Preorganized Metallomacrocycles: Improved Binding and Selectivity of NH3 over Primary Amines

Article abstract of DOI:10.1021/jo971943q

Three new metallohosts (2-4) were synthesized by the uranyl-templated macrocyclization of the appropriate dialdehydes 13a-c and 1,2-phenylenediamine in methanol. Cyclization of the dialdehyde 13b led to two isomers (3a, major, and 3b, minor), which only differed in the orientation of the isopropyl substituents on the outer phenolic rings of the terphenyl moiety. The binding constants in CDCl₃ of ammonia and benzylamine were determined by ¹H NMR spectroscopy. The binding of ammonia by these hosts follows 1 ≈ 2 ≈ 4 < 3b < 3a. The binding of benzylamine follows 1 ≈ 2 ≈ 4 ≈ 3b ≈ 3a. Monte Carlo free-energy perturbation calculations reproduced the binding of ammonia and n-propylamine quite well. The much larger binding of benzylamine by host 3a was rationalized by a much deeper positioning of the guest in the cavity of the host compared to 3b.

Full text of DOI:10.1021/jo971943q

3554
J . Org. Chem. 1998, 63, 3554-3559
P r eor ga n ized Meta llom a cr ocycles: Im p r oved Bin d in g a n d
Selectivity of NH3 over P r im a r y Am in es
Frank C. J . M. van Veggel,*,† Hettie G. Noorlander-Bunt,† William L. J orgensen,‡ and
David N. Reinhoudt†
Laboratories of Supramolecular Chemistry and Technology and MESA Research Institute, Faculty of
Chemical Technology, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands, and
Department of Chemistry, Yale University, P.O. Box 208107, New Haven, Connecticut 06520-8107
Received October 21, 1997
Three new metallohosts (2-4) were synthesized by the uranyl-templated macrocyclization of the
appropriate dialdehydes 13a -c and 1,2-phenylenediamine in methanol. Cyclization of the
dialdehyde 13b led to two isomers (3a , major, and 3b, minor), which only differed in the orientation
of the isopropyl substituents on the outer phenolic rings of the terphenyl moiety. The binding
1
constants in CDCl3 of ammonia and benzylamine were determined by H NMR spectroscopy. The
binding of ammonia by these hosts follows 1 ≈ 2 ≈ 4 < 3b < 3a . The binding of benzylamine
follows 1 ≈ 2 ≈ 4 ≈ 3b , 3a . Monte Carlo free-energy perturbation calculations reproduced the
binding of ammonia and n-propylamine quite well. The much larger binding of benzylamine by
host 3a was rationalized by a much deeper positioning of the guest in the cavity of the host compared
to 3b.
Ch a r t 1
In tr od u ction
In a previous article1 we have shown that the preor-
ganized metallomacrocycle 1 (Chart 1) binds ammonia
in chloroform with a higher association constant than
n-propylamine and benzylamine. The maximum selec-
tivity (KNH /KRNH ) obtained was 33. The binding of
3
2
ammonia or a primary amine is actually a substitution
reaction of a water that is bound in the cavity. The
presented molecular mechanics calculations suggested
that host 1 did not use all available binding sites of the
ammonia. Beside the binding of the lone pair of ammonia
to the uranyl cation, only two of the hydrogens were used.
This observation has stimulated us to improve the
binding and selectivity of ammonia over primary amines
by two ways. The first was the introduction of bulkier
substituents on the outer phenol moieties of the terphenyl
unit. The inner anisole moiety was not changed because
the above-mentioned calculations showed that this ac-
ceptor site was not used. The second approach was the
introduction of an extra binding site by bridging the outer
phenol groups with a diethylene glycol chain.
the inner phenol moiety according to a literature proce-
dure (Scheme 1).3
Formylation of 6 was performed with hexamethylene-
tetraamine in CF3COOH at 90 °C.1 The remaining outer
phenol moieties were alkylated with EtI, i-PrI, and
diethylene glycol ditosylate, respectively, with K2CO3 as
a base in CH3CN. The formyl groups were subsequently
reduced with NaBH4, followed by conversion of the
hydroxymethyl groups into bromomethyl groups with
PBr3 leading to 10. The dialdehydes 13a ,b, required for
the formation of the hosts 2-4, were obtained by reaction
of 10 with the allyl-protected 2,3-dihydroxybenzaldehyde
11,4 followed by palladium-catalyzed deallylation. The
macrocyclization of 13 with 1,2-phenylenediamine was
carried out in the presence of UO2(OAc)2, giving the
desired hosts in 7-35% yield.5 The hosts 2-4 gave a
characteristic absorption for the imine bond in the IR
spectrum (νNdC 1603-1604 cm-1), and the imine proton
In this article the syntheses of three new hosts and
the experimental and calculated Gibbs free-energy change
of binding of NH3, PrNH2, and PhCH2NH2 are described.
These calculations were complemented with some ad-
ditional simulations and gave a good rationale of the
observed binding and selectivity.
Resu lts a n d Discu ssion
Syn th esis. The synthesis of the hosts 2-4 starts from
the well-known terphenyl 5,2 which was methylated on
* Corresponding author: tel, +31 53 4892987; fax, +31 53 4894645;
e-mail, F.C.J .M.vanVeggel@ct.utwente.nl.
† University of Twente.
(3) Cram, D. J .; Kaneda, T.; Helgeson, R. C.; Brown, S. B.; Knobler,
C. B.; Maverick, E.; Trueblood, K. N. J . Am. Chem. Soc. 1985, 107,
3654.
(4) van Staveren, C. J .; van Eerden, J .; van Veggel, F. C. J . M.;
Harkema, S.; Reinhoudt, D. N. J . Am. Chem. Soc. 1988, 110, 4994.
(5) Similar to the synthesis of 1,1 the Ba2+-templated macrocycliza-
tion of 13c with 1,2-phenylenediamine is possible. However, this
reaction failed for 13b.
‡ Yale University.
(1) van Veggel, F. C. J . M.; Chiosis, G.; Cameron, B. R.; Reinhoudt,
D. N. Supramol. Chem. 1995, 4, 177.
(2) Koenig, K. E.; Lein, G. M.; Stuckler, P.; Kaneda, T.; Cram, D. J .
J . Am. Chem. Soc. 1979, 101, 3553.
S0022-3263(97)01943-9 CCC: $15.00 © 1998 American Chemical Society
Published on Web 05/07/1998

Preorganized Metallomacrocycles
J . Org. Chem., Vol. 63, No. 11, 1998 3555
Sch em e 1
K
1
shows the expected absorption in the H NMR spectrum
L‚H2O + RNH2 L‚RNH + H O
(1)
d
2
2
at 9.27-9.43 ppm.1,4 The H NMR spectra further show
1
an AB system for the benzylic protons, confirming the
cyclic nature of the hosts. The positive and negative FAB
mass spectra exhibit distinct peaks for M+ and M- (2);
(M + H)+, (M + Na)+, and M- (3a ,b); and (M + Na)+ and
M- (4). The cyclization of 13b with 11 gave two cyclic
products which could be separated by column chroma-
The ∆G° values of binding (CDCl3, 293 K) of ammonia
and benzylamine were determined as previously de-
scribed1 and are given in Table 1.
The binding of ammonia and benzylamine by host 2 is
essentially the same as by host 1. This is also the case
for host 4 with respect to ammonia, but the binding of
benzylamine by host 4 is reduced compared to host 1.
The result is a small increase in the selectivity of
ammonia over benzylamine (Kammonia/Kbenzylamine) from 17
for host 1 to 52 for host 4. It is apparent that the
introduction of an extra acceptor atom in host 4 is hardly
beneficial for the selective binding of ammonia compared
to host 1. The hosts 3a ,3b, however, show interesting
behavior. Host 3a shows a large increase of binding
affinity of ammonia compared to host 1, but it also
strongly binds benzylamine. The resulting selectivity
Kammonia/Kbenzylamine is only 2. The binding affinity of
ammonia by host 3b is smaller than by host 3a , but
larger than by host 1. The binding affinity of benzyl-
1
tography. Both gave H NMR spectra that show a set of
signals in agreement with a plane of symmetry through
the inner phenol ring of the terphenyl moiety, indicating
that the two isopropyl groups are mirror-related. Both
isomers gave two doublets for the two methyl groups of
the isopropyl substituents, as expected because there is
no symmetry relation. Standard 2D ROESY and NOESY
NMR spectroscopy on the major isomer 3a showed that
the CH groups of the isopropyl substituents point out-
ward, whereas the minor isomer 3b has the CH groups
pointing inward. Host 3a has a clear NOE contact
between the CH group of the isopropyl substituent with
the axial hydrogen of the benzylic group. The methyl
group of the isopropyl substituent closest to the uranyl
cation also has a NOE contact with the axial hydrogen
of the benzylic group, whereas the methyl group of the
isopropyl substituent closest to the terphenyl unit has a
NOE contact with hydrogen of the inner phenyl ring. The
minor isomer 3b showed a NOE contact of the CH and
one of the methyl groups of the isopropyl substituents
with the hydrogen of the inner phenyl ring of the
terphenyl unit, whereas the other methyl group has a
NOE contact with the axial benzylic hydrogen. These
two isomers are most likely conformational isomers. The
fact that they could be separated implies that the
activation barrier must be in the order of 20-25 kcal/
mol.
amine by host 3b is small, leading to a selectivity Kammonia
/
Kbenzylamine of 2000. These results imply that the arrange-
ment of the isopropyl group has a determining effect on
the binding and selectivity, which will be addressed on
the basis of (Gibbs free-energy) calculations in the next
section.
Gibbs F r ee-En er gy Ca lcu la tion s. The Gibbs free-
energy calculations6 are accessible by Monte Carlo free-
energy perturbation (MC-FEP) simulations,7 which rely
on a thermodynamic cycle. It basically comes down to
(6) J orgensen, W. L. Acc. Chem. Res. 1989, 22, 184. Duffy, E. M.;
J orgensen, W. L. J . Am. Chem. Soc. 1994, 116, 6337. Kaminski, G.;
Duffy, E. M.; Matsui, T.; J orgensen, W. L. J . Phys. Chem. 1994, 98,
13077. Kollman, P. Chem. Rev. 1993, 93, 2395. Gunsteren, W. F. v.;
Berendsen, H. J . C. Angew. Chem., Int. Ed. Engl. 1990, 29, 992.
Lipkowitz, K. B., Boyd, D. B., Eds. Reviews in Computational Chem-
istry; VCH: Weinheim, 1990; Vol. 1-9.
Bin d in g of NH3 a n d P h CH2NH2. The binding of
ammonia or a primary amine in host 1-4 is a substitu-
tion reaction of a bound water (eq 1).1,4
(7) J orgensen, W. L. BOSS version 3.5; Yale University: New
Haven, CT, 1994.

3556 J . Org. Chem., Vol. 63, No. 11, 1998
van Veggel et al.
Ta ble 1. Exp er im en ta la a n d Ca lcu la ted b Bin d in g F r ee
En er gies (k ca l/m ol) of 1-4 w ith NH3 a n d RNH2
(Ch lor ofor m , 293 K)
NH3
n-PrNH2
PhCH2NH2 MeNH2 EtNH2
host ∆G°exp ∆G°calcd
∆G°exp
∆G°calcd ∆G°calcd ∆G°exp ∆G°calcd
-2.47 -2.74 -2.60 -2.38
1
2
-3.47 -4.82
-3.54
-1.78
-1.47
-5.47
-1.36
-1.23
3a -5.91 -7.14
-0.19
-0.84
3b -4.51 -6.13
4
-3.58 -4.84
a
Errors are in the order of <(0.10 kcal/mol, as determined from
b
duplicate experiments. Errors are in the order of (0.10-0.26
kcal/mol (see Experimental Section).
the calculation of the Gibbs free-energy change of per-
turbing guest 1 into guest 2, both bound to a particular
host and in neat solvent. Proper subtraction of the two
∆G’s gives, via ∆G° ) -RT ln K, the association constant
K in eq 1. The required perturbations are for the
conversion of the water molecule to the amine. All MC-
FEP calculations have been performed with the BOSS
program.7 However, before setting out the CPU intensive
calculations, the ‘balance’ between the water, modeled
as TIP3P,8 and ammonia was addressed. Ammonia was
modeled with point charges qN ) -1.05, qH ) 0.35 and
the Lennard-J ones parameters σ ) 2.940 Å and ꢀ )
0.150 kcal/mol for N and σ ) 1.425 Å and ꢀ ) 0.0498 kcal/
mol for H, respectively.9 The interaction energy with Na+
is experimentally known (-24.0 and -29.1 kcal/mol for
one water and one ammonia, respectively)10 and has been
calculated with the Åqvist model11 of Na+ (-22.16 and
-26.84 kcal/mol for water and ammonia, respectively).
This gives a difference of 4.68 kcal/mol, which is close to
the experimental value of 5.10 kcal/mol. Hence, the
balance between the two models seems adequate for our
purpose. The calculated ∆G°’s are summarized in Table
1. The calculated binding affinity of ammonia follows
the experimental data, i.e., 1 ≈ 2 ≈ 4 < 3b < 3a , but is
in absolute sense too negative. However, the difference
between the experimental and calculated numbers is
rather constant, suggesting that a systematic error is
made in the calculations. A likely source is the simple
description of the electrostatic interactions by point
charges, neglecting, e.g., polarization effects, etc. The
similar binding constants for 1 and 4 were studied in
some more detail with a Monte Carlo run in chloroform.
The average interaction energy between host and am-
monia was -33.90 ( 0.09 for host 1 and -37.00 ( 0.15
kcal/mol for 4, respectively. This small favoring of 4 is
apparently offset by unfavorable changes in intramolecu-
lar free-energy components. Snapshots of the ammonia
complexes of 1 and 4 are shown in Figure 1. Visual
checking a number of saved configurations showed no
four-point binding of ammonia in host 4, suggesting that
the extra oxygen atom is not positioned correctly for the
formation of a third hydrogen bond. Occasionally, hy-
drogen bonding of ammonia to this oxygen atoms occurs,
but at the expense of another hydrogen bond.
F igu r e 1. Snapshots of 1‚NH3 (top) and 4‚NH3 (bottom).
FEP calculations, i.e., H2O T NH3, NH3 T MeNH2,
MeNH2 f EtNH2, and EtNH2 f n-PrNH2. The calcu-
lated ∆G° of binding of n-propylamine by host 1 is -2.38
kcal/mol, in good agreement with the experimental value.
These calculations also show that the binding of MeNH2,
EtNH2, and n-PrNH2 does not differ significantly, sug-
gesting that the steric demands imposed by the host are
similar for all three primary amines. This is probably
also true for benzylamine, and therefore no calculations
have been performed. A second reason to refrain from
such calculations is the formidable perturbations needed.
It would require probably 20 or more windows and long
equilibrations to get reliable results.
As a model for the binding of benzylamine, the ∆G° of
binding of methylamine by the hosts 3a ,3b was calcu-
lated. The ∆G° of binding of methylamine by the minor
isomer 3b was calculated as -0.84 kcal/mol, which is the
same order of magnitude as the experimental binding of
benzylamine (∆G° ) -1.36 kcal/mol). The conclusion
that host 3b does not impose special steric constraints
on the binding of primary amines is probably also valid
here. The minimized structure12 of 3b‚PhCH2NH2 is
shown in Figure 2 and shows that the guest is not deeply
bound by the host. In fact, the phenyl moiety is out of
the cavity. This picture also suggests that here methyl-
amine is a good model for benzylamine.
The interaction energy between host and guest is
-15.85 kcal/mol (van der Waals and electrostatic part
are -10.19 and -5.66 kcal/mol, respectively; the U‚‚‚N
distance is 2.710 Å). However, the calculated binding
affinity of methylamine by host 3a is only -0.19 kcal/
mol, which is much too small compared to the experi-
mental value (∆G° ) -5.47 kcal/mol). Here, obviously
methylamine does not suffice as a model for benzylamine.
The minimized structure12 of 3a ‚PhCH2NH2 is shown in
Figure 2 and shows that the guest is bound more deeply
The ∆G° of binding of n-propylamine by host 1 is
experimentally -2.60 ( 0.10 kcal/mol. The ∆G of binding
was also calculated through a series of individual MC-
(8) J orgensen, W. L.; Chandrasekhar, J .; Madura, J . D. J . Chem.
Phys. 1983, 79, 926. J orgensen, W. L.; Madura, J . D. Mol. Phys. 1985,
56, 1381.
(9) Quanta 3.3; Molecular Simulations: Waltham.
(10) Castleman, A. W.; Keesee, R. G. Chem. Rev. 1986, 86, 589.
(11) Åqvist, J . J . Phys. Chem. 1990, 94, 8021.
(12) The minimization was carried out as described in ref 1, but with
a constant dielectric constant.

Preorganized Metallomacrocycles
J . Org. Chem., Vol. 63, No. 11, 1998 3557
solid: 1H NMR (CDCl3) δ 11.12 (s, 2 H), 9.53 (s, 2 H), 7.45 (d,
2 H, J ) 2.1 Hz), 7.37 (d, 2 H, J ) 2.1 Hz), 7.15 (s, 2 H), 3.23
(s, 3 H), 2.37 (s, 9 H); MS (FAB, NBA) m/z 388.9 (M+, calcd
for C24H22O5 390.4).
Gen er a l P r oced u r e for th e Syn th esis of 8a ,b. A mix-
ture of the appropriate starting material (7-10 mmol), 4-5
equiv of iodoethane or 2-iodopropane, and 4-5 equiv of K2-
CO3 in 80 mL of CH3CN was refluxed for 5 h, after which it
was cooled to room temperature. After removal of the salts,
the crude product was purified by column chromatography.
3,3′′-(2,2′′-Diet h oxy-2′-m et h oxy-5,5′,5′′-t r im et h yl[1,1′:
3′,1′′]ter p h en yl)d ia ld eh yd e (8a ): (SiO2, CH2Cl2-ethyl ac-
etate ) 99:1) colorless foam, yield 71%; 1H NMR (CDCl3) δ
10.47 (s, 2 H), 7.86-7.25 (m, 6 H), 3.78 (q, 4 H, J ) 7.0 Hz),
3.22 (s, 3 H), 2.42 (s, 3 H), 2.40 (s, 6 H), 1.18 (t, 6 H, J ) 7.0
Hz); 13C NMR (CDCl3) δ 190.7, 138.6, 131.7, 127.6 (d), 71.3
(t), 60.7, 20.7, 20.7, 15.3 (q); IR (KBr) 1688 (CdO) cm-1
.
3,3′′-(2,2′′-Diisop r op oxy-2′-m et h oxy-5,5′,5′′-t r im et h yl-
[1,1′:3′,1′′]ter p h en yl)d ia ld eh yd e (8b): (SiO2, CH2Cl2) yield
1
55%; H NMR (CDCl3) δ 10.49 (s, 2 H), 7.67 (d, 2 H, J ) 2.2
Hz), 7.50 (d, 2 H, J ) 2.2 Hz), 7.27 (s, 2 H), 3.93 (heptet, 2 H,
J ) 6.2 Hz), 3.24 (s, 3 H), 2.40 (s, 3 H), 2.38 (s, 6 H), 1.09 (d,
12 H, J ) 6.2 Hz); 13C NMR (CDCl3) δ 191.2, 138.8, 131.7,
127.3, 76.3 (d), 60.7, 21.9, 20.7 (q); IR (KBr) 1685 cm-1; MS
(FAB, NBA) m/z 475.5 [(M + H)+, calcd for C30H35O5 475.2],
497.8 (M + Na)+.
3,3′′-(2,2′′-Bis((2,1-eth an ediyloxy)oxy)-2′-m eth oxy-5,5′,5′′-
tr im eth yl[1,1′:3′,1′′]ter p h en yl)d ia ld eh yd e (8c). A solution
of 7 (4.85 g, 12.5 mmol) and diethylene glycol ditosylate (5.15
g, 12.5 mmol) in 80 mL of CH3CN was added to a refluxing
mixture of K2CO3 (3.80 g, 27.5 mmol) in 420 mL of CH3CN
over a period of 10 h, after which the mixture was cooled to
room temperature. The salts were filtered off, and the crude
product was purified by column chromatography (SiO2, CH2-
Cl2-Et2O ) 94:6) to give 8c in 46% yield: mp 218-220 °C; 1H
NMR (CDCl3) δ 10.37 (s, 2 H), 7.60 (d, 2 H, J ) 2.2 Hz), 7.31
(d, 2 H, J ) 2.2 Hz), 7.07 (s, 2 H), 3.80-3.69 (m, 6 H), 3.39-
3.36 (m, 2 H), 2.91 (s, 3 H), 2.37 (s, 3 H), 2.33 (s, 6 H); 13C
NMR (CDCl3) δ 190.7, 138.4, 131.4, 127.5 (d), 75.7, 69.0 (t),
61.1, 20.9, 20.7 (q); IR (KBr) 1684 (CdO) cm-1; MS (FAB, NBA)
m/z 460.4 (M+, calcd for C28H28O6 460.2).
F igu r e 2. Minimized structures of 3a ‚PhCH2NH2 (top) and
3b‚PhCH2NH2 (bottom).
in the cavity of the host, compared to 3b‚PhCH2NH2. The
interaction energy between host and guest is -28.28 kcal/
mol (van der Waals and electrostatic part are -13.03 and
-15.24 kcal/mol, respectively; the U‚‚‚N distance is 2.848
Å), reflecting this deeper binding. The much larger
interaction between host 3a and PhCH2NH2 compared
to host 3b might be the origin of the higher binding free
energy.
Gen er a l P r oced u r e for th e Syn th esis of 9a -c. The
appropriate dialdehyde (6 mmol) was dissolved in a mixture
of MeOH and THF (1:1), and 2 mol equiv of NaBH4 was added.
After stirring for 3 h at room temperature, 120 mL of water
and 240 mL of CH2Cl2 were added. The layers were separated,
and the aqueous layer was extracted twice with 240 mL of
CH2Cl2. The combined organic layers were washed twice with
a saturated solution of NH4Cl and subsequently dried with
MgSO4. After evaporation of the solvent, the product was
obtained in >90% yield and used without purification.
3,3′′-(2,2′′-Diet h oxy-2′-m et h oxy-5,5′,5′′-t r im et h yl[1,1′:
Con clu sion s
The binding affinity of ammonia and the selectivity of
ammonia over primary amines can be improved by
changing the methyl substituents in host 1, leading to
hosts 2-4. The calculated free energies of binding are
in good (quantitative) agreement with experiment. The
analysis of the experimental observations was comple-
mented by addition calculations, suggesting that predic-
tions based on such simulations should be possible.
1
3′,1′′]ter p h en yl)d im eth a n ol (9a ): colorless foam; H NMR
(CDCl3) δ 7.18-7.14 (m, 6 H), 4.74 (s, 4 H), 3.65 (q, 4 H, J )
7.0 Hz), 3.23 (s, 3 H), 2.36 (s, 3 H), 2.33 (s, 6 H), 1.12 (t, 6 H,
J ) 7.0 Hz); 13C NMR (CDCl3) δ 131.7, 131.5, 128.8 (d), 69.1,
61.9 (t), 60.4, 20.8, 20.7, 15.7 (q).
3,3′′-(2,2′′-Diisop r op oxy-2′-m et h oxy-5,5′,5′′-t r im et h yl-
[1,1′:3′,1′′]ter p h en yl)d im eth a n ol (9b): colorless foam; 1H
NMR (CDCl3) δ 7.21 (s, 2 H), 7.15 (d, 2c H, J ) 2.0 Hz), 7.14
(d, 2 H, J ) 2.0 Hz), 4.74 (s, 4 H), 3.91 (heptet, 2 H, J ) 6.2
Hz), 3.25 (s, 3 H), 2.35 (s, 3 H), 2.33 (s, 6 H), 1.05 (d, 12 H, J
) 6.2 Hz); 13C NMR (CDCl3) δ 132.0, 131.3, 128.6, 59.9 (d),
74.4, 22.2, 20.8, 20.6 (q), 62.5 (t).
Exp er im en ta l Section
Syn th esis. 3,3′′-(2,2′′-Dih yd r oxy-2′-m eth oxy-5,5′,5′′-tr i-
m eth yl[1,1′:3′,1′′]ter p h en yl)d ia ld eh yd e (7). A solution of
compound 6 (5.0 g, 15 mmol) and HMTA (6.3 g, 45 mmol) in
90 mL of CF3COOH was heated at 60 °C for 3.5 days, followed
by the addition of 180 mL of water. After stirring for 3 h at
60 °C the mixture was poured into 600 mL of ethyl acetate
and neutralized with a saturated solution of NaHCO3. The
aqueous layer was extracted once with ethyl acetate. The
combined organic layers were washed with water and dried
with MgSO4, followed by evaporation of the solvent. The
desired compound was purified by column chromatography
(SiO2, CH2Cl2) to give 7 in 93% yield as a slightly yellow
3,3′′-(2,2′′-Bis((2,1-eth an ediyloxy)oxy)-2′-m eth oxy-5,5′,5′′-
tr im eth yl[1,1′:3′,1′′]ter p h en yl)d im eth a n ol (9c): mp 207-
1
210 °C; H NMR (CDCl3) δ 7.21 (d, 2 H, J ) 2.0 Hz), 7.07 (s,
2 H), 7.03 (d, 2 H, J ) 2.0 Hz), 4.83 and 4.64 (AB-q, 4 H, J )
12.3 Hz), 3.89-3.85 (m, 2 H), 3.66-3.62 (m, 4 H), 3.34-3.30
(m, 2 H), 2.96 (s, 3 H), 2.39 (s, 3 H), 2.35 (s, 6 H); 13C NMR
(CDCl3) δ 131.5, 131.3, 129.1 (d), 73.0, 69.1, 61.1 (t), 60.9, 20.9,
20.8 (q).
Gen er a l P r oced u r e for th e Syn th esis of 10a -c. Con-
version of the hydroxymethyl groups into bromomethyl groups
was carried out by reacting the corresponding compound 9 (5.5

3558 J . Org. Chem., Vol. 63, No. 11, 1998
van Veggel et al.
mmol) with 1.5 equiv of PBr3 in 150 mL of dry toluene. During
the addition of PBr3 to the solution of 9 in toluene, the
temperature was kept at 0 °C with an ice bath. After the
addition the reaction mixture was stirred for 4-5 h, after
which the mixture was washed with brine (2 × 150 mL) and
a saturated solution of NaHCO3 (150 mL). After the organic
layer was dried with MgSO4, the solvent was evaporated under
reduced pressure (yield > 80%).
(PPh3)4-catalyzed (5-8 mol %) reaction with Et3N‚HCOOH (3
equiv) in a mixture of DMF (50 mL) and water (8 mL). After
stirring for 1 h the mixture was poured into 60 mL of ethyl
acetate and washed three times with a saturated solution of
NH4Cl (100 mL). The organic layer was dried with MgSO4
and concentrated to dryness.
((2,2′′-Dieth oxy-2′-m eth oxy-5,5′,5′′-tr im eth yl[1,1′:3′,1′′]-
t e r p h e n y l-3,3′′-d iy l)b is (m e t h y le n e o x y ))d ih y d r o x y -
ben za ld eh yd e (13a ): crude product was purified by column
chromatography (SiO2, CH2Cl2-MeOH ) 99:1) to give a yellow
3,3′′-Diet h oxy-2′-m et h oxy-5,5′,5′′-t r im et h yl[1,1′:3′,1′′]-
1
ter p h en yl (10a ): colorless foam; H NMR (CDCl3) δ 7.19-
1
7.14 (m, 6 H), 4.63 (s, 4 H), 3.71 (q, 4 H, J ) 7.0 Hz), 3.22 (s,
3 H), 2.36 (s, 3 H), 2.33 (s, 6 H), 1.16 (t, 6 H, J ) 7.0 Hz); 13C
NMR (CDCl3) δ 133.1, 131.5, 130.9 (d), 69.0, 29.2 (t), 60.5, 20.7,
15.7 (q); MS (FAB, NBA) m/z 576.6 (M+, calcd for C28H32Br2O3
576.1).
foam, yield 63%; H NMR (CDCl3) δ 11.06 (s, 2 H), 9.93 (s, 2
H), 7.31 (d, 2 H, J ) 2.0 Hz), 7.26-7.19 (m, 8 H), 6.94 (dd, 2
H, J ) 7.9 Hz, J ) 7.9 Hz), 5.26 (s, 4 H), 3.68 (q, 4 H, J ) 7.0
Hz), 3.25 (s, 3 H), 2.37 (s, 3H), 2.33 (6 H), 1.08 (t, 6 H, J ) 7.0
Hz); 13C NMR (CDCl3) δ 196.5, 132.4, 131.5, 129.3, 125.0,
121.0, 120.4, 119.6 (d), 69.5, 66.9 (t), 60.5, 20.9, 20.7, 15.6, 15.5
3,3′′-(br om om eth yl)-2,2′′-diisopr opoxy-2′-m eth oxy-5,5′,5′′-
tr im eth yl[1,1′:3′,1′′]ter p h en yl (10b): a sample was recrys-
tallized from petroleum ether for a complete characterization,
mp 138-140 °C; 1H NMR (CDCl3) δ 7.16 (d, 2 H, J ) 2.0 Hz),
7.12 (s, 2 H), 7.07 (d, 2 H, J ) 2.0 Hz), 4.57 (s, 4 H), 3.85
(heptet, 2 H, J ) 6.1 Hz), 3.17 (s, 3 H), 2.27 (s, 3 H), 2.23 (s,
6 H), 0.98 (d, 12 H, J ) 6.2 Hz); 13C NMR (CDCl3) δ 133.2,
131.4, 131.1, 60.0 (d), 74.5, 22.4, 20.7, 20.6 (q), 29.6 (t); MS
(EI) m/z 604.102 (M+, calcd for C30H36Br2O3 604.101).
3,3′′-Bis(br om om eth yl)-2,2′′-bis((2,1-eth an ediyloxy)oxy)-
2′-m eth oxy-5,5′,5′′-tr im eth yl[1,1′:3′,1′′]ter p h en yl) (10c): a
solid was obtained by addition of petroleum ether to a solution
of 10b in CH2Cl2; 1H NMR (CDCl3) δ 7.21 (d, 2 H, J ) 2.0 Hz),
7.10 (s, 2 H), 7.04 (d, 2 H, J ) 2.0 Hz), 4.96 and 4.38 (AB-q, 4
H, J ) 9.8 Hz), 4.20-4.08 (m, 2 H), 3.78-3.62 (m, 4 H), 3.36-
3.28 (m, 2 H), 2.92 (s, 3 H), 2.41 (s, 3 H), 2.33 (s, 6 H). 13C
NMR (CDCl3) δ 132.9, 131.2, 131.1 (d), 72.7, 69.9, 29.0 (t), 60.9,
21.0, 20.7 (q). MS (FAB, NBA) m/z 613.3 [(M + Na)+, calcd for
(q); IR (KBr) 1657 (CdO) cm-1
.
((2,2′′-Diisop r op oxy-2′-m et h oxy-5,5′,5′′-t r im et h yl[1,1′:
3′,1′′]ter p h en yl-3,3′′-d iyl)bis(m eth ylen eoxy))d ih yd r oxy-
ben za ld eh yd e (13b): crude product was purified by column
chromatography (SiO2, CH2Cl2-MeOH ) 99:1), yield 57%; 1H
NMR (CDCl3) δ 11.04 (s, 2 H), 9.94 (s, 2 H), 7.34 (d, 2 H, J )
2.0 Hz), 7.26-7.19 (m, 8 H), 6.93 (dd, 2 H, J ) 7.9 Hz, J ) 7.9
Hz), 5.28 (s, 4 H), 3.68 (heptet, 4 H, J ) 6.1 Hz), 3.27 (s, 3 H),
2.37 (s, 3H), 2.32 (6 H), 1.08 (d, 12 H, J ) 6.1 Hz); 13C NMR
(CDCl3) δ 196.5, 132.4, 131.4, 128.8, 124.8, 120.4, 119.6, 74.8
(d), 66.8 (t), 60.5, 22.2, 20.9, 20.7 (q); IR (KBr) 1658 (CdO)
cm-1
.
((2,2′′-Bis((2,1-et h a n ed iyloxy)oxy)-2′-m et h oxy-5,5′,5′′-
tr im eth yl[1,1′:3′,1′′]ter ph en yl-3,3′′-diyl)bis(m eth ylen eoxy))-
d ih yd r oxyben za ld eh yd e (13c): crude product was purified
by column chromatography (SiO2, CH2Cl2), yield 47%; 1H NMR
(CDCl3) δ 11.05 (bs, 2 H), 9.92 (s, 2 H), 7.31 (d, 2 H, J ) 2.0
Hz), 7.26-7.09 (m, 8 H), 6.95 (dd, 2 H, J ) 7.9 Hz, J ) 7.9
Hz), 5.38 and 5.07 (AB-q, 4 H, J ) 11.2 Hz), 3.95-3.90 (m, 2
H), 3.72-3.53 (m, 4 H), 3.26-3.21 (m, 2 H) 3.02 (s, 3 H), 2.41
(s, 3H), 2.35 (s, 6 H); 13C NMR (CDCl3) δ 196.7, 132.3, 131.4,
129.8, 125.0, 120.3, 119.6 (d), 73.2, 68.9, 67.2 (t), 60.5, 20.8
C
28H30Br2O4‚Na 613.0].
Gen er a l P r oced u r e for th e Syn th esis of 12a -c.
A
solution of 10 in some CH3CN and CH2Cl2 (1:1) was added to
a mixture of 2 equiv of 11 and 4 equiv of K2CO3 in CH3CN
(140 mL) that was refluxed prior to this addition for 15-30
min. The resulting mixture was refluxed for 2-2.5 h, after
which it was cooled to room temperature. The salts were
filtered off and the solvent was evaporated under reduced
pressure. The crude product was purified by column chroma-
tography (SiO2, CH2Cl2).
(q); IR (KBr) 1658 (CdO) cm-1
.
Gen er a l P r oced u r e for th e Syn th esis of 2-4. A solution
of 1,2-phenylenediamine (0.7 mmol) in 50 mL of MeOH and a
solution of 13 (0.7 mmol) in 50 mL of CH2Cl2 were simulta-
neously added to a refluxing solution (450 mL) of UO2(OAc)2‚
2H2O (0.7 mmol) over a period of 3 h, after which the solution
was cooled to room temperature and concentrated under
reduced pressure.
((2,2′′-Dieth oxy-2′-m eth oxy-5,5′,5′′-tr im eth yl[1,1′:3′,1′′]-
ter p h en yl-3,3′′-d iyl)bis(m eth ylen eoxy))bis(2-(2-p r op en -
1
yloxy)ben za ld eh yd e) (12a ): yield 37%, colorless foam; H
NMR (CDCl3) δ 10.47 (s, 2 H), 7.47-7.39 (m, 2 H), 7.24-7.08
(m, 10 H), 6.18-5.96 (m, 2 H), 5.24 (s, 4 H), 5.48-5.20 (m, 4
H), 4.73 (d, 4 H, J ) 6.0 Hz), 3.67 (q, 4 H, J ) 7.0 Hz), 3.26 (s,
3 H), 2.38 (s, 3 H), 2.36 (s, 6 H), 1.10 (t, 6 H, J ) 7.0 Hz); 13C
NMR (CDCl3) δ 190.5, 133.3, 132.2, 131.5, 129.0, 124.2, 119.8,
119.4 (d), 118.9, 75.2, 69.3, 66.7 (t), 60.5, 20.9, 20.7, 15.7 (q);
(39,41-Diet h oxy-40-m et h oxy-12,17,22-t r im et h yl-25H -
3,7:10,14:15,19:20,24:27,31-p en t a m et h en o-9H -8,26,1,33-
b en zod ioxa d ia za cyclop en t a t r ia con t in e-38,42-d iola t o-
(2-)-N1,N33,O38,O42)d ioxou r a n iu m (2): crude product was
purified by column chromatography (SiO2, CH2Cl2-acetone )
98:2), yield 35%, mp 270 °C dec; 1H NMR (CDCl3) δ 9.35 (s, 2
H), 7.55-7.48 (m, 2 H), 7.46-7.39 (m, 8 H), 7.32-7.29 (m, 2
H), 7.23-7.18 (m, 4 H), 6.65 (dd, 2 H, J ) 7.8 Hz, J ) 7.8 Hz),
5.75 and 5.02 (AB-q, 4 H, J ) 8.8 Hz), 3.92 (ABX3, 4 H), 3.06
(s, 3 H), 2.45 (s, 3 H), 2.43 (s, 6 H), 0.99 (t, 6 H, J ) 7.1 Hz);
13C NMR (CDCl3) δ 165.6, 132.8, 132.2, 130.6, 129.9, 128.7,
125.1, 119.9, 116.7 (d), 71.1, 70.5 (t), 60.9, 21.0, 20.9, 15.2 (q);
IR (KBr) 1603 (NdC) cm-1; MS (FAB, NBA) m/z 1030.9 (M+,
calcd for C48H44N2O9U 1030.4), 1032.0 [(M + H)+], 1053.8 [(M
+ Na)+], 1030.6 (M-). No satisfactory elemental analysis was
obtained for the bishydrate.
IR (KBr) 1688 (CdO) cm-1
.
((2,2′′-Diisop r op oxy-2′-m et h oxy-5,5′,5′′-t r im et h yl[1,1′:
3′,1′′]ter p h en yl-3,3′′-d iyl)bis(m eth ylen eoxy))bis(2-(2-p r o-
p en yloxy)ben za ld eh yd e) (12b). yield 70%, colorless foam;
1H NMR (CDCl3) δ 10.47 (s, 2 H), 7.47-7.39 (m, 2 H), 7.24-
7.08 (m, 10 H), 6.20-5.98 (m, 2 H), 5.30 (s, 4 H), 5.45-5.20
(m, 4 H), 4.75 (d, 4 H, J ) 6.0 Hz), 3.90 (heptet, 2 H, J ) 6.1
Hz), 3.28 (s, 3 H), 2.37 (s, 3 H), 2.35 (s, 6 H), 1.04 (d, 12 H, J
) 6.1 Hz); 13C NMR (CDCl3) δ 190.6, 133.3, 132.4, 131.4, 128.6,
124.2, 119.7, 119.3, 74.6 (d), 118.9, 75.2, 66.6 (t), 22.2, 20.9,
20.7 (q); IR (KBr) 1689 (CdO) cm-1
.
(39,41-Diisop r op oxy-40-m et h oxy-12,17,22-t r im et h yl-
25H-3,7:10,14:15,19:20,24:27,31-p en ta m eth en o-9H-8,26,1,
33-ben zod ioxa d ia za cyclop en ta tr ia con tin e-38,42-d iola to-
((2,2′′-Bis((2,1-et h a n ed iyloxy)oxy)-2′-m et h oxy-5,5′,5′′-
tr im eth yl[1,1′:3′,1′′]ter ph en yl-3,3′′-diyl)bis(m eth ylen eoxy))-
bis(2-(2-p r op en yloxy)ben za ld eh yd e (12c): yield 47%, col-
(2-)-N1,N33,O38 42)d ioxou r a n iu m (3a ,b): crude product was
,
1
orless foam; H NMR (CDCl3) δ 10.45 (s, 2 H), 7.47-7.39 (m,
purified by column chromatography (SiO2, CH2Cl2-acetone )
99:1), major isomer 3a yield 14%, mp 280 °C dec, 1H NMR
(CDCl3) δ 9.27 (s, 2 H), 7.49-7.33 (m, 10 H), 7.24-7.14 (m, 4
H), 6.66 (dd, 2 H, J ) 7.8 Hz, J ) 7.8 Hz), 5.53 and 4.93 (AB-
q, 4 H, J ) 10.1 Hz), 3.71 (heptet, 2 H, J ) 6.0 Hz), 3.00 (s, 3
H), 2.43 (s, 3 H), 2.39 (s, 6 H), 1.04 (d, 6 H, J ) 6.0 Hz), 0.93
(d, 6 H, J ) 6.0 Hz); 13C NMR (CDCl3) δ 165.3, 132.7, 131.6,
130.8, 129.9, 128.7, 126.0, 119.7, 116.4, 115.1, 76.1 (d), 65.1
(t), 62.5, 22.6, 21.9, 20.8 (q); IR (KBr) 1604 (NdC) cm-1; MS
2 H), 7.24-7.08 (m, 10 H), 6.14-5.98 (m, 2 H), 5.45-5.10 (m,
4 H), 4.80-4.62 (m, 4 H), 3.93-3.85 (m, 2 H), 3.71-3.58 (m, 4
H), 3.33-3.21 (m, 2 H), 3.02 (s, 3 H), 2.42 (s, 3 H), 2.37 (s, 6
H); 13C NMR (CDCl3) δ 190.5, 133.2, 132.1, 131.4, 129.1, 124.3,
119.9, 118.4 (d), 118.9, 75.2, 73.1, 69.0, 66.8 (t), 61.0, 20.9 (q);
IR (KBr) 1685 (CdO) cm-1
.
Gen er a l P r oced u r e for th e Syn th esis of 13a ,b. Deal-
lylation of compound 12 (2 mmol) was performed by the Pd-

Preorganized Metallomacrocycles
J . Org. Chem., Vol. 63, No. 11, 1998 3559
(FAB, NBA) m/z 1059.3 [(M + H)+], 1081.1 [(M + Na)+], 1058.0
(M-). Anal. Calcd for C50H48N2O9U‚2H2O: C, 54.84; H, 4.79;
N, 2.56. Found: C, 54.42; H, 4.65; N, 2.51.
Monte Carlo free-energy perturbation (MC-FEP), or impor-
tance sampling, simulations were performed with the BOSS
program,7 using the OPLS chloroform model.17 Details are as
follows. The appropriate z-matrices for 1‚H2O/NH3, 3a ‚H2O/
NH3, 3b ‚H2O/NH3, 4‚H2O/NH3, 1‚NH3/MeNH2, 1‚MeNH2/
EtNH2, 1‚EtNH2/PrNH2, 3a ‚NH3/MeNH2, and 3b‚NH3/MeNH2
were constructed from the minimized structures. The com-
plexes were placed in a box of 33.0 × 33.0 × 49.5 Å dimension,
initially filled with 400 CHCl3s. On the basis of the worst
interaction energies, 17 molecules of CHCl3 were removed at
the start of the simulations. A cutoff of 11 Å was used for the
nonbonded interactions, which were quadratically smoothed
to zero between the cutoff and the cutoff - 0.5 Å. The ligand
and coordinated guest were sampled independently. Trans-
lational and rotational sampling was applied to the ligand
(0.02 Å and 2.00°, respectively), in addition to sampling of the
phenolic substituents through their dihedrals (sampling range
was 5°). The sampling of the bridge in host 4 also included
angles and bonds (treated automatically). This gave an
acceptance ratio of approximately 35%. The translational and
rotational sampling ranges of H2O/NH3 were set to 0.15 Å and
15.0°, respectively, giving an acceptance ratio of roughly 40%.
The dihedral sampling of RNH2 was 5.0°, and the translational
and rotational sampling ranges were somewhat smaller than
with H2O/NH3 such that an acceptance ratio of roughly 40%
was obatined. A solute move was attempted every 25 solvent
moves. Preferential sampling was used.18 The perturbations
were carried out in five equally spaced, double-wide windows.
The calculations on H2O T NH3 were equilibrated for 1 million
configurations, followed by averaging over 2 million configura-
tions. The calculations on R1NH2/R2NH2 in the absense of a
host were equilibrated for 1 million configurations and aver-
aged over 4 million configurations. The perturbations of NH3
into MeNH2 and the reverse were equilibrated for 3 million
configurations and averaged over 4 million configurations. The
perturbation of MeNH2 into EtNH2 in host 1 was equilibrated
for 4 million configurations and averaged over 3 million
configurations. The NPT ensemble at 1 atm and 298 K was
used. Full periodic boundary conditions were imposed. This
was done by making 26 images in the (x, (y, and (z
directions. The standard deviations were between 0.03 and
0.50 kcal/mol for all runs, except for the perturbation of NH3
into MeNH2 and the reverse which gave standard deviations
of 0.69 and 0.78 kcal/mol. As a lower bound estimation of the
error, the average of the forward and backward runs was
taken, if applicable. Calculations were run on Silicon Graphics
workstations and on a Pentium 133 personal computer.19
Min or isom er 3b: yield 7%, mp 280 °C dec; 1H NMR
(CDCl3) δ 9.38 (s, 2 H), 7.58-7.30 (m, 12 H), 7.24-7.14 (m, 2
H), 6.66 (dd, 2 H, J ) 7.8 Hz, J ) 7.8 Hz), 5.93 and 4.87 (AB-
q, 4 H, J ) 9.0 Hz), 4.10 (heptet, 2 H, J ) 6.2 Hz), 2.92 (s, 3
H), 2.47 (s, 3 H), 2.41 (s, 6 H), 1.27 (d, 6 H, J ) 6.2 Hz), 0.99
(d, 6 H, J ) 6.2 Hz); 13C NMR (CDCl3) δ 165.5, 132.4, 132.1,
130.4, 129.6, 128.7, 124.5, 119.8, 116.6, 78.4 (d), 70.2 (t), 60.3,
22.9, 22.6, 21.1, 20.8 (q); IR (KBr) 1603 (NdC) cm-1; MS (FAB,
NBA) m/z 1060.0 [(M + H)+], 1081.9 [(M + Na)+], 1058.6 (M-).
(39,41-Bis((2,1-et h a n ed iyloxy)oxy)-40-m et h oxy-12,17,
22-tr im eth yl-25H-3,7:10,14:15,19:20,24:27,31-pen tam eth en o-
9H-8,26,1,33-ben zodioxadiazacyclopen tatr iacon tin e-38,42-
d iola to(2-)-N1,N33,O38,O42)d ioxou r a n iu m (4): crude prod-
uct was purified by column chromatography (SiO2, CH2Cl2-
ethyl acetate ) 95:5), yield 35%, mp 280 °C dec; 1H NMR
(CDCl3) δ 9.43 (s, 2 H), 7.60-7.30 (m, 10 H), 7.24-7.20 (m, 4
H), 6.68 (dd, 2 H, J ) 7.7 Hz, J ) 7.7 Hz), 5.74 and 5.04 (AB-
q, 4 H, J ) 8.7 Hz), 4.02-3.97 (m, 2 H), 3.71-3.65 (m, 4 H),
3.20-3.15 (m, 2 H), 3.03 (s, 3 H), 2.47 (s, 3 H), 2.45 (s, 6 H);
13C NMR (CDCl3) δ 165.7, 132.9, 132.0, 130.7, 130.5, 128.8,
126.8, 119.9, 116.9 (d), 75.4, 70.5, 68.7 (t), 60.7, 21.2, 20.9 (q);
IR (KBr) 1603 (NdC) cm-1; MS (FAB, NBA) m/z 1068.4 [(M +
Na)+], 1045.1 (M-). Anal. Calcd for C48H42N2O10U‚2H2O: C,
53.34; H, 4.29; N, 2.59. Found: C, 53.36; H, 4.25; N, 2.53.
Bin d in g of NH3, P r NH2, a n d P h CH2NH2. These were
performed as described before1 except that the concentration
of NH3 in CHCl3 was determined titrimetrically. A sample
was taken from the solution and put into 20 mL of ethanol,
followed by tritration with trifluoromethanesulfonic acid,
dissolved in 2-propanol.
Gibbs F r ee-En er gy Ca lcu la tion s. Minimized structures
were obtained as previously described for 1,1 and charges were
assigned with the charge equilibration method13 as imple-
mented in the Cerius2 package14 (version 1.5). Symmetry-
related charges were averaged. The charge on the uranium
was set to +2.0.15 Except for the substituents on the phenol
moieties and the p-methyl groups, all hydrogens were treated
explicitly. See the Supporting Information for a set of typical
charges.16 The Lennard-J ones parameters for uranium were
σ ) 2.6727 Å and ꢀ ) 1.0 kcal/mol.15 All other Lennard-J ones
parameters were taken from the BOSS parameter file.7 The
methylene and/or methyl groups of the guests were treated
as united atoms. The (coordinated) water was simulated with
the TIP3P model8 and the (coordinated) ammonia with qN
)
-1.05, qH ) 0.35, σ ) 2.940 Å, and ꢀ ) 0.150 kcal/mol for N
and σ ) 1.425 Å and ꢀ ) 0.0498 kcal/mol for H, respectively.9
See Results and Discussion for an evaluation of the balance
of these models. The (coordinated) MeNH2 was simulated with
qN ) -0.90, qH ) 0.35, and qMe ) 0.20,7 with the Lennard-
J ones parameters for N and H as for NH3 and the united
methyl group as in the BOSS parameter file.7 The methyl
group in EtNH2 and the methyl and ‘second’ methylene group
of PrNH2 were not charged. The other charges were as in
MeNH2.
Ack n ow led gm en t. NATO is gratefully acknowl-
edged for a collaborative research grant (SRG 940060).
Su p p or tin g In for m a tion Ava ila ble: 13C NMR spectra of
compounds 2-4 and 7-13; also a set of typical point charges
(21 pages). This material is contained in libraries on micro-
fiche, immediately followes this article in the microfilm version
of the journal, and can be ordered from the ACS; see any
current masthead page for ordering information.
(13) Rappe´, A. K.; Goddard, W. A., III. J . Phys. Chem. 1991, 95,
3358.
(14) Cerius2 version 1.5; Molecular Simulations: Waltham.
(15) van Doorn, A. R.; Rushton, D. J .; van Straaten-Nijenhuis, W.
F.; Verboom, W.; Reinhoudt, D. R. Recl. Trav. Chim. Pays-Bas 1992,
111, 421.
J O971943Q
(17) J orgensen, W. L.; Briggs, J . M.; Contreras, M. L. J . Phys. Chem.
1990, 94, 1683.
(18) J orgensen, W. L. J . Phys. Chem. 1983, 87, 5304.
(19) Tirado-Rives, J .; J orgensen, W. L. J . Comput. Chem. 1996, 17,
1385.
(16) In principle, a new minimization is required, but only marginal
changes are expected due to the fact that the hosts are preorganized.
The small expected changes of the phenolic substituents will be taken
care of in the equilibration phase of the MC-FEP calculations.