis of phosphorochloridates
3
1-3
0.24 – 0.41
hydrolysis of phosphates and
phosphinates
5
1
0.14 – 0.35
hydrolysis of sulphonyl esters
6
1-3
0.25 – 0.96
acidic hydrolysis of esters
6
1-3
0.09 – 0.24
neutral hydrolysis of esters
3
1
0.33 – 0.74
alkaline hydrolysis of esters
3
1-3
0.11 – 0.62
other reactions
8
1-4
-0.62 – 0.29
Total
70c
Complete tables containing similarity coefficients, range of cosolvents in mol%, intercepts of the similarity correlations with statistical errors, correlation
coefficients, logarithms of partition coefficients of compounds between water and 1-octanol (log P) with statistical errors, and the list of references can
be found in Supplemental Materials. b The reference reaction is solvolysis of tert-butyl chloride (2-chloro-2-methylpropane). c Total number of similarity
correlations is 161.
a
reactions of haloalkanes but also for some sulphonyl
chlorides and esters of sulphonic acids. The values of
similarity coefficients are large or even close to unity
for haloalkanes. For reactions like ester hydrolyses,
solvolysis of sulphonyl chlorides, and some others, the
similarity coefficients are small or even negative and the
linear correlation holds in the range from pure water up
to 20 or even 55 mol% of the cosolvent. Evidently, the
water-rich region of the binary aqueous solvents can
be extended to systems containing 20 mol% or more
of the cosolvent (cf. also [19,20]). The striking similarity
observed for all the reactions studied in mixtures containin
water and an organic cosolvent is certainly caused to
some extent by the preferential solvation of the reacting
substrates by water, independently of the specific watercosolvent mixtures used [45]. If not caused by changes
in the reaction mechanism, the region where the linear
correlations cease to hold can possibly be related to the
changes in the molecular structure of the solvent system
and accordingly in the solute-solvent interactions.
Spectroscopic, X-ray diffraction, neutron diffraction,
and mass spectrometric experimental studies have shed
light on the structure of water-organic binary mixtures
[7,46-52]. Experimental evidence has been found for
incomplete mixing at the microscopic level occurring in
these binary solvents. This means that water-organic
binary mixtures contain microscopic phase separation
at the cluster level [7,46-49,51,52], i.e., the water-rich
clusters and the cosolvent-rich clusters coexist showing
microscopic phase separation in wide mixing ratios.
Water molecules promote self-association of organic
molecules as a balance of interactions controlling the
microscopic structure in the solution [51].
Recent experimental data suggested that the
microscopic solvent structure is the major factor
governing solvation of added solutes [7]. It has been
found that solvation of the solute in a binary aqueousorganic solvent mixture is strongly related to the mixed
solvent cluster structure, e.g., alcohols preferentially
interact with the water clusters when those are stabilised
in the mixed solvent, however; phenol is excluded from
clustering with water molecules [7]. These and related
results complement and suitably extend the concept of
preferential solvation of reacting substrates.
An impressive conclusion reached from the results of
our LFE analysis is that statistically reliable similarities
have been found between significantly different reactions
(cf. Table 1). There seems to be a very specific feature
affecting the reactions occurring in aqueous binary
mixtures, as the striking difference between solvent
1603
Unauthenticated
Download Date | 6/15/17 6:39 PM
Role of water in determining organic
reactivity in aqueous binary solvents
Table 2. Correlations of similarity coefficients for reactions in binary aqueous mixtures with those for reactions in water-ethanol mixtures.
cosolvent
slope
intercept
number of points
correl. coeff. R
methanol
0.941 ± 0.029
0.022 ± 0.025
27
0.989
acetone
0.974 ± 0.050
0.057 ± 0.040
32
0.959
1,4-dioxane
1.124 ± 0.070
-0.045 ± 0.052
13
0.979
acetonitrile
0.927 ± 0.042
0.037 ± 0.023
8
0.995
All cosolventsa
0.994 ± 0.027
0.017 ± 0.022
82
0.972
a
Points for DMSO included.
effects, e.g. in SN1 and SN2 reactions, is one of the
fundamental principles of physical organic chemistry.
In fact, both these mechanisms are represented in
the LFE analysis. The linearity of these plots within a
large range of water content indicates that independent
of the reaction mechanism and type of solvent added,
similar interactions between substrates and the reaction
medium govern these processes.
3. Role of cosolvents in determining
the reactivity
Numerical values of the similarity coefficients
determined in this work reflect the susceptibility of
reaction rates to changes in the solvent composition
relative to the solvolysis of tert-butyl chloride (2-chloro2-methylpropane) in the same binary solvent mixtures.
Examination of the similarity coefficients revealed
little differences between the similarities for reactions
carried out in different water-organic binary solvents.
The similarity coefficients for reactions in the watermethanol, water-acetone, water-1,4-dioxane, and
water-acetonitrile mixtures were plotted against the
similarities for the water-ethanol solvent system. The
results of these correlations are collected in Table 2.
The binary solvent mixture water-ethanol was chosen
as the reference system because the most numerous
data were determined in this solvent.
It is remarkable that similarity coefficients for
different reactions lay well on a common straight line
with the slope close to unity (Fig. 2). The intercepts of
similarity plots give similar values for all cosolvents (See
Supplemental Materials and Fig. 3). This is in a good
agreement with their physical meaning as the reaction
rate in water.
Noteworthy, previous attempts have been made
to clarify the role of water and cosolvents. Fainberg
and Winstein found that the plots of log kRBr vs
log kRCl for solvolyses of tert-butyl and α-phenylethyl
halides in mixtures of water with ethanol, methanol,
and 1,4-dioxane “formed a good single straight line” [53].
The same was found for the two pairs of organic halides
- neophyl bromide and neophyl chloride, as well as for
α-phenylethyl chloride and neophyl chloride [54]. Bentley
et al. reported very similar solvent effects for solvolyses
of 4-methoxybenzoyl and 4-methoxybenzyl chlorides.
Remarkable features of the correlations reported for five
aqueous binary mixtures were the slopes close to unity
and the small range of values for the intercepts [55].
Later on, the group of Bentley and Koo [56] concluded
that the main effects of added organic cosolvents are
due to changes in water concentration.
Correlations presented in Table 2 confirm and
generalise the conclusions drawn by Bentley et al. [56]
since in the correlations determined by us a substantially
larger set of very different reactions was investgated.
It is evident that organic cosolvents virtually dilute the
aqueous solution, at least in the water-rich region, thus
affecting the solvation of substrates. This conclusion
may seem rather trivial: if one adds an organic cosolvent
to water, the overall water concentration decreases.
However, what is remarkable, the same effect is
exerted by cosolvents of different polarity, acidity or
hydrophobicity. Although effects of cosolvents on
polarity and nucleophilicity of the solvent system cannot
be ignored, in our similarity model, log k vs log kt-BuCl,
these effects seem to be mutually cancelled to some
extent.
4. Reactivity vs. hydrophobicity of
reagents
The results of our LFE analysis (Supplemental
Materials) showed that the similarity coefficient Csim
significantly depended upon the structure of the reacting
compounds. At the same time this value was not much
affected by the reaction mechanism, e.g. similar values
of the similarity coefficients were determined for acidic
and alkaline hydrolysis of ethyl acetate in aqueous
ethanol (0.24 and 0.31), acetone (0.12 and 0.14) and
1604
Unauthenticated
Download Date | 6/15/17 6:39 PM
S. Salmar et al.
Figure 2.
Similarity coefficients, Csim, for reactions in other five water-organic mixtures plotted against the similarities for water-ethanol solvent
system. ○- water-methanol; ●- water-acetone; +- water-1,4-dioxane; ■-water-acetonitrile; □- water-DMSO.
Figure 3. Intercepts of Eq. 2 for
systems.
other five water - cosolvent systems (with error limits) plotted vs the intercepts for corresponding water-ethanol
1,4-dioxane (0.09 and 0.11) solutions. This means that
the major part of the solvent effect observed in the binary
water-solvent mixtures was determined by a uniform
interaction between the reacting compound and the
reaction medium. A similar conclusion was drawn from
the results of sonochemical experiments [29,30], which
inferred additionally that this interaction could include a
contribution from the hydrophobic effect.
To test this assumption we compared the values
of the similarity factors with hydrophobicity of reacting
compounds. It is evident that this comparison could be
sound in cases where the same or very similar second
reagent is involved in the reaction. In most of the cases
analysed in this study either water plays the role of
the second reagent or the reaction is unimolecular
(the solvolysis reactions). For these reactions the
hydrophobicity parameters of substrates [57,58], i.e.,
the logarithms of partition coefficients of compounds
between water and 1-octanol (log P) were calculated
[59] (See Supplemental Materials). These partition
parameters have been extensively used as a measure of
the hydrophobic character of compounds [4,58], although
the log P values seem to comprise contributions of molar
refractivity, polarisability, hydrogen bond acidity and
basicity, etc. [60]. The log P value for trimethylsulfonium
ion was calculated on the basis of the experimental data
published by Sikk et al. [61].
In the context of solvent effects in binary solvent
mixtures, attempts have been made to quantitatively
describe the contribution of organic cosolvents by using
their log P values [16,62,63]. However, unlike in those
experiments, we have used hydrophobicity parameters
of solutes in this study. It can be seen in Fig. 4 that
the intensity of the relative solvent effect, quantified
by the value of Csim, correlates with the hydrophobicity
parameters of certain organic reagents. The set of
points plotted in Fig. 4 comprises of those recorded for
hydrolysis of esters of carbonic and sulphonic acids, and
1605
Unauthenticated
Download Date | 6/15/17 6:39 PM
Role of water in determining organic
reactivity in aqueous binary solvents
Figure 4. Plot of Csim vs log P. ○- hydrolyses
of esters of sulphonic acids; ●- hydrolyses of esters of carbonic acids; □-solvolyses of sulfonyl
chlorides; ■-solvolyses of chloroformates; +-solvolyses of carbamoyl chlorides; ∆- decomposition of trimethylsulfonium ion.
solvolysis of sulphonyl chlorides and chloroformates.
The correlation with the log P values is quite good as
seen in Eq. 6.
(6)
This calculation did not include the data-point for
decomposition of the sulfonium ion. However, this point
lies very close to the correlation line in Fig. 4, although
hydrophobicity of this cationic reagent is dramatically
different from the hydrophobicity of all other reagents
under consideration. In spite of that, the hydrophobicity
plot summarised by Eq. 6 also holds for this compound.
According to the general principle of correlation
analysis, Eq. 4 reveals that over 80% of changes in Csim
(R2 = 0.817) could be described by the properties of
substrates expressed in log P terms and thus pointing
to the involvement of hydrophobic phenomena in these
cases. This result cannot be ignored in any further
analysis of solvent effects in binary water-organic
systems.
The remainder of the data is almost independent on
the log P values (not shown in Fig. 4) and consists of the
Csim values for solvolysis of haloalkanes, most of which
exceed 0.8. A more detailed examination of these data
revealed some regularities in distribution of Csim values.
Similarities for solvolysis reactions of haloalkanes, except
those for adamantyl halides, were scattered around
Csim = 1 for tert-butyl chloride (2-chloro-2-methylpropane).
Keeping in mind the basic principles of the similarity
analysis, it can be concluded that in these cases
(Csim close to unity) many effects have been mutually
compensated, including those jointly described by the
log P parameters [61]. However, some properties were
obviously not entirely compensated, as Csim for bromoand iodoalkanes were systematically lower than those
for chloro derivatives (See Supplemental Materials).
Most probably this can be related to differences in
polarisabilities of the substrates. Solvolyses of carbamoyl
chlorides exhibit lower similarity coefficients and form a
separate dependence on the log P values as can be
seen in Fig. 4.
In summary, while the similarity model used in this
study holds independent of the reaction mechanisms
(See Sections 2 and 3), the Csim values themselves
seem to be determined additionally by the reaction
mechanism, while the general pattern of solvation
effects seems to be similar for different reactions.
Our approach is focused on the solutes instead
of the solvents, thus susceptibilities of reactions
to changes in solvation of substrates should be
considered. As the log P values include good portions
of substrate electrophilicity [61] the dependence of the
Csim values on log P for reactions represented in Fig. 4
reflects greater susceptibilities to solvent nucleophilicity
inherent to these reactions. In contrast to this, reactions
more similar to the solvolysis of tert-butyl chloride are
typical for substrates of relatively high hydrophobicity
and lesser demand for the nucleophilic assistance when
being solvolysed. Independence of Csim on the log P
values can be related to the solvation patterns being very
similar to those for tert-butyl chloride. On the other hand,
results of our sonochemical experiments have pointed to
possible dissimilarities in solvation of tert-butyl chloride
and alkyl esters in binary aqueous organic solvent
mixtures [29,30]. Similar and more general conclusions
can be inferred from the results of the current study.
Clearly, the data need an interpretation in terms of the
transition state theory [20], however currently there is no
sufficient information to support such analysis, including
the values of activation parameters.
Correlations with hydrophobicity parameters are
intriguing; however, they cannot provide an unambiguous
1606
Unauthenticated
Download Date | 6/15/17 6:39 PM
S. Salmar et al.
proof for dominating hydrophobic interactions in all
the cases. As currently reported [10,11], hydrophobic
effects are important over only relatively small cosolvent
concentration ranges. At larger cosolvent concentrations
hydrophobic hydration shells are broken down and
many complicated intermolecular interactions occur. In
this context, it is interesting to conclude that for all the
reactions studied, a similar mixture of intermolecular
interactions is playing a role with different sensitivities
for the different reactions. Moreover, considering the
extensive stretch of most similarity plots over solvent
compositions, it can be assumed that even at significant
“dilution” by an organic cosolvent, water retains its
specific solvating properties.
The relationships discussed above can also be
explained by partition of reacting compounds between
the solvent-rich and the water-rich clusters, assuming
that reaction rates for these micro-media are different
[7,10,11]. This conceivable model is implicitly related
to the quantitative approach used by the Engberts
group [19,64]. However, this approach was specifically
developed for low concentrations of cosolvents.
5. Conclusions
Similarity, defined in terms of LFE relationships between
kinetic data for the tert-butyl chloride (2-chloro-2methylpropane) solvolysis and reactions of largely
different mechanisms, was observed for a variety
of binary water-organic solvents. These linear plots
indicated that similar solute-solvent interactions should
govern these processes. Further, these interactions
were associated with hydrophobicity of substrates,
as the similarity coefficients were correlated with the
hydrophobicity parameters of reagents. This means that
the influence of organic cosolvents could be explained
by dilution of water in the binary solvent mixtures that
changes the solvation effect. This mechanism should be
valid at least in the water-rich regions where the linear
similarity plots can be observed.
Acknowledgements
This work was financially supported by the Estonian
Ministry of Education and Research (Grant
SF0180064s08) and by Estonian Science Foundation
(Grant ETF7498 and ETF9268).
Supplemental Materials Available
Tables containing similarity coefficients, ranges
of cosolvents in mol%, intercepts of the similarity
correlations, correlation coefficients, logarithms of
partition coefficients of compounds between water and
1-octanol (log P) with statistical errors, and the list of
references are in Supplemental Materials, which can be
found in the online version of this paper.
References
[1] R. Breslow, In: M. Lindström (Ed.), Organic
Reactions in Water (Blackwell Publishers, Oxford,
2007) Chapter 1
[2] T.J. Mason, P. Cintas, In: J. Clark, D. Macquarrie
(Eds), Handbook of Green Chemistry (Blackwell
Publishers, Oxford, 2000)
[3] F. Franks (Ed.), Water: a Comprehensive Treatise
(Plenum Press, New York, 1972-1982) Vols. 1-7
[4] C. Reichardt, T. Welyon, Solvents and Solvent
Effects in Organic Chemistry, 4th edition (WileyVCH, Weinheim, 2011)
[5] J.B.F.N. Engberts, In: F. Franks (Ed.) Water: a
Comprehensive Treatise (Plenum Press, New York,
1979) Vol. 6, Chapter 4
[6] M.J. Blandamer, J. Burgess, J.B.F.N. Engberts,
P. Warrick Jn., J. Mol. Liq. 52, 15 (1992)
[7] D.N. Shin, J.W. Wijnen, J.B.F.N. Engberts,
A. Wakisaka, J. Phys. Chem. B 106, 6014 (2002)
[8] O.A. El Seoud, Pure Appl. Chem. 81, 697, (2009)
[9] P.L. Silva, M.A.S. Trassi, C.T. Martins,
O.A. El Seoud, J. Phys. Chem. B 113, 9512 (2009)
[10] (a) W. Blokzijl, J.B.F.N. Engberts, Angew. Chem.
105, 1610 (1993); (b) W. Blokzijl, J.B.F.N. Engberts,
Angew. Chem. Int. Ed. Engl. 32, 1545 (1993)
[11] S. Otto, J.B.F.N. Engberts, Org. Biomol. Chem. 1,
2809 (2003)
[12] C. Tanford, The Hydrophobic Effect, 2nd edition
(Wiley, New York, 1980)
[13] N.T. Southall, K.A. Dill,A. D.J. Haymet, J. Phys.
Chem. B 106, 521 (2002)
[14] U.M. Lindström, Chem. Rev. 102, 2751 (2002)
[15] C. Li, Chem. Rev. 105, 3095 (2005)
[16] C.T. Martins, M.S. Lima, O.A. El Seoud, J. Org.
Chem. 71, 9068 (2006)
[17] K.A. Connors, In: G. Wypych (Ed.), Handbook of
Solvents (ChemTec Publishing, Toronto, 2001) 281
[18] W. Blokzijl, J.B.F.N. Engberts, J. Jager,
M.J. Blandamer, J. Phys. Chem. 91, 6022 (1987)
[19] T. Rispens, C. Caballeiro-Lago, J.B.F.N. Engberts,
Org. Biomol. Chem. 3, 597 (2005)
1607
Unauthenticated
Download Date | 6/15/17 6:39 PM
Role of water in determining organic
reactivity in aqueous binary solvents
[20] T. Rispens, J.B.F.N. Engberts, J. Phys. Org. Chem.
18, 725 (2005)
[21] T.W. Bentley, H.C. Harris, Z.H. Ryu, G.T. Lim,
D.D. Sung, S.R. Szajda, J. Org. Chem. 70, 8963
(2005)
[22] T.W. Bentley, M.S. Garley, J. Phys. Org. Chem. 19,
341 (2006)
[23] M.J. D’Souza, M.J. McAneny, D.N. Kevill,
J.B. Kyong, S.H. Choi, Beilstein J. Org. Chem. 7,
543 (2011), and references therein.
[24] D.N. Kevill, K.H. Park, H.J. Koh, J. Phys. Org.
Chem. 24, 378 (2011)
[25] K.H. Park, D.N. Kevill, J. Phys. Org. Chem. 25, 267
(2011)
[26] R. Breslow, J. Phys. Org. Chem. 19, 813 (2006)
[27] A. Tuulmets, S. Salmar, H. Hagu, J. Phys. Chem. B
107, 12891 (2003)
[28] A. Tuulmets, H. Hagu, S. Salmar, G. Cravotto,
J. Järv, J. Phys. Chem. B 111, 3133 (2007)
[29] A. Tuulmets, J. Järv, S. Salmar, G. Cravotto,
J. Phys. Org. Chem. 21, 1002 (2008)
[30] A. Tuulmets, S. Salmar, J. Järv, Sonochemistry in
Water Organic Solutions (Nova Science Publishers,
New York, 2010)
[31] A. Tuulmets, J. Järv, T. Tenno, S. Salmar, J. Mol.
Liq. 148, 94 (2009)
[32] A. Williams, Free Energy Relationships in Organic
and Bio-organic Chemistry, (RSC, Cambridge,
2003)
[33] E. Grunwald, S. Winstein, J. Am. Chem. Soc. 70,
846 (1948)
[34] R.I. Zalewski, T.M. Krygowski, J. Shorter, Similarity
Models in Organic Chemistry, Biochemistry and
Related Fields (Elsevier, New York, 1991)
[35] A.H. Fainberg, S. Winstein, J. Am. Chem. Soc. 78,
2770 (1956)
[36] R.E. Robertson, Can. J. Chem. 50, 1353 (1972)
[37] K. Heinonen, E. Tommila, Suomen Kem. B38, 9
(1965)
[38] S. Winstein, E. Grunwald, H.W. Jones, J. Am.
Chem. Soc. 73, 2700 (1951)
[39] D.N. Kevill, M.J. D’Souza, J. Chem. Res. 61-66
(2008)
[40] E. Tommila, A. Koivisto, J.P. Lyyra, K. Antell,
S. Heimo, Ann Acad. Sci. Fennicae A11, Nr. 47
(1952)
[41] A.H. Fainberg, S. Winstein, J. Am. Chem. Soc. 79,
1602 (1957)
[42] D.N. Kevill, M.J. D’Souza, J. Org. Chem. 69, 7044
(2004)
[43] M. Harati, M.R. Gholami, Int. J. Chem. Kinetics
37(2), 90 (2005)
[44] A.A. Sosunov, A.P. Kilimov, V.V. Smirnov, Zh. Obch.
Khimii 40, 2688 (1970) (in Russian)
[45] Y. Marcus, Solvent Mixtures-Properties and
Selective Solvation (M. Dekker, New York, Basel,
2002)
[46] K. Egashira, N. Nishi, J. Phys. Chem. B 102, 4054
(1998)
[47] T. Takamuku, T. Yamaguchi, M. Asato,
M. Matsumoto, N. Nishi, Z. Naturforsch., Part A
55a, 513 (2000)
[48] A. Wakasaka, S. Komatsu, Y. Usui, J. Mol. Liq. 90,
175 (2001)
[49] S. Dixit, J. Crain, W.C.K. Poon, J.L. Finney,
A.K. Soper, Nature 416, 829 (2002)
[50] Y. Liu, X. Luo, Z. Shen, J. Lu, X. Ni, Optical Rev. 13,
303 (2006)
[51] A. Wakisaka, K. Matsura, J. Mol. Liq. 129, 25
(2006)
[52] N.P. Hu, D. Wu, K. Cross, S. Burikov, T. Dolenko,
S. Patsaeva, D.W. Schaefer, J. Agric. Food Chem.
57, 7394 (2010)
[53] A.H. Fainberg, S. Winstein, J. Am. Chem. Soc. 79,
1602 (1957)
[54] A.H. Fainberg, S. Winstein, J. Am. Chem. Soc. 79,
1608 (1957)
[55] T.W. Bentley, I.S. Koo, S.J. Norman, J. Org. Chem.
56, 1604 (1991)
[56] T.W. Bentley, I.S. Koo, H. Choi, G. Llewellyn,
J. Phys. Org. Chem. 21, 251 (2008)
[57] A. Leo, C. Hansch, D. Elkins, Chem. Rev. 71, 525
(1971)
[58] C. Hansch, Acc. Chem. Res. 2, 232 (1969)
[59] ACD/ChemSketch Freeware (Advanced Chemistry
Development, Inc., Canada, 2010)
[60] M.H. Abraham, J.C. Dearden, G.M. Bresnen,
J. Phys. Org. Chem. 19, 242 (2006)
[61] P. Sikk, A. Aaviksaar, A. Abduvahhabov, Eesti NSV
Tead. Akad. Toim., Keem., Geol. 26(3), 242 (1977)
(in Russian)
[62] K.A. Connors, M.J. Mulski, A. Paulson, J. Org.
Chem. 57, 1794 (1992)
[63] J.M. LePree, M.J. Mulski, K.A. Connors, J. Chem.
Soc. Perkin Trans. 2 1491 (1994)
[64] N.J. Buurma, L. Pastorello, M.J. Blandamer,
J.B.F.N. Engberts, J. Am. Chem. Soc. 123, 11848
(2001)
1608
Unauthenticated
Download Date | 6/15/17 6:39 PM