Base strengths of substituted tritylamines, N-alkylanilines, and tribenzylamine in aqueous solution and the gas phase: Steric effects upon solvation and resonance interactions

Article abstract of DOI:10.1002/ejoc.200400497

The dissociation constants of the conjugate acids of N-tritylacetamide (1h; pK_BH+ = 3.81) and N-benzyl-N-methyl-4,4′, 4″- trimethoxytritylamine (4i; pK_BH+ = 9.86) have been measured in aqueous acetonitrile at 25 °C and at other temperatures to determine the enthalpies and entropies of reaction. For 1h, ΔH = 40.7 kJ·mol^-1 and ΔS = 64 J·K ^-1·mol^-1, and for 4i ΔH = 9.1 kJ·mol^-1 and ΔS = -159 J·K^-1·mol^-1. In addition, gas-phase base strengths at 25 °C (GB values in kJ·mol^-1) of TrNH ₂ (1a; 902.1), TrNHPh (1c; 926.3), TrNHAc (1h; 929.7), TrNHC ₆H₄(o-NO₂) (1i; 895.0), DMTrNH₂ (3a; 921.3), DMTrNHCH₂CO₂Me (3b; 879.1), and DMTrNH(p-NO ₂Bn) (3d; 886.6) have been determined by ICR measurements. The GB of TrNHAc corresponds to protonation at oxygen and B3LYP/6-31G* calculations indicate that the N-protonated isomer is 46.4 kJ·mol^-1 less stable, i.e. the GB value for N-protonation is 883.3 kJ·mol^-1. Correspondingly, the literature GB value of 857.6 5 kJ·mol^-1 for N-methylacetamide corresponds to protonation at oxygen, and B3LYP/6-31G* calculations indicate that the N-protonated isomer is 58.1 kJ·mol^-1 less stable, i.e. the GB value for N-protonation of MeNHAc is 799.5 kJ·mol^-1. The GB of PhNH(tBu) (5; 920.1 kJ·mol^-1) has been measured and compared with values for other N-alkylanilines, PhNHR, including PhNHTr; the results indicate that the increasing GB values as R increases in size are due solely to the increasing polarisability of R. This indicates that the increasing solution base strength of PhNHR as R increases in size is a solvation effect and is not due to decreasing resonance interactions between the nitrogen lone-pair and the phenyl ring. Similarly, the base-strengthening effect in solution of the (substituted) trityl in TrNHZ, where Z is an alkyl with an electron-withdrawing group, is shown to be due to solvation phenomena as it is absent in the gas phase; for one such compound, TMTrNHCH₂CO₂Me (4b; pK_BH+ = 9-30), ΔH = 17.9 kJ·mol^-1 and ΔS = -118 J·K^-1·mol ^-1. In contrast, the difference in solution base strengths between MeNHAc (pK_BH+ = -0.56) and TrNHAc (pK_BH+ = 3.81) is attributed, at least in part, to a reduced base-weakening resonance interaction between the lone pair on N and the acetyl group in TrNHAc, as the effect is also evident in the gas phase. The GB value for tribenzylamine (6) has also been measured (965.2 kJ·mol^-1) and is unexceptional; this indicates that the low base strength of 6 in aqueous solution (pK_BH+ = 4.90 at 25 °C) is a solvation effect which is expressed mainly through an abnormally large positive entropy of reaction (ΔS = 76 J·K^-1·mol^-1), the value of ΔH (50.5 kJ·mol^-1) being only slightly larger than normal for tertiary amines. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200400497

FULL PAPER
Base Strengths of Substituted Tritylamines, N-Alkylanilines, and
Tribenzylamine in Aqueous Solution and the Gas Phase: Steric Effects Upon
Solvation and Resonance Interactions
[a]
[b]
[a]
[c]
´
Moises Canle L., Ibrahim Demirtas, Antonio Freire, Howard Maskill,* and
Masaaki Mishima[d]
Keywords: Amines / Basicity / Density functional calculations / Solvent effects / Steric effects
The dissociation constants of the conjugate acids of N-trityl-
of R. This indicates that the increasing solution base strength
+
acetamide (1h; pKBH
=
3.81) and N-benzyl-N-methyl- of PhNHR as R increases in size is a solvation effect and is
+
4,4Ј,4ЈЈ-trimethoxytritylamine (4i; pKBH = 9.86) have been
measured in aqueous acetonitrile at 25 °C and at other tem-
peratures to determine the enthalpies and entropies of reac-
not due to decreasing resonance interactions between the ni-
trogen lone-pair and the phenyl ring. Similarly, the base-
strengthening effect in solution of the (substituted) trityl in
TrNHZ, where Z is an alkyl with an electron-withdrawing
group, is shown to be due to solvation phenomena as it is
ᮎ
ᮎ
tion. For 1h, ∆H = 40.7 kJ·mol−1 and ∆S = 64 J·K−1·mol−1
,
ᮎ
ᮎ
and for 4i ∆H = 9.1 kJ·mol−1 and ∆S = −159 J·K−1·mol−1. In
addition, gas-phase base strengths at 25 °C (GB values in
kJ·mol−1) of TrNH2 (1a; 902.1), TrNHPh (1c; 926.3), TrNHAc
(1h; 929.7), TrNHC6H4(o-NO2) (1i; 895.0), DMTrNH2 (3a;
921.3), DMTrNHCH2CO2Me (3b; 879.1), and DMTrNH(p-
absent in the gas phase; for one such compound,
ᮎ
TMTrNHCH2CO2Me (4b; pKBH = 9.30), ∆H = 17.9 kJ·mol−1
+
ᮎ
and ∆S = −118 J·K−1·mol−1. In contrast, the difference in so-
lution base strengths between MeNHAc (pKBH = −0.56) and
+
+
NO2Bn) (3d; 886.6) have been determined by ICR measure- TrNHAc (pKBH = 3.81) is attributed, at least in part, to a re-
ments. The GB of TrNHAc corresponds to protonation at oxy- duced base-weakening resonance interaction between the
gen and B3LYP/6-31G* calculations indicate that the N-pro-
tonated isomer is 46.4 kJ·mol−1 less stable, i.e. the GB value
for N-protonation is 883.3 kJ·mol−1. Correspondingly, the lit-
lone pair on N and the acetyl group in TrNHAc, as the effect
is also evident in the gas phase. The GB value for tribenzyl-
amine (6) has also been measured (965.2 kJ·mol−1) and is un-
erature GB value of 857.6 5 kJ·mol−1 for N-methylacetamide exceptional; this indicates that the low base strength of 6 in
+
corresponds to protonation at oxygen, and B3LYP/6-31G*
calculations indicate that the N-protonated isomer is
58.1 kJ·mol−1 less stable, i.e. the GB value for N-protonation
of MeNHAc is 799.5 kJ·mol−1. The GB of PhNH(tBu) (5; ∆H (50.5 kJ·mol−1) being only slightly larger than normal for
aqueous solution (pKBH = 4.90 at 25 °C) is a solvation effect
which is expressed mainly through an abnormally large pos-
ᮎ
itive entropy of reaction (∆S = 76 J·K−1·mol−1), the value of
ᮎ
920.1 kJ·mol−1) has been measured and compared with
values for other N-alkylanilines, PhNHR, including PhNHTr;
the results indicate that the increasing GB values as R in-
creases in size are due solely to the increasing polarisability
tertiary amines.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
Figure 1] have been widely used in synthesis as protecting
groups for hydroxyl and amino functions.[1] We have re-
The trityl group (triphenylmethyl, Tr) and its methoxy- ported previously that (substituted) tritylamines under
substituted analogues [4-monomethoxy- (MMTr), 4,4Ј-di- acidic conditions, where they are present as the correspond-
methoxy- (DMTr), and 4,4Ј,4ЈЈ-trimethoxytrityl (TMTr); ing (substituted) tritylammonium cations, undergo deamin-
ation reactions, and that these are specific acid catalysed.[2,3]
[a]
Chemical Reactivity and Photoreactivity Group, Department This was an exceptional finding as the tritylammonium cat-
of Physical Chemistry and Chemical Engineering I, University
ion has no basic site at the reaction centre to accept a pro-
of A Corun˜a,
ton, and a mechanism involving intermediate ion-molecule
pairs was proposed. Methoxy substituents at the para posi-
tions of the phenyl groups are, as expected, rate enhancing,
and these rate effects are observed in both the catalysed
and uncatalysed reaction channels. Remarkably, however, R
substituents on the nitrogen have much more dramatic ef-
fects upon the deamination rate constants, and these effects
15071 A Corun˜a, Galicia, Spain
[b]
Present address: Department of Chemistry, University of
Gaziosmanpasa,
60250, Tokat, Turkey
[c]
Chemistry Department, University of Newcastle,
Newcastle upon Tyne, NE1 7RU, UK
Institute for Materials Chemistry and Engineering, Kyushu
University,
[d]
Hakozaki, Higashi-ku, Fukuoka, Japan
Eur. J. Org. Chem. 2004, 5031Ϫ5039
DOI: 10.1002/ejoc.200400497
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5031

M. Canle L., I. Demirtas, A. Freire, H. Maskill, M. Mishima
FULL PAPER
Figure 1. Structures of substituted amines used as protection
are not very sensitive to the nature of R.[4] In addition to
The modern measure of base strength in aqueous solu-
ϩ
these kinetic effects, we have also observed unexpected base tion is the pKBH , i.e. the pKA of the conjugate acid of the
strengths in a range of substituted tritylamines.[5] For ex- base, and this may be related by a familiar thermodynamic
ᮎ
ample, N-tritylarylamines are strong bases like alkylamines relationship to the standard molar free energy, ∆G , for the
rather than weak bases like arylamines, and substituents in proton transfer reaction between the protonated base and
the aromatic ring of the arylamine have very little effect water (the solvent), usually at 298 K [Equation (1)].[11,12]
upon the base strength. We speculated whether the base
strength of a substituted N-tritylalkylamine was directly re-
ᮎ
BH ϩ H2O Ǟǟ B ϩ H3Oϩ; ∆G
(1)
ϩ
ϩ
ᮎ
ᮎ
lated to the rate constants for its deamination, but this ap-
pears not to be the case.[4] Nevertheless, it seems improbable
that, for example, there is no connection at all between the
large rate enhancements in the deaminations of N-tritylaryl-
amines and their unusually high base strengths, both being
characteristic of the same structural features. We have now
taken a step back to investigate whether the anomalously
high base strengths of some substituted tritylamines are due
to intrinsic steric or electronic molecular effects, or peculi-
arities arising out of aqueous solvation phenomena.[6]
We also became aware of the anomalously low base
strength in aqueous solution of tribenzylamine Bn3N
ϩ
ϩ
∆G ϭ ϪRT·lnKA; KA of BH ϭ KBH of B; ∆G ϭ 2.303 RT·pKBH
In the gas phase, however, base strengths are conven-
tionally expressed as GB values where GB is the negative of
the free energy of reaction between the base and proton;
i.e. it is the free energy of deprotonation of the conjugate
acid of the base in the gas phase, usually at 298 K, and
values are always positive.[10,13]
BH Ǟǟ B ϩ Hϩ; ∆G° ϭ GB of B
(2)
ϩ
ϩ
(pKBH ഠ 5 compared with values of about 10 for typical
trialkylamines).[7,8] Here also, we sought to establish
Direct comparison between gas phase and solution is
whether this anomaly is an intrinsic molecular effect (steric further complicated by the conventions that in the gas
or electronic) or due to aqueous solvation phenomena phase the standard state is the pure gas at 1 atm (and the
(whilst acknowledging that solvation itself is subject to thermochemical standard state superscript is used) whereas
steric constraints).[6,9,10] For selected substituted tritylam- in solution an activity of 1 mol·dmϪ3 is usually more con-
ines and related compounds, and for tribenzylamine, there- venient for solutes and mol fraction of unity for the solvent
fore, we have measured base strengths in the gas phase for (and the defined standard state superscript is used). Oc-
comparison with results in solution. And, to establish casionally, the molal scale is used for solutions but, for di-
whether unusual base strengths in solution are entropy or lute aqueous solutions at 25 °C, the numerical difference
enthalpy effects, we have measured equilibrium constants between molal and molar values is minute.[14] In this report,
for some proton-transfer equilibria at different tempera- we are comparing differential effects within the gas phase
tures.
with differential effects in solution, rather than making di-
5032
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 5031Ϫ5039

Base Strengths of Substituted Amines
FULL PAPER
rect comparisons of absolute base-strength values in the two results for substituted tritylamines are given in Table 1, and
phases; consequently, there is no need to convert reactions in Table 2 for other compounds; computational results are
in the two phases to a common standard state. However, it given below these tables.
ϩ
is occasionally convenient to convert solution pKBH values
into standard free energies, and these are referred to as SB
Base Strengths of Substituted Tritylamines
ᮎ
values, where SB ϭ ∆G [Equation (1)]; so, SB is the stand-
ϩ
Simple alkylamines have pKBH values in the region of
ard molar free energy for proton transfer from the pro-
tonated base to water in aqueous solution, usually at 25 °C,
the standard state for solutes being an activity of 1
mol·dmϪ3, and values are positive for bases stronger than
water.
9Ϫ11.[8,11,14] The results in Table 1 for the base strengths of
TrNH2 (1a), MMTrNH2 (2a), DMTrNH2 (3a), and
TMTrNH2 (4a) in aqueous solution containing acetonitrile
confirm that: (i) all are typical values for simple alkylam-
ines, (ii) the modest differences between the solvent compo-
sitions have minimal effect, and (iii) the methoxy substitu-
ents have no appreciable effect. In contrast, GB values for
simple primary amines, RNH2, increase from the value for
methylamine with increasing molecular weight (Table 2). If
we consider the sequence of primary amines RNH2 for R ϭ
Me, Et, nPr, nBu, Bn, cyclohexyl, and n-C10H21NH2, i.e.
an arbitrary sequence of primary amines with hydrocarbon
residues of increasing molecular weight, an upper limit for
GB of about 900 kJ·molϪ1 is indicated (Table 2).[13] That
this is largely a polarisability effect is indicated by the ap-
It is well-known that the dissociation of some acids in
solution is dominated by entropy effects; for example, for a
ᮎ
wide range of polar-substituted carboxylic acids ∆H ഠ 0
and substituent effects are due to changes in the strongly
ᮎ
negative ∆S .[12] In the case of sterically hindered alkyl-sub-
stituted carboxylic acids, however, small changes in acid
strength are due to complex entropic and enthalpic solv-
ation effects for both the neutral acid and its conjugate
base, the anion.[15] For dissociation of protonated amines,
ᮎ
∆H values are usually moderately positive (typically
ᮎ
40Ϫ50 kJ·molϪ1), but ∆S values may vary widely (though
most are negative in the range Ϫ5 to Ϫ30 J·KϪ1·molϪ1).[8]
In order to identify causes of particular effects in some of
Table 1. Base strengths of substituted tritylamines in aqueous
the present cases, we have measured pK values at different acetonitrile and in the gas phase[a]
ᮎ
temperatures in order to resolve SB values [∆G , Equa-
ᮎ
ᮎ
ϩ
Base
pKBH (% CH3CN) GB (kJ·molϪ1
)
tion (1)] into ∆H and ∆S components. In contrast, struc-
tural effects upon gas-phase base strengths are invariably
energy (enthalpy) effects, and it is seldom necessary to re-
solve the GB term [∆G°, Equation (2)] into the proton affin-
ity (PA) and the entropy of proton transfer.[13]
TrNH2 (1a)
TrNHCH2CO2Me (1b)
TrNHPh (1c)
TrNHAc (1h)
TrNHC6H4(2-NO2) (1i)
MMTrNH2 (2a)
9.2(16)[b] 25 °C
10.1(22)[c]
Ϫ
902.1
Ϫ
926.3
929.7[e]
895.0[f]
Ϫ
3.81(37)[d] 25 °C
Ϫ
9.3(10)[c]
MMTrNH(p-NO2Bn) (2d) 10.5(38)[c]
Ϫ
DMTrNH2 (3a)
DMTrNHCH2CO2Me (3b)
DMTrNHPh (3c)
DMTrNH(p-NO2Bn) (3d)
DMTrNHBn (3f)
N-DMTr-pyrrolidine (3g)
N-DMTr-imidazole (3h)
9.5(28)[b]
921.3
879.1
Ϫ
9.7(21)[c]
Results and Discussion
8.9(39)[c]
Ϫ
886.6
Ϫ
9.9(30)[g]
8.7(21)[c]
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Preparations and Methods
7.5 (37)[h] 25 °C
9.3(26)[b]
Most amines were available from our previous TMTrNH2 (4a)
studies;[2Ϫ5] additionally, N-tert-butylaniline (5) and N-tri-
tylacetamide (1h) were prepared by methods based upon
literature descriptions,[16,17] and their spectral and other
properties were in agreement with assigned structures.[16Ϫ19]
The new compound N-(4,4Ј-dimethoxytrityl)imidazole (3h)
was also prepared in the obvious way from imidazole and
dimethoxytrityl tetrafluoroborate.[4] Our methods for base-
strength measurements in solution have already been de-
TMTrNHCH2CO2Me (4b)
TMTrNHC3H7 (4e)
TMTrNHBn (4f)
N-TMTr-pyrrolidine (4g)
TMTrNHC4H9 (4h)
TMTrN(Me)Bn (4i)
9.30(15)[i] 25 °C
9.7(26)[b]
9.4(32)[b] 25 °C
8.2(21)[c]
9.7(21)[c] 22 °C
9.86(20)[j] 25 °C
Ϫ
[a]
ϩ
pKBH values of the bases, i.e. the pKA values of the conjugate
acids, were measured in acetonitrile/water solution at 21 °C, ionic
strength (NaClO4) ϭ 1 mol·dmϪ3 unless otherwise indicated; the
scribed and require no further comment.[4,5] Our gas-phase volume % of acetonitrile is given in brackets. GB values are at
ᮎ
Present work. ∆H ϭ 40.7 kJ·molϪ1
[b]
[c]
[d]
298 K.
Ref.[5]
Ref.[4]
measurements followed conventional ICR (Ion Cyclotron
Resonance) practice,[20] and aspects of the method have al-
ready been described.[21] The GB values of amides have
been calculated at the B3LYP/6-31G* level of theory using
the Gaussian 98 program suite to obtain separate values for
protonation at nitrogen and at oxygen.[22] Our computed
result for protonation of N-methylacetamide at oxygen
(867.5 kJ·molϪ1) is in satisfactory agreement with the litera-
ᮎ
and ∆S ϭ 64 J·KϪ1·molϪ1
.
[e] Corresponds to protonation at oxy-
gen; B3LYP/6-31G* calculations indicate that the N-protonated
isomer is 46.4 kJ·molϪ1 less stable, i.e. GB ϭ 883.3 kJ·molϪ1 for N-
[f]
protonation. Corresponds to protonation at the amino nitrogen;
B3LYP/6-31G* calculations indicate that the O-protonated isomer
is 23.0 kJ·molϪ1 less stable, i.e. GB ϭ 872.0 kJ·molϪ1 for pro-
[g]
tonation at an oxygen of the nitro group. Ref.[3] [h] Approximate
ᮎ
ᮎ
[i]
value.
Present work. ∆H ϭ 17.9 kJ·molϪ1 and ∆S ϭ Ϫ118
ᮎ ᮎ
[j]
J·KϪ1·molϪ1
.
Present work. ∆H ϭ 9.1 kJ·molϪ1 and ∆S
ϭ
ture experimental GB value (857.6 kJ·molϪ1). Experimental Ϫ159 J·KϪ1·molϪ1
.
Eur. J. Org. Chem. 2004, 5031Ϫ5039
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5033

M. Canle L., I. Demirtas, A. Freire, H. Maskill, M. Mishima
FULL PAPER
expectation, GB values for DMTrNH(p-NO2Bn) (3d) and
DMTrNHCH2CO2Me (3b) are 886.6 and 879.1 kJ·molϪ1
compared with 921.3 kJ·molϪ1 for DMTrNH2 (3a; Table 1).
This effect is not observed in aqueous solution, however,
where TrNHCH2CO2Me (1b), MMTrNH(p-NO2Bn) (2d),
DMTrNHCH2CO2Me (3b), and TMTrNHCH2CO2Me (4b)
Table 2. Base strengths of other amines in aqueous solution and in
the gas phase[a]
ϩ
Base
pKBH
GB (kJ·molϪ1
)
MeNH2
EtNH2
nPrNH2
nBuNH2
BnNH2
10.65[b]
864.5
878.0
883.9
886.6
879.4
899.9
899.6
896.5
857.6[e]
850.6
890.1
892.9
Ϫ
920.1
918.1
951
960.1
967.6
965.2[i]
10.68[b]
10.57[b]
10.64[b]
9.34[c]
ϩ
are all strong bases (pKBH values of 10.1, 10.5, 9.7, and
9.3, respectively; Table 1), comparable with DMTrNHBn
(3f), TMTrNHC3H7 (4e), TMTrNHC4H9 (4h), and
tBuNH2
10.68[c]
10.58[b]
ca. 10.6
Ϫ0.56[d]
4.63[f]
ϩ
TMTrNHBn (4f) (pKBH values of 9.9, 9.7, 9.7, and 9.4,
cyclo-C6H11NH2
n-C10H21NH2
MeNHAc
PhNH2
PhNHMe
PhNHEt
PhNH(iPr)
PhNH(tBu) (5)
Me3N
Et3N
n-Pr3N
nBu3N
Bn3N (6)
respectively; Table 1). Looked at the other way round,
ϩ
NH2CH2CO2Me (pKBH ϭ 6.84Ϫ7.59 depending upon the
experimental conditions)[8,23] and p-NO2BnNH2 (pKBH
ϭ
ϩ
4.85[c]
8.50)[24] are both weaker bases than simple alkylamines in
water on account of the electron-withdrawing effects of the
methoxycarbonyl and the nitrophenyl groups, but addition
of the trityl appears to cancel these inductive base-weaken-
ing effects in solution. In other words, introduction of the
sterically demanding N-Tr group increases the base
strengths of NH2CH2CO2Me and p-NO2BnNH2 in solu-
tion. This is unlikely to be a molecular strain effect as this
is invariably base-weakening for monoamines.[25] The effect
is parallel with, though larger than, increased steric hin-
drance enhancing the base strength of alkyl-substituted car-
boxylate and (in methanol) phenolate anions,[15,26] bases of
a different charge type, where the effect was attributed to
steric hindrance of solvation of the (anionic) bases. In con-
trast, increased steric hindrance leads to reduced solution
base strength in ketones,[27] pyridines,[28] piperidines,[29] and
5.12[c]
5.77[c]
7.0[g]
9.80[b]
10.72[b]
10.66[b]
9.93[b]
4.90[h]
[a] pKBH values of the bases are the pKA values of the correspond-
ing alkylammonium ions; all results are at 298 K. GB values are at
298 K and taken from reference 12 except where otherwise indi-
ϩ
ᮎ
cated. [b] Ref.[14] [c] Ref.[11] [d] ∆H ϭ Ϫ2.9 kJ·molϪ1 and ∆Sᮎ ϭ 1.3
[e]
J·KϪ1·molϪ1 [35]
.
Experimental value corresponding to pro-
tonation at oxygen (see ref.[13]); our B3LYP/6-31G* calculations
give a value of 867.5 kJ·molϪ1 and indicate that the N-protonated
isomer is 58.1 kJ·molϪ1 less stable, i.e. the GB value for N-pro-
tonation, based upon the experimental result for O-protonation, is
[f]
[g]
[h]
799.5 kJ·molϪ1
.
Ref.[34]
Ref.[43]
Present work by interp-
olation from results at other temperatures in 12% acetonitrile/ anilines,[30] where the effect has been attributed to steric
water; ∆Hᮎ ϭ 50.5 kJ·molϪ1 and ∆S ϭ 76 J·KϪ1·molϪ1; see
ᮎ
hindrance of solvation of the (cationic) conjugate acid,
though it may be expressed through either the enthalpy or
also ref.[7] [i] Present work.
the entropy of dissociation. There is clearly an unresolved
preciably larger value for tert-butylamine (899.9 kJ·molϪ1
)
paradox here: in some cases increased steric hindrance leads
than for n-butylamine (886.6 kJ·molϪ1). We appear to be to enhanced base strength, and in others to reduced base
observing, therefore, the asymptotic value for the GB of strength.
alkylamines, RNH2, with TrNH2 (GB ϭ 902.1 kJ·molϪ1
)
For 4b, a representative member of this group of com-
and it seems unlikely that the difference between this value pounds of enhanced base strength through steric hindrance,
ϩ
and the appreciably higher result for DMTrNH2 (GB ϭ the pKBH was measured at different temperatures to give
ᮎ
ᮎ
921.3 kJ·molϪ1) is due simply to the changed molecular ∆H ϭ 17.9 kJ·molϪ1 and ∆S ϭ Ϫ118 J·KϪ1·molϪ1 for
weight/polarisibility attributable to the addition of two the reaction of Equation (1). These results are broadly simi-
methoxy substituents. More likely, the induced positive lar to those for TMTrN(Me)Bn (4i, Table 1), which has a
ϩ
charge at the carbon to which the protonated nitrogen is similar pKBH at 25 °C, but characteristically different from
bonded in the conjugate acid is stabilised by delocalisation results for simple secondary amines of similar base strength
ᮎ
ᮎ
of electron density from the two para-methoxy substituents. where ∆H is typically about 50Ϫ60 kJ·molϪ1 and ∆S
ϭ
This is entirely compatible with our report that protonation Ϫ12 to Ϫ30 J·KϪ1·molϪ1 [8]
The relatively low positive en-
.
of the nitrogen of the parent tritylamine leads to elongation thalpy and large negative entropy of reaction are compat-
of the CϪN bond and some planarisation of the central ible with a greater increase in solvation for Equation (1)
carbon;[4] it represents a mixing of resonance and polaris- with base 4b than with simpler analogues of similarly high
ation electronic effects normally regarded as separate. On base strength. So, compared with more typical amines, B,
this basis, we predict that TMTrNH2 will have a slightly 4b (like 4i) must be unusually strongly solvated or its pro-
ϩ
larger GB value, and that the result for MMTrNH2 will lie tonated form, BH , unusually weakly solvated.
between those for TrNH2 and DMTrNH2.
From a consideration of intrinsic molecular properties
Base Strengths of N-Alkylanilines PhNHR
alone, addition of an inductively electron-withdrawing alkyl
Arylamines are generally weaker bases in water than
substituent onto the nitrogen of a tritylamine would be ex- alkylamines and this is ascribed to the delocalisation of the
pected to reduce the gas-phase base strength, even though lone pair on nitrogen into the arene ring, i.e. a reson-
the molecular weight is increased. In agreement with this ance effect. However, for N-alkylanilines, PhNHR, the base
5034
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 5031Ϫ5039

Base Strengths of Substituted Amines
FULL PAPER
ϩ
strength increases from pKBH ϭ 4.63 for aniline through ing less basic than a simple amine. Thermochemical data
ϩ
pKBH ϭ 4.85, 5.12, 5.77, 7.0, and 8.9 (corresponding to a for proton-transfer reactions of amides in solution are
range of 23.1 kJ·molϪ1 in SB values) as R increases from sparse, however, and results in water have to be obtained
R ϭ Me, Et, iPr, tBu, to DMTr (Tables 1 and 2), and indirectly, for example by acidity function techniques in
PhNHDMTr is virtually as strong a base as a simple alkyl- concentrated acids. For three substituted benzamides, Ar-
ϩ
amine. We suggested previously that the increasing steric CONH2, which have typically low pKBH values of Ϫ1.54
ᮎ
bulk of R increasingly inhibits the delocalisation of the lone to Ϫ2.49 in water at 25 °C, ∆H ϭ Ϫ1 to Ϫ4 kJ·molϪ1
ᮎ
pair from nitrogen into the benzene ring,[4] a previously and ∆S ϭ 25Ϫ36 J·KϪ1·molϪ1 [38] For a range of acylated
.
ᮎ
identified phenomenon,[30] in which case the same trend methylamines, RCONHMe, ∆H ranges between 0 and
ᮎ
should be observed in the gas phase. For aniline, GB is only Ϫ10.9 kJ·molϪ1
,
and ∆S
between 20.9 and Ϫ6.7
850.6 kJ·molϪ1 compared with 899.6 for cyclohexylamine, J·KϪ1·molϪ1, with the values for MeNHAc (pKBH
ϭ
ϩ
ᮎ
ᮎ
a primary alkylamine of similar molecular weight but with- Ϫ0.56) being ∆H ϭ Ϫ2.9 kJ·molϪ1 and ∆S ϭ 1.3
out the resonance possibility, and the GB then increases J·KϪ1·molϪ1 [35]
.
It appears, therefore, that the enhanced
along the series 890.1, 892.9, and 920.1, to 926.3 for base strength of N-tritylacetamide in solution is expressed
PhNHR for R ϭ Me, Et, tBu, and Tr (Tables 1 and 2). both in exceptional enthalpy and entropy terms compared
However, this increase in GB (36.2 kJ·molϪ1) with increas- with normal amides.
ing size of R in PhNHR is only about the same as the in-
In the gas phase, as in solution,[32] protonation of amides
crease in GB along the series RNH2 for R ϭ Me, Et, nPr, occurs at oxygen rather than nitrogen (Figure 2). Calcu-
nBu, cyclohexyl, and Tr (37.6 kJ·molϪ1) which can only be lations at the B3LYP/6-31G* level, indicated below Tables 1
due to the increasing polarisability of R. It appears, there- and 2, lead to GB values for MeNHAc (799.5 kJ·molϪ1
)
fore, that there is no evidence of an attenuation of the base- and TrNHAc (883.3 kJ·molϪ1) for protonation at nitrogen,
weakening resonance effect in PhNHR in the gas phase as i.e. TrNHAc is a stronger gas phase base by 83.8 kJ·molϪ1
.
R increases in size from Me to trityl. In accord with this The free-energy diagrams for protonation of TrNHAc and
interpretation, introduction of an ortho-nitro substituent MeNHAc are combined in Figure 3 with superposition of
into N-tritylaniline (1c) to give 1i causes an appreciable re- the levels for the two N-protonated species for which there
duction in GB from 926.3 to 895.0 kJ·molϪ1 (Table 1).[31] can be no resonance in the functional group. If we assume
The effect in solution, therefore, must be a sterically in- that the difference in GB values of 37.6 kJ·molϪ1 between
duced solvation phenomenon. And, here again, it is base- methylamine and tritylamine is due to the increased polaris-
strengthening, as for alkyl-substituted carboxylate and (in ability of the alkyl residue on the nitrogen that becomes
methanol) phenolate anions,[15,26] rather than base-weaken- protonated (see above), and that this may be transferred
ing as in ketones,[27] pyridines,[28] piperidines,[29] and ring- to other pairs of compounds bearing identical structural
alkylated anilines.[30]
Base Strengths of N-Acetylamines
The base strengths of amines in solution are generally
much reduced by acetylation, and the site of protonation is
usually oxygen rather than nitrogen.[32] For example,
ϩ
pKBH ϭ 0.61 for acetanilide compared with 4.63 for ani-
line (δSB ϭ 22.9 kJ·molϪ1),[33,34] and Ϫ0.56 for N-methyl-
acetamide compared with 10.65 for methylamine (δSB ϭ
64.0 kJ·molϪ1).[35,14] In line with our observations above, N-
ϩ
tritylacetamide (pKBH ϭ 3.81, Table 1) is not as weak a
base as N-methylacetamide, but is still much weaker than
Figure 2. Resonance in acetamides and their conjugate acids
ϩ
tritylamine (pKBH ϭ 9.2). It compares with, and supports,
ϩ
the result (pKBH ϭ 5.3) estimated by Pracejus for 6,6-di-
methyl-1-azabicyclo[2,2,2]octan-2-one (7a), an amide whose
bicyclic structure prevents appreciable resonance between
the nitrogen and the carbonyl.[36] This difference of 4.37 in
ϩ
pKBH values between N-tritylacetamide and N-methyl-
acetamide corresponds to δSB ϭ 24.9 kJ·molϪ1, and makes
N-tritylacetamide one of the strongest carbonyl bases
known (assuming protonation is at the oxygen rather than
ᮎ
at the nitrogen).[37] We found that ∆H ϭ 40.7 kJ·molϪ1 for
N-tritylacetamide in Equation (1) (see also footnote d to
Table 1), a result not dissimilar from values for much more
strongly basic amines,[8] and it is only the substantially posi-
ᮎ
tive ∆S value of 64 J·KϪ1·molϪ1 [corresponding to desolv-
ation in Equation (1)] which leads to N-tritylacetamide be- TrNHAc and MeNHAc in the gas phase
Figure 3. Combined free-energy diagrams for protonation of
Eur. J. Org. Chem. 2004, 5031Ϫ5039
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5035

M. Canle L., I. Demirtas, A. Freire, H. Maskill, M. Mishima
FULL PAPER
features, the higher GB value of TrNHAc compared with Experimental difficulties allowed determination of only an
MeNHAc for N-protonation, which is not attributable to approximate result for N-(4,4Ј-dimethoxytrityl)imidazole
the increased alkyl polarisability, drops to 46.2 kJ·molϪ1
.
(3h, pKBH ϭ 7.5, 25 °C, 37% aqueous acetonitrile). The
ϩ
We suggest that this, or an appreciable proportion of it, is provisional conclusion is that the N-trityl group has a
the difference in gas-phase base-strength between TrNHAc minor base-strengthening effect upon imidazole, a base of
ϩ
and MeNHAc for N-protonation due to a reduced reson- moderate strength in solution (pKBH ϭ 6.99 for imidaz-
ance interaction between the lone pair on nitrogen and the ole),[8] compared with no effect for strongly basic primary
acetyl group in TrNHAc. The resonance energy (RE) in and secondary alkylamines, and a large base-strengthening
normal amides is about 85 kJ·molϪ1 [39]
so the reduction effect for weakly basic arylamines, alkylamines, and aceta-
,
being proposed here is not more than about 50%. Such a mide. These matters warrant further exploration since the
reduction would not be expected to be detectable in the IR steric features introduced by the trityl group are the same
stretching vibration of the carbonyl bond as 1-azabicyclo- in all cases.
[3.3.1]nonan-2-one (8), whose cage structure leads to a re-
Base Strength of Tribenzylamine (6)
duction in RE of about 60%, still has an amide carbonyl
vibration at 1680 cmϪ1 [40]
,
i.e. within the normal range for
The anomalously low base strength of tribenzylamine (6)
a tertiary amide. The value for N-tritylacetamide is 1655
cmϪ1, i.e. fairly normal for a secondary amide in accord
with a proposed reduction in RE of up to about 50%. In
contrast, the CO stretching band for the unstable 1-aza-
bicyclo[2.2.2]octan-2-one (7b) and its alkylated derivatives,
which have almost no RE, has been estimated to be at 1733
in aqueous solution has been known in the literature for
ϩ
many years (pKBH ca. 5 compared with ca. 9 for typical
trialkylamines).[7,8,11] It is also known to be a very weak
hydrogen-bond acceptor, so weak that the π-systems of the
phenyl groups, rather than the lone pair on nitrogen, are
the basic site.[42] Our new result of pKBH ϭ 4.90 at 25 °C
ϩ
cmϪ1 [39]
antanone (9) is 1732 cmϪ1 [41]
;
the experimental value for the stable 1-azaadam-
in 12% acetonitrile/water confirms the modest solvent ef-
fect. We have now measured its base strength in the gas
phase for comparison with other trialkylamines. First, we
note in Table 2 the increasing base strengths of simple alkyl-
amines, RNH2, with increasing polarizabilty of the R
groups (864.5, 878, 883.9, and 886.6 kJ·molϪ1 for R ϭ Me,
Et, nPr, and nBu).[13] The result for benzylamine is
879.4 kJ·molϪ1, i.e. it lies between EtNH2 and nPrNH2 in
base strength in the gas phase. We also observe a gradual
increase in base strength with the size of R in simple tri-
alkylamines, R3N (918.1, 951, 960.1, and 967.6 kJ·molϪ1
.
A similar analysis for O-protonatation is more compli-
cated for two reasons. First, resonance is feasible in both
the O-protonated and the unprotonated forms; indeed, as
it does not involve charge separation, one might expect (in
the absence of other effects) a greater degree of resonance
in the O-protonated form. Secondly, less than 37.6 kJ·molϪ1
due to the polarisability difference between Me and Tr in
amines may be transferable because, in O-protonation, the
alkyl residue is not now on the atom suffering protonation.
We observe that, for O-protonation, TrNHAc is more basic
than MeNHAc by 72.1 kJ·molϪ1 (results under Tables 1
and 2, illustrated in Figure 3). It follows, therefore, that,
depending upon how much of the 37.6 kJ·molϪ1 is transfer-
able, the difference in gas-phase base-strength at oxygen be-
tween TrNHAc and MeNHAc attributable to reduced res-
onance stabilisation in the O-protonated species lies be-
for
R ϭ Me, Et, nPr, and nBu). Our result of
965.2 kJ·molϪ1 for Bn3N puts it between tripropylamine
and tributylamine. It appears that there is nothing excep-
tional about the base strength of Bn3N in the gas phase,
and the abnormal result in aqueous solution must, there-
fore, be
a
solvation effect. By measuring its
ϩ
pKBH over the temperature range 7.9 to 45.4 °C, we have
tween 11.7 and 49.3, i.e. between about 12 and 50 kJ·molϪ1
.
ᮎ
ᮎ
determined ∆H and ∆S for Equation (1) (Figure 4). The
This resonance effect will contribute towards the base
strength of TrNHAc being greater than that of MeNHAc
ϩ
in solution considered above, i.e. δpKBH ϭ 4.37. However,
in view of the sterically induced solvation effects identified
above for other N-tritylamine derivatives, it would be sur-
prising if there were not a similar contribution here also.
Other Tritylamines
Substituted N-trityl tertiary amines [N-DMTr-pyrrolidine
(3g), N-TMTr-pyrrolidine (4g), and TMTrN(Me)Bn (4i)]
are comparable in base strength with simple secondary am-
ines and other tertiary amines (Table 1), so addition of an
N-trityl group to a secondary amine has no base-enhancing
effect when the secondary amine is already strongly basic.
ᮎ
For TMTrN(Me)Bn, however, the positive ∆H is numeri-
ᮎ
cally small (9.1 kJ·molϪ1) and ∆S unusually large and
ᮎ
negative (Ϫ159 J·KϪ1·molϪ1). So, although ∆G (pKBH ) is
unexceptional, it is the result of untypical component terms. ous acetonitrile
ϩ
ϩ
Figure 4. vanЈt Hoff plot for pKBH of tribenzylamine in 12% aque-
5036  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjoc.org
Eur. J. Org. Chem. 2004, 5031Ϫ5039

Base Strengths of Substituted Amines
FULL PAPER
ᮎ
N-Dimethoxytritylimidazole (3h): A solution of dimethoxytrityl
tetrafluoroborate (0.300 g, 0.769 mmol) in dichloromethane (5 mL)
was added dropwise to a stirred solution of imidazole (0.16 g,
2.35 mmol) in dichloromethane (5 mL).[4] After about 5 min, the
reaction mixture was added to water (ca. 25 mL) and worked up
in the normal way to give a pale-yellow oil which gave crystals from
diethyl ether/pentane (1:1). Recrystallisation from heptane gave
colourless crystals (160 mg, 0.432 mmol, 56%). M.p. 115Ϫ116 °C.
1H NMR: δ ϭ 3.82 (s, 6 H, OMe), 6.8Ϫ7.45 (m incl. AAЈBBЈq, 16
H, arom) ppm. 13C NMR: δ ϭ 55.27 (OMe), 74.41 (C-N), 113.17
substantial positive enthalpy of reaction (∆H
ϭ
50.5 kJ·molϪ1) is typical of results for simple alkylamines,
though perhaps slightly large, but the large positive entropy
ᮎ
of reaction (∆S ϭ 76 J·KϪ1·molϪ1) is abnormal; for com-
ᮎ
ᮎ
parison, ∆H ϭ 47.4 and 54.3 kJ·molϪ1, and ∆S ϭ Ϫ4
and Ϫ3 J·KϪ1·molϪ1, for dibenzylamine (pKBH ϭ 8.52)
ϩ
[8]
ϩ
and benzylamine (pKBH ϭ 9.35). Clearly, the large posi-
tive entropy of deprotonation of Bn3NHϩ is principally
what causes tribenzylamine to be so anomalously weak a
base in aqueous solution. This cannot be due to abnormal (C3,5) 121.55 (Im), 127.81 (C4Ј), 127.91 (C3Ј,5Ј), 128.2 1 (Im),
129.53 (C2Ј,6Ј), 131.00 (C2,6), 134.89 (Im), 138.93 (C1), 143.12
(C1Ј), 159.02 (C4) ppm.
intrinsic entropic differences between protonated and un-
protonated forms otherwise the reduced base strength
would also have been seen in the gas phase. We conclude,
therefore, that either Bn3N is exceptionally weakly solvated,
or that Bn3NHϩ is unusually strongly solvated, and it is
Solution Base-Strength Measurements: Base-strength determi-
nations in solution were made in the conventional way usually both
by titration of the free base against standard acid and by titration
the shedding of solvent molecules in the deprotonation of of the conjugate acid of the base against standard sodium hydrox-
ᮎ
ide, as described previously.[4] Microcal Origin (Microcal Software
Inc.) was used to analyze the potentiometric data. Multiple deter-
minations were usually made at each of a range of temperatures
for each compound in a cell thermostatted by water circulating
Bn3NHϩ which causes ∆S to be so positive (and, corre-
ᮎ
spondingly, ∆H to be somewhat more endothermic) and
Bn3N such a weak base. The evidence that Bn3N is a very
weak hydrogen-bond acceptor suggests that it is the free
base which is weakly solvated rather than the conjugate acid
which is strongly solvated.[42]
ϩ
from a thermostatted water bath. Average pKBH values are as fol-
lows: TrNHAc (1h; 37% acetonitrile/water): 3.81 (298 K), 3.60
(308 K), and 3.26 (323 K); TMTrN(Me)Bn (4i; 20% acetonitrile/
water): 10.01 (283 K), 9.95 (289 K), 9.87 (302 K), 9.83 (316 K), 9.79
(323 K); TMTrNHCH2CO2Me (4b; 15% acetonitrile/water): 9.53
(283 K), 9.30 (298 K), 9.25 (303 K), 9.17 (318 K); Bn3N (6; 12%
acetonitrile/water): 5.47 (281 K), 5.30 (285 K), 5.24 (288 K), 4.95
(296 K), 4.77 (305 K), 4.33 (318 K).
Experimental Section
N-tert-Butylaniline (5):[16,43] n-Butyllithium (Aldrich, 100 mL of a
2.5 mol·dmϪ3 solution in hexanes, 0.25 mol) was added by cannula
dropwise to magnetically stirred tert-butylamine (Avocado, 98%,
500 mL) under a stream of nitrogen in a three-necked flask at Ϫ78
°C. After the initially formed suspension cleared, the temperature
was allowed to rise first to Ϫ50 and then to 0 °C as another suspen-
sion formed. Iodobenzene (Avocado, 98%, 10 mL) was added over
10 min by syringe whereupon the colour changed to yellow and the
solid dissolved; the reaction was allowed to warm to room tempera-
ture as a solid reappeared. Analysis by TLC after ca. 40 min indi-
cated no iodobenzene, so a further amount (5 mL) was added (to-
tal ϭ 28 g, 0.137 mol). After 1 h, further analysis indicated the
presence of iodobenzene; water (10 mL) was added dropwise after
a further 12 h and the reaction was worked up in the normal way
to give a crude oil (18 g) which was fractionally distilled at oil-
pump pressure to give the product as a colourless oil [b.p. 54Ϫ55
Gas-Phase Base-Strength Measurements: An Extrel FTMS 2001
Fourier transform mass spectrometer was used for base-strength
measurements in the gas phase.[21] An experiment was initiated by
a 5 ms pulse of an electron beam (20 eV) through the ICR cell in
which the total pressure of the bases was maintained at lower than
1 ϫ 10Ϫ6 Torr. The proton-transfer equilibrium was achieved
within several hundred ms of the initiation of the reaction (de-
pending on the pressure of the neutrals), and the equilibrium con-
stant for each proton-transfer reaction was evaluated from K ϭ
ϩ
ϩ
ϩ
[Bo][BH ]/[B][BoH ]. The relative abundances of ions BH and
BoHϩ were determined by the relative intensities of ICR mass spec-
tra at equilibrium. The pressures of the neutral reactants were
measured by a BayardϪAlpert-type ionization gauge applying ap-
propriate correction factors to correct the gauge readings for the
different ionization cross-sections of the various compounds,[44]
and experiments were performed at several ratios of partial press-
ures and at different overall pressures. The arithmetic means of the
values of K were used to calculate ∆G°, the average uncertainty
being Ϯ0.8 kJ·molϪ1 in most cases, and each value was measured
with two reference bases. Gas-phase basicity values for reference
compounds were taken from the literature.[13] Ion-eject experiments
using the SWIFT technique[45] were also carried out to ensure pro-
ton-transfer reactions occurred. For compounds of low volatility
investigated in this study, the solid sample direct-inlet system was
used and all vacuum chamber systems were kept at 60Ϫ80 °C.
1
°C; 14.0 g, 0.094 mol, 69%. H NMR: δ ϭ 1.25 (s, 9 H), 3.31 (br.,
exchangeable, 1 H), 6.6Ϫ6.7 (m, 3 H) 7.0Ϫ7.1 (m, 2 H) ppm] which
was stored under nitrogen.
N-Tritylacetamide (1h):[17Ϫ19] Sulfuric acid (96%, 0.2 mL) was ad-
ded to a stirred solution of trityl alcohol (0.84 g, 3.23 mmol) in
acetonitrile (10 mL). The clear red solution was heated gently for
1 h then allowed to cool and the stirring was continued for several
days, after which time water was added and the precipitated colour-
less solid (0.92 g, 3.06 mmol, 95%) was filtered at the pump and
washed with copious amounts of water; a sample for measurements Calculations: Conformational searches were carried out using the
was recrystallised from aqueous ethanol. M.p. 215.5Ϫ216 °C Spartan 02 program (Wavefunction Inc.) and several lowest-energy
(ref.[17Ϫ19] 206.5Ϫ207.2, 211, 218Ϫ219 °C). H NMR: δ ϭ 2.06 (s, conformers were further optimized at the RHF/3-21G* level of the-
3 H, Me), 6.68 (br, 1 H, NH), 7.2Ϫ7.4 (m, 15 H, arom) ppm. 13C
ory to identify each global minimum. Finally, the geometries were
NMR: δ ϭ 24.63 (Me), 70.57 (C-N), 127.00 (C4Ј), 127.91 (C3Ј,5Ј), fully optimized at the B3LYP/6-31G* level with normal conver-
128.66 (C2Ј,6Ј), 144.68 (C1Ј), 169.0 (CϭO) ppm. IR (KBr): ν˜ ϭ
gence using the Gaussian 98 program.[21] Vibrational normal-mode
3250 (br), 3200 (sh), 1655 (s), 1533 (m), 1491 (m), 1447 (m), 745 analyses were performed at the same level to ensure that each opti-
(m), 700 (s) cmϪ1
mized structure was a true minimum on the potential energy sur-
1
.
Eur. J. Org. Chem. 2004, 5031Ϫ5039
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5037

M. Canle L., I. Demirtas, A. Freire, H. Maskill, M. Mishima
FULL PAPER
[17]
[18]
[19]
[20]
J. A. Sanguigni, R. Levine, J. Med. Chem. 1964, 7, 573Ϫ574.
R. Fosse, Bull. Soc. Chim. Fr. 1931, 49, 159Ϫ173.
S. Dayagi, J. Israel Chem. 1968, 6, 477Ϫ483.
D. H. Aue, M. T. Bowers, Gas-Phase Ion Chemistry (Ed.: M.
T. Bowers), Academic Press, New York, 1979, Vol. 2, pp. 2Ϫ51.
M. Mishima, Mustanir, M. Fujio, Y. Tsuno, Bull. Chem. Soc.
Jpn. 1996, 69, 2009Ϫ2018.
Gaussian 98, Revision A.5, M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G.
Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Bu-
rant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin,
M.C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R.
Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J.
Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman,
J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J.
Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
C. Gonzalez, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-Gor-
don, E. S. Replogle, J. A. Pople, Gaussian, Inc., Pittsburgh
PA, 1998.
face and to calculate the thermal correction needed to obtain the
Gibbs free energies. The zero-point energies used for the thermal
correction were unscaled.
[21]
[22]
Acknowledgments
We thank Dr. M. H. Abraham for bringing the tribenzylamine ano-
maly to our attention, Dr. A. P. Henderson for the preparation of
TMTrN(Me)Bn, Dr. C. Laurence for a helpful exchange of infor-
mation, and the Spanish Ministry of Education and Science for a
visiting professorship at the University of A Corun˜a for HM.
[1]
B. Helferich, Adv. Carbohydr. Chem. 1948, 3, 79Ϫ111; J. F. W.
McOmie, Protective Groups in Organic Chemistry, Plenum
Press, London, New York, 1973; T. W. Greene, P. G. M. Wuts,
Protective Groups in Organic Synthesis, Wiley-Interscience, 3rd
edition, New York, 1999.
[2]
C. Bleasdale, B. T. Golding, W. H. Lee, H. Maskill, J. Risebor-
[23]
[24]
P. Monjoint, M. F. Ruasse, Tetrahedron Lett. 1984, 25,
3183Ϫ3186; J. T. Edsall, M. H. Blanchard, J. Am. Chem. Soc.
1933, 55, 2337Ϫ2353.
L. F. Blackwell, A. Fischer, I. J. Miller, R. D. Topsom, J.
Vaughan, J. Chem. Soc. 1964, 3588Ϫ3591.
ough, E. Smits, Chem. Commun. 1994, 93Ϫ94; J. Crugeiras, H.
Maskill, J. Chem. Soc., Perkin Trans. 2 1998, 1901Ϫ1907.
M. C. Lopez, J. C. Martinez, I. Demirtas, H. Maskill, E. Stix,
[3]
Org. React. 1997, 30, 71Ϫ77.
M. Canle L, W. Clegg, I. Demirtas, M. R. J. Elsegood, H.
[4]
[25]
[26]
R. W. Alder, Chem. Rev. 1989, 89, 1215Ϫ1223.
Maskill, J. Chem. Soc., Perkin Trans. 2 2000, 85Ϫ92.
M. Canle L., I. Demirtas, H. Maskill, J. Chem. Soc., Perkin
[5]
C. H. Rochester, B. Rossall, Trans. Faraday Soc. 1969, 65,
1004Ϫ1013; P. D. Bolton, C. H. Rochester, B. Rossall, Trans.
Faraday Soc. 1970, 66, 1348Ϫ1350.
A. Bagno, R. L. Boso, N. Ferrari, G. Scorrano, Eur. J. Org.
Chem. 1999, 1507Ϫ1515.
H. P. Hopkins, D. V. Jahagirdar, P. S. Moulik, D. H. Aue, H.
M. Webb, W. R. Davidson, M. D. Pedley, J. Am. Chem. Soc.
1984, 106, 4341Ϫ4348.
J. C. Day, J. Am. Chem. Soc. 1981, 103, 7355Ϫ7357.
S. Böhm, M. Decouson, O. Exner, J.-F. Gal, P.-C. Maria, J.
Org. Chem. 1994, 59, 8127Ϫ8131.
Trans. 2 2001, 1748Ϫ1752.
The investigation of solvation of dilute solutes remains a major
[6]
[27]
[28]
issue. See: Z. Cao, M. Lin, Q. Zhang, Y. Mo, J. Phys. Chem.
A 2004, 108, 4277Ϫ4282 for recent relevant computational
work and references to the earlier literature.
[7]
ϩ
pKBH ϭ 5.38 in water is reported for tribenzylamine by: S.
Bruckenstein, I. M. Kolthoff, J. Am. Chem. Soc. 1956, 78,
[29]
[30]
2974Ϫ2979, and 4.00 in 80:20 methylcellosolve/H2O: H. Dahn,
ˆ
J. C. Farine, T. T. T. Nguyen, Helv. Chim. Acta 1980, 63,
780Ϫ787.
[8]
[9]
[31]
J. J. Christensen, R. M. Izatt, D. P. Wrathall, L. D. Hansen, J.
Chem. Soc. (A) 1969, 1212Ϫ1223.
In principle, protonation of N-trityl-2-nitroaniline could occur
at the amino nitrogen or at an oxygen of the nitro group in the
gas phase. Our B3LYP/6-31G* calculations indicate that the
For relevant previous comparative work on base strengths in
the gas phase and solution, and the relationship between base
strength and molecular structure, see: H. K. Hall, J. Am. Chem.
Soc. 1957, 79, 5441Ϫ5444; F. E. Condon, J. Am. Chem. Soc.
1965, 87, 4481Ϫ4484; F. E. Condon, J. Am. Chem. Soc. 1965,
87, 4485Ϫ4490; D. H. Aue, H. M. Webb, W. R. Davidson, P.
Toure, H. P. Hopkins, S. P. Moulik, D. V. Jahagirdar, J. Am.
former is favoured by 23.0 kJ·molϪ1
.
[32]
R. Stewart, The Proton: Applications to Chemistry, Chapter 5,
Academic Press, London, 1985; H. M. Grant, P. McTigue, D.
G. Ward, Austr. J. Chem. 1983, 36, 2211Ϫ2218; G. Wado, T.
Takenaka, Bull. Chem. Soc. Jpn. 1971, 44, 2877; R. L. Adel-
man, J. Org. Chem. 1964, 29, 1837Ϫ1844; A. R. Goldfarb, A.
Mele, N. Gutstein, J. Am. Chem. Soc. 1955, 77, 6194Ϫ6196.
Handbook of Chemistry and Physics, The Chemical Rubber
Publishing Company, Cleveland, 1951.
G. N. Okafo, R. Brown, P. Camilleri, J. Chem. Soc., Chem.
Commun. 1991, 864Ϫ866; A. I. Biggs, R. A. Robinson, J.
Chem. Soc. 1961, 388Ϫ393.
A. Bagno, G. Lovato, G. Scorrano, J. Chem. Soc., Perkin Trans.
2 1993, 1091Ϫ1098.
´
Chem. Soc. 1991, 113, 1770Ϫ1780; I. Tun˜on, E. Silla, J. Tom-
asi, J. Phys. Chem. 1992, 96, 9043Ϫ9048 in addition to refer-
[33]
[34]
ences given below.
[10]
[11]
D. H. Aue, H. M. Webb, M. T. Bowers, J. Am. Chem. Soc.
1976, 98, 318Ϫ329.
A. Albert, E. P. Serjeant, The Determination of Ionization Con-
stants. A Laboratory Manual, 3rd ed., Chapman and Hall,
1984.
H. Maskill, The Physical Basis of Organic Chemistry, Oxford
University Press, Oxford, 1985.
[35]
[12]
[13]
[36]
[37]
H. Pracejus, Chem. Ber. 1959, 92, 988Ϫ998.
The possibility that N-tritylacetamide undergoes protonation
at nitrogen in aqueous acids is currently under investigation.
S. A. Attiga, C. H. Rochester, J. Chem. Soc., Perkin Trans. 2
1975, 1411Ϫ1114.
A Greenberg, D. T. Moore, D. D. DuBois, J. Am. Chem. Soc.
1996, 118, 8658Ϫ8668; A Greenberg, C. A. Venanzi, J. Am.
Chem. Soc. 1993, 115, 6951Ϫ6957.
H. K. Hall, A. El-Shekeil, Chem. Rev. 1983, 83, 549Ϫ555; H.
K. Hall, R. G. Shaw, A Deutschmann, J. Org. Chem. 1980,
45, 3722Ϫ3724.
A. J. Kirby, I. V. Komarov, P. D. Wothers, N. Feeder, Angew.
Chem. Int. Ed. 1998, 37, 785Ϫ786.
E. P. L. Hunter, S. G. Lias, J. Phys. Chem. Ref. Data 1998, 27,
413Ϫ457; E. P. Hunter, S. G. Lias in NIST Chemistry Web-
Book, NIST Standard Reference Database, 69, Ed. W. G. Mal-
lard, National Institute of Standards and Technology, Gai-
thersburg, MD, 2003 (http://webbook.nist.gov).
F. M. Jones, E. M. Arnett, Prog. Phys. Org. Chem. 1974, 11,
263Ϫ322; see also E. M. Arnett, Prog. Phys. Org. Chem. 1963,
1, 223Ϫ403.
B Wilson, R. Georiadis, J. E. Bartmess, J. Am. Chem. Soc.
1991, 113, 1762Ϫ1769.
J. F. Bunnett, R. G. Scamehorn, R. P. Traber, J. Org. Chem.
1976, 41, 3677Ϫ3682.
[38]
[39]
[14]
[40]
[41]
[15]
[16]
5038
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 5031Ϫ5039

Base Strengths of Substituted Amines
FULL PAPER
[42]
[45]
J. Graton, F. Besseau, M. Berthelot, E. D. Raczynska, C. Laur-
A. G. Marshall, T.-C. L. Wang, T. L. Ricca, J. Am. Chem. Soc.
1985, 107, 7893Ϫ7897; R. B. Cody, R. E. Hein, S. D. Good-
man, Rapid Commun. Mass Spectrom. 1987, 1, 99Ϫ102; S.
Guan, A. G. Marshall, Int. J. Mass Spectrom. Ion Process 1997,
157/158, 5Ϫ37.
ence, Can. J. Chem. 2002, 80, 1375Ϫ1385.
[43]
´
A. Castro, P. Sanchez, J. A. Santaballa, Bull. Soc. Chim. Fr.
1988, 364Ϫ366.
[44]
J. E. Bartmess, R. M. Georgiadis, Vacuum 1983, 33, 149Ϫ153;
K. J. Miller, J. Am. Chem. Soc. 1990, 112, 8533Ϫ8542.
Received July 15, 2004
Eur. J. Org. Chem. 2004, 5031Ϫ5039
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5039