Competitive homolytic and heterolytic dediazoniation mechanisms: Rate constants and product distribution of methoxy-, hydroxy-, and hydro-dediazoniation of 3- and 4-methylbenzenediazonium salts in acidic MeOH/H2O mixtures

Article abstract of DOI:10.1002/(SICI)1097-4601(2000)32:4<210::AID-KIN3>3.0.CO;2-8

The rates and product distribution for methoxy-, hydroxy- and hydro-dediazoniation and the rate constants for disappearance of 3- and 4-methylbenzenediazonium tetrafluoroborate in acidic MeOH/H₂O mixtures, in the presence and absence of electrolytes like HCl, NaCl, and CuCl₂, are reported. Data were obtained by using a combination of VIS-UV and HPLC techniques. The kinetics and product distributions are completely consistent with competitive homolytic and heterolytic mechanisms, the heterolytic one being predominant at any solvent composition. Heterolytic data are in agreement with the predictions of a D_N+A_N mechanism; that is, rate determining formation of an aryl cation that reacts immediately with available nucleophiles. Selectivity values, determined from product yields, are low and independent of solvent composition. Product formation is discussed in terms of a preassociation step between aryl cations and the nucleophile, which does not account for much of the trapping, and a nucleophilic attack on a `free' arenediazonium cation. Activation parameters were also determined at 99.5% MeOH: enthalpies of activation are high and entropies of activation are positive, and they are similar to those reported for pure water.

Full text of DOI:10.1002/(SICI)1097-4601(2000)32:4<210::AID-KIN3>3.0.CO;2-8

Competitive Homolytic and
Heterolytic Dediazoniation
Mechanisms: Rate
Constants and Product
Distribution of Methoxy-,
Hydroxy-, and Hydro-
Dediazoniation of 3- and
4-Methylbenzenediazonium
Salts in Acidic MeOH/H2O
Mixtures
ROMAN PAZO-LLORENTE,1 ELISA GONZALEZ-ROMERO,2 CARLOS BRAVO-DIAZ1
1Universidad de Vigo, Facultad, de Ciencias, Dpto. Quimica Fisica y Organica, As Lagoas-Marcosende, 36200 Vigo-
Pontevedra, Spain
2Universidad de Vigo, Facultad, de Ciencias, Dpto. Quimica Analitica y Alimentaria, As Lagoas-Marcosende, 36200 Vigo-
Pontevedra, Spain
Received 3 May 1999; accepted 22 December 1999
ABSTRACT: The rates and product distribution for methoxy-, hydroxy- and hydro-dediazonia-
tion and the rate constants for disappearance of 3- and 4-methylbenzenediazonium tetrafluo-
roborate in acidic MeOH/H2O mixtures, in the presence and absence of electrolytes like HCl,
NaCl, and CuCl2, are reported. Data were obtained by using a combination of VIS-UV and
Correspondence to: Carlos Bravo-Diaz
E-mail: cbravo@uvigo.es)
Contract grant sponsor: Ministerio de Educacio´n y Cultura
Contract grant number: DGICYT, PB94-0741
Contract grant sponsor: Xunta de Galicia
Contract grant number: XUGA 38305A94
Contract grant sponsor: Universidad de Vigo
᭧ 2000 John Wiley & Sons, Inc.

COMPETITIVE HOMOLYTIC AND HETEROLYTIC DEDIAZONIATION MECHANISMS
211
HPLC techniques. The kinetics and product distributions are completely consistent with com-
petitive homolytic and heterolytic mechanisms, the heterolytic one being predominant at any
solvent composition. Heterolytic data are in agreement with the predictions of a DN ϩ AN
mechanism; that is, rate determining formation of an aryl cation that reacts immediately with
available nucleophiles. Selectivity values, determined from product yields, are low and inde-
pendent of solvent composition. Product formation is discussed in terms of a preassociation
step between aryl cations and the nucleophile, which does not account for much of the trap-
ping, and a nucleophilic attack on a “free” arenediazonium cation. Activation parameters were
also determined at 99.5% MeOH: enthalpies of activation are high and entropies of activation
are positive, and they are similar to those reported for pure water. ᭧ 2000 John Wiley & Sons,
Inc. Int J Chem Kinet 32: 210–220, 2000
INTRODUCTION
nors, heterolytic dediazoniations mainly take place in
solvents of low nucleophilicity like H2O or TFE;
meanwhile, homolytic dediazoniations are favored by
an increase in the nucleophilicity (HMPT, Pyridine).
MeOH, EtOH, and DMSO have been reported as bor-
derline solvents where both mechanisms can be ob-
served, depending on several factors like pH [5], sub-
stituents of the aromatic ring [7,8], or even the
atmosphere [8] (O2 or N2). The source of electrons is
thought to be the solvent (i.e., D ϭ MeOH), although
the mechanism and the nature of the initiation step is
unclear [5]. Bunnet et al. [7,8] postulated that, in al-
kaline methanol solutions, the source of radicals is the
homolytic decomposition of a (Z)-diazo methyl ether,
which is formed rapidly, yielding the aryl radical, N2,
and the methoxy radical; meanwhile, under acidic con-
ditions they favor initiation by direct electron transfer
from MeOH to the arenediazonium ion [7]. A partic-
ularly interesting aspect is the gradual changeover
from heterolytic to homolytic dediazoniation, which
led to the question of a common intermediate for these
ϩ
The reactions of arenediazonium ions, ArN2 , Scheme
I, have attracted considerable attention [1–3], both
from a synthetic and from a mechanistic point of view,
because of their extraordinary sensitivity to environ-
mental changes [4,5]. In spite of the fact that a sub-
stantial body of knowledge about their reactions is
available, some of their mechanisms are not com-
pletely understood. Electron density analysis [6] of a
number of diazonium ions has shown that both ali-
phatic and aromatic diazonium ions are best thought
of as carbenium ions closely associated with an inter-
nally polarized N2 molecule, which implies a dative
C9N bond (N:C ␴ donation and C:N ␲ backdo-
nation).
Solvolytic dediazoniations have been studied in a
number of solvents [4,5] and, to date, it is believed
that they take place through two main mechanisms,
Scheme II, heterolytic, and homolytic. Literature re-
ports [5] indicate that, in the absence of electron do-
Scheme I Reactions of arenediazonium ions. (a) Dediazoniation. (b) Representative nucleophilic
addition: coupling reaction with 2-napthol-6-sulphonic, sodium salt.

212
PAZO-LLORENTE, GONZALEZ-ROMERO, AND BRAVO-DIAZ
Scheme II Dediazoniation mechanisms. (a) Heterolytic. (b) Homolytic.
two types of mechanisms [9]. Investigations devel-
oped by Bunnet et al. [7,8] and by Kuokkanen [10]
suggest that there is no need to invoke such a common
intermediate and that the formation of both homolytic
and heterolytic products can be explained in terms of
competitive mechanisms that are taking place simul-
taneously.
the relatively few data published on rates of product
formation and on alcohol/water selectivities for the
highly unstable arenediazonium cations, encouraged
us to examine the kinetics and mechanism of the de-
diazoniation of 3- and 4-methylbenzenediazonium te-
trafluoroborate (MMBDand PMBD, respectively) in
binary methanol-water acid solutions over the whole
composition range. The aim of the manuscript is two-
fold: to complement studies concerning nucleophilic
attack on the short-lived aryl cations and to use it as a
starting point for future solvolytic dediazoniations in
binary mixtures and in colloidal systems [21–23] and
in macromolecular systems [4,5,24]. Colloidal sys-
tems show significant gradients of polarity on going
from the bulk phase to the core of the aggregate; mean-
while, macromolecular systems show relatively hydro-
phobic microscopic environments where arenediazon-
ium ions can be located [21,25], thus competing with
solvents for the macromolecular cavity or micellar
sites. Therefore, those systems can determine the ar-
enediazonium chemical reactivity by governing their
contact with other substrates, which include the nucle-
ophilic solvent molecules, incorporated in the system
[21–23] and by governing the polarity of their im-
mediate environment, which can be decisive for uni-
molecular decomposition reactions [24,26]. Few ref-
erences about activation parameters of dediazoniations
in methanol [9,10] can be found in the literature, al-
though a number of them involve other solvents
[9,27–29], so we have also determined their activation
parameters to get insights into the nature of the tran-
sition state.
Evidence for the involvement of free radicals in
dediazoniations has been obtained primarily by prod-
uct analyses (primarily by HPLC and GC-MS) and
from EPR measurements [11–13], and evidence for
aryl cation intermediates in dediazoniations has been
reported by several authors [5,14,15]. Experiments de-
veloped by Swain et al. [16] suggest that the aryl cat-
ion is formed reversibly, and a molecular orbital study
of the benzenediazonium cation in water [17] is con-
sistent with the formation of stable molecule-ion pairs.
Zollinger and coworkers [18] concluded that molecu-
lar nitrogen reacts reversibly with the phenyl cation
intermediate, and a number of flash photolysis studies
[19] confirm that aryl cations are real intermediates in
heterolytic dediazoniations although their lifetimes are
extremely short, on the order of a few nanoseconds.
The selectivity, S, of the reaction toward nucleophiles
can be defined by Eq. 1, where C stands for percent
of conversion to the corresponding dediazoniation
product, and nucleophile concentrations are assumed
to refer to the reaction site.
kNu
CR Nu [Nu2]
Ϫ
1
1
Nu1
S
ϭ
ϭ
(1)
Nu2
k
Nu2
CR [Nu1]
Nu2
Ϫ
EXPERIMENTAL
Instrumentation
Dediazoniations in MeOH [7–9] were mainly stud-
ied under alkaline conditions, usually at one fixed sol-
vent composition, in contrast to the relatively few
number of studies done under acidic conditions [20].
The lack of studies under such acidic conditions, and
UV-VIS spectra and some kinetic experiments were
followed on a Beckman DU-640 UV-VIS spectropho-

COMPETITIVE HOMOLYTIC AND HETEROLYTIC DEDIAZONIATION MECHANISMS
213
tometer equipped with a thermostated cell carrier at-
tached to a computer for data storage. Product analysis
was carried out on a WATERS HPLC system, which
included a 510 pump, a 717 automatic injector, a 486
VIS-UV detector, and a computer for data storage.
Products were separated on a Microsorb-MV C-18
(Rainin) reverse-phase column (25 cm length, 4.6 mm
internal diameter, and 5 ␮m particle size) using a mo-
All runs were done at T ϭ 35 Ϯ 0.1ЊC (MMBD)
and at T ϭ 60 Ϯ 0.1ЊC (PMBD) with diazonium salts
as the limiting reagents. Duplicate or triplicate exper-
iments gave average deviations lower than 5%.
Spectrophotometric kinetic data were obtained by
following the disappearance of diazonium salt at an
appropriate wavelength to minimize interference
mainly by chlorocuprate(II) complexes. Preliminary
HPLC experiments showed that only four decompo-
sition products are formed: ArOH, ArCl, ArOMe, and
ArH. Linear (cc Ͼ 0.999) calibration curves for con-
verting HPLC peak areas, A, into concentrations were
obtained for these products by employing commercial
samples. The equations used were for MMBD( ␭ ϭ
220 nm): A ϭ 2.06 ϫ 109 [ArH], A ϭ 8.78 ϫ 109
[ArOH], A ϭ 1.30 ϫ 1010 [ArOMe], A ϭ 9.16 ϫ 109
[ArCl]; and for PMBD( ␭ ϭ 210 nm): A ϭ 8.46 ϫ
109 [ArH], A ϭ 9.78 ϫ 109 [ArOH], A ϭ 1.24 ϫ 1010
[ArOMe], A ϭ 1.31 ϫ 1010 [ArCl].
Chromatographic kinetic data for all dediazoniation
products were obtained following a published proce-
dure [30] by quenching the dediazoniation reaction at
convenient times with an aliquot of a stock quenching
solution prepared by dissolving 2N6S in a solution
containing TRIS buffer ([TRIS] ϭ 0.05 M) to give,
after mixing, final 2N6S concentrations about 20-fold
excess over that of arenediazonium salt and final pH
about pH ϭ 8. Details of the method are given else-
where [30]. Extra precautions have been taken to min-
imize MeOH evaporation when working at tempera-
tures close to its normal boiling point or when studying
the effect of temperature. Auxiliary experiments in-
dicate that negligible evaporation, if any, takes place
under our experimental conditions. Percentages of
MeOH will be given hereafter by volume, and molar
concentrations were calculated by ignoring the small
excess volume of mixed solvents [36].
4
bile phase of 70/30 v/v MeOH/H2O containing 10Ϫ
M HCl. The injection volume was 25␮L in all runs,
and the UV detector was set at 210 nm (PMBD) or
220 nm (MMBD).
Materials
Reagents were of the maximum purity available and
were used without further purification. Toluene,
ArH, cresols, ArOH, chlorotoluenes, ArCl, anisoles,
ArOMe, copper (II) chloride (99.999%), and the rea-
gents used in the preparation of diazonium salts (as
tetrafluoroborates) were purchased from Aldrich. 2-
Naphthol-6-sulfonic acid, sodium salt, (2N6S) was
purchased from Pfaltz & Bauer. Other materials em-
ployed were from Riedel de Ha¨en. All solutions were
prepared by using Milli-Q grade water.
Diazonium salts were prepared under nonaqueous
conditions [30] and were stored in the dark at low
temperature to minimize its decomposition and re-
crystallyzed periodically. Stock solutions were pre-
pared dissolving the diazonium salt in the appropriate
acidic (HCl) mixture to minimize diazotate formation
[31], to give final concentrations of about 1 ϫ 10Ϫ4 M
3
and [HCl] ϭ 3.6 ϫ 10Ϫ M and were generally used
immediately or within a short period of time with stor-
age in an ice bath to minimize decomposition. Beer’s
law plots (not shown) for MMBDand PMBDaqueous
4
and methanolic solutions up to 2.00 ϫ 10Ϫ M are
linear (cc ϭ 0.999).
RESULTS
Methods
(a) Effects of Solvent Composition and
Added Electrolytes on the Observed Rate
Constant k0
Kinetic data were obtained from spectrophotometric
and chromatographic (HPLC) data by employing a
well-established methodology [21,30,32–35]. Ob-
served rate constants were obtained by fitting the ab-
sorbance-time, percent yield-time data to the inte-
grated first-order equation (2) using a nonlinear least
squares method provided by a commercial computer
program, where M is the measured magnitude.
The effect of solvent composition on ko, in the absence
of added electrolytes, was investigated by changing
the percentage of MeOH in the reaction mixture, Fig-
ure 1. For both MMBDand PMBD, ko increases
smoothly from ko ϭ 7.97 ϫ 10Ϫ4 sϪ1 and ko ϭ 6.11 ϫ
10Ϫ4 sϪ1 up to ko ϭ 14.1 ϫ 10Ϫ4 sϪ1 and ko ϭ 14.2 ϫ
4
1
10Ϫ sϪ at 99.6 % MeOH/H2O, respectively. The ko
values at 0% MeOH are in agreement with those ob-
tained with different techniques [30,32,34].
Mt Ϫ Mϱ
ln
ϭ Ϫkot
(2)
ͩ ͪ
M0 Ϫ Mϱ

214
PAZO-LLORENTE, GONZALEZ-ROMERO, AND BRAVO-DIAZ
Figure 1 Variation of the observed dediazoniation rate constant, ko, with solvent composition.
Ϫ
4
᭺ MMBD, ᭹ PMBD. [MMBD] ϭ [PMBD] ϭ 1.0-2.0 ϫ10 M, [HCl] ϭ 0.01 M, T ϭ 35ЊC
(MMBD), T ϭ 60ЊC (PMBD)
The rate constants for product formation at different
the parent arenediazonium ion and the corresponding
aryl cation [32].
solvent compositions were obtained by HPLC. Results
in Figures 2 and 3 are representative. Figure 2 shows
the variation in the percent of conversion of PMBDto
ArOMe (2A), ArOH, and ArH (2B) with time and the
corresponding first-order plots in 98% MeOH: ko
The effect of added electrolytes like HCl (0–1.0
M), NaCl (0–1.0 M), and CuCl2 (0–0.01 M) on ko
was investigated at different MeOH/H2O mixtures (20,
40, 80 and 99.5% v/v). Tabulated data are given as
supplementary material. The observed rate constants
do not change significantly upon addition of any of
those electrolytes, exhibiting a similar variation as that
observed for other arenediazonium ions in water
[30,32]. It has been reported that CuCl2 has a negli-
gible effect on the observed rate constants of methyl-
benzenediazonium ions [30,32] but catalyzes the spon-
taneous decomposition of p-nitrobenzenediazonium in
aqueous solutions [33]. CuCl2 forms a number of chlo-
rocomplexes in solution whose relative concentration
depends on a number of factors such as halide con-
centration, temperature, and solvent [41], and up to
four complexes can be formed in aqueous solutions
values for product formation are the same, kArOMe
ϭ
1
13.8 ϫ 10Ϫ4 sϪ , kArOH ϭ 13.2 ϫ 10Ϫ4 sϪ1 and kArH ϭ
4
1
13.7 ϫ 10Ϫ sϪ , thus indicating that both homolytic
and heterolytic mechanisms are competitive, in agree-
ment with previous reports. Figure 3A shows the vari-
ation in the amount of ArOMe formed with time for
MMBDand the corresponding first-order plot, which
4
1
yields ko ϭ 13.60 ϫ 10Ϫ sϪ , a value equal, within
experimental error, to that obtained monitoring ArOH
formation (not shown) and to that obtained for MMBD
loss from the variation of the absorbance of the cor-
responding azo dye with time (Fig. 3B).
The sensitivity of the dediazoniation to changes in
solvent polarity can be shown by means of the Grun-
wald-Winstein equation [37–40], Figure 4, yielding
slopes of Ϫ0.037 (MMBD) and Ϫ0.07 (PMBD), con-
sistent with the low selectivity of dediazoniations to
medium effects [4,5], which has been attributed to the
similarity in structure and charge distribution between
Ϫ
containing ClϪ ions [41]: CuClϩ, CuCl2, CuCl3 and
CuCl4 Ϫ, CuCl3 and CuCl4 being predominant at
high chloride ion concentrations [42]. However, the
main copper(II) complexes formed in pure methanol
[43] are CuClϩ and CuCl2. Therefore, our results are
consistent with those expected from the interaction be-
2
Ϫ
2Ϫ

COMPETITIVE HOMOLYTIC AND HETEROLYTIC DEDIAZONIATION MECHANISMS
215
MMBD, in agreement with literature reports [5]. Re-
sults in pure water for OMBDand PMBDare in agree-
ment with published data [30,32,34].
Table I shows that the selectivity of the MMBD
and PMBDdediazoniation, SMWeOH, as defined by Eq.
1, is low and essentially independent of the percentage
of MeOH in the reaction mixture, suggesting that the
ratio of rates involved in Eq. 1 is insensitive to solvent
composition and polarity. The average S values,
SMWeOH ϭ 0.63 (MMBD) and SMWeOH ϭ 0.74 (PMBD),
are similar to those reported for other dediazoniations
in butanol [22], SWA ϭ 0.28 at T ϭ 40ЊC.
The effect of added electrolytes like HCl (0–1.0
M), NaCl (0–1.0 M), and CuCl2 (0–0.01 M) on prod-
uct yields was investigated at different MeOH/H2O
mixtures up to 75% MeOH v/v. Tabulated data are
given as supplementary material. Three heterolytic de-
diazoniation products, ArOH, ArCl, and ArOMe, are
obtained. Upon increasing the percentage of MeOH,
ArOMe yield increases with a concomitant decrease
Figure 2 (a) Variation in the yield of PMB-OMe with time
(᭺) and first-order plot (᭹). (b) Variation in the percent of
conversion to PMB-OH (᭺) and PMB-H (ٗ) and first-order
Ϫ
4
plots (᭹ and ᭿ respectively). [PMBD] ϭ 1.3 ϫ 10 M,
[HCl] ϭ 0.01 M, 98% MeOH, T ϭ 60ЊC
tween positive or neutral Cu(II) chlorocomplexes with
ArN2ϩ ions.
(b) Effects of Solvent Composition and
Added Electrolytes on Dediazoniation
Product Yields
Figure 5 shows the effects of solvent composition on
MMBDand PMBDdediazoniation product yields in
the absence of added electrolytes. Figure 5A shows
that for MMBDonly two dediazoniation products,
ArOH and ArOMe, are formed in significant yields.
The reduced ArH dediazoniation product is detected
in highly alcoholic solutions, but its yield is very low
compared with those of heterolytic products. Quanti-
tative conversion to products is achieved in all com-
position ranges. For PMBD(Fig. 5B), quantitative
conversion to products is achieved in all composition
ranges and product distribution is similar to that for
MMBD, but the formation of the corresponding re-
duced product, ArH, is a little more favored than for
Figure 3 (a) Variation in the yield of MMB-OME with
time (᭺) and ln plot (᭹). (b) Variation in the absorbance of
the azo dye (see text) with time (᭺) and ln plot (᭹) for
Ϫ
MMBD. [MMBD] ϭ 1.5 ϫ 10 4 M, [HCl] ϭ 0.01 M, 99%
MeOH, T ϭ 35ЊC.

216
PAZO-LLORENTE, GONZALEZ-ROMERO, AND BRAVO-DIAZ
Figure 4 Winstein-Grunwald plots for MMBD( ᭺) and PMBD( ᭹). Y values from reference 37.
in ArOH yield. Conversions to ArCl as low as 8% are
obtained at [ClϪ] ϭ 1.0 M (75% MeOH/H2O), and the
reduction product ArH is only detected at high per-
centages of MeOH, its yield being essentially inde-
pendent of electrolyte concentration.
For the sake of comparison, we have included pub-
lished activation parameters in pure water and MeOH
and ko values for the three methylbenzene derivatives
at one fixed temperature (T ϭ 40ЊC) at 99.5 % MeOH.
DISCUSSION
(c) Effect of Temperature on k0: Activation
Parameters
Observed rate constants of formation of homolytic and
heterolytic dediazoniation products are equal and
equal to those obtained for diazonium ion loss (Figs.
1, 2, and 3), thus the products are generated with the
same half-life. As noted, there is no need to invoke a
common intermediate [7,8] for homolytic and heter-
olytic dediazoniation mechanisms; therefore, our re-
sults are consistent with the competitive dediazonia-
tion mechanism indicated in Scheme II. HPLC data
(Fig. 5), show that quantitative conversion to products
is achieved in the whole composition range. At low
percentages of MeOH, two heterolytic products are
formed: ArOH and ArCl; however, at high methanol
content the reduction product ArH is detected in rel-
atively significant amounts, its yield being very low
compared to that of ArOMe. ArH formation is favored
by a para-methyl compared with a meta-methyl and
Experimental activation parameters were determined
by measuring ko at different temperatures for MMBD
at 99.5% MeOH, (Fig. 6A), and for PMBDat a num-
ber of solvent compositions, (Fig. 6B). Activation pa-
rameters were obtained according to the theory of ab-
solute rates by means of Eq. 3, where kB and h are the
Boltzmann and Planck constants, respectively.
kobs
kB
⌬S# ⌬H#
ln
ϭ ln
ϩ
Ϫ
(3)
ͩ ͪ ͩ ͪ
T
h
R
RT
Table II shows the activation energy and the acti-
vation parameters for the three methylbenzenediazon-
ium ions at 99.5% MeOH: values of ⌬H# are relatively
high compared with those for bimolecular reactions
[44,45], and the entropic term is substantially positive.

COMPETITIVE HOMOLYTIC AND HETEROLYTIC DEDIAZONIATION MECHANISMS
217
Table I Selectivity Values According to Equation 1 for
MMBD and PMBD at Different Solvent Compositions.
Data from Figure 5
% MeOH
SMWeOH (MMBD)
SMWeOH (PMBD)
10
20
30
40
50
60
70
80
90
0.63
0.62
0.65
0.65
0.66
0.63
0.63
0.62
0.60
0.67
0.77
0.77
0.75
0.75
0.73
0.73
0.70
0.70
Figure 5 Influence of solvent composition on dediazon-
iation product yields. ᭹ ArOH, ᭺ Ar-OME, ٗ Ar-H, ᭿
Total yield.
ortho-methyl, in agreement with literature reports
[5,11], but for both MMBDand PMBDdiazonium
salts, ArH yields are very low compared with those of
the heterolytic products (ArOH or ArOMe), and there-
fore the predominant mechanism is the heterolytic one
in the whole composition range. Therefore, assuming
the competitive mechanism indicated in Scheme II,
ko ϭ kHOM ϩ kHET. For PMBD, the ratio of the rate
constants for the homolytic and heterolytic mecha-
nisms at 100% MeOH is given by kHET/kHOM
ϭ
(CArOH ϩ CArOMe) /CArH Ϸ 14 (Fig. 5B). The value of
the kHET/kHOM ratio increases upon decreasing the per-
centage of MeOH, and because of this, we considered
that the homolytic process is not significant for these
arenediazonium salts at any solvent composition and
focused our attention on the heterolytic mechanism by
Figure 6 Arrhenius plots for MMBDat 99.5% MeOH and
for PMBDat: ᭺ 0%, ᭹ 25%, ٗ 50%, ᭿ 75%, ᭝ 99.5%
assuming that ko Ϸ kHET
.
Ϫ
MeOH. [MMBD] ϭ [PMBD] ϭ 1.0 Ϫ 1.5 ϫ 10 4 M, [HCl]
Spectrophotometric kinetic data (Fig. 1) indicate
that the rates of dediazoniation do not change signifi-
ϭ 0.01 M.

218
PAZO-LLORENTE, GONZALEZ-ROMERO, AND BRAVO-DIAZ
Table II Activation Parameters for the Three Methyl Substituted Benzenediazonium Ion in Water and 99.5% MeOH
and Dediazoniation k0 Values at T ϭ 40ЊC
Ϫ
1
Ϫ
Ϫ
1
Ϫ
1
Ϫ
1
Ϫ
Ϫ1
T Range/ЊC
104 k40/s
10 15 A/s
EA/kJ mol
⌬H#/kJ mol
⌬S#/J mol 1 K
0% MeOH
BDa
8.50
3.23
113
OMBD26.5–41.0
MMBD26.5–41.0
PMBD40.0–60.0
99.5% MeOH
BD8.5
14.88
12.55
0.52
2.1
2.8
1.7
108.2
109.5
112.4
106
107
110
40.4
43.0
21.6
OMBD25–40
MMBD25–40
PMBD40–65
23.8
24.05
0.9e1T(1a.r0))(D
0.79
0.94
105
105
Ϯ 1
Ϯ 1
Ϯ 2
105 Ϯ 2
106 Ϯ 1
109 Ϯ 2
40.8
42.5
25.6
0.79
114
* BDstands for benzenediazonium. 9
cantly on going from water to methanol. A unimolecu-
lar reaction in which the nucleophilic attack of the
solvent is rate-determining would lead to a strong de-
pendence of ko on nucleophile concentration, which is
not observed. On the other hand, dediazoniation rate
constants are insensitive to added electrolytes, and
HPLC kinetic data (Figs. 2 and 3) indicate that the rate
constant of formation of dediazoniation products is the
same as that of disappearance of diazonium ion.
Therefore, our results are consistent with a rate-deter-
mining step formation of a highly reactive aryl cation
intermediate.
Selectivity values, as defined by Eq. 1, remain es-
sentially constant on going from water to methanol
(Table I), the average value for MMBDbeing
SMWeOH ϭ 0.63 and that for PMBD SMWeOH ϭ 0.74. Such
values are orders of magnitude lower than those ob-
served for anionic nucleophiles competing with water
in reactions with stabilized carbocations [38,39,46] but
very similar to those reported for other arenediazon-
ium ions in alcohols [22,23]. Low selectivity values
are consistent with the formation of a highly reactive
cation, that is, with the reactivity-selectivity principle.
The nucleophilic attack on carbocations [38] may oc-
cur on “free” carbocations, on solvent-separated ion
pairs, or on “intimate” contact ion pairs. The low S
values obtained are consistent with a preassociation
stepwise mechanism, Scheme III, in which the aryl
cation, which has a short but finite lifetime, forms a
“complex” with the nucleophile before the product is
formed [47]. The rates of nucleophilic attack on aryl
cations [19] have been reported to be close to the dif-
fusion-control limit (i.e., ϳ109 MϪ sϪ ) and, since
lifetimes of aryl cation intermediates [19], although
currently unknown, are Յ2–4 nanoseconds, most of
the reaction of the aryl cation with the added nucleo-
phile is occurring through the “free” ion, and preas-
sociation of the nucleophile with the aryl cation does
not account for much of the trapping. Therefore, all
evidence is consistent with a heterolytic DN ϩ AN
mechanism; that is, rate-determining formation of a
“free” aryl cation, which reacts immediately, with very
low selectivity, with available nucleophiles.
The low variation in ko on going from water to
methanol (Fig. 1) suggests that the solvation of the
ground state is essentially the same as that in the prox-
imity of the transition state; that is, the free energy of
activation is not significantly affected by solvent com-
position. Analysis of activation parameters (Table II)
indicates that ⌬H# values are as high as in many uni-
molecular reactions [38,45], suggesting a transition
state that has undergone bond breaking with little com-
pensating bond making. Solvolytic unimolecular re-
actions can exhibit both positive and negative ⌬S# val-
ues [39,46]. Positive ⌬S# values, as we have found,
suggest that the transition state has a greater structural
freedom than reactants. These results are consistent
with the reported positive volumes of activation, ⌬V#,
for a number of dediazoniations in methanol [28,48–
50], providing further support to the Dn ϩ An mech-
anism. On the other hand, since ⌬S# values are posi-
tive, they compensate the large enthalpy term, making
dediazoniations proceed at a considerable rate com-
1
1
Scheme III

COMPETITIVE HOMOLYTIC AND HETEROLYTIC DEDIAZONIATION MECHANISMS
219
pared with other unimolecular reactions, as found for
REFERENCES
a number of dediazoniations in different solvents [5].
This term, however, does not contribute significantly
to ⌬G# because dediazoniation, as described in
Scheme 3, suggests that formation of the aryl cation
does not involve separation of charge but its redistri-
bution, in agreement with electronic structure analysis
studies [6]; thus, the parent arenediazonium ion and
the aryl cation polarize to a similar extent the solvent.
Consequently, dediazoniations show a much lower
solvent dependence [5] than typical unimolecular re-
actions in agreement with our data and with indepen-
dent experimental observations: in 19 solvents [51] the
rates of heterolytic dediazoniation vary by a factor of
only 9.
1. Klamann, D. Methoden der Organischen Chemie, 4th
ed.; Houben-Weyl: Thieme, Stuttgart, 1990; Vol. E16a,
Part II.
2. Hertel, H. Ullman’s Encyclopedia of Industrial Chem-
istry, 5th ed.; VCH: Weinheim, 1987; Vol. A8.
3. Herbs, W.; Hunger, K. Industrial Organic Pigments;
VCH: Weinheim, 1993.
4. Saunders, K. H.; Allen, R. L. M. Aromatic Diazo Com-
pounds; 3rd ed.; Edward Arnold:, 1985.
5. Zollinger, H. Diazo Chemistry I, Aromatic and Heter-
oaromatic Compounds; VCH: Weinheim, 1994.
6. Glaser, R.; Horan, C. J.; Lewis, M.; Zollinger, H. J Org
Chem 1999, 64, 902.
7. Broxton, T. J.; Bunnet, J. F.; Paik, C. H. J Org Chem
1977, 42, 643.
8. Bunnet, J. F.; Yijima, C. J Org Chem 1977, 42, 639.
9. DeTar, D. F.; Kosuge, T. J Am Chem Soc 1958, 80,
6072.
CONCLUSIONS
10. Kuokkanen, T. Finn Chem Lett 1984, 38.
11. Galli, C. Chem Rev 1988, 88, 765.
We have studied the solvolysis of 3- and 4-methyl-
benzenediazonium ions over the whole methanol-
water composition range employing a methodology
that allows simultaneous determination of rate con-
stants for product formation and diazonium ion loss
and estimations of product yields. Observed rate con-
stants and product yields are completely consistent
with competitive homolytic and heterolytic dediazon-
iation mechanisms, although the heterolytic one is pre-
dominant at any solvent composition. Enthalpies of
activation are high and entropies of activation are pos-
itive, making dediazoniations proceed at a consider-
able rate compared with typical unimolecular reac-
tions. Their values are very similar to those found in
pure water, consistent with the slight increase in de-
diazoniation rate constants observed on going from
water to methanol. Therefore, all evidence is consist-
ent with a DN ϩ AN mechanism; that is, rate-deter-
mining formation of a “free” aryl cation that reacts
immediately with available nucleophiles.
12. Hanson, P.; Jones, J. R.; Gilbert, B. C.; Timms, A. W.
J Chem Soc, Perkin Trans. 2 1991, 1009.
13. Hanson, P.; Hammond, R. C.; Goodracre, P. R.; Purcell,
J.; Timms, A. W. J Chem Soc, Perkin Trans 2, 1994,
691.
14. Scaiano, J. C.; Kim-Thuan, J. J Photochem 1983, 23,
269.
15. Cacace, F. Science 1990, 250, 292.
16. Swain, C. G.; Sheats, J. E.; Harbison, K. G. J Am Chem
Soc 1975, 97, 783.
17. Gamba, A.; Simonetta, M.; Suffritti, G. B.; Szele, I.;
Zollinger, H. J Chem Soc, Perkin Trans 2 1980, 493
18. Ravenscroft, M. D.; Tagaki, K.; Weiss, B.; Zollinger,
H. Gazz Chim Italiana 1978, 117, 353.
19. Steenken, S.; Ashokkumar, M.; Maruthamuthu, P.;
McClelland, R. A. J Am Chem Soc 1998, 120, 11925.
20. Pazo-LLorente, R.; Sarabia-Rodriguez, M. J.; Gonzalez-
Romero, E.; Bravo-D´ıaz, C. Int J Chem Kin 1999, in
press.
21. Bravo-Diaz, C.; Soengas-Fernandez, M.; Rodriguez-
Sarabia, M. J.; Gonzalez-Romero, E. Langmuir 1998,
14, 5098.
22. Chauduri, A.; Loughlin, J. A.; Romsted, L. S.; Yao, J.
J Am Chem Soc 1993, 115, 8351.
SUPPLEMENTARY MATERIAL
23. Chauduri, A.; Romsted, L. S.; Yao, J. J Am Chem Soc
1993, 115, 8362.
Tabulated data of the observed rate constants and of
the dediazoniation product yields at a number of sol-
vent compositions, in the absence and presence of
added electrolytes (4 pages), are available. Ordering
information is provided at any current masthead.
24. Bravo-D´ıaz, C.; Sarabia-Rodriguez, M. J.; Barreiro-Sio,
P.; Gonzalez-Romero, E. Langmuir 1999, 15, 2823.
25. Bartsch, R. A. Complexation of Aryl Diazonium ions
by polyethers, in The Chemistry of Triple-Bonded
Functional Groups; J. Wiley & Sons: New York, 1983;
Vol. 2.
26. Bravo-D´ıaz, C.; Romero-Nieto, M. E.; Gonzalez-Rom-
ero, E. Langmuir 1999, accepted for publication.
27. Ishida, K.; Kobori, N.; Kobayashi, M.; Minato, H. Bull
Chem Soc Jpn 1970, 43, 285.
This work was supported by the following institutions: Min-
isterio de Educacio´n y Cultura (DGICYT, PB94-0741),
Xunta de Galicia (XUGA 38305A94) and Universidad de
Vigo.
28. Kuokkanen, T. Finn Chem Lett 1981, 52.

220
PAZO-LLORENTE, GONZALEZ-ROMERO, AND BRAVO-DIAZ
29. Kuokkanen, T. Finn Chem Lett 1981, 81.
30. Garcia-Meijide, M. C.; Bravo-Diaz, C.; Romsted, L. S.
Int J Chem Kinet 1998, 30, 31.
31. Zollinger, H.; Wittwer, C. Helv Chim Acta 1952, 35,
1209.
32. Pazo-LLorente, R.; Rodriguez-Sarabia, M. J.; Gonzalez-
Romero, E.; Bravo-D´ıaz, C. Int J Chem Kinet 1999, 31,
73.
39. Pross, A. Theoretical & Physical Principles of Organic
Reactivity; J. Wiley & Sons: New York, 1995.
40. Grunwald, E.; Winstein, S. J Am Chem Soc 1948, 70,
846.
41. Schwing-Weill, M. J. Bull Soc Chim France 1973, 3,
823–830.
42. Salmon, P. S.; Neilson, G. W.; Enderby, J. E. J Phys C,
Solid State Phys 1988, 21, 1335–1349.
33. Bravo-Diaz, C.; Romsted, L. S.; Harbowy, M.; Romero-
Nieto, M. E.; Gonzalez-Romero, E. J Phys Org Chem
1999, 12, 130.
34. Bravo-Diaz, C.; Gonzalez-Romero, E. Anal Chim Acta
1999, 385, 373.
35. Romero-Nieto, M. E.; Bravo-Diaz, C.; Gonzalez-Rom-
ero, E. Submitted for publication (Int J Chem Kin) 1998.
36. Yamamoto, H.; Ichikawa, K.; Tokunaga, J. J Chem Eng
Data 1994, 39, 155.
43. Manahan, S. E.; Iwamoto, R. T. J Electroanal Chem
1967, 13, 411–417.
44. Isaacs, N. Physical Organic Chemistry, 2nd ed.; Long-
man Scientific & Technicals:, 1995.
45. Maskill, H. The Physical Basis of Organic Chemistry;
Oxford Univ. Press: Oxford, 1995.
46. Ritchie, C. D. Physical Organic Chemistry: The Fun-
damental Concepts, 2nd ed.; Marcel Dekker: New York,
1990.
37. Leffler, J. E.; Grunwald, E. Rates and Equilibria of Or-
ganic Reactions; Dover: New York, 1989.
38. Lowry, T. H.; Richardson, K. S. Mechanism and Theory
in Organic Chemistry, 3rd ed.; Harper-Collins Pub.:
New York, 1987.
47. Jenks, W. P. Chem Soc Rev 1981, 10, 345.
48. Brower, K. R. J Am Chem Soc 1960, 82, 4535–4537.
49. Kuokkanen, T. Finn Chem Lett 1980, 189.
50. Kuokkanen, T. Finn Chem Lett 1980, 192.
51. Szele, I.; Zollinger, H. Helv Chim Acta 1978, 61, 1721.