Dramatic acceleration of the Menschutkin reaction and distortion of halide leaving-group order

Article abstract of DOI:10.1021/jo702090p

(Chemical Equation Presented) Compared to structurally related linear trialkylamines, a simple macrocyclic amine with an anion-binding cavity exhibits very large rate enhancements (>10⁵) for stoichiometric N-alkylation with primary alkyl, allyl

Full text of DOI:10.1021/jo702090p

Dramatic Acceleration of the Menschutkin Reaction and Distortion
of Halide Leaving-Group Order
Keith J. Stanger, Jung-Jae Lee, and Bradley D. Smith*
Department of Chemistry and Biochemistry and Walther Cancer Research Center,
UniVersity of Notre Dame, Notre Dame, Indiana 46556
smith.115@nd.edu
ReceiVed September 25, 2007
Compared to structurally related linear trialkylamines, a simple macrocyclic amine with an anion-binding
cavity exhibits very large rate enhancements (>105) for stoichiometric N-alkylation with primary alkyl,
allyl, and benzyl halides in the weakly polar solvent CDCl3. There is also a major distortion of the halide
leaving-group order. For example, with benzyl halides the relative leaving-group order with a control
amine is Cl (1) < Br (71) < I (160), whereas the leaving-group order with the macrocyclic amine is I
(0.4) < Cl (1) < Br (8.5). Reaction with the macrocyclic amine is inhibited by the addition of DMSO,
which is unusual because the Menschutkin reaction is normally enhanced by the presence of a polar
aprotic solvent. Competitive inhibition studies indicate that the reaction proceeds through a prereaction
complex. Effective molarities for the subsequent unimolecular N-alkylation step with 4-t-butylbenzyl
halides are 4-t-BuBnCl (62 000 M) > 4-t-BuBnBr (2200 M) > 4-t-BuBnI (35 M); thus, the free energy
of activation is selectively decreased for organohalides having smaller and more charge dense leaving
groups. Likely reasons for this selective enhancement effect are: (a) increased transition-state stabilization
due to hydrogen bonding in the macrocyclic pocket and (b) reduced entropic penalty in the transition
state due to an increased fraction of prereaction complexes that are oriented in a near attack conformation.
The study suggests that it should be possible to develop highly reactive macrocyclic amines that selectively
sense or scavenge carcinogenic haloalkanes from the atmosphere.
Introduction
undergo an SN2 reaction to produce an ion pair, a process that
is accelerated by polar aprotic solvents, increased pressure,
elevated temperature, and increased leaving-group ability.4-6
The literature contains a handful of structured molecular
systems that exhibit an accelerated Menschutkin reaction.7-9
The quaternization of a trialkylamine is known as the
Menschutkin reaction.1 The reaction is formally an alkyl group
transfer, a process of central importance in biochemistry and a
widely used strategy in organic synthesis. Numerous physical
organic studies have been conducted on the Menschutkin
reaction,2 and it is a useful test system for theoretical methods.3
To briefly summarize the basic features, two neutral molecules
(3) Castejon, H.; Wiberg, K. B. J. Am. Chem. Soc. 1999, 121, 2139-
2146.
(4) Wang, T.; Iou, Q. Chem. Eng. J. 2002, 87, 197-206.
(5) Abraham, M. H. In Progress in Physical Organic Chemistry;
Streitwieser, A., Taft, R. W., Eds.; Wiley: New York, 1974; Vol. 11, pp
1-87.
(6) Taft, R. W.; Pienta, N. J.; Kamlet, M. J.; Arnett, E. M. J. Org. Chem.
1981, 46, 661-667.
(7) Purse, B. W.; Gissot, A.; Rebek, J., Jr. J. Am. Chem. Soc. 2005, 127,
11222-11223.
(1) Menschutkin, N. Z. Phys. Chem. 1890, 6, 41-57. Note that the
chemistry literature contains a substantial number of citations with the
German spelling Menschutkin, and also the English spelling Menshutkin.
(2) Abboud, J. M.; Notario, R.; Berra´n, J.; Sola, M. H. In Progress in
Physical Organic Chemistry; Taft, R. W., Ed.; Wiley: New York, 1993;
Vol. 19, pp 1-182.
10.1021/jo702090p CCC: $37.00 © 2007 American Chemical Society
Published on Web 11/03/2007
J. Org. Chem. 2007, 72, 9663-9668
9663

Stanger et al.
to product precipitation and product inhibition (i.e., noncovalent
association of the ion-pair product with the starting amine).12
Each reaction system exhibited bimolecular kinetics, and
pseudo-first-order kinetics was observed when the electrophile
was present in large excess. Thus, the reactions are first-order
in nucleophilic amine and first-order in electrophilic organo-
halide.
Here, we describe a relatively simple bifunctional macrocycle,
1, that produces very large rate enhancements. Recently, we
discovered that 1 is N-alkylated by methylene chloride solvent
with a half-life of 2 min at 298 K, a rate that is about 50 000
times faster than that observed with structurally related, linear
amines.10,11 We have now measured reaction rates with other
organohalides, and we find that macrocycle 1 reacts very rapidly
with primary alkyl, allyl, and benzyl halides (i.e., RCH2X).
Furthermore, when compared to control amines 2-4 there is a
major distortion of the halide leaving-group order. Although
this nucleophilic substitution reaction is stoichiometric, the
unprecedented reactivity of 1 is attributed to its hydrogen-
bonding cavity, which produces some enzyme-like properties
such as the formation of a prereaction complex and selective
rate enhancement for organohalide substrates having smaller
and more charge dense leaving groups. The small size of the
reactants and the fact that the product is an internally solvated
contact ion pair in a weakly polar solvent make it an experi-
mentally and theoretically tractable bimolecular model for
testing concepts in organic and enzyme catalysis.
In the Supporting Information (Table S1) is a listing of all
the second-order rate constants that were measured in this study.
As expected for an SN2 process, the electrophile reactivity trend
was 4-t-BuBn halide > allyl halide >1-propyl halide.13 Control
amines 3 and 4 have very similar steric and electronic environ-
ments as macrocycle 1, but they lack a hydrogen-bonding
pocket. They are relatively unreactive at 298 K, and measurable
rates could only be obtained with 4-t-BuBn halide as the
electrophile. Therefore, the more reactive quinuclidine, 2, was
employed as the control amine with allyl and 1-propyl halides.
Tables 1 and 2 show a selection of kinetic data that illustrate
the major points. The most obvious result is the extremely high
reactivity of macrocycle 1. For example, reaction of 4-t-BuBnCl
with macrocycle 1 is 160 000 times faster than that with control
amine 3; smaller rate enhancements are observed with 4-t-
BuBnBr (12 000) and 4-t-BuBnI (220). Listed in Table 2 are
leaving-group propensities, as measured by the ratio of rate
constants with the different organohalides. The control amines
2-4 exhibit typical nucleophilic reactivity, with normal halide
leaving-group orders. An example is the reactivity of amine 2;
with 4-t-BuBn halides, the relative leaving-group order is Cl
(1) < Br (71) < I (160); with allyl halides, the relative leaving-
group order is Cl (1) < Br (41) < I (110), and with 1-propyl
halides, the relative leaving-group order is Cl (1) < Br (84) <
I (140). In contrast, the leaving-group order with macrocycle 1
is very different. For example, with 4-t-BuBn halides the
leaving-group order I (0.4) < Cl (1) < Br (8.5), which is similar
to the order obtained when 1 reacts with allyl halides and
1-propyl halides. Inspection of the rate data shows that mac-
rocycle 1 alters the inherent halide leaving-group order by
selectively improving the leaving-group ability of Cl > Br > I
(see, for example, the rate constant ratio k(1)/k(3) in Table 1).14,15
Although leaving-group tendencies in an SN2 reaction are
known to change with nucleophilic strength (which alters the
structure of the transition state16), the large differences in
Results and Discussion
Kinetic Studies. The reaction kinetics with 1-propyl, allyl,
and 4-tert-butylbenzyl (4-t-BuBn) halides were monitored by
1H NMR spectroscopy in CDCl3 at 298 K. In each case, the
corresponding quaternary ammonium salt was the only product
formed, as determined by standard spectroscopic methods and
X-ray crystallography. The starting amine concentrations were
kept below 10 mM, which avoided kinetic complications due
(12) The alkylation products with unsubstituted benzyl halides tended
to precipitate very quickly during the reaction; therefore, the more soluble
4-tert-butylbenzyl halides were employed instead.
(13) The calculational and kinetic isotope data reported in ref 10 are
consistent with a classical SN2 process, but with some of the organohalides
we cannot rule out the possibility that the mechanism may involve two
SET steps.
(14) Another example of selective improvement of leaving-group ability
in the order Cl > Br > I is the ratio k(1)/k(2). With 4-t-BuBn halides, the
values of k(1)/k(2) are 4-t-BuBnCl (55), 4-t-BuBnBr (6.5), and 4-t-BuBnI
(0.15). In other words, 4-t-BuBnCl reacts with macrocycle 1 much faster
than with 2, whereas the ratio is reversed with 4-t-BuBnI.
(15) The leaving-group ability of fluoride in SN2 reactions is about 200
times less than that of chloride. (McMurry, J. Organic Chemistry; 6th ed.,
Thomson-Brooks/Cole: Belmont, CA, 2004). The main reason is the high
strength of the C-F bond (e.g., bond strengths are CH3-F 108 kcal/mol,
CH3-Cl 84 kcal/mol, CH3-Br 70 kcal/mol, and CH3-I 56 kcal/mol).
t-BuBnF reacts with macrocycle 1 in CDCl3 with a rate constant of (5.63
( 0.20) × 10-5 M-1 s-1 at 298 K. In contrast, no reaction with 2 could be
detected at 298 K. To provide a point of comparison, the experiments with
t-BuBnF were repeated at 323 K, where krel(1)/krel(2) ) 100, which means
that the rate enhancement for fluoride leaving group is greater than that for
chloride (100- vs 55-fold).
(8) (a) Heemstra, J. M.; Moore, J. S. J. Am. Chem. Soc. 2004, 126, 1648-
1649. (b) Heemstra, J. M.; Moore, J. S. J. Org. Chem. 2004, 69, 9234-
9237. (c) Smaldone, R. A.; Moore, J. S. J. Am. Chem. Soc. 2007, 129,
5444-5450.
(9) McCurdy, A.; Jimenez, L.; Stauffer, D. A.; Dougherty, D. A. J. Am.
Chem. Soc. 1992, 114, 10314-10321.
(10) Lee, J.-J.; Stanger, K. J.; Noll, B. C.; Gonzalez, C.; Marquez, M.;
Smith, B. D. J. Am. Chem. Soc. 2005, 127, 4184-4185.
(11) For examples of N-alkylation with methylene chloride, see: (a)
Nevstad, G. O.; Songstad, J. Acta Chem. Scand. B 1984, 38, 469-477. (b)
Mills, J. E.; Maryanoff, C. A.; Cosgrove, R. M.; Scott, L.; McComsey, D.
F. Org. Prep. Proced. Int. 1984, 16, 97-114. (c) Tal, S.; Salman, H.;
Abraham, Y.; Botoshansky, M.; Eichen, Y. Chem.-Eur. J. 2006, 12, 4858-
4862. (d) Ernst, R. D.; Harvey, B. G.; Oteri, O. J.; Arif, A. M. Z. Kristallogr.
2005, 220, 373-375.
(16) Lowry, T. H.; Richardson, K. Mechanism and Theory in Organic
Chemistry, 3rd ed.; Harper: Cambridge, 1987; p 374.
9664 J. Org. Chem., Vol. 72, No. 25, 2007

Menschutkin Reaction and Halide LeaVing-Group Order
TABLE 1. Second-Order Rate Constants (k) and Free Energies of Activation (∆Gq) for the Reaction of Macrocycle 1 and Control Amine 3
with 4-tert-Butylbenzyl Halidesa
q
q
k(1)
∆G1
k(3)
∆G3
∆∆Gq
(kcal/mol)
4-t-BuBnhalide
(M-1 s-1
)
(kcal/mol)
(M-1 s-1
)
(kcal/mol)
k(1)/k(3)
4-t-BuBnCl
4-t-BuBnBr
4-t-BuBnI
1.02 ( 0.12
8.70 ( 0.43
0.44 ( 0.03
17.4
16.2
17.9
(6.58 ( 0.62) × 10-6
(7.35 ( 0.92) × 10-4
(1.98 ( 0.25) × 10-3
24.5
21.7
21.1
160 000
12 000
220
7.1
5.5
3.2
a In CDCl3 at 298 K.
FIGURE 1. Proposed mechanism for reaction of RCH2X with macrocycle 1.
TABLE 2. Second-Order Rate Constants (k) and Leaving-Group
Propensitiesa
k for
organochloride
ratio of rate constants
Cl/Cl:Br/Cl:I/Cl
amine
organohalide
(M-1 s-1
)
1
1
1
2
2
2
3
4
4-t-BuBn halides 1.02 ( 0.12
1.0/8.5/0.4
1.0/8.4/2.5
1.0/4.5/2.9
1.0/71/160
1.0/41/110
1.0/84/140
1.0/110/300
1.0/73/b
allyl halides
(9.79 ( 0.12) × 10-2
1-propyl halides (4.20 ( 0.73) × 10-5
4-t-BuBn halides (1.87 ( 0.48) × 10-2
allyl halides
(5.70 ( 0.87) × 10-3
1-propyl halides (6.40 ( 0.09) × 10-6
4-t-BuBn halides (6.58 ( 0.62) × 10-6
4-t-BuBn halides (1.95 ( 0.62) × 10-6
a In CDCl3 at 298 K. b Not measured.
leaving-group order cannot be attributed solely to altered
nucleophilicities, because each nucleophile is an sp3 hybrid-
ized trialkylamine (this assumption is verified by other studies
in the literature4). A more likely explanation for the observed
change in reactivity is that the binding pocket in macrocycle 1
accelerates the reaction and distorts the halide leaving-
group order by the mechanism shown in Figure 1. A central
feature with this mechanism is association of the electrophilic
organohalide with the macrocycle to form a prereaction
complex. Evidence for this process is provided in the next
section.
Evidence for a Prereaction Complex. The mechanism in
Figure 1 involves a prereaction complex and therefore raises
the possibility of saturation kinetics at high concentrations of
electrophile. However, this effect was not observed with
macrocycle 1. For example, identical second-order rate constants
were obtained for reaction of 1 (5.0 mM) with various
concentrations of allyl chloride (60-500 mM). The fact that
the rate does not plateau can be explained if the organohalide/1
association constant is very low, which means that 1 is not
saturated even when the concentration of electrophile is as high
as experimentally possible. Thus, in this concentration regime,
the involvement of a prereaction complex needs to be inferred
by other methods,17 such as a demonstration of competitive
FIGURE 2. Effect of added DMSO inhibitor on pseudo-first-order
rate constant for the reaction of macrocycle 1 (5 mM) with 4-t-BuBnCl
(200 mM) in CDCl3 at 298 K. Inset: Reaction of control 3 (5 mM)
and 4-t-BuBnCl (200 mM) in CDCl3 with increasing amounts of
DMSO.
inhibition kinetics. This situation was recently encountered in
related study by Heemstra and Moore,8b and we have employed
a similar inhibition strategy to prove that macrocycle 1 forms
a prereaction complex. In our case, the inhibitor is DMSO. As
shown in Figure 2, the rate constant for the reaction of
macrocycle 1 with 4-t-BuBnCl in CDCl3 decreases with
increasing amounts of added DMSO. This DMSO inhibition
effect is unusual because the Menschutkin reaction is normally
enhanced by the presence of a polar aprotic solvent.4 For
example, the reaction of 4-t-BuBnCl with control amine 3 is
21 times faster in DMSO than in CDCl3 (inset in Figure 2).
However, DMSO forms a 1:1 hydrogen-bonded complex with
macrocycle 1 (observed in the solid and solution states18), and
the association constant for 1‚DMSO is 160 M-1 in CDCl3 at
298 K.10 Thus, DMSO is an inhibitor that competes for the
binding pocket in macrocycle 1, and a computer-based nonlinear
(18) An X-ray crystal structure of 1 as its DMSO adduct (see the
Supporting Information in ref 10) shows that the DMSO oxygen atom forms
hydrogen bonds with the two NH residues in the macrocyclic cavity.
(17) Prereaction complexes can also be identified using non-steady state
kinetic analysis. Parker, V. D. Pure Appl. Chem. 2005, 77, 1823-1833.
J. Org. Chem, Vol. 72, No. 25, 2007 9665

Stanger et al.
TABLE 3. Kinetic Data from DMSO Inhibition Studiesa
b
c
d
Ka
(M-1
ku
kin
(M-1 s-1
k(3)e
EMf
(M)
organohalide
)
(s-1
)
)
(M-1 s-1
)
4-t-BuBnCl
4-t-BuBnBr
4-t-BuBnI
allyl Cl
allyl Br
allyl I
3.3 ( 0.1
7.5 ( 1.4
17.3 ( 1.0
1.3 ( 0.1
6.4 ( 1.2
19.0 ( 1.0
(4.1 ( 0.4) × 10-1
1.6 ( 0.3
(1.7 ( 0.5) × 10-2
(9.1 ( 2.3) × 10-2
(1.4 ( 0.4) × 10-2
(1.4 ( 0.4) × 10-3
(2.8 ( 0.1) × 10-2
(2.3 ( 0.7) × 10-3
(6.6 ( 0.6) × 10-6
(7.4 ( 0.9) × 10-4
(2.0 ( 0.3) × 10-3
62 000
2200
35
(7.0 ( 0.5) × 10-2
(8.3 ( 0.6) × 10-2
(2.1 ( 0.4) × 10-1
(1.4 ( 0.1) × 10-2
a In CDCl3 at 298 K. b 1/Organohalide association constant. c Unimolecular rate constant for the prereaction complex. d Second-order rate constant for
inhibited 1/DMSO complex. e Second-order rate constant for reaction of organohalide with control 3. f Effective molarity or ku/k(3).
weak, which explains the absence of saturation kinetics. In
addition, the trend of Ka is 4-t-BuBnCl (3.3 M-1) < 4-t-BuBnBr
(7.5 M-1) < 4-t-BuBnI (17.3 M-1) and allyl Cl (1.3 M-1) <
allyl Br (6.4 M-1) < allyl I (19.0 M-1). In other words,
macrocycle 1 associates with organoiodides the strongest and
with organochlorides the weakest, which is opposite to the trend
for relative rate enhancement (i.e., Cl > Br > I). Thus, the
source of the selective acceleration is not the initial bimolecular
association, but instead the subsequent unimolecular N-alkyla-
tion step. The data in Table 3 can be used to construct
quantitative reaction coordinate diagrams. For example, Figure
3 shows the reactions of t-BuBnCl with macrocycle 1 and
control amine 3. The top pathway represents the SN2 reaction
with 3 and its single transition state (TS3), whereas macrocycle
1 proceeds along the bottom pathway, through a prereaction
complex, before crossing the transition state (TS1). The observed
acceleration of 1 over 3 (∆∆Gq ) 7.1 kcal mol-1) is the
difference between TS3 and TS1.
Also provided in Table 3 are second-order rate constants k(3)
for control amine 3 and the calculated values of effective
molarity (EM). EM is a measure of the enhancement that is
gained when a bimolecular reaction is compared to an “in-
tramolecular” process that uses the same mechanism.19 Dividing
the unimolecular rate constant, ku, by the control amine second-
order rate constant, k(3), gives the following values of EM for
reaction with 4-t-BuBn halides: 4-t-BuBnCl (62 000 M) > 4-t-
BuBnBr (2200 M) > 4-t-BuBnI (35 M). Thus, the hydrogen-
bonding macrocycle in 1 decreases ∆Gq for the unimolecular
alkylation step in the order Cl > Br > I.20
FIGURE 3. Reaction coordinate diagram for the reaction of 4-t-
BuBnCl with macrocycle 1 or control amine 3.
least-squares method was used to fit the curve in Figure 2 with
the following reaction scheme for competitive inhibition. In
this scheme, Ka is the association constant for formation of a
1:1 complex of 1 with the electrophilic organohalide (prereaction
complex in Figure 1), ku is the unimolecular rate constant for
conversion of this complex into N-alkylated product 5, and
kin is the second-order rate constant for reaction of the in-
hibited complex (1‚DMSO, Ki ) 160 M-1) with the organo-
halide.
How is ∆Guq selectively decreased with organohalides having
smaller and more charge dense leaving groups? Elucidation of
the fundamental factors that produce enzymatic catalysis remains
a major intellectual topic in bioorganic chemistry.21 Most likely,
the major contribution is transition-state stabilization. Even
though the solvent is weakly polar (CDCl3), the macrocycle
provides its own preorganized, polar solvation sphere,22 and the
energy of the dipolar transition state is lowered by stabilizing
hydrogen bonds between the leaving halide and the macrocycle’s
two NH residues (Figure 4).23 Furthermore, the strength of the
In addition to the 4-t-BuBnCl curve in Figure 2, DMSO
inhibition studies were conducted with the other two 4-t-BuBn
halides and also the three allyl halides (Supporting Information).
The kinetic variables, extracted by curve-fitting, are listed in
Table 3. The data trends for the two electrophilic systems are
very similar, and taken together they present compelling
evidence for a prereaction complex.
Reaction Mechanism and Source of Acceleration. The
results of this study support the mechanism proposed in
Figure 1; that is, in the weakly polar solvent, CDCl3, the
electrophilic organohalide associates with macrocycle 1 to
form a prereaction complex, which then undergoes unimole-
cular nucleophilic attack to give N-alkylated product, 5, as a
stabilized contact ion pair. The rate enhancement is clearly due
to the ability of the hydrogen-bonding macrocycle in 1 to lower
the free energy of activation. The kinetic data in Table 3 provide
additional detail for the enhanced reactivity and altered leaving-
group order. The organohalide affinities for 1, Ka, are all quite
(19) Kirby, A. J. Angew. Chem., Int. Ed. Engl. 1996, 35, 707-724.
(20) A similar but less dramatic correlation is seen when comparing
leaving-group abilities for the Menschutkin reaction in aprotic and protic
solvents. For example, moving from aprotic DMF to protic methanol
enhances leaving-group ability in the order Cl > Br > I. Parker, A. J. Chem.
ReV. 1969, 69, 1-32.
(21) Gao, J.; Ma, S.; Major, D. T.; Nam, K.; Pu, J.; Truhlar, D. G. Chem.
ReV. 2006, 106, 3188-3209.
(22) The transition state of a typical Menschutkin reaction at ambient
temperature in moderately polar solvents is characterized by T∆S‡ values
of -9 to -11 kcal/mol (see ref 2). The highly negative ∆S‡ is attributed to
ordering of the solvent cage and a negative volume of activation, which
explains why the reaction is promoted by high pressure.
9666 J. Org. Chem., Vol. 72, No. 25, 2007

Menschutkin Reaction and Halide LeaVing-Group Order
Summary
While the literature contains many examples of bifunctional
metal coordination complexes that can activate and attack
phosphate and carboxylate esters,27 there are very few uncharged
organic molecules that exhibit such dramatic reaction enhance-
ments (>105) as macrocyclic amine 1.28 Although stoichiomet-
ric, the Menschutkin reaction with 1 exhibits several kinetic
features that are reminiscent of enzymes, namely, formation of
a prereaction complex and substrate selectivity. There is a major
distortion of the halide leaving-group order because of selective
improvement of leaving-group ability in the order Cl > Br >
I, and the EM for reaction with 4-t-BuBn halides is 4-t-BuBnCl
(62 000 M) > 4-t-BuBnBr (2200 M) > 4-t-BuBnI (35 M).
Likely reasons for the selective enhancement are: (a) increased
transition-state stabilization due to hydrogen bonding in the
macrocyclic pocket and (b) reduced entropic penalty in the
transition state due to an increased fraction of prereaction
complexes that are oriented in a near attack conformation. The
small molecular size and structural simplicity of this bimolecular
reaction system means that it should be possible to devise
experimental and theoretical studies that determine the relative
contribution of these fundamentally important reaction accelera-
tion factors. A more practical goal is to develop highly reactive
macrocyclic amines that selectively sense and scavenge carci-
nogenic haloalkanes from the atmosphere.
FIGURE 4. Structures of NAC and TS.
hydrogen bonds increases in the order Cl- > Br- > I-.24 There
also may be differences in the entropy of activation. For
example, the near attack conformation (NAC) hypothesis states
that all enzyme-catalyzed reactions must proceed through a
ground state NAC that approximates the structure of the
transition state, but has not yet started bond forming and
shortening.25 Although formation of a prereaction complex (1‚
organohalide) is favored in the halide order of I > Br > Cl
(see values for Ka in Table 3), it is possible that the fraction of
prereaction complexes that adopt an NAC is higher when the
organohalide has a smaller and more polar C-X bond (i.e., Cl
> Br > I). Evidence for this statement are the results of high-
level molecular modeling, which indicate that upon binding to
macrocycle 1, the organohalide is oriented by weak intermo-
lecular interactions between its halogen atom and the macro-
cycle’s NH residues, and also its CH2 residues and the
macrocycle’s ether oxygens (Figure 4).10,26 This would favor
an NAC where the macrocycle’s tertiary nitrogen is poised to
attack the electrophilic CH2 with a classic SN2 trajectory. The
relative contributions of these energy components to the overall
value of ∆Guq cannot be determined from the experimental data,
but they can be accessed with modern molecular dynamics
simulations,25 an exercise that is beyond the scope of this current
experimental study.
Experimental Section
Materials. The synthesis and characterization of compounds 1,
3, and 4 is described in the Supporting Information. Compound 2
was purchased from a commercial supplier and used as supplied.
Kinetic Measurements. A 5-mm NMR tube, containing a
solution of macrocycle 1 in CDCl3 (5 mM, 750 µL), was placed in
an NMR spectrometer (500 MHz) and allowed to reach thermal
equilibrium (298.0 ( 0.1 K). In certain cases, the solution also
contained a specific concentration of DMSO inhibitor. After being
shimmed, a starting spectrum was acquired, a very small aliquot
of organohalide electrophile was added, and the reaction was
monitored by periodic acquisition of a spectrum. The changes in
peak intensity were monitored over time. Usually, the NH and
N-methyl peaks of starting material (1) and product were well
resolved, but changes in other peaks were also measured if possible.
The concentrations of starting material and product at any time
were calculated from the ratio of peak integrations. For relatively
fast reactions (>95% complete within 4 h), the NMR tube remained
in the NMR spectrometer throughout the run. For slower reactions,
the NMR tube was stored at 298 K in an incubator oven and
removed for periodic NMR acquisition. The curve fitting and
extraction of kinetic data are described in the Supporting Informa-
tion.
(23) Activation of a halide leaving group by hydrogen bonding is a
strategy that is employed by the family of enzymes known as haloalkane
dehydrogenases. The first catalytic step in the enzymatic process involves
attack of an organohalide by an active-site carboxylate nucleophile, and a
significant fraction of the rate enhancement is attributed to two NH residues
in the enzyme active site that form stabilizing hydrogen bonds with the
halide leaving group. Dev-Kesavan, L.; Gao, J. J. Am. Chem. Soc. 2003,
125, 1532-1540.
(24) The halide association constants for macrocycle 1 in CDCl3
at 298 K are F- (1900 ( 400 M-1) > Cl- (620 ( 50 M-1) > Br-
(140 ( 30 M-1) > I- (90 ( 15 M-1), as determined by standard
NMR titration experiments using tetrabutylammonium halides. The
interaction is even stronger after the ion-pair product, 5, is formed.
Anion Association with Macrocycle 1. A 10 mM solution of
macrocycle 1 in CDCl3 was prepared in a 5-mm NMR tube (solution
volume 750 µL). Small aliquots of tetrabutylammonium halide stock
solution (0.75 M) were added, and an 1H NMR spectrum was
acquired after each addition. Care was taken to avoid water
absorption from the atmosphere. The changes in NH chemical
Consider the salts obtained by N-alkylation of
1 with 4-t-BuBn
halides; the ion-pair association constants in highly competitive
DMSO-d6 at 298 K are Cl- (1360 ( 280 M-1) > Br- (136 (
32 M-1) > I- (<10 M-1) as determined by NMR dilution experi-
ments.
(25) For discussions of near attack conformation, see: (a) Hur, S.;
Kahn, K.; Bruice, T. C. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
2215-2219. (b) Schowen, R. L. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
1191-11932. (c) Hur, S.; Bruice, T. C. J. Am. Chem. Soc. 2003,
125, 10540-10542. (d) Shurki, A.; Strajbl, M.; Villa, J.; Warshel, A. J.
Am. Chem. Soc. 2002, 124, 4097-4107. (e) Zhang, X.; Zhang, X.;
Bruice, T. C. Biochemistry 2005, 44, 10443-10448. (f) Guo, H.; Cui, Q.;
Lipscomb, W. N.; Karplus, M. Angew. Chem., Int. Ed. 2003, 42, 1508-
1511. (g) Ranaghan, K. E.; Mulholland, A. J. Chem. Commun. 2004, 1238-
1239.
(27) (a) Bazzicalupi, C.; Bencini, A.; Bianchi, A.; Danesi, A.;
Giorgi, C.; Lodeiro, C.; Pina, F.; Santarelli, S.; Valtancoli, B. Chem.
Commun. 2005, 2630-2632. (b) Garcia-Espana, E.; Gavina, P.; Latorre,
J.; Soriano, C.; Verdejo, B. J. Am. Chem. Soc. 2004, 126, 5082-5083. (c)
O’Neil, E. J.; Smith, B. D. Coord. Chem. ReV. 2006, 250, 3068-3080.
(28) (a) Artificial Enzymes; Breslow, R., Ed.; Wiley: New York, 2005.
(b) Cacciapaglia, R.; Di Stefano, S.; Mandolini, L. Acc. Chem. Res. 2004,
37, 113-122. (c) Motherwell, W. B.; Bingham, M. J.; Six, Y. Tetrahedron
2001, 57, 4663-4686. (d) Murakami, Y.; Kikuchi, J.; Hisaeda, Y.;
Hayashida, O. Chem. ReV. 1996, 96, 721-758.
(26) Haeffner, F.; Marquez, M.; Gonzalez, C. J. Phys. Chem. A 2007,
111, 268-272.
J. Org. Chem, Vol. 72, No. 25, 2007 9667

Stanger et al.
Ka ) R/(1 - R)2[c]
shift were used to generate titration isotherms that were fitted
to a 1:1 binding model using an iterative curve-fitting method.29
In each case, the titration was repeated an average of three
times.
Ion-Pair Association of the N-Alkylation Product of Macro-
cycle 1. General method: A stock solution of the N-alkylation
product of 1 in DMSO-d6 was prepared, and serial dilutions were
made to cover the concentration range from 0.02 to 0.0002 M. The
changes in NH chemical shift were used to generate titration
isotherms that were fitted to the following equation:29
where R ) (δ - δ0)/(δmax - δ0), δ0 is the initial chemical shift, δ
is the chemical shift at each titration point, δmax is the saturated
chemical shift, and [c] is concentration of N-alkylation product.
Acknowledgment. Financial support for this work was
provided by the ACS-Petroleum Research Fund and the Walther
Cancer Institute.
Supporting Information Available: Synthetic methods and
spectral data, kinetic data, and curve fittings. This material is
available free of charge via the Internet at http://pubs.acs.org.
(29) Kelly, T. R.; Kim, M. H. J. Am. Chem. Soc. 1994, 116, 7072-
7080.
JO702090P
9668 J. Org. Chem., Vol. 72, No. 25, 2007