Silver(I)-Diene complexes as versatile catalysts for the C -Arylation of N -Tosylaziridines: Mechanistic insight from in situ diagnostics

Article abstract of DOI:10.1021/jo100260k

(Figure presented) Silver(I) complex [Ag(diene)₂] ⁺Y^- (where diene = cyclooctadiene, norbornadiene, and 1,3-cyclohexadiene; Y^- = PF₆^-, BF ₄^-) efficiently catalyzes the

Full text of DOI:10.1021/jo100260k

pubs.acs.org/joc
Silver(I)-Diene Complexes as Versatile Catalysts for the C-Arylation
of N-Tosylaziridines: Mechanistic Insight from In Situ Diagnostics
Milan Bera† and Sujit Roy*,†,‡
†Organometallics & Catalysis Laboratory, Chemistry Department, Indian Institute of Technology,
Kharagpur 721302, India, and ‡Organometallics & Catalysis Laboratory, School of Basic Sciences,
Indian Institute of Technology, Bhubaneswar 751013, India
royiitkgp@gmail.com; sroychem@iitbbs.ac.in
Received February 27, 2010
Silver(I) complex [Ag(diene)2]þY- (where diene = cyclooctadiene, norbornadiene, and 1,3-cyclo-
hexadiene; Y- = PF6-, BF4-) efficiently catalyzes the arylation of N-tosylaziridines with arenes and
heteroarenes under ambient condition to provide the corresponding β-aryl amine derivatives with
excellent regioselectivity. To understand the nature of substrate activation, and initial bond break-
ing/making steps, the following studies were conducted with the help of in situ NMR (1H, 31P, 109Ag)
and ESI-MS probe: (I) evaluation of Hammett reaction constant (F); (II) correlation of initial rate (k)
versus cone angle (θ) of ligand L for reactions mediated by [Ag(COD)2]PF6/L (where L is a phosphine
or a phosphite ligand); (III) identification of silver-arene intermediates in solution; and (IV) correla-
tion of initial rate (k) with ΔHOMO-LUMO of [Ag(diene)2]PF6 obtained from preliminary DFT studies.
Study I led to a F-value of -0.586, indicating that the extent of electrophilic perturbation is
considerably less than a typical Lewis acid catalyzed process. Study II indicated that initial rate (k)
increases with concomitant increase in θ, as well with Δ31P(complex-ligand), which corroborates to a
mechanism involving prior ligand dissociation. Study III showed the plausible formation of
[Ag(diene)(arene)]þ and [Ag(arene)2]þ as reactive species in solution. Study IV showed that the
dependence of initial rate (k) with diene ligand is in the order COD > NBD > CHD; which
corresponds well with the order of hardness of the respective Ag(I) complexes.
1. Introduction
and pseudocorhydrin as well as bioactive molecules such as
HIV-1 protease inhibitors and tryptophan.2 Nucleophile-
assisted regioselective ring-opening of aziridines has been
Aziridines are important building blocks for the synthesis
of various nitrogen-containing organic architectures includ-
ing natural products and bioactive molecules.1 A few chosen
examples include alkaloids like pumiliotoxin-C, verruculotoxin,
(2) (a) Oppolzer, W.; Flaskamp, E. Helv. Chim. Acta 1977, 60, 204.
(b) Oppolzer, W.; Flaskamp, E.; Bieber, L. W. Helv. Chim. Acta 2001, 84,
141. (c)Martens, J.; Scheunemann, M. Tetrahedron Lett. 1991, 32, 1417. (d) Hu,
X. E. Tetrahedron 2004, 60, 2701 and references cited therein. (e) Saha, B.;
Nandy, J. P.; Shukla, S.; Siddiqui, I.; Iqbal, J. J. Org. Chem. 2002, 67, 7858.
(f) Nishikawa, T.; Ishikawa, M.; Wada, K.; Isobe, M. Synlett 2001, 945.
(1) (a) Tanner, D. Angew. Chem., Int. Ed. Engl. 1994, 33, 599. (b) McCoull,
W.; Davis, F. A. Synthesis 2000, 1347.
4402 J. Org. Chem. 2010, 75, 4402–4412
Published on Web 06/03/2010
DOI: 10.1021/jo100260k
r
2010 American Chemical Society

Bera and Roy
JOCArticle
one of the trustworthy routes to access these architectures.3
Among the organometallic reagents, those of Li, Cu, and
Mg are well-known in promoting facile ring-opening of
aziridines.4 Various silyl nucleophiles such as TMSX (X= Cl,
CN, N3) also effectively react with aziridines in the presence
of transition metal catalyst and with organocatalysts.5 Aziri-
dine carboxylate reacts with carbonyl-stabilized Wittig re-
agents to provide an isolable phosphorus ylide, and the latter
reacts with carbonyl compounds giving rise to optically pure
unsaturated amino acids.6 A variety of Lewis acids such as
Yb(OTf)3, Cu(OTf)2, and Sn(OTf)2 and organocatalysts are
used in aziridine ring-opening with amine nucleophile to afford
the synthetically important diamine compounds.7 In some cases
stoichiometric amount of MgBr2 or ZnX2 (X = Cl, Br, I)
are used as Lewis acid as well as the source for halogen
nucleophile for aziridine ring-opening.8 Alkyl-aryl ethers in-
cluding amino ethers are important constituents of many
pharmacologically active molecules, which can be accessed
via the ring-opening of aziridine by O-nucleophile.9 Both
arylboronic acid and borate behave as activated nucleophile
for aziridine ring-opening. For example, palladium-pincer
complex catalyzes the aziridine ring-opening with arylboro-
nic acid, but even without the assistance of a transition metal,
aryl borate reacts with aziridine effectively.10 The cycloaddi-
tion of aziridines to dipolarophiles provides a facile entry to
heterocycles such as oxazolidine11 (when carbonyl com-
pound acts as a dipolarophile), imidazoline12 (when nitrile
acts as a dipolarophile), and pyrrolidine13 (in the case of
nonactivated alkenes). The cascade reaction between azir-
idines and propargylic alcohol containing internal or term-
inal acetylenic linker affords various N,O-heterocycles,
namely oxazine, oxazepine, and oxazocine.14 In contrast
to the above, there are fewer reports on the regioselective
ring-opening of aziridines by an arene nucleophile, which are
mostly Lewis acid mediated. To our knowledge the known
Lewis acids used include stoichiometric AlCl3 (1 equiv),15 BF3.
Et2O (1-3 equiv),16 catalytic In(OTf)3 (5-10 mol %),17 Cu-
(OTf)2 (10 mol %),18 AuCl3 (5%),19 and FeCl3 (5%).20
Our continuing effort in late transition metal catalysis
toward aromatic alkylation reactions21 prompted us to in-
vestigate the reactivity of N-tosylaziridines with arene. This
endeavor led to the novel finding22 that under ambient
condition, only 1-2% of AgPF6 promotes a highly regiose-
lective ring-opening of N-tosylaziridines with a variety of
arenes and heteroarenes giving rise to the corresponding
β-aryl amine derivatives in excellent yields. Equipped with
the above success we next aimed to explore the mechanism of
catalysis to better our understanding of the initial bond-
breaking and bond-making process and the factors that
control these steps. We reasoned that such an exercise would
take us to the next logical portal of mechanism-based
rational design of ligand to improve the efficiency and selec-
tivity of the catalyst. In the course of these efforts, we found
that Ag(I)-diene complexes are equally efficient in promot-
ing the C-arylation of aziridines. The Ag(I)-diene com-
plexes are easy to synthesize and purify, have good shelf
life, and have turned out to be good candidates for mechan-
istic studies. In the present article, we delineate the details of
the synthetic and mechanistic studies. In the mechanistic
front, the following studies were conducted with the help of
in situ NMR (1H, 31P, 109Ag) and ESI-MS probe: (I) evalua-
tion of Hammett reaction constant (F) to determine the
extent of electrophilic perturbation; (II) correlation of initial
rate (k) versus cone angle (θ) of ligand L for reactions
mediated by [Ag(COD)2]PF6/L to understand the ligand
influence; and (III) identification of reactive species in solu-
tion using ESI-MS and 109Ag NMR as probe to understand
Ag-ligand/Ag-arene binding. Aided by preliminary DFT
studies we also aimed to correlate the initial rate (k) with
(3) (a) Sweeney, J. B. Chem. Soc. Rev. 2002, 31, 247. (b) Padwa, A.;
Murphree, S. S. ARKIVOC 2006, 6. (c) Lu, P. Tetrahedron. 2010, 66, 2549.
(d) Bergmeier, S. C., Lapinsky, D. J. Three Membered Ring Systems. In Prog.
Heterocycl. Chem. 2009, 20, 47; 2009, 21, 69.
(4) (a) Kozikowski, A.; Ishida, H.; Isobe, K. J. Org. Chem. 1979, 44, 2788.
(b) Ibuka, T.; Nakai, K.; Habashita, H.; Fujii, N.; Garrido, F.; Mann, A.;
Chounan, Y.; Yamamoto, Y. Tetrahedron Lett. 1993, 34, 7421. (c) Rodrigo,
L. O. R. C.; Diego, D. J.; Simonelli, F.; Comasseto, J. V. Tetrahedron Lett.
2005, 46, 2539.
ΔHOMO-LUMO of [Ag(diene)2]PF6 in order to understand the
electronic influence of the diene ligand.
2. Results and Discussion
2.1. Efficacy of [Ag(COD)2]PF6 as a Catalyst. The initial
reaction of N-tosyl-2-phenylaziridine 1a with anisole as the
arene as well as the solvent, and [Ag(COD)2]PF6 (1 mol %)
as catalyst at room temperature did not proceed to comple-
tion even after 5 h (vide TLC), and workup afforded 50% of
N-tosyl-2-phenyl-2-(4-methoxyphenyl)ethylamine 2. In con-
trast, by using dichloroethane (DCE) as the solvent, the yield
of 2 rose to 65% after 5 h (Scheme 1). Gratifyingly, by
(5) (a) Wu, J.; Hou, X.-L.; Dai, L.-X. J. Org. Chem. 2000, 65, 1344. (b)
Wu, J.; Sun, X.; Ye, S.; Sun, W. Tetrahedron Lett. 2006, 47, 4813. (c)
Minakata, S.; Okada, Y.; Oderaotoshi, Y.; Komatsu, M. Org. Lett. 2005,
7, 3509.
(6) Baldwin, J. E.; Adlington, R. M.; Robinson, N. G. J. Chem. Soc.,
Chem. Commun. 1987, 153.
(7) (a) Meguro, M.; Asao, N.; Yamamoto, Y. Tetrahedron Lett. 1994, 35,
7395. (b) Sekar, G.; Singh, V. K. J. Org. Chem. 1999, 64, 2537. (c) Wu, J.; Sun,
X.; Li, Y. Eur. J. Org. Chem. 2005, 4271. (d) Cresty, F.; Witt, M.;
Frydenvang, K.; Staerk, D.; Jaroszewski, J. W.; Franzik, H. J. Org. Chem.
2008, 73, 3566.
(8) (a) Righi, G.; Franchini, T.; Bonini, C. Tetrahedron Lett. 1998, 39,
2385. (b) Ghorai, M. K.; Das, K.; Kumar, A.; Ghosh, K. Tetrahedron Lett.
2005, 46, 4103. (c) Ghorai, M. K.; Kumar, A.; Tiwari, D. P. J. Org. Chem.
2010, 75, 137.
(15) (a) Stamm, H.; Sommer, A.; Woderer, A.; Wiesert, W.; Mall, T.;
Assithianakis, P. J. Org. Chem. 1985, 50, 4946. (b) Stamm, H.; Onistschenko,
A.; Buchholz, B.; Mall, T. J. Org. Chem. 1989, 54, 193.
(9) (a) Prasad, B. A. B.; Sekar, G.; Singh, V. K. Tetrahedron Lett. 2000,
41, 4677. (b) Li, P.; Forbeck, E. M.; Evans, C. D.; Joullie, M. M. Org. Lett.
2006, 22, 5105.
(16) (a) Bergmeier, S. C.; Kltz, S. J.; Huang, J.; Mcpherson, H.; Donoghue,
P. J.; Reed, D. D. Tetrahedron. Lett. 2004, 45, 5011. (b) Schneider, M. R.;
Mann, A.; Taddei, M. Tetrahedron Lett. 1996, 37, 8493.
(10) (a) Aydin, J. K. J.; Wallner, O. A.; Saltanova, I. V.; Szabo, J. K.
Chem.;Eur. J. 2005, 11, 5260. (b) Pineschi, M.; Bertolisi, B.; Haak, R. M.;
Crotti, P.; Macchia, F. Chem. Commun. 2005, 1426.
(11) (a) Ghorai, M. K.; Ghosh, K. Tetrahedron Lett. 2007, 48, 3191. (b)
Kang, B.; Miller, A. W.; Goyal, S.; Nguyen, S. T. Chem. Commun. 2009,
3928.
(12) Ghorai, M. K.; Ghosh, K.; Das, K. Tetrahedron Lett. 2006, 47, 5399.
(13) (a) Ungureanu, I.; Koltz, P.; Mann, A. Angew. Chem., Int. Ed. Engl.
2000, 41, 4615. (b) Loh, T. P.; Hu, Q.-Y.; Ma, L. T. J. Am. Chem. Soc. 2001,
123, 2450. (c) Dauban, P.; Malik, G. Angew. Chem., Int. Ed. 2009, 48, 9026.
(14) Bera, M.; Roy, S. J. Org. Chem. 2009, 74, 8814.
(17) Yadav, J. S.; Reddy, B. V. S.; Rao, R. S.; Veerendhar, G.; Nagaiah,
K. Tetrahedron Lett. 2001, 42, 8067.
(18) Hajra, S.; Maji, B.; Mal, D. Adv. Synth. Catal. 2009, 351, 859.
(19) Sun, X.; Sun, W.; Fan, R.; Wu, J. Adv. Synth. Catal. 2007, 349, 2151.
(20) Wang, Z.; Sun, X.; Wu, J. Tetrahedron 2008, 64, 5013.
(21) (a) Choudhury, J.; Podder, S.; Roy, S. J. Am. Chem. Soc. 2005, 127,
6162. (b) Podder, S.; Choudhury, J.; Roy, S. J. Org. Chem. 2007, 72, 3129. (c)
Podder, S.; Choudhury, J.; Roy, U. K.; Roy, S. J. Org. Chem. 2007, 72, 3100.
(d) Podder, S.; Roy, S. Tetrahedron. 2007, 63, 9146. (e) Choudhury, J.; Roy,
S. J. Mol. Catal. A: Chem. 2008, 279, 37.
(22) Bera, M.; Roy, S. Tetrahedron Lett. 2007, 48, 7144.
J. Org. Chem. Vol. 75, No. 13, 2010 4403

Bera and Roy
JOCArticle
SCHEME 1. Model Reaction Study
SCHEME 2. C-Arylation of N-Tosylaziridine
TABLE 1. [Ag(COD)2]PF6-Catalyzed Regioselective Ring-Opening of
N-Tosylaziridine with Arene as Nucleophilea
employing 2 mol % of the catalyst, the reaction proceeded to
completion within 3 h and amine 2 was isolated in 82% yield.
Screening of solvents demonstrated the superiority of DCE
over others. Notably the catalytic activity of [Ag(COD)2]-
PF6 toward C-arylation was inhibited in water, ethanol,
and tetrahydrofuran as solvents. Instead, side reactions
were predominant. Among the other silver salts and com-
plexes screened for their catalytic activity, only AgBF4 and
[Ag(COD)2]BF4 were promising (Table S1, Supporting
Information). The generality of the reaction was successfully
tested under the optimized conditions, with N-tosylaziri-
dines 1a-d and various arenes and heteroarenes as nucleo-
philes, giving rise to the corresponding β-substituted
ethylamines 2-21 (Scheme 2, Tables 1 and 2). The products
were fully characterized by 1H, 13C, IR, HRMS, and X-ray
crystallography (in case of 6). The salient remarks are high-
lighted below.
(1) Arene and heteroarenes always attacked the benzylic
position of N-tosylaziridines to afford the corresponding
product with high regioselectivity.
(2) For ring-activated arenes, reactions were complete within
shorter reaction times, and product yields were very good. In
contrast, benzene and ring-deactivated arenes were inactive.
(3) The ring-opening reaction was also facilitated with
furan and thiophene derivatives, as indicated by shorter
reaction time and good yields (Table 2). Clearly, in all cases
the substitution took place exclusively at the 2-position of the
heteroarene.
(4) The nature of the substituent on the aromatic ring of
N-tosylaziridine had some effect on the conversion, and the
reactivity trend was found to be 1c>1a>1b>1d.
(5) Attempted arylation with aziridine bearing an alkyl
substituent such as N-tosyl-2-hexylaziridine and N-benzyl-2-
phenylaziridine failed.
aConditions: aziridine (1 mmol), arene (2 mmol), [Ag(COD)2]PF6
(2 mol %), DCE (3 mL).
we next undertook the following mechanistic studies with the
help of in situ NMR (1H, 31P, 109Ag) and ESI-MS probe: (I)
evaluation of Hammett reaction constant (F) to determine
the extent of electrophilic perturbation; (II) correlation of
initial rate (k) versus cone angle (θ) of ligand L for reactions
mediated by [Ag(COD)2]PF6/L to understand the ligand
influence; (III) identification of reactive species in solu-
tion using ESI-MS and 109Ag NMR as probe to understand
Ag-ligand/Ag-arene binding; and (IV) correlation of initial
rate (k) with ΔHOMO-LUMO of [Ag(diene)2]PF6 to understand
2.2. Mechanistic Studies. Inspired by the efficacy of the
[Ag(COD)2]PF6 catalysts, and in keeping with our original goal,
4404 J. Org. Chem. Vol. 75, No. 13, 2010

Bera and Roy
JOCArticle
TABLE 2. [Ag(COD)2]PF6-Catalyzed Regioselective Ring-Opening of
N-Tosylaziridine with Heteroarene as Nucleophilea
SCHEME 3
TABLE 3. Rate Constant (kR) for the Reaction of Anisole with Various
Para-Substituted Phenyl N- Tosylaziridines
entry
R
σ
104kR (s-1 a
)
kR/kH
log(kR/kH)
1
2
3
4
OMe
Me
H
-0.27
-0.17
0
2.9
2.6
2.1
1.5
1.380
1.238
1
0.1398
0.0927
0
Cl
0.22
0.714
-0.1463
aError <5% from 2 runs.
FIGURE 1. log(kR/kH) vs σ of para-substituted phenyl N-tosyla-
ziridine.
aConditions: aziridine (1 mmol), heteroarene (2 mmol), [Ag(COD)2]-
PF6 (2 mol %), DCE (3 mL).
the electronic influence of the diene ligand. The results of these
studies are presented below.
Hammett Correlation Study with Substituted Aziridine
(Silver-Aziridine Interaction). The effect of substituent on
the reaction kinetics is often derived from the Hammett
linear free energy relationship [log(kR/kH) = Fσ]. From the
Hammett plot of relative rate constant values (kR/kH) versus
the substituent constant (σ), one can obtain the Hammett
reaction constant F. The latter is a measure of the sensitivity
of a reaction to electronic perturbation due to substituent
effect. In other words, it is a measure of the relative suscepti-
bility of a reaction to the electron-donating or electron-
withdrawing effect exerted by a substituent. We undertook
the Hammett study to look into the influence of substituent
in aziridine on C-arylation. This was attempted by monitor-
ing the product (vide 1H NMR, with diphenylmethyl toluene
as the internal reference) for the initial part of the reaction
between anisole and para-substituted phenyl N-tosylaziri-
dine in the presence of 1 mol % [Ag(COD)2]PF6 (Scheme 3,
Table 3). The data fitted well with pseudo-first-order rate
plots, from which rate constants were evaluated. The linear
free energy relationship between the relative rate constant
(kR/kH) and substituent constant (σ) (Figure 1) resulted in a
negative Hammett reaction constant F-value (-0.586) indi-
cating an electrophilic mechanism for the present reaction.
FIGURE 2. Silver-aziridine coordination.
The small negative F-value suggests that the catalyst-sub-
strate interaction leads to the generation of a weak positive
charge (δþ) at the benzylic (-CH2) carbon center of the
aziridine (Figure 2). Absence of a large negative reaction
constant F-value also indicates that, unlike in the case of
strong Lewis acid mediated reactions, in our system “distinct
free benzyl cation” is not formed.23,24
(23) (a) Sykes, P. A Guidebook to Mechanism in Organic Chemistry, 6th ed.;
Pearson Education. Inc.: Singapore, 1986. (b) Frost, A. A.; Pearson, R. G. Kinetics
and Mechanism, A Study of Homogeneous Chemical Reactions; Wiley Inter-
science: New York, 1961. (c) Laidler, K. J. Chemical Kinetics, 3rd ed.; Pearson
Education Inc.: Singapore, 1987. (d) March, J. Advanced Organic Chemistry,
Reactions, Mechanisms, and Structure, 4th ed.; Wiley-Interscience: New York,
1992. (e) Clayden, J.; Greeves, N.; Warren, S.; Wothers, P. In Organic Chemistry;
Oxford University Press: Oxford, U.K., 2001; pp 1090-1100.
(24) A small Hammett reaction constant F-value of -0.586 (within the
range from -1 to þ1) indicates that the aryl-ring on aziridine is further away
from the reactive site. In other words, the δþ charge at the site of attack is
expected to be less affected by the substituent on the arene ring. To
accommodate this view, we prefer the initial binding of aziridine with Ag(I)
via the nitrogen center. An alternate binding mode involving Ag(I)-arene π-
interaction would markedly alter the F-value, hence is less likely. However, a
conclusive judgment on this aspect is not possible at this stage. We thank one
of the reviewers for raising this issue.
J. Org. Chem. Vol. 75, No. 13, 2010 4405

Bera and Roy
JOCArticle
TABLE 4. Changes (Δ, δ, ppm) in the 31P Chemical Shiftsa of [Ag(PR3)2]PF6
ligands (PR3)
δ (ppm) free ligand
δ (ppm) complex
Δ (ppm)b
cone angle (θ) (deg)
P(OEt)3
P(OPh)3
PPh3
P(p-tol)3
P(cy)3
142.249
13.432
-1.999
-4.431
14.155
-26.246
-32.998
125.039
2.112
-17.21
-11.32
19.723
24.086
32.294
13.67
109
127
145
143
170
194
212
17.724
19.585
46.449
-12.576
-26.126
P(o-tol)3
P(mes)3
6.872
aMeasurements made in DCE-CDCl3 solvent system, with all the values relative to external H3PO4. bΔ = δcomplex - δfree ligand
.
TABLE 5. Changes (Δ, δ, ppm) in the 31P Chemical Shifts for a Mixture of [Ag(COD)2]PF6, PR3, and 2-Phenyl N-Tosylaziridinea
ligands (PR3)
δ (ppm) free ligand
δ (ppm) complex
Δ (ppm)
cone angle (θ) (deg)
P(OEt)3
P(OPh)3
PPh3
P(p-tol)3
P(cy)3
142.495
12.007
-1.757
-4.183
14.411
-26.237
-32.703
130.572
2.301
-11.923
-9.706
18.421
23.405
31.801
13.108
6.582
109
127
145
143
170
194
212
16.664
19.222
46.212
-13.129
-26.121
P(o-tol)3
P(mes)3
aMeasurements made in DCE-CDCl3 solvent system, with all the values relative to external H3PO4.
FIGURE 3. Profile of coordination chemical shift (Δ) vs cone angle of phosphorus ligand in [Ag(COD)2]PF6 catalyst.
SCHEME 4
the steric effect of a ligand is equally pertinent as its electronic
effect. Therefore, the concept of cone angle is of practical
importance in homogeneous catalysis. Indeed, in a number
of cases, reaction rates, equilibrium constants, and NMR
chemical shifts have been found to correlate well with ligand
cone angle.25
Keeping the above in view, we desired to confirm whether
there is any ligand cone angle dependency on the metal
center in the C-arylation reaction for Ag(I) catalysts bearing
phosphine/phosphites ligands. For this study, we have cho-
sen seven phosphine/phosphite ligands encompassing a cone
angle range of 109-212°. The corresponding silver complexes
[Ag(PR3)2]PF6 were generated in situ by mixing [Ag(COD)2]-
PF6 with 2 equiv of phosphine ligand in a DCE-CDCl3
mixture. In each case 31P NMR spectra were recorded for (i)
the free ligand, (ii) the free ligand in the presence of aziridine 1a,
(iii) the in situ generated complex [Ag(PR3)2]PF6, and (iv) a
mixture of [Ag(PR3)2]PF6 and aziridine. From the chemical
shift data we determined the coordination chemical shifts Δ31P
which is equal to (δcomplex - δfree ligand) (Tables 4 and 5). The
corresponding plots of Δ31P versus θ are shown in Figure 3. In
Relating Ligand Cone Angle with 31P NMR Chemical Shift
and Initial Rate. Correlating ligand effect with efficiency is an
important exercise in the development of a catalyst. In the
case of a catalyst bearing phosphine and phosphite ligands it
has been found that the steric effect of the ligand influences
the catalytic reactivity. Ligand cone angle (also known as
Tolman cone angle, and represented by θ) is an important
parameter in understanding the steric effect. Ligand cone
angle is a measure of the size of a coordinated ligand and is
defined as the solid angle (angle in three-dimensional spaces)
formed with the metal at the vertex and the hydrogen atoms
at the perimeter of the cone. Tertiary phosphine ligands are
commonly classified by using this steric parameter. Many
studies have established that in metal-mediated coupling
reactions, where substrate-binding is an important event,
(25) (a) Tolman, C. A. Chem. Rev. 1977, 77, 313. (b) Bunten, K. A.; Chen,
L.; Fernandez, A. L.; Poe, A. J. Coord. Chem. Rev. 2002, 233, 41.
4406 J. Org. Chem. Vol. 75, No. 13, 2010

Bera and Roy
JOCArticle
FIGURE 4. Profile of (i) k vs θ and (ii) log k vs θ (θ is the cone angle of phosphorus ligand).
TABLE 6. Rate Constant for the C-Arylation of Aziridine with
N-Tosylphenylaziridine, Anisole, and Different Phosphorus Ligands
SCHEME 5. Ligand Substitution and Dissociation Steps
Involving [Ag(COD)2]PF6 and PPh3 in Solution
ligand
cone angle (deg)
rate constant 104k (min-1 a
)
log k
P(OEt)3
P(OPh)3
P(Ph)3
P(p-tol)3
P(Cy)3
P(mes)3
109
127
145
143
170
212
75.3
79.6
83.5
98.3
100.1
122.7
-2.12
-2.09
-2.07
-2.00
-1.99
-1.91
and P(mes)3 ligand (4 mol %) afforded 2 in 65% yield
after 5 h.
To gain further evidence of an initial ligand dissociation
in the early part of the mechanism, we have subjected a
mixture of [Ag(COD)2]PF6 and PPh3 (1:2 ratio) in methanol
to ESI-MS study under ES(þ) mode. The resulting spectrum
clearly indicated the formation of species [Ag(PPh3)2]þ at
m/z 631, [Ag(PPh3)(MeOH)]þ at m/z 401, and [Ag(PPh3)]þ at
m/z 368.9. This is indicative of (i) the facile replacement of
cyclooctadiene by phosphine ligands and (ii) the possibility
of [Ag(PPh3)]þ as a reactive species in solution (Scheme 5).
Diagnosis of Ag(I)-Arene Intermediate in Solution via
ESI-MS and 109Ag NMR Probe. Being a d10 system Ag(I)
is well-known to form π-complexes with olefin, alkyne, and
arene.28 In the case of an olefin, formation of a 1:1 complex
[olefin-Ag]þ is predominant; however, complexes such as
[olefin-Ag2]2þ and [(olefin)2-Ag]þ are also known in cer-
tain cases. The Ag(I)-olefin bonding follows a η2-mode and
consists of a σ-bond formed by the overlap of a vacant 5s
orbital of silver with a filled π-2p orbital of the olefin, and a
π-bond formed by the overlap of a filled 4d orbital of the
metal with a vacant π*-2p antibonding orbital of olefin.29
One may note that the metal olefin bond in Ag[(COD)2]PF6
aError <5% from 2 runs.
parallel, the reactivity of the silver(I)-phosphine complexes has
been monitored through kinetic study for the reaction of phenyl
N-tosylaziridine and anisole at 1% catalyst loading (Scheme 4).
The rate constant (k) values thus obtained (Table 6) were
plotted against cone angle (Figure 4). The pertinent results
from the above studies are highlighted below.
(1) For the Ag(I) complexes bearing phosphine ligand,
there is a positive coordination chemical shift (Δ31P). More-
over, the shift Δ31P decreases with increase in cone angle (θ).
This is consistent with a steric dependence on NMR;the
bulkier ligand being away from the metal center. In contrast,
the phosphite ligands show the opposing trend reflecting the
dependence on electronic perturbation.26
(2) For all the ligands, the rate constants for C-arylation
(k) increases with simultaneous increase in cone angle of the
phosphorus ligand (Figure 4). It is therefore evident that
the steric effect of the ligand is dominant in these reactions.
We therefore conclude that the mechanism is initiated by
prior ligand dissociation.27
(3) Since kinetic data indicated that among all the phos-
phorus ligands P(mes)3 is the most efficient, therefore a
bench-scale reaction was attempted with the corresponding
Ag(I) complex. The reaction of N-tosyl-2-phenylaziridine
with anisole in the presence of [Ag(COD)2]PF6 (2 mol %)
˚
˚
(2.50 A) is longer than that in Ni(COD)2 (2.16 A),
[Cu(COD)Cl]2 (2.05 A), and [Pt(COD)2] (2.221 A).30-32 As a
result olefin-silver(I) complexes are often more labile com-
pared to other d10 analogues. Moreover as the double bond
character in the silver olefin complexes are less perturbed,
therefore 1H NMR or vibrational spectra of the complexes do
not differ greatly from those of the free olefins.33 Keeping the
above in view, we attempted mass spectral and 109Ag NMR
˚
˚
(26) (a) Kargol, A. J.; Crecely, R. W.; Burmeister, P. J. Inorg. Chim. Acta
1980, 40, 79. (b) Miiller, T. E.; Green, J. C.; Mingos, D. N. P.; Mepartlin,
C. M.; Whittingham, C.; Williams, D. J.; Woodroffe, T. M. J. Organomet.
Chem. 1998, 551, 313.
(27) (a) Tolman, A. C. J. Am. Chem. Soc. 1970, 92, 2956. (b) Tolman,
A. C. J. Am. Chem. Soc. 1970, 92, 2953. (c) Moser, R. W.; Marshik-Guerts,
J. B.; Okrasinski, J. S. J. Mol. Catal. A: Chem. 1999, 143, 71. (d) Niyomura,
O.; Iwasawa, T.; Sawada, N.; Tokunaga, M.; Obora, Y.; Tsuji, Y. Organo-
metallics 2005, 24, 3468. (e) Socol, S. M.; Verkade, J. G. Inorg. Chem. 1984,
23, 3487. (f) Ohta, H.; Tokunaga, M.; Obora, Y.; Iwai, T.; Iwasawa, T.;
Fujihara, T.; Tsuji, Y. Org. Lett. 2007, 9, 89.
(28) Bennett, A. M. Chem. Rev. 1961, 313.
(29) Such synergistic bonding is common in compounds bearing a d-rich
metal and a π-ligand. For some historically relevant notes, please see: (a)
Winstein, S.; Lucas, H. J. J. Am. Chem. Soc. 1938, 60, 836. (b) Dewar, M. J. S.
Bull. Soc. Chim. Fr. 1951, 18, C79.
(30) Schunn, R. A.; Ittel, S. D.; Cushing, M. A. Inorg. Synth. 1990, 28, 94.
(31) Vander, J.; Hende, H.; Baird, W. C. J. Am. Chem. Soc. 1963, 85,
1003.
(32) Howard, J. A. K. Acta Crystallogr. 1982, B38, 2896.
(33) Powell, D. B.; Sheppard, N. J. Chem. Soc. 1960, 2519.
J. Org. Chem. Vol. 75, No. 13, 2010 4407

Bera and Roy
JOCArticle
SCHEME 6. Reactive Ag(I)-Arene Species in Reaction Medium
SCHEME 7
TABLE 7. Variation of Reaction Rate (k) with Diene Ligands
no.
diene ligands
104ka (min-1
)
log k
1
2
3
1,5-cyclooctadiene (COD)
2,5-norbornadiene (NBD)
1,3-cyclohexadiene (CHD)
127.1
115.1
106.6
-1.89
-1.93
-1.97
aError <5% from 2 runs.
studies to determine the nature of the species present in solution
when AgPF6 and [Ag(COD)2]PF6 interact with arene.
ESI-MS Study. Electrospray ionization mass spectrome-
try is a soft technique to investigate species in solution. By
virtue of having two naturally occurring isotopes (107Ag,
109Ag in 51:49 ratio) silver complexes also allow for easy dia-
gnosis via isotopic mass distribution pattern matching. In the
present study trimethoxybenzene (C9H12O3) was chosen as the
representative arene. The studies were conducted with metha-
nolic solutions of AgPF6 or [Ag(COD)2]PF6 mixed with tri-
methoxybenzene in 1:2 molar ratio at different cone voltage.
The ESþ mass spectrum (at cone voltage of 30 V) of a
solution of AgPF6 and trimethoxybenzene (1:2) is dominated
by a peak at m/z 274.8 corresponding to the ion [Ag-
(C9H12O3)]þ; there is also a relatively weak peak at m/z
292.8 conforming to the species [Ag(C9H12O3)(H2O)]þ. On
the contrary, the spectrum recorded at lower cone voltage
(15 V) shows a peak at m/z 442.8 corresponding to the ion
[Ag(C9H12O3)2]þ beside the peak at m/z 274.8. These results
demonstrate that, in the absence of other ligand, the bis-
arene complex ion [Ag(arene)2]þ is labile in solution and
dissociates to monoarene complex ion [Ag(arene)]þ.
FIGURE 5. Comparison of (log k) for different diene ligands
(COD, NBD, CHD).
benzene. In the presence of trimethoxybenzene, the 109Ag
signal remarkably shifted to 95 ppm.35 The shift in the Ag
NMR indicates that trimethoxybenzene forms a complex
with the Ag(I).
Effect of Diene Ligand (COD, NBD, 1,3-CHD) on the
Efficiency of Ag(I) Catalyst. Since [Ag(COD)2]PF6 proved to
be a good catalyst, we explored the efficacy of other diene
ligands on reaction medium. Accordingly, we attempted the
kinetic study for the reaction of phenyl N-tosylaziridine and
anisole in the presence of 1 mol % AgPF6 and 2 mol % diene
ligand (Scheme 7). The study was conducted under pseudo-
The ESþ mass spectra of [Ag(COD)2]PF6 and trimethoxy-
benzene (1:2) at high cone voltage (30 V) show a major peak
at m/z 274.8 and a weak peak at m/z 214.9 and are assigned to
the species [Ag(C9H12O3)]þ and [Ag(C8H12)]þ, respectively,
whereas the spectrum at low cone voltage (15 V) shows
a major peak at 383.1 due to Ag[(C8H12)(C9H12O3)]þ, and
weak peaks at 323.1 and 443.1 assigned to [Ag(C8H12)2]þ and
Ag[(C9H12O3)]2þ respectively.
1
first-order condition and by monitoring the product by H
NMR for the initial part of the reaction (with diphenyl-
methyltoluene as an internal reference). The rate data
(Table 7, Figure 5) demonstrate that the catalytic efficiency
decreases in the order COD>NBD>CHD.
As in the case of COD, the formation of silver-NBD and
CHD complex was confirmed by ESI-MS study, which was
carried out by mixing AgPF6 and the diene ligand in 1:2
molar ratio in methanol solvent at different cone voltages. In
case of 1,3-cyclohexadiene (CHD) and at high cone voltage
From the above we conclude that Ag(I)-arene binding
remains dominant even in the presence of a diene, with the
reactive species in solution being [Ag(diene)(arene)]þ and
[Ag(arene)2]þ. It is reasonable to suppose that in solution the
Ag(I)-arene species (Scheme 6) could remain in exchange.
However, from ESI-MS studies alone, the position of these
equilibria cannot be estimated.34
109Ag NMR Study. To further look into the Ag(I)-arene
binding, we recorded the 109Ag NMR of saturated solutions
AgPF6 in CDCl3 before and after the addition of trimethoxy-
(35) It is noteworthy that among the two NMR active nuclei of silver,
namely 107Ag and 109Ag, the latter is more sensitive; hence it is a useful
mechanistic tool. The experiment was conducted with saturated solutions of
trimethoxybenzene and AgPF6 in CDCl3. The 109Ag spectrum of AgPF6
showed a single peak that was set as zero. Upon addition of trimethoxyben-
zene to the above, the 109Ag signal shifted to 95 ppm. We could not record the
109Ag NMR spectrum of [Ag(COD)2]PF6 due to solubility issues. Please see:
(a) Goodfellow, R. J., In Multinuclear NMR; Mason, J., Ed.; Plenum Press:
New York, 1983; pp 563-589. (b) Letinois-Halbes, U.; Pale, P.; Berger, S. J.
J. Org. Chem. 2005, 70, 9185.
(34) One may recollect that ESI-MS studies only detect the ionic species in
a solution. However, there may well be equilibria between ionic and neutral
components. Please see: Canty, J. A.; Colton, R. Inorg. Chim. Acta 1994,
220, 99.
4408 J. Org. Chem. Vol. 75, No. 13, 2010

Bera and Roy
JOCArticle
SCHEME 8. Catalytic Cycle of C-Arylation of N-Tosylaziridine
(30 V), the spectrum showed a base peak at m/z 267.3 for
[Ag(CHD)2]þ, along with peaks at 187.3 and 219.3 for
[Ag(CHD)]þ and [Ag(CHD)(MeOH)]þ species, respectively.
But at low cone voltage (15 V) the most dominant peak
appeared at 187.3 corresponding to [Ag(CHD)]þ along with
a minor peak due to [Ag(CHD)(MeOH)]þ. It may be noted
that for 2,5-norbornadiene (NBD), the spectrum obtained at
both high and low cone voltage were similar, and consisted of
a dominant peak at 199.3 for [Ag(NBD)]þ along with a
minor peak due to [Ag(MeOH)(NBD)]þ (please see the
Supporting Information).
Proposed Catalytic Cycle. By correlating k, σ, θ, and δ,
and from the MS and NMR diagnostics, a mechanism is
proposed for the present Ag(I) catalyzed arylation reaction
(Scheme 8). A few important features of the mechanism are
highlighted below.
(1) In the reaction medium phosphine ligand (PR3) re-
places the COD ligand from [Ag(COD)2]þ species to afford
[Ag(PR3)2]þ. The latter undergoes dissociation to generate
the monocoordinated [Ag(PR3)]þ species.
(2) In reactions where [Ag(COD)2]PF6 alone is used as the
catalyst, [Ag(COD)]þ is the reactive species.
(3) The Ag(I) species having coordinative unsaturation
(such as [Ag(PR3)]þ) is capable of further coordination with
both aziridine and arene.
(4) The coupling between the two organic partners is
facilitated by their close proximity in the Ag-bound activated
complex.
FIGURE 6. Molecular orbital picture of Ag(I)-diene complexes.
DFT Study. To rationalize the observed reactivity pattern
of the Ag(I)-diene complexes as discussed above we have
performed DFT-calculation to find their HOMO-LUMO
energy gap, the latter being a measure of the chemical
hardness (η) of the respective complexes. The DFT calcula-
tions for the three different Ag(I)-diene complexes were
performed with the Gaussian 03 package.36 The molecular
structure and orbital shapes were visualized by using Gaussian
view 03 software. The ground state geometries of Ag(I)-
diene complexes were optimized by using a PBE1PBE approx-
imation with 6-31G basis set for C and H and sdd basis set for
Ag. The geometry of the Ag(COD)2þ system was taken from
experimental X-ray data.
The values of chemical hardness (η) depend on HOMO-
LUMO energy gap.37 Hard molecules have a larger HOMO-
LUMO gap and soft molecules have a small HOMO-LUMO
(36) Frisch, M. J.; et al. Gaussion 03, Revision C.02; Gaussion, Inc.:
Wallingford, CT, 2004.
(37) Pearson, G. R. Acc. Chem. Res. 1993, 26, 250 and references cited
therein.
J. Org. Chem. Vol. 75, No. 13, 2010 4409

Bera and Roy
JOCArticle
gap. The HOMO-LUMO orbital picture and energy plot of
the Ag(I)-diene complexes are shown in Figure 6, from which
it is evident that η decreases in the order [Ag(COD)2]þ > [Ag-
(NBD)2]þ>Ag(CHD)2þ. Notably the hardness order also reci-
procates the order of catalytic efficiency. One may appreciate
the relationship, since coordination of both aziridine and arene
at the Ag(I) metal center is an important event in the mecha-
nistic cycle.
General Procedure for the Preparation of N-Tosylaziridine
(1a-d). To a mixture of styrene (1.16 mL, 10 mmol) and
chloramin-T (3.098 g, 11 mmol) in acetonitrile (50 mL) was
added phenyltrimethylammonium tribromide (376 mg, 1 mmol)
at room temperature. After vigorous stirring for 12 h, the
reaction mixture was diluted with ethyl acetate (100 mL) and
water (60 mL). The organic layer was separated, washed with
brine, and dried over anhydrous Na2SO4. After evaporation of
the excess solvent, the product was purified by column chroma-
tography on silica gel.
Preparation of Ag[(COD)2]PF6. To an aqueous solution of
AgPF6 (0.5 mmol in 3 mL of water) was added a methanolic
solution of 1,5-cyclooctadiene (1 mmol in 0.5 mL of methanol)
at room temperature in the absence of light. The reaction
completed just after 5 min and the residue was filtered and
washed with water and hexane. The white solid was dried under
vacuum and was further recrystallized from dichloromethane/
benzene solvent system.
3. Summary
In summary, we have demonstrated that a number
of silver(I) complexes of the type [Ag(diene)2]þY- (where
diene = cyclooctadiene, norbornadiene, and 1,3-cyclohexa-
diene; Y- = PF6 -, BF4-) act as efficient catalysts for the
C-arylation of N-tosylaziridines with arenes and heteroarenes
under ambient condition to provide the corresponding β-aryl
amine derivatives with excellent regioselectivity. To understand
the nature of substrate activation, and initial bond breaking/
making steps, we have performed the following studies with the
help of in situ NMR (1H, 31P, 109Ag) and ESI-MS probe: (I)
evaluation of Hammett reaction constant (F); (II) correlation of
initial rate (k) versus cone angle (θ) of ligand L for reactions
mediated by [Ag(COD)2]PF6/L (where L is a phosphine or
phosphite ligand); (III) identification of silver-arene inter-
mediates in solution; and (IV) correlation of the initial rate
(k) with ΔHOMO-LUMO of [Ag(diene)2]PF6 obtained from
preliminary DFT studies. The proposed catalytic cycle invokes
a ligand dissociation step as a key step to generate the active
catalytic species; the latter then binds both the aziridine and
the arene and acts as an anchor for the final coupling stage. The
various correlations involving k, σ, θ, and δ values and the
results from the MS and NMR studies are in conformity to this
proposal. We believe that the mechanistic description presented
in this account will aid in future development of catalysts having
superior efficiency and selectivity.
Anal. Calcd for C16H26OAgPF6: C, 39.44; H, 5.38. Found: C,
39.46; H, 5.04.
MS (m/z): Ag[(C8H12)2]þ at 323 at low cone voltage (15 V),
Ag(C8H12)þ at 215 at high cone voltage (35 V).
Preparation of [Ag(COD)2]BF4. To a suspension of AgBF4
(1 g, 5 mmol) in 40 mL of dry dichloromethane under nitrogen at
-10 °C was added 12 mL of 1,5-cyclooctadiene through a side
arm fitted with a serum cap. After 12 h a white crystalline
precipitate was formed. This was filtered off and dried under
vacuum. The mother liquor was concentrated under a gentle
flow of nitrogen to get a second crop of crystals.
General Procedure for the C-Arylation of Tosylaziridine. A
mixture of N-tosylaziridine (1 mmol), arene or heteroarene
(2 mmol), and [Ag(COD)2]PF6 (2 mol %) in 3 mL of dry
dichloroethane was stirred at room temperature. Following
completion (vide TLC) the reaction mixture was diluted with
water (5 mL) and extracted with dichloromethane. The com-
bined organic layer was dried over anhydrous sodium sulfate
then concentrated under vacuum, and the resulting product was
purified by column chromatography on silica gel (100-200
mesh, ethyl acetate-petroleum ether, 1:10) to afford pure β-aryl
ethylamine derivatives.
Spectral Data of All Compounds. N-[2-(4-Methoxyphenyl)-2-
phenylethyl]-4-methylbenzenesulfonamide (2): 1H NMR (400
MHz, CDCl3) δ 2.46 (s, 3H), 3.46-3.57 (m, 2H), 3.78 (s, 3H),
4.02 (t, J = 8 Hz, 1H), 4.29 (t, J = 6 Hz, 1H), 6.82 (d, J = 8.8
Hz, 2H), 7.01 (d, J = 8.4 Hz, 2H), 7.09 (d, J = 6.8 Hz, 2H),
7.20-7.33 (m, 5H), 7.74 (d, J = 8.4 Hz, 2H); 13C NMR
(100 MHz, CDCl3) δ 21.6, 47.4, 49.7, 55.3, 114.3, 127.1, 127.2,
127.8, 128.9, 128.9, 129.8, 132.6, 136.7, 141.0, 143.6, 158.6;
HRMS (ESI) calcd for C22H23NO3S [M þ Na]þ 404.1296,
found 404.1293; IR (KBr) 3283, 2932, 1512, 1324, 1153, 1092,
4. Experimental Section
1
General Comments. H NMR spectra were recorded at 400
and 200 MHz. Chemical shifts are reported in ppm from
tetramethylsilane with the solvent resonance as the internal
standard (chloroform: δ 7.26 ppm). Data are reported as
follows: chemical shifts, multiplicity (s = singlet, d = doublet,
t = triplet, q = quartet, br = broad, m = multiplet), integra-
tion, coupling constant (Hz). 13C NMR spectra were recorded at
100 MHz and 54.6 MHz with proton decoupling. Chemical
shifts are reported in ppm from tetramethylsilane with the
solvent resonance as the internal standard (chloroform: δ 77.0
ppm). ESI-MS experiments were conducted at þve ionization mode.
All reactions were carried out under an argon atmosphere in
flame-dried glassware, using Schlenk techniques. Chromato-
graphic purifications were done with either 60-120 or 100-200
mesh silica gel. For reaction monitoring, precoated silica gel
60 F254 TLC sheets were used. Petroleum ether refers to the
fraction boiling in the range 60-80 °C. Dichloroethane was
dried and distilled prior to use.
The silver salt of hexafluorophosphate was prepared by a
metathetic reaction between equimolar amounts of potassium
hexafluorophosphate and silver nitrate in dry acetonitrile med-
ium at room temperature. The reaction completed in just 5 min
after which the residue was filtered under argon atmosphere.
The filtrate was slowly concentrated under vacuum, when the
product separated out as a white salt. It was further recrystal-
lized from acetone/methanol mixture.
820 cm-1
.
4-Methyl-N-(2-phenyl-2-p-tolylethyl)benzenesulfonamide (3):
1H NMR (400 MHz, CDCl3) δ 2.28 (s, 3H), 2.45 (s, 3H), 3.52
(t, J = 7.6 Hz, 2H), 4.00 (d, J = 7.6 Hz, 1H), 4.26 (s, 1H), 6.97 (d,
J = 8.0 Hz, 2H), 7.07 (d, J = 7.6 Hz, 2H), 7.20-7.31 (m,
5H), 7.29 (d, J = 8.0 Hz, 2H), 7.69 (d, J = 8.0 Hz, 2H); 13C
NMR (100 MHz, CDCl3) δ 20.9, 21.5, 47.2, 50.1, 126.9,
127.1, 127.8, 127.9, 128.8, 129.5, 129.7, 131.1, 136.7, 137.5,
140.8, 143.5; HRMS (ESI) calcd for C22H23NO2S [M þ H]þ
366.1528, found 366.1507; IR (KBr) 3287, 2924, 1598, 1322,
1154, 1093, 816 cm-1
.
4-Methyl-N-[2-(3,4-dimethylphenyl)-2-phenylethyl]benzene-
sulfonamide (4): 1H NMR (400 MHz, CDCl3) δ 2.20 (s, 3H), 2.24
(s, 3H), 2.45 (s, 3H), 3.46-3.55 (m, 2H), 3.98 (t, J= 8 Hz, 1H), 4.28
(br s, 1H), 6.79-6.88 (m, 2H), 7.02-7.09 (m, 3H), 7.18-7.32 (m,
5H), 7.68 (d, J = 8 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 19.3,
19.8, 21.5, 47.2, 50.0, 125.1, 126.9, 127.1, 128.1, 128.7, 128.7, 129.2,
129.7, 135.4, 136.7, 137.0, 137.9, 138.0, 140.8; DEPT-135 NMR δ
4410 J. Org. Chem. Vol. 75, No. 13, 2010

Bera and Roy
JOCArticle
19.3 (CH3), 19.9 (CH3), 21.6 (CH3), 47.3 (CH2), 50.1 (CH), 125.2
(CH), 126.9 (CH), 127.8 (CH), 128.2 (CH), 128.8 (CH), 129.3
(CH), 129.7 (CH), 130.1 (CH); HRMS (ESI) calcd for C23H25-
NO2S [M þ Na]þ 402.1504, found 402.1489; IR (KBr) 3281, 2919,
39.1, 45.0, 55.2, 55.4, 91.0, 108.4, 127.0, 128.0, 129.1, 129.4,
131.5, 136.9, 140.4, 142.9, 159.1, 160.4; DEPT-135 NMR δ 21.5
(CH3), 39.2 (CH), 45.1 (CH2), 55.3 (CH3), 55.5 (CH3), 91.1
(CH), 127.1 (CH), 128.1 (CH), 129.2 (CH), 129.5 (CH); HRMS
(ESI) calcd for C24H26NO5SCl [M þ H]þ 476.1298, found
476.1292; IR (KBr) 3257, 2941, 2841, 1608, 1327, 1158, 1092,
1598, 1327, 1156, 1092, 817 cm-1
.
N-[2-(2,5-Dimethoxyphenyl)-2-phenylethyl]-4-methylbenzene-
sulfonamide (5): H NMR (400 MHz, CDCl3) δ 2.43 (s, 3H),
1
814 cm-1
.
3.46-3.57 (m, 2H), 3.67 (s, 3H), 3.69 (s, 3H), 4.46 (t, J = 8 Hz,
2H), 6.50 (d, J = 2.8 Hz, 1H), 6.68-6.77 (m, 2H), 7.10 (d, J =
7.2 Hz, 2H), 7.17-7.29 (m, 5H), 7.65 (d, J = 8 Hz, 2H); 13C
NMR (100 MHz, CDCl3) δ 21.5, 43.5, 46.2, 55.5, 55.9, 111.5,
111.7, 114.9, 126.8, 127.1, 128.1, 128.6, 129.6, 130.3, 136.7,
140.3, 143.2, 151.2, 153.5; DEPT-135 NMR 21.6 (CH3), 43.5
(CH), 46.3 (CH2), 55.6 (CH3), 56.0 (CH3), 111.6 (CH), 111.8
(CH), 115.0 (CH), 126.9 (CH), 127.2 (CH), 128.2 (CH), 128.7
(CH), 129.7 (CH); HRMS (ESI) calcd for C23H25NO4S [M þ
Na]þ 434.1402, found 434.1386; IR (KBr) 3261, 2928, 2835,
N-[2-(4-Methoxyphenyl)-2-p-tolylethyl]-4-methylbenzenesul-
fonamide (11): 1H NMR (400 MHz, CDCl3) δ 2.29 (s, 3H), 2.44
(s, 3H), 3.48 (t, J = 6.8 Hz, 2H), 3.76 (s, 3H), 3.96 (t, J = 8 Hz,
1H), 4.25 (t, J = 6 Hz, 1H), 6.79 (d, J = 7.6 Hz, 2H), 6.94-7.00
(m, 4H), 7.07 (d, J = 7.2 Hz, 2H), 7.30 (d, J = 7.6 Hz, 2H), 7.67
(d, J = 7.6 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 20.9, 21.5,
47.4, 49.2, 55.2, 114.1, 127.1, 127.6, 128.8, 129.5, 129.5, 132.8,
136.6, 136.7,137.9, 143.4, 158.5; HRMS (ESI) calcd for
C23H25NO3S [M þ H]þ 396.1633, found 396.1624; IR (KBr)
3284, 2932, 2813, 1508, 1322, 1153, 1094, 819 cm-1
.
1596, 1329, 1162, 1092, 817 cm-1
.
4-Methyl-N-(2-phenyl-2-thiophen-2-ylethyl)benzenesulfonamide
(12): 1H NMR (400 MHz, CDCl3) δ 2.43 (s, 3H), 3.46-3.60 (m,
2H), 4.28 (t, J = 8 Hz, 1H), 4.40 (s, 1H), 6.78 (s, 1H), 6.92 (s, 1H),
4-Methyl-N-[2-phenyl-2-(2,4,6-trimethoxyphenyl)ethyl]benze-
nesulfonamide (6): 1H NMR (400 MHz, CDCl3) δ 2.42 (s, 3H),
3.61 (s, 6H), 3.66-3.76 (m, 2H), 3.79 (s, 3H), 4.38 (br s, 1H), 4.65
(t, J = 7.2 Hz, 1H), 6.05 (s, 2H), 7.09-7.20 (m, 5H), 7.25 (d, J =
6.5 Hz, 2H), 7.66 (d, J = 8 Hz, 2H); 13C NMR (100 MHz,
CDCl3) δ 21.4, 39.6, 45.2, 55.2, 55.4, 91.0, 108.8, 125.9, 127.1,
127.7, 127.9, 129.4, 137.0, 141.8, 142.8, 159.2, 160.3; HRMS
(ESI) calcd for C24H27NO5S [M þ Na]þ 464.1508, found
464.1499; IR (KBr) 3289, 2939, 2839, 1610, 1325, 1164, 1094, 809
7.13-7.21 (m, 2H), 7.23-7.37 (m, 6H), 7.70 (d, J = 8 Hz, 2H); 13
C
NMR (100 MHz, CDCl3) δ 21.5, 46.4, 48.4, 124.6, 124.9, 126.3,
127.2, 127.7, 127.8, 128.9, 129.7, 136.7, 140.4, 143.6, 144.2; DEPT-
135 NMR δ 21.6 (CH3), 46.5 (CH), 48.5 (CH2), 124.7 (CH), 125.0
(CH), 126.4 (CH), 127.3 (CH), 127.8 (CH), 127.9 (CH), 128.9
(CH), 129.8 (CH); HRMS (ESI) calcd for C19H19NO2S2 [M þ
Na]þ 380.0755, found 380.0762; IR (KBr) 3292, 2923, 2842, 1598,
cm-1
.
N-[2-(4-Methoxynaphthyl)-2-phenylethyl]-4-methylbenzene-
1321, 1154, 1092, 816 cm-1
.
4-Methyl-N-(2-phenyl-2-furan-2-ylethyl)benzenesulfonamide
(13): 1H NMR (400 MHz, CDCl3) δ 2.44 (s, 3H), 3.37-3.44 (m,
1H), 3.55-3.61 (m, 1H), 4.09 (t, J = 7.6 Hz, 1H), 4.44 (s, 1H), 6.01
(d, J = 3.2 Hz, 1H), 6.27 (d, J = 3.2 Hz, 1H), 7.12 (t, J = 8 Hz,
2H), 7.23-7.36 (m, 6H), 7.69 (d, J = 8.4 Hz, 2H); 13C NMR (100
MHz, CDCl3) δ 21.5, 45.1, 46.5, 107.0, 110.2, 127.0, 127.5, 127.9,
128.8, 129.8, 136.8, 138.6, 142.0, 143.5, 154.0; DEPT-135 NMR δ
21.5 (CH3), 45.2 (CH), 46.6 (CH2), 107.1 (CH), 110.3 (CH), 127.1
(CH), 127.6 (CH), 128.1 (CH), 128.9 (CH), 129.8 (CH), 142.1
(CH); HRMS (ESI) calcd for C19H19NO3S [M þ Na]þ 364.0983,
1
sulfonamide (7): H NMR (200 MHz, CDCl3) δ 2.42 (s, 3H),
3.64-3.66 (d, J = 3.6 Hz, 2H), 3.98 (s, 3H), 4.61 (br s, 1H),
4.71-4.79 (t, J = 8 Hz, 1H), 6.71-6.75 (d, J = 8 Hz, 1H),
7.11-7.30 (m, 7H), 7.35-7.45 (m, 3H), 7.56-7.61 (d, J = 8.4 Hz,
2H), 7.68-7.77 (m, 1H), 8.26-8.30 (m, 1H); 13CNMR(54.6MHz,
CDCl3) δ 21.6, 45.7, 47.2, 55.5, 103.0, 122.8, 123.1, 124.4, 125.1,
126.3, 126.9, 127.0, 127.2, 128.1, 128.1, 128.9, 129.8, 132.6, 136.9,
141.2, 143.4, 154.9; HRMS (ESI) calcd for C26H25NO3S[Mþ Na]þ
454.1453, found 454.1452.
N-[2-(3,4,5-Trymethoxyphenyl)-2-phenylethyl]-4-methylben-
zenesulfonamide (8): 1H NMR (200 MHz, CDCl3) δ 2.43 (s, 3H),
3.47 (d, J = 8 Hz, 2H), 3.60 (s, 3H), 3.81 (s, 6H), 4.33 (t, J = 8
Hz, 1H), 6.54 (d, J = 8.6 Hz, 1H), 6.68 (d, J = 8.6 Hz, 1H), 7.07
(d, J = 9.4 Hz, 2H), 7.13-7.30 (m, 5H), 7.65 (d, J = 8.2 Hz, 2H);
13C NMR (54.6 MHz, CDCl3) δ 21.5, 43.7, 46.7, 55.9, 60.7,
107.3, 122.0, 126.8, 127.2, 128.0, 128.7, 129.7, 136.9, 141.3,
142.5, 143.4, 151.9, 152.8; DEPT-135 NMR δ 21.5 (CH3), 43.7
(CH), 46.7 (CH2), 55.9 (OCH3), 60.7 (OCH3), 107.3 (CH), 122.0
(CH), 126.8 (CH), 127.1 (CH), 128.0 (CH), 128.7 (CH), 129.7
(CH); HRMS (ESI) calcd for C24H27NO5S [M þ H]þ 442.1683,
found 442.1684.
found 364.0977; IR (KBr) 3448, 1637, 1328, 1160, 1093, 814 cm-1
.
4-Methyl-N-[2-(5-methylthiophen-2-yl)-2-phenylethyl]benzene-
1
sulfonamide (14): H NMR (400 MHz, CDCl3) δ 2.39 (s, 3H),
2.44 (s, 3H), 3.50-3.59 (m, 2H), 4.11 (t, J = 8 Hz, 1H), 4.42 (s,
1H), 6.54 (s, 2H), 7.21 (d, J = 8.8 Hz, 2H), 7.24-7.41 (m, 5H),
7.73 (d, J = 8.4 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 15.2,
21.5, 46.5, 48.3, 124.8, 127.1, 127.6, 127.8, 128.7, 129.7, 136.7,
139.2, 140.6, 141.8, 143.5; DEPT-135 NMR δ 15.4 (CH3), 21.6
(CH3), 46.6 (CH), 48.4 (CH2), 124.9 (CH), 127.1 (CH), 127.5
(CH), 127.7 (CH), 128.7 (CH), 128.8 (CH), 129.8 (CH); HRMS
(ESI) calcd for C20H21NO2S2 [M þ H]þ 372.1092, found
372.1075; IR (KBr) 3291, 2919, 2815, 1598, 1322, 1154, 1093,
816 cm-1
.
N-[2-(4-Chlorophenyl)-2-(4-methoxyphenyl)ethyl]-4-methyl-
benzenesulfonamide (9): 1H NMR (400 MHz, CDCl3) δ 2.44 (s,
3H), 3.45-3.51 (m, 2H), 3.77 (s, 3H), 4.00 (t, J = 8 Hz, 1H), 4.26
(t, J = 6 Hz, 1H), 6.81 (d, J = 8.4 Hz, 2H), 6.96-7.04 (m, 4H),
7.19-7.31 (m, 6H), 7.67 (d, J = 8.4 Hz, 2H); 13C NMR (100
MHz, CDCl3) δ 21.5, 47.2, 49.1, 55.3, 114.3, 127.0, 129.1, 129.4,
129.6, 129.7, 132.0, 132.6, 136.6, 139.6, 143.5, 158.7; DEPT-135
NMR δ 21.6 (CH3), 47.3 (CH2), 49.2 (CH), 55.3 (CH3), 114.4
(CH), 127.1 (CH), 128.9 (CH), 128.9 (CH), 129.2 (CH),
129.8 (CH); HRMS (ESI) calcd for C22H22ClNO3S [MþH]þ
416.1087, found 416.1095; IR (KBr) 3286, 2932, 2835, 1512,
4-Methyl-N-[2-(5-methylfuran-2-yl)-2-phenylethyl]benzene-
sulfonamide (15): 1H NMR (400 MHz, CDCl3) δ 2.21 (s, 3H),
2.44 (s, 3H), 3.27-3.41 (m, 1H), 3.46-3.56 (m, 1H), 4.01 (t,
J = 8 Hz, 1H), 4.46 (s, 1H), 5.85 (s, 2H), 7.10 (d, J = 8.4 Hz,
2H), 7.16-7.40 (m, 5H), 7.71 (d, J = 8.4 Hz, 2H); 13C NMR
(100 MHz, CDCl3) δ 13.8, 21.5, 45.1, 46.6, 106.0, 107.7, 127.0,
127.4, 127.9, 128.8, 129.7, 136.9, 138.9, 143.4, 151.7, 152.0;
DEPT-135 NMR δ 13.6 (CH3), 21.6 (CH3), 45.2 (CH), 46.7
(CH2), 106.1 (CH), 107.8 (CH), 127.1 (CH), 127.5 (CH), 128.0
(CH), 128.8 (CH), 129.8 (CH); HRMS (ESI) calcd for C20H21
-
1322, 1152, 1093, 819 cm-1
.
NO3S [M þ Na]þ 378.1140, found 378.1134; IR (KBr) 3298,
2921, 2815, 1598, 1320, 1153, 1093, 816 cm-1
.
N-[2-(4-Chlorophenyl)-2-(2,4,6-trymethoxyphenyl)ethyl]-4-
methylbenzenesulfonamide (10): 1H NMR (400 MHz, CDCl3) δ
2.42 (s, 3H), 3.61 (s, 6H), 3.64-3.68 (m, 2H), 3.78 (s, 3H), 4.38 (t,
J = 6 Hz, 1H), 4.61 (t, J = 8 Hz, 1H), 6.04 (s, 2H), 7.06 (d, J =
8.4 Hz, 2H), 7.13 (d, J = 8.4 Hz, 2H), 7.25 (d, J = 8 Hz, 2H),
7.63 (d, J = 8.4 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 21.4,
4-Methyl-N-(2-phenyl-2-benzothiophen-2-ylethyl)benzene-
sulfonamide (16): 1H NMR (200 MHz, CDCl3) δ 2.43 (s, 3H),
3.54-3.74 (m, 2H), 4.35-4.53 (m, 1H), 7.01-7.53 (m, 10H),
7.65-7.70 (m, 1H), 7.70-7.84 (m, 3H); 13C NMR (54.6 MHz,
CDCl3) δ 21.5, 45.1, 47.3, 122.4, 122.8, 124.1, 124.6, 127.1,
J. Org. Chem. Vol. 75, No. 13, 2010 4411

Bera and Roy
JOCArticle
127.5, 127.9, 128.1, 128.9, 129.0, 129.8, 135.4, 136.9, 139.6,
NMR spectroscopy we monitored the benzylic protons and area
under benzylic protons with respect to standard was measured
under several time intervals. We have taken 6 points at 20 min
intervals, from which the k-values were determined.
Hammett Study with Respect to Aziridine: Experimental Pro-
cedure. In an NMR tube 0.08 mmol of para-substituted N-
tosylaziridine was mixed with 0.0008 mmol of [Ag(COD)2]PF6
in 400 μL of CDCl3 solvent. Then 100 μL of anisole was added.
Here also diphenylmethyltoluene is used as internal standard
(for each experiment a fixed amount 4 mg was added). The
sample was analyzed by 1H NMR spectroscopy by monitoring
the benzylic protons as described before.
140.5, 143.6; LRMS (ESI)
430.0491.
N-[2-(4-Chlorophenyl)-2-furan-2-ylethyl]-4-methylbenzene-
C
23H21NO2S2 [M þ Na]þ
1
sulfonamide (17): H NMR δ 2.44 (s, 3H), 3.35-3.42 (m, 1H),
3.51-3.58 (m, 1H), 4.09 (t, J = 7.6 Hz, 1H), 4.43 (t, J = 6.4 Hz,
1H), 6.01 (d, J = 3.2 Hz, 1H), 6.29 (s, 1H), 7.05 (d, J = 8.4 Hz,
2H), 7.23-7.32 (m, 5H), 7.67 (d, J = 8.4 Hz, 2H); 13C NMR
(100 MHz, CDCl3) δ 21.5, 44.6, 46.4, 107.2, 110.3, 127.0, 128.9,
129.3, 129.7, 133.3, 136.8, 137.2, 142.2, 143.6, 153.4; DEPT-135
NMR δ 21.6 (CH3), 44.7 (CH), 46.5 (CH2), 107.3 (CH), 110.4
(CH), 127.1 (CH), 129.0 (CH), 129.3 (CH), 129.8 (CH), 142.3
(CH); HRMS(ESI) calcdfor C19H18NO3SCl [M þ H]þ 376.0774,
found 376.0772; IR (KBr) 3288, 2925, 1597, 1322, 1153, 1093,
ESI-MS Study. Electrospray mass spectra were obtained by
using a methanol mobile phase. The solutions of the compounds
were prepared as follows:
819 cm-1
.
N-[2-(4-Chlorophenyl)-2-thiophen-2-ylethyl]-4-methylbenze-
nesulfonamide (18): 1H NMR (400 MHz, CDCl3) δ 2.44 (s, 3H),
3.43-3.57 (m, 2H), 4.29 (t, J = 7.6 Hz, 1H), 4.38 (t, J = 6.4 Hz,
1H), 6.77 (d, J = 3.2 Hz, 1H), 6.92 (d, J = 3.6 Hz, 1H), 7.08 (d,
J = 8.4 Hz, 2H), 7.18-7.33 (m, 5H), 7.67 (d, J = 8.4 Hz, 2H);
13C NMR (100 MHz, CDCl3) δ 21.5, 45.9, 48.3, 124.8, 125.0,
126.9, 127.0, 128.9, 129.2, 129.8, 133.3, 136.7, 138.9, 139.0,
143.6; HRMS (ESI) calcd for C19H18ClNO2S2 [M þ H]þ
392.0546, found 392.0556; IR (KBr) 3285, 1597, 1321, 1153,
(i) For silver-phosphine complex: First 5 ꢀ 10-3 mmol (2.5
mg) of [Ag(COD)2]PF6 and 10 ꢀ 10-3 mmol (2.62 mg) of PPh3
were added to 1 mL of methanol (HPLC grade). Then 100 μL of
this stock solution was diluted with methanol in 1:10 v/v ratio.
The resulting solution was injected directly into the mass
spectrometer.
(ii) For silver-trimethoxy benzene complex: (a) Set-1: First 6 ꢀ
10-3 mmol (1.5 mg) of AgPF6 and 12 ꢀ 10-3 mmol (2 mg) of
C9H12O3 were added to 1 mL of methanol (HPLC grade). Then
100 μL of this stock solution was diluted with methanol in
1:10 v/v ratio. The resulting solution was injected directly into
the mass spectrometer. (b) Set-2: First 5 ꢀ 10-3 mmol (2.5 mg) of
[Ag(COD)2]PF6 and 10 ꢀ 10-3 mmol (1.5 mg) of C9H12O3 were
added to 1 mL of methanol (HPLC grade). Then 100 μL of this
stock solution was diluted with methanol in 1:10 v/v ratio. The
resulting solution was injected directly into the mass spectro-
meter.
1093, 819 cm-1
.
N-[2-(4-Chlorophenyl)-2-benzofuran-2-ylethyl]-4-methylben-
zenesulfonamide (19): 1H NMR (400 MHz CDCl3) δ 2.42 (s, 3H),
3.54-3.64 (m, 2H), 4.15-4.17(t, J = 5.2 Hz, 1H), 4.34 (br s,
1H), 6.68-6.72 (d, J = 16 Hz, 1H), 6.91-7. 04 (m, 1H), 7.06-
7.12 (m, 2H), 7.15-7.33 (m, 7H), 7.60-7.67 (m, 2H); 13C NMR
(100 MHz CDCl3) δ 21.5, 47.4, 50.0, 106.3, 110.6, 121.5, 122.8,
127.0, 128.0, 128.9, 129.1, 129.2, 129.8, 136.6, 139.4, 143.6,
145.0, 145.4, 145.7; HRMS (ESI) calcd for C23H20ClNO3S
[M þ Na]þ 448.0750, found 448.0750.
The solutions of the compounds were injected directly into the
spectrometer via a Rheodyne injector equipped with a 10 μL
loop. A 20 micro LC syringe pump delivered the solution to the
vaporization nozzle of the electrospray ion source at a flow rate
of 3 μL min-1. Nitrogen was used both as a drying gas and
for nebulization with flow rates of approximately 3 L min-1 and
100 mL min-1, respectively.
N-(2-Furan-2-yl-2-p-tolylethyl)-4-methylbenzenesulfonamide
(20): 1H NMR (400 MHz, CDCl3) δ 2.31 (s, 3H), 2.43 (s, 3H),
3.34-3.41 (m, 1H), 3.53-3.60 (m, 1H), 4.04 (t, J = 7.6 Hz, IH),
4.41 (t, J = 6.4 Hz, 1H), 5.98 (s, 1H), 6.27(s, 1H), 6.96 (d, J = 8.4
Hz, 2H), 7.09 (d, J = 8 Hz, 2H), 7.26-7.35 (m, 3H), 7.68 (d, J =
8 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 20.9, 21.5, 44.7, 46.5,
106.8, 110.2, 127.0, 127.8, 129.5, 129.7, 135.5, 136.9, 137.2,
141.9, 143.4, 154.2; HRMS (ESI) calcd for C20H21NO3S [M þ
H]þ 356.1320, found 356.1347; IR (KBr) 3448, 2923, 1638, 1513,
109Ag NMR Study. The 109Ag spectra were recorded on a
400 MHz spectrometer, using a 5 TBI probe head at a frequency
of 18.622 MHz to obtain the signal of the desired intensity; 1 M
solution of AgPF6 in CDCl3 was used.
1328, 1160, 1093, 813 cm-1
.
Kinetic Study by Varying Diene Ligands: Experimental Pro-
cedure. In an NMR tube 0.08 mmol of N-tosylaziridine 1a (22
mg) was mixed with 0.0008 mmol of AgPF6 (0.20 mg, which was
taken from a stock solution of catalyst in CDCl3) in 400 μL of
CDCl3. To this solution was added 0.0016 mmol of the respec-
tive diene ligand, 100 μL of anisole (to maintain pseudo-first-
order condition), and diphenylmethyltoluene as as internal
reference (in each kinetic run a fixed amount of 4 mg of the
N-(1,2-Diphenyl-2-thiophen-2-ylethyl)-4-methylbenzenesulfo-
namide (21): 1H NMR (400 MHz, CDCl3) δ 2.36 (s, 3H), 4.15 (d,
J = 10 Hz, 1H), 4.69 (d, J = 4.8 Hz, 1H), 5.39 (dd, J = 5.2 Hz,
J = 4.8 Hz, 1H), 6.48 (d, J = 3.2 Hz, 1H), 6.56 (t, J = 3.6 Hz,
1H), 6.96 (d, J = 4.4 Hz, 1H), 7.02-7.13 (m, 5H), 7.23-7.29 (m,
7H), 7.32 (d, J = 7.6 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ
21.4, 57.0, 59.3, 124.8, 125.8, 126.7, 126.9, 127.1, 127.6, 127.7,
128.24, 128.4, 128.4, 129.0, 137.1, 139.5, 140.2, 142.9, 143.1; HRMS
(ESI) calcd for C25H23NO2S2 [MþNa]þ 456.1068, found 456.1058;
1
standard was added). The sample was analyzed by H NMR
spectroscopy by monitoring the benzylic protons as described
before.
IR (KBr) 3449, 1638, 1560, 1325, 1158, 1093, 810 cm-1
.
Kinetics Study by Varying Phosphorus Ligands in NMR:
Experimental Procedure. In an NMR tube 0.08 mmol of N-tos-
ylaziridine 1a (22 mg) was mixed with 0.0008 mmol of [Ag-
(COD)2]PF6 (0.37 mg, which is taken from stock solution of
catalyst in CDCl3 solvent) in 400 μL of CDCl3 solvent. Then
0.0016 mmol of PR3 ligand (for PPh3 ligand 0.42 mg is required,
which is added by preparing stock solution in CDCl3 solvent)
was added. Into this mixture then 100 μL of anisole was added.
Diphenyl methyl toluene is used as an internal reference (in each
kinetic run a fixed amount of 4 mg of the standard was added).
The sample was analyzed by 1H NMR spectroscopy. In 1H
Acknowledgment. We thank DST (financial support to
S.R.), and CSIR (fellowship to M.B.). M.B. especially thanks
Shrabani for helpful discussion on the DFT calculation.
Supporting Information Available: Tables, characteriza-
tion data, kinetic plots, MS, 1H and 13C NMR spectra for
all compounds, and crystallographic data for 6. This mater-
ial is available free of charge via the Internet at http://pubs.
acs.org.
4412 J. Org. Chem. Vol. 75, No. 13, 2010