Acid-catalysed hydrolysis of trityl derivatives in strongly acidic aqueous media

Article abstract of DOI:10.1002/poc.3230

The kinetics of hydrolysis (deamination or dealcoholation) of tritylamines and 2,2,2-trifluoroethyl ethers and their corresponding 4-methoxy, 4,4′-dimethoxy-, and 4,4′,4″-trimethoxy-substituted analogues in aqueous solutions up to 3.5 mol · dm^-3 in strong acid have been investigated at constant ionic strength. In all cases, acid-catalysed hydrolytic processes have been observed, with finite reactivity at [H ₃O⁺] = 0. Strong upward curvature has been observed for k_obs versus [HClO₄]. Analysis of this dependence in terms of the H_R acidity function and the X₀ excess acidity scale allow explanation of the observed behaviour in terms of the increasing differences between concentrations and activities of the various species involved in the processes, including water, for which the activity coefficient strongly diverges from its standard state value as the acidity increases. This analysis has shown that, by taking account of the effect of the ionic strength, the same mechanistic models proposed for mildly acidic solutions are valid in more highly concentrated acid media. These comprise (i) protonation of the trityl ether followed by C-O bond heterolysis to give a carbenium ion-alcohol (ion-molecule) pair which can separate, and (ii) C-N bond heterolysis of the protonated tritylamine to give a carbenium ion-amine (ion-molecule) pair followed by separation of the fragments or protonation of the amine and subsequent separation of the ions. Each separated (substituted) trityl carbenium ion, regardless of its provenance, is invariably captured by a solvent molecule (water). Copyright 2013 John Wiley & Sons, Ltd. Strong upward curvatures observed for k_obs versus [H₃O⁺] kinetics results in the hydrolysis of tritylamines and trityl ethers in strongly acidic aqueous solutions are entirely understandable as medium effects and accurately described in terms of excess acidities. The mechanisms previously proposed for mild acidic conditions have been generalized to more strong acid media. Copyright

Full text of DOI:10.1002/poc.3230

Special Issue Article
Received: 30 March 2013,
Revised: 25 July 2013,
Accepted: 2 September 2013,
Published online in Wiley Online Library: 5 November 2013
(wileyonlinelibrary.com) DOI: 10.1002/poc.3230
Acid-catalysed hydrolysis of trityl derivatives
in strongly acidic aqueous media†
Moisés Canle L.a* and Howard Maskillb**
The kinetics of hydrolysis (deamination or dealcoholation) of tritylamines and 2,2,2-trifluoroethyl ethers and their
corresponding 4-methoxy, 4,4′-dimethoxy-, and 4,4′,4″-trimethoxy-substituted analogues in aqueous solutions up
to 3.5 mol · dmꢀ3 in strong acid have been investigated at constant ionic strength. In all cases, acid-catalysed
hydrolytic processes have been observed, with finite reactivity at [H3O+] = 0. Strong upward curvature has been
observed for kobs versus [HClO4]. Analysis of this dependence in terms of the HR acidity function and the X0 excess
acidity scale allow explanation of the observed behaviour in terms of the increasing differences between
concentrations and activities of the various species involved in the processes, including water, for which the activity
coefficient strongly diverges from its standard state value as the acidity increases. This analysis has shown that, by
taking account of the effect of the ionic strength, the same mechanistic models proposed for mildly acidic solutions
are valid in more highly concentrated acid media. These comprise (i) protonation of the trityl ether followed by C–O
bond heterolysis to give a carbenium ion–alcohol (ion–molecule) pair which can separate, and (ii) C–N bond
heterolysis of the protonated tritylamine to give a carbenium ion–amine (ion–molecule) pair followed by separation
of the fragments or protonation of the amine and subsequent separation of the ions. Each separated (substituted)
trityl carbenium ion, regardless of its provenance, is invariably captured by a solvent molecule (water). Copyright
© 2013 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this paper.
Keywords: trityl cations; acid catalysis; hydrolysis; deamination; acidity functions; excess acidities
For the reverse, studied directly by generating the trityl cation
INTRODUCTION
by laser flash photolysis, kr = (1.68 0.06) · 105 sꢀ1. Both rate
The trityl (triphenylmethyl) group (Tr) and its derivatives receive
ongoing attention for several reasons including their use as
easily removable protecting groups for alcohols and amines in
syntheses of a wide range of compounds.[1,2] Early investigations
of the deprotection of tritylated alcohols[3] suggested that
they take place by a simple SN1 heterolysis of the C–O bond
following a pre-equilibrium protonation via a carbenium ion,
analogous to the one shown in Scheme 1 (R = H) by which trityl
alcohols ionize.
constants were measured at 298.0 K, the first using 5% MeCN
as co-solvent and no control of the ionic strength.[10] These
results show that the C–O bond cleavage in the alcohol to give
the Tr carbenium ion in Scheme 1 is rate determining and
þ
allowed estimation of pKR = (ꢀ6.46 0.03), in good agreement
with previously reported literature values.
The higher basicity of amines compared with alcohols, and the
greater ease of C–N bond breaking relative to the C–O bond,[11]
In the corresponding mechanism for trityl ethers (Scheme 1,
R = alkyl), no appreciable extent of internal return from the trityl
cation and the alcohol was initially expected as the departing
nucleofuge (ROH) is a poor nucleophile in dilute aqueous
solution. Also, no reaction was expected at [H3O+] = 0, a
prediction we have already shown to be correct.[4] Some internal
return has been observed, however, presumably assisted by
H-bonding between the departing ROH and a solvent molecule.
This finding is not wholly unexpected considering the product
selectivities observed for solvolyses of p-methoxybenzoyl
chloride in 2,2,2-trifluoroethanol/water mixtures.[5] Detailed
results also require the involvement of ion–molecule pairs[6] in
the formation of separated trityl carbenium ions.[7–9]
* Correspondence to: Moisés Canle L., Chemical Reactivity and Photoreactivity
Group, Department of Physical Chemistry and Chemical Engineering I,
University of A Coruña, E-15071 A Coruña, Galicia, Spain.
E-mail: mcanle@udc.es
** Correspondence to: Howard Maskill, School of Natural Sciences, University of
Newcastle, Newcastle upon Tyne, NE1 7RU, UK.
E-mail: h.maskill@newcastle.ac.uk
†
This article is published in Journal of Physical Organic Chemistry as a Special
issue: In Memoriam edited by Michael Page (University of Huddersfield,
Chemistry, Queensgate, Huddersfield, HD1 3DH, United Kingdom).
In memoriam: Prof. Dr. Rory A. More O’Ferrall.
Rate constants for the reactions of Eqn 1 have been measured
in both directions.
a M. Canle L.
Chemical Reactivity and Photoreactivity Group, Department of Physical
Chemistry and Chemical Engineering I, University of A Coruña, E-15071 A
Coruña, Galicia, Spain
Ph3COH þ Hþ ⇄ Ph3CþþH2O
(1)
b H. Maskill
For the forward reaction, measured by following 18O exchange
in a labelled alcohol in dilute HCl, kf = (5.8 0.3) · 10ꢀ2 Mꢀ1 sꢀ1
School of Natural Sciences, University of Newcastle, Newcastle upon Tyne,
NE1 7RU, UK
.
J. Phys. Org. Chem. 2013, 26 1016–1022
Copyright © 2013 John Wiley & Sons, Ltd.

ACID-CATALYSED HYDROLYSIS OF TRITYL DERIVATIVES IN STRONGLY ACIDIC AQUEOUS MEDIA
0.0020
0.0015
0.0010
0.0005
Scheme 1. Mechanism for hydrolysis of (substituted) trityl alkyl ethers in
dilute aqueous acidic solution
0.0000
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
suggested that tritylamines should readily undergo reactions
similar to those of the alcohols and ethers.[12,13] Although
tritylamines, in contrast to alcohols, are wholly protonated in
dilute aqueous strong acid, detailed studies showed that
ion–molecule pairs are involved in these reactions, which
turned out to be general for tritylamines[14–17] and N-
tritylhydroxylamines.[18] In all reactions investigated in this study,
the final product is an equilibrium mixture of substituted trityl
carbenium ion and solvolysis product (overwhelmingly the trityl
alcohol) as determined by the composition of the solvent and
-3
.
[HCIO4] / mol dm
Figure 1. Kinetics results for the hydrolysis of monomethoxytritylamine
in aqueous acid up to 3.5 M; I (NaClO4) = 3.50 mol · dmꢀ3, 4% CH3CN,
T = 323.08 K
The strong curvature exemplified by the kobs versus [H3O+]
dependence illustrated in Fig. 1 has been lacking a proper
interpretation.[4,16] Our current working hypothesis is that we
are observing a medium effect[4], and, in this paper, we deal with
it in terms of acidity functions and excess acidities.[20–22] Trityl
alcohols have been used to define the widely employed HR
acidity function based on their protonation and dissociation to
yield the easily observable trityl carbenium ions[23,24] according
to Eqn 1 (the reversible reaction related to Scheme 1). A recent
comprehensive study of the stabilities and reactivities of trityl
þ
relevant equilibrium constants, e.g. KR in Scheme 2 below.
A generic mechanism, which accounts for the unprecedented
acid catalysis observed in the reaction of a substrate (the
substituted ammonium ion) lacking a basic centre at the reaction
site, is illustrated in Scheme 2 for tritylamines. The rate constants
for reactions between different primary amines and the
diarylmethyl/triarylmethyl cations have already been measured
directly by generating the cations by laser flash photolysis. They
are close to the diffusion controlled limit in MeCN but become
increasingly lower as the water content increases.[19]
The mechanism in Scheme 2 leads to a simple linear relationship
between the observed pseudo-first-order rate constant, kobs, and
[H3O+], which is generally observed for hydrolysis of (substituted)
tritylamines in aqueous strong acids up to 1 mol · dmꢀ3 (limited
by a constant ionic strength, I, of 1 mol · dmꢀ3). However, upward
curvatures were observed for plots of kobs versus [H3O+] when
higher acid concentrations were used ([H3O+] ≤ 3 M), as illustrated
in Fig. 1; such acidities were necessary to achieve appreciable
reactivity of the less reactive tritylamines, i.e. those with fewer
rate-enhancing methoxy groups, for example, in the Tr residue.
þ
cations which range over 21 pKR units compared the ionization
processes which yield them.[25]
HR is defined according to Eqn 2 in terms of [H+] or Eqn 3 in
terms of a+H.
½R3Cþꢁ
þ
HR ¼ pKR ꢀ log
½R3COHꢁ
þ
γ
ꢂγH
R3COH
¼ ꢀ log½Hþꢁ ꢀ log
þ logaH O
(2)
2
þ
γ
R3C
½R3Cþꢁ
þ
HR ¼ pKR ꢀ log
½R3COHꢁ
γ
R3COH
þ
¼ ꢀ logaH ꢀ log
þ logaH O
(3)
2
þ
γ
R3C
In Eqns 2–3, γi is the molar activity coefficient for each
different species involved, taken to be γi = 1 at infinite dilution.
Rigorously, this condition should be fulfilled in a reference
hypothetical ideal solution 1 mol· dmꢀ3 in the acid used (as for
the pH scale). Such conditions introduce a related change in the
activity of water, aH2O, which would be unity in its standard state
– an infinitely dilute solution of water (the product of Eqn 1) in
itself (the solvent). The common standard state aH2O = 1 for water
as a solvent relates to the mole fraction scale, as opposed to the
molarity scale which is more conveniently used for solutes in reac-
tivity and mechanistic studies. The standard state aH2O = 1 leads to
the pH scale ranging from 0 to 14, but does not easily allow
consideration either of acids which are too weak to protonate wa-
ter to any appreciable extent, or bases which are so strong that
Scheme 2. Mechanism for hydrolytic deamination of substituted N-
tritylalkylamines in dilute aqueous acid, showing the involvement of
ion–molecule pairs
J. Phys. Org. Chem. 2013, 26 1016–1022
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

M. CANLE L. AND H. MASKILL
they are completely protonated by water; both are commonly
used in reactivity studies. The effect of water activity becomes
increasingly important as the hydration of H+ changes, i.e. as the
acid concentration increases. This has been analysed by Perrin
who proposed a model of the hydration for H+ which allows
reasonable predictions of values of acidity functions except for
very high ionic strengths.[26]
Table 1. Trityl derivatives used in this study
Compound
Abbreviation
TrNH
2
Equation 2 can be reorganized as in Eqns 4, 5:
TrOTFE
logð½R3Cþꢁ=½R3COHꢁÞ–log½Hþꢁ–log ɑH2O
À
Á
þ
þ
þ
¼ m·log γR COH·γH =γR C þ pKR
(4)
3
3
logð½R3Cþꢁ=½R3COHꢁÞ–log½Hþꢁ–log ɑH O ¼ m·X þ pKR
(5)
(6)
þ
2
MMTrNH
2
where
À
Á
þ
þ
X ¼ log γR COH·γH =γR C
:
3
3
The parameter X in Eqns 5 and 6, known as the “excess
acidity”, represents the disparity between the acidity of the
solution and the stoichiometric concentration of H+ (H3O+).[21,22]
The empirical factor “m” in Eqns 4 and 5 needs to be introduced to
account for the experimental fact that, although Eqn 2 is fulfilled in
the case of the Hammett acidity function (H0), this is not true for many
other compounds upon which acidity functions are based, such
as carbocations as in our case.[23,24] Values of X have been
calculated and tabulated as a function of wt% acid or molar
concentrations.[22]
MMTrOTFE
DMTrNH
2
There have been attempts to replace the concept of chemical
activity in the context of organic reaction mechanisms by
empirical kinetic measures of medium effects; for example,
solvent ionizing power and nucleophilicity have been used,[27]
the latter being linked to excess acidity.[28] However, the activity
of a species is strictly defined in terms of the difference between
its chemical potential (μ) under the conditions of the reaction in
question and its standard chemical potential (μ0).[29] Conse-
quently, we decided to work within this quasi-thermodynamic
framework as we are not comparing reactivities of different
families of compounds or members of a family, but the reactiv-
ities of a single compound in media of different compositions,
and its reactivity is altered only by the composition of the media.
We now report kinetics results and their corresponding
mechanistic analysis for the hydrolysis of 4,4′,4″-trimethoxytrityl,
4,4′-dimethoxytrityl, 4-methoxytrityl, and tritylamines, and the
analogous 2,2,2-trifluoroethyl ethers of some of these (substituted)
trityl groups in concentrated aqueous perchloric acid.
DMTrOTFE
TMTrNH
2
TMTrOTFE
grade CH3CN were used as solvents. Reactions were initiated by injecting
the trityl compound (dissolved in CH3CN) into a UV cell so that the final
solution was always lower than 5% in CH3CN (details are given in
headings of Tables in each case). The ionic strength was kept constant
in all cases using NaClO4. The reactions were monitored in the cell
compartment of a double beam UV spectrophotometer by following
the increase in absorption at the wavelengths shown in Table 2 due to
the formation of the equilibrium mixture containing the (substituted)
trityl carbenium ion.
EXPERIMENTAL
The compounds used in this study and their abbreviations are compiled
in Table 1.
In all cases, the temperature was kept constant to within 0.05 K and a
minimum of three half-lives of reaction were followed.
Primary tritylamines were obtained by reaction of the appropriately
substituted trityl bromide or trityl tetrafluoroborate with liquid NH3,
while the N-substituted derivatives and ethers were synthesized by
reaction of the tetrafluoroborate salts with the corresponding amines or with
CF3CH2OH (in the presence of Et3N), respectively. All compounds whose
kinetics were investigated were purified by column chromatography;
melting points, elemental analyses, 1H and 13C NMR, IR and mass spectra
agreed with the assigned structures and/or literature data; preparative
details have been given elsewhere.[4,14–16]
RESULTS AND DISCUSSION
Hydrolytic deamination of tritylamines
The influence of the acidity on the deamination of tritylamines
was investigated at aqueous concentrations of HClO4 up to
3.5 mol dmꢀ3. Tritylamines (represented generically as Tr′–NH2)
are strong bases and virtually wholly protonated in dilute
aqueous strong acid; the rate law for deamination of their
All reagents used in the kinetics experiments were of the maximum
purity available. Acid concentrations were repeatedly determined by ti-
tration with standard aqueous NaOH. Glass distilled water and HPLC
wileyonlinelibrary.com/journal/poc
Copyright © 2013 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2013, 26 1016–1022

ACID-CATALYSED HYDROLYSIS OF TRITYL DERIVATIVES IN STRONGLY ACIDIC AQUEOUS MEDIA
Table 2. λmax corresponding to the different carbenium ions
0.0030
0.0025
0.0020
0.0015
0.0010
0.0005
0.0000
Carbenium ion
λ / nm
Tr+
432
471
498
483
MMTr+
DMTr+
TMTr+
protonated forms was previously found to be specific acid catalysed
below about [H3O+] ~ 1 M and described by Eqn 7.[14–16]
rate ¼ kobs·½Tr’NH3þꢁ; where kobs ¼ ko þ kH·½H3Oþꢁ
(7)
0
1000 2000 3000 4000 5000 6000 7000
10-H
R
ꢀHR
,
þ
Those results were interpreted in terms of the mechanism
shown in Scheme 2; the protonated tritylamine Tr′–NH2R+
undergoes C–N bond heterolysis yielding an ion–molecule pair
(Tr′+ · NH2R) which may dissociate to a carbenium ion and
an amine or, if time allows, be protonated on the amine. The
trityl carbenium ion is then reversibly captured by water in a
post-equilibrium.
Figure 3. Plot of kobs versus acid activity (a
), approximated by 10
H30
for the hydrolytic deamination of TMTrNH2 in aqueous perchloric acid;
[TMTrNH2]0 =1.41ꢂ10ꢀ5 mol · dmꢀ3; I (NaClO4)=3.50mol·dmꢀ3; T = 298.00 K,
0.5% CH3CN
of choice, considering the structure of the trityl compounds, is HR
(Eqn 2),[31–35] and we have adopted here the procedure
developed by Kresge et al.[36] to extrapolate its values to
dilute solution. This method is reasonable and satisfactory as
long as the ratios of activity coefficients are not far from unity
As stated above, upward curvatures of the kobs versus [H3O+]
dependences were observed when more concentrated strong
aqueous acids were used, as shown in Figs. 1 and 2; the lines
through the points do not fit any mechanistic model and are
included simply to emphasize the tendency. These curved plots,
more or less pronounced depending on the structure of the
compound, correspond to deviations from the linear behaviour
described by Eqn 7 and require a proper explanation.
A common cause of upwards curvature in plots analogous to
the ones in Figs. 1 and 2 is the intrusion of a second (parallel)
mechanism with a higher order term in the rate law.[30] This
seems improbable in the present case. Increasing disparities
between activity and concentration as the medium becomes
more concentrated in acid seemed a more likely cause, i.e.
activity coefficients increasingly deviate from unity as the
concentration of the strong acid increases. One way to handle
this is to interpolate an appropriate acidity function for each of
the acid concentrations used and then take the antilogarithm
of the acidity function as the H3O+ activity. The acidity function
þ
(γR COH=γR C ≈1) and aH O may be taken to be unity. The validity
2
3
3
of the use of the HR acidity function is based on the assumption
that the ratio of the activity coefficients of R3COH and the
carbenium ion, R3C+, does not change for molecules of similar
structure. The HR acidity function is based on the protonation
and dissociation of aryl methanols, which are slightly different
from the tritylamines and ethers we have worked with, but the
work of Deno[23] indicates it is applicable.
As shown in Fig. 3 for TMTrNH2, this approach is successful
only in the lower range of HClO4 concentrations, with
strong deviations in the higher range (not all the points are
shown, in order to visualize the deviations better). A similar
situation is observed for the other tritylamines used. A probable
defect of this approach is that the assumptions upon which it
is based are not valid, i.e. γR3COH/γR3C+ and/or aH2O are not con-
stant at approximately unity.
Despite the incomplete adequacy of this approach, its partial
success in describing the behaviour in the lower range of strong
acid concentrations supports our hypothesis that curvature in
plots of kobs versus [H3O+] is indeed due to medium effects
quantifiable by activity coefficients but that the activity
0.004
0.003
0.002
0.001
0.000
þ
coefficient ratio (γR COH=γR C ) and/or aH O in Eqn 3 deviates from
2
3
3
unity. A logical improvement is to use excess acidities (Eqns 5, 6)
rather than the HR acidity function. Our choice is to use the X0
scale for HClO4, which is derived using only primary aromatic
amines rather than a broad mix of compounds as for the X
scale.[22] When this is done, we obtain good linear fits for plots
of kobs versus X0 for all the tritylamines over the whole range
of concentrations we used, as exemplified for TMTrNH2 in Fig. 4.
Thus, the hydrolytic deamination of tritylamines can be
described by the same mechanism in mildly and strongly acidic
media, and the observed deviations (upward curvature of kobs
versus [H3O+] plots) are mainly medium effects expressed
through activity coefficients. Consequently, the mechanism in
Scheme 2 can be extended up to at least 3.5 M HClO4, and the
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
-3
.
[HCIO4] / mol dm
Figure 2. Kinetics results for the hydrolysis of TMTrNH2 in acidic media up
to 3.5M; [TMTrNH2]0 = 1.41ꢂ10ꢀ5 mol · dmꢀ3; I (NaClO4) = 3.50mol ·dmꢀ3
;
T = 298.00 K, 0.5% CH3CN
J. Phys. Org. Chem. 2013, 26 1016–1022
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

M. CANLE L. AND H. MASKILL
TMTrNH2 and DMTrNH2 is very satisfactory. Agreement is
expected for the common proposed mechanism corresponding
to the kuncat step in Scheme 2 with an initial unimolecular
fragmentation of the ammonium ion into a trityl cation and a
neutral molecule. As there is no change in charge during this
rate-limiting step, the medium effect should be mimimal.[37]
The observed difference in reactivity between DMTrNH2 and
TMTrNH2 is attributed to the rate-enhancing resonance effect
of an extra 4-methoxy substituent which further stabilizes the
developing charge of the nascent carbenium ion.
0.004
0.003
0.002
0.001
0.000
Effect of ionic strength, I
0.0
0.3
0.6
0.9
As shown in Fig. 5, the observed rate constant increases linearly
with ionic strength. This enhancing effect depends on the nature
of the cation of the electrolyte used to maintain the ionic
strength: the reactivity increases when Na+ is replaced by Li+
even using the same counter anion, ClOꢀ4 . As the acid
concentration increases in our deaminative hydrolysis
experiments, Na+ is replaced by hydrated H+ with a consequent
effect on the reaction rate. It follows that the results obtained in
concentrated acid media cannot be simply compared with those
obtained in dilute acids.
X
0
Figure 4. Plot of kobs versus excess acidity (X0) for the hydrolysis of
TMTrNH2 in strongly acidic media; [TMTrNH2]0 = 1.41ꢂ10ꢀ5 mol · dmꢀ3
;
I (NaClO4) = 3.50 mol · dmꢀ3; T = 298.00 K, 0.5% CH3CN
values obtained for intercepts and slopes of kobs versus X0
correlations for the different tritylamines are compiled in Table 3.
Since X0 does not have dimensions, the slope of a plot of kobs
versus X0 has units of sꢀ1 – the same as for the intercept. Conse-
quently, the slope cannot be interpreted as corresponding exactly
þ
The equilibrium constants KR for the hydration of the
carbenium ion (Eqn 8) were obtained from absorbance
measurements for the different carbenium ions.[13]
to the kcat step in Scheme 2, which has units of molꢀ1 · dm3 · sꢀ1
,
although the two must be related. There is a small but discernable
trend in the values of the slope for the different reactants, and two
possible causes are (i) the small changes in experimental conditions
(temperature, ionic strength, and the composition of the medium),
and/or (ii) the structural differences between the reactants (the dif-
ferent numbers of methoxy substituents). Within the framework of
the proposed mechanism (Scheme 2), we would not expect an ap-
preciable medium effect on the bimolecular step since protonation
of the ion–molecule pair by H3O+ followed by fast diffusion apart of
the cation-cation pair does not involves a significant change in
charge redistribution.[37,38] Consequently, no appreciable medium
effect is expected on this step for any of the four tritylamines inde-
pendently of their structural differences. Similarly, all the ion–ion
pairs generated upon protonation of the ion–molecule pairs are
highly unstable and rapidly become separately solvated trityl and
ammonium cations. However, the minor differences between the
slopes for the different substrates could be attributable to different
degrees of stabilization/delocalization of the differently substituted
trityl cations as their immediate environment changes upon be-
coming separately solvated cations.
R
þ þ 2 H2O K⇌Rþ ROH þ H3Oþ
(8)
This was done at different ionic strengths for TMTr+ and Fig. 6
þ
shows the complex dependence of KR upon I with a maximum
value around I = 0.35–0.40 mol · dmꢀ3. As KR is used to define
þ
the HR acidity function and there is no acknowledgement of ionic
[LiClO4] / M
0.0
0.7
1.4
2.1
2.8
0.00040
0.00032
0.00024
0.00016
0.00040
0.00032
0.00024
0.00016
The observed intercepts now determined from reactions in
solutions of perchloric acid up to 3.5 M may be compared with
the values previously obtained from reactions in dilute acids:
k0 = 3.81 · 10ꢀ4 sꢀ1 (TMTrNH2)[16] versus 3 · 10ꢀ4 sꢀ1 (this study)
and 3.49 · 10ꢀ5 sꢀ1 (DMTrNH2)[16] versus 3.7 · 10ꢀ5 sꢀ1 (this study).
The agreement from quite different experimental protocols for
0.0
0.7
1.4
2.1
2.8
[NaClO4] / M
Figure 5. Influence of ionic strength and of the replacement of Na+ by
Li+ in the hydrolysis of DMTrNH2; [HClO4] = 0.9955 mol · dmꢀ3; I (MClO4) =
3.50 mol dmꢀ3, M = {Na, Li}; 1% CH3CN, T = 298.00 K
Table 3. Results obtained from kobs versus X0 fits for tritylamines
Compound
k0 / sꢀ1
Slope
Conditions
TMTrNH2
DMTrNH2
MMTrNH2
TrNH2
3 · 10ꢀ4
3.7 · 10ꢀ5
9 · 10ꢀ5
4.1 · 10ꢀ3
6.3 · 10ꢀ4
8 · 10ꢀ4
298.0 K, I = 3.5 mol · dmꢀ3, 0.5 % CH3CN
298.0 K, I = 3.5 mol · dmꢀ3, 2 % CH3CN
323.1 K, I = 3.5 mol · dmꢀ3, 4 % CH3CN
339.8 K, I = 3.5 mol · dmꢀ3, 4 % CH3CN
7.3 · 10ꢀ5
1.6 · 10ꢀ4
wileyonlinelibrary.com/journal/poc
Copyright © 2013 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2013, 26 1016–1022

ACID-CATALYSED HYDROLYSIS OF TRITYL DERIVATIVES IN STRONGLY ACIDIC AQUEOUS MEDIA
0.10
structure of the compound, corresponds to a deviation from
the linear behaviour described by Eqn 7 and, as for the
results for the corresponding deamination reactions, requires a
mechanistic explanation.
0.08
0.06
0.04
0.02
0.00
As for tritylamines, the deviation from linearity is attributed to
concentration–activity differences in strongly acidic solutions,
and the activity of H3O+ was initially approximated by the
antilogarithm of HR for each acid concentration. Plots of kobs
versus 10ꢀH obtained in this way for all the trityl ethers used
R
are linear only in the lower range of HClO4 concentrations, with
increasingly large deviations as [HClO4] increases. In contrast,
good linear fits are obtained when excess acidities are used;
the plot of kobs versus X0 for DMTrOTFE in Fig. 8 is typical of
results for all the trityl ethers investigated.
Thus, as for tritylamines, the hydrolysis of trityl ethers can be
described by the same mechanism in mildly and strongly acidic
media, and the observed deviations (upward curvature of kobs
versus [H3O+] plots) are attributable to medium effects. In other
words, the mechanism in Scheme 1 can be extended at least
up to about 3 M aqueous HClO4, and values obtained for slopes
of the kobs versus X0 correlations for the different trityl ethers are
compiled in Table 4.
0.0
0.2
0.4
0.6
0.8
1.0
I / mol dm-3
Figure 6. Dependence of KR for the hydration of TMTr+ upon the ionic
strength at 298.00 K
þ
strength in its definition, HR is expected to work less well at higher
ionic strengths (therefore at higher acidities) owing to larger
deviations of (γR COH=γR C ) and aH O from unity; consequently,
þ
2
3
3
there will be increasing divergences between real activity values
and those calculated as the antilogarithm of HR as the acidity/ionic
strength increases. These ionic strength factors contribute to-
wards the deviations from linearity in the kobs versus [HClO4] plots
found in this study and discussed above (Fig. 2).
The slopes here are appreciably larger than those for the
deaminations, which indicates greater medium effects. Within
the framework of the proposed mechanism (Scheme 1), the
observation must be interpreted in the light of the very low
pKa values of protonated trityl 2,2,2,-trifluoroethyl ethers
(we have estimated pKa(DMTrO(H)TFE+) ≈ –9).[4] These results
indicate that the rate-limiting step is the generation of the
protonated trityl ether intermediate. In this step, positive charge
is being developed with an attendant increase in solvation.[37,38]
This will be true for all four trityl ethers studied, regardless of
minor structural details, and (in turn) is the main cause of
Hydrolysis of trityl 2,2,2-trifluoroethyl ethers
The influence of the acidity on the dealcoholation of trityl 2,2,
2-trifluoroethyl ethers was investigated at concentrations of
HClO4 up to 3 mol dmꢀ3. A specific acid catalysis rate law was
found (Eqn 7),[4] and interpreted in terms of the mechanism
shown in Scheme 1. The protonated trityl ether undergoes C–O
bond heterolysis yielding a trityl carbenium ion and the
nucleofuge (2,2,2-trifluoroethyl alcohol) which undergo some
extent of internal return. The trityl carbenium ion is predomi-
nantly reversibly captured by water in a post-equilibrium.
Upward curvature of the plots of kobs versus [H3O+] was
observed at higher concentrations of acid, as shown in Fig. 7.
The curvature, more or less pronounced depending on the
þ
deviations of (γR COH=γR C ) and aH O from unity.
2
3
3
The observed intercepts cannot be compared with those
previously obtained for the reactions in dilute acid,[4] as kobs
versus X0 plots generate negative values in all cases. Whilst an
uncatalysed pathway is always evident for protonated
tritylamines (Scheme 2), and typically accounts for as much of
the overall reaction as the acid-catalysis path in dilute acid,[16]
a corresponding uncatalysed path for trityl ethers is absent
from Scheme 1. However, it remains possible that accumulated
0.004
0.003
0.002
0.001
0.000
0.015
0.010
0.005
0.000
0.0
0.5
1.0
1.5
2.0
-3
2.5
3.0
0.10
0.15
0.20
0.25
.
X
[HCIO4] / mol dm
0
Figure 7. Kinetics results for the hydrolysis of MMTrOTFE in acidic
Figure 8. Plot of kobs versus excess acidity (X0) for the hydrolysis of
media; [MMTrOTFE]0 = 4.43ꢂ10ꢀ5 mol · dmꢀ3; I (NaClO4) = 3.00 mol · dmꢀ3
T = 298.00 K, λ = 471 nm; 1% CH3CN
;
DMTrOTFE in acidic aqueous solution; [DMTrOTFE]0 = 2.82ꢂ10ꢀ5 mol ·dmꢀ3
;
I (NaClO4) = 1.00 mol · dmꢀ3; T = 298.00 K, λ = 498 nm; 1% CH3CN
J. Phys. Org. Chem. 2013, 26 1016–1022
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

M. CANLE L. AND H. MASKILL
Table 4. Results obtained from kobs versus X0 correlations for trityl ethers
Compound
k0 / sꢀ1
Slope
Conditions
TMTrOTFE
DMTrOTFE
MMTrOTFE
TrOTFE
–
–
–
–
1.79 · 10ꢀ1
8.5 · 10ꢀ2
3.8 · 10ꢀ3
3.4 · 10ꢀ2
298.0 K, I = 1.0 mol · dmꢀ3, 0.7 % CH3CN
298.0 K, I = 1.0 mol · dmꢀ3, 1 % CH3CN
298.0 K, I = 3.0 mol · dmꢀ3, 1 % CH3CN
298.0 K, I = 6.0 mol · dmꢀ3, 3 % CH3CN
[8] J. Crugeiras, H. Maskill, Can J Chem 1999, 77, 530.
experimental errors leading to imprecision in the correlations
have prevented detection of a minor uncatalysed reaction
pathway for the hydrolysis of trityl ethers.
[9] J. Crugeiras, H. Maskill, J Chem Soc, Perkin Trans 2 2000, 441.
[10] N. Mathivanan, R. A. McClelland, S. Steenken, J Am Chem Soc 1990,
112, 8454.
[11] J. A. Kerr, Chem Rev 1966, 66, 465.
[12] C. Bleasdale, B. T. Golding, W. H. Lee, H. Maskill, J. Riseborough,
E. Smits, J Chem Soc, Chem Commun 1994, 93.
[13] W. H. Lee, H. Maskill, J Chem Soc, Perkin Trans 2 1994, 1463.
[14] M. Canle López, J. Crugeiras Martinez, I. Demirtas, H. Maskill, E. Stix,
Org React 1997, 31, 71.
[15] M. Canle, I. Demirtas, M. Elsegood, H. Maskill, J Chem Soc, Perkin
Trans 2 2000, 85.
[16] M. Canle, I. Demirtas, H. Maskill, J Chem Soc, Perkin Trans 2 2001,
1748.
[17] M. Canle, I. Demirtas, A. Freire, H. Maskill, M. Mishima, Eur J Org
Chem 2004, 24, 5031–5039.
[18] M. Canle, W. Clegg, I. Demirtas, M. Elsegood, J. Haider, H. Maskill, P.
Miatt, J Chem Soc, Perkin Trans 2 2001, 1742.
[19] R. A. McClelland, J. V. M. Kanagasabapathy, N. S. Banait, S. Steenken,
J Am Chem Soc 1992, 114, 1816.
[20] R. A. Cox, K. Yates, J Am Chem Soc 1978, 100, 3861.
[21] A. Bagno, G. Scorrano, R. A. M. O’Ferrall, Res Chem Intermediates
1987, 7, 313.
[22] R. A. Cox, Adv Phys Org Chem 2000, 35, 1.
[23] N. C. Deno, J. J. Jaruzelski, A. Schriesheim, J Am Chem Soc 1955,
77, 3044.
[24] E. M. Arnett, R. D. Bushick, J Am Chem Soc 1964, 86, 1564.
[25] M. Horn, H. Mayr, J Phys Org Chem 2012, 25 979.
[26] C. Perrin, J Am Chem Soc 1964, 86, 256.
[27] T. W. Bentley, S. Jurczyk, K. Roberts, D. J. Williams, J Chem Soc, Perkin
Trans 2 1987, 293.
[28] T. W. Bentley, Can J Chem 2008, 86, 277.
[29] A. D. McNaught, A. Wilkinson, Compendium of Chemical Terminology
(the "Gold Book"), 2nd ed, Blackwell Scientific Publications, Oxford,
1997.
[30] M. Canle López, H. Maskill, J. A. Santaballa, Experimental Methods
for Investigating Kinetics, in: The Investigation of Organic Reactions
and their Mechanisms, H. Maskill (Ed.), Blackwell Publishing, Oxford,
2006.
CONCLUSION
We have explained the upward curvatures observed for newly
reported kobs versus [H3O+] kinetics results in the hydrolysis of
tritylamines and trityl ethers in aqueous perchloric acid up to
3.5 mol dmꢀ3. Such previously unexplained and ill-understood
phenomena are shown to be satisfactorily interpreted as
medium effects which can be described to a degree in terms
of the HR acidity function or (better and more appropriately) in
terms of the excess acidity parameter, X0. In particular, our results
þ
show that the HR acidity function (which is defined from KR
values without any consideration of ionic strength effects) is
inadequate at high acidities (and hence high ionic strengths).
The mechanisms previously proposed for the hydrolytic
deamination of protonated tritylamines[16] and dealcoholation
trityl ethers[4] under more mildly acidic conditions account fully
for reactions in the more strongly acidic media described here
once medium effects are acknowledged by using X0 correlations.
Acknowledgements
The authors thank the University of A Coruña, the Autonomous
Government of Galicia (Consellería de Educación e Ordenación
Universitaria) and the Spanish Government (Ministerio de Educación,
Cultura y Deporte, grant SAB2003-0058) for financial support. We
also thank Prof. Ibrahim Demirtas (Univ. of Çankiri Karatekin,
Turkey) who provided compounds, the preparations of which have
already been published, and referees for helpful comments.
[31] N. C. Deno, H. E. Berkheimer, W. L. Evans, H. J. Peterson, J Am Chem
Soc 1959, 81, 2344.
[32] R. A. Cox, K. Yates, Can J Chem 1983, 61, 2225.
[33] E. M. Arnett, Prog Phys Org Chem 1963, 1, 223.
[34] J. F. Bunnett, F. P. Olsen, Can J Chem 1966, 44, 1899.
[35] N. C. Marziano, A. Tomasin, P. G. Traverso, J Chem Soc, Perkin Trans II
1981, 1070.
REFERENCES
[1] J. McOmie, Protective Groups in Organic Chemistry, Plenum Press,
London and New York, 1973.
[2] P. Wuts, T. Greene, Protective Groups in Organic Synthesis, 4th ed,
Wiley, New York, 2006.
[36] A. J. Kresge, H. J. Chen, G. L. Capen, M. F. Powell, Can J Chem 1983,
61, 249.
[37] E. D. Hughes, C. K. Ingold, J Chem Soc 1935, 244.
[38] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, Wiley-
VCH, Berlin, 2003.
[3] M. J. Postle, P. A. H. Wyatt, J Chem Soc, Perkin Trans 2 1972, 474.
[4] M. Canle, I. Demirtas, M. Maskill, M. Mishima,J Phys Org Chem2008, 21, 614.
[5] T. W. Bentley, D. N. Ebdon, E.-J. Kim, I. S. Koo, J Org Chem 2005, 70, 1647.
[6] K. Okamoto, Adv Carbocation Chem 1989, 1, 171.
[7] A. Thibblin, J Chem Soc, Perkin Trans 2 1992, 1195.
wileyonlinelibrary.com/journal/poc
Copyright © 2013 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2013, 26 1016–1022