NMR studies of hydrogen bonding inte1ractions with secondary amide and urea groups

Article abstract of DOI:10.1002/poc.391

The cis/trans ratios for six model secondary amides were determined by ¹H NMR in a range of solvent systems. The trans to cis equilibrium in chloroform is only slightly affected by addition of the hydrogen bond donor, trifluorethanol, but the c

Full text of DOI:10.1002/poc.391

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2001; 14: 463–467
DOI:10.1002/poc.391
NMR studies of hydrogen bonding interactions with
secondary amide and urea groups
Martin J. Deetz, Julianne E. Fahey and Bradley D. Smith*
Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556-5670, USA
Received 20 October 2000; Revised 21 March 2001; Accepted 1 April 2001
ABSTRACT: The cis/trans ratios for six model secondary amides were determined by 1H NMR in a range of solvent
systems. The trans to cis equilibrium in chloroform is only slightly affected by addition of the hydrogen bond donor,
trifluorethanol, but the cis rotamer is stabilized by an average of 0.7 kcal molÀ1 when acetic acid is used as an
intermolecular donor–acceptor template. Conversely, amide interaction with anionic hydrogen bond acceptors
decreases the percentage of cis rotamer. 15N NMR spectroscopy was used to determine the effect of hydrogen bonding
on the trans amide structure. The direction and the magnitude of 15N complex-induced-shifts indicate that both
hydrogen bond donors and acceptors raise the secondary amide rotational barrier by increasing the C—N bond order.
The relationship of these results to protein structure is discussed. Copyright  2001 John Wiley & Sons, Ltd.
KEYWORDS: 15N NMR; hydrogen bonding; secondary amide groups; secondary urea groups; cis/trans ratio
INTRODUCTION
of cis amide for dipeptides drops to around 0.4% for
amino acids with Ca side chains.6 Furthermore, peptide
elongation eliminates the electrostatic stabilization and
lowers the amount of cis rotamer to 0.11%. This is still
almost four times higher than that seen in the data bank
surveys. A possible explanation is that many cis amide
bonds have been misassigned because X-ray refinement
programs automatically refine amides as trans rotamers.
Evidence in favor of this view is the observation that
Surveys of the Protein Structure Data Bank show that
around 0.03% of the secondary amide bonds are assigned
as cis rotamers.1 These cis amides are usually located
near the functional site in a protein2,3 and thus play an
important role in protein function. For example, the
binding ability of the saccharide-binding protein con-
canavalin A is controlled by a calcium-induced cis/trans
isomerization.4 It is speculated that the cis bond is an
energy source that can be utilized during protein
function.1,5
˚
high-resolution structures (<2.0 A) have 0.04% cis
˚
whereas low-resolution structures (>2.5 A) have 0.01%
cis.1 Two other explanations for the scarcity of cis amides
in proteins are: (a) restrictions in conformational space
lower the likelihood of a cis amide; (b) non-covalent
stabilization of trans amide bonds.
There is literature evidence that secondary amide
conformations can be affected by non-covalent interac-
tions. For example, protein structures show a preference
for one of the amino acid residues involved in a cis amide
bond to be an aromatic group, suggesting that aromatic
residues are able to stabilize the cis conformation via C—
HÁÁÁp interactions.5 With regard to model studies,
Wolfenden and coworkers examined solvent effects and
found that the fraction of cis isomer for N-methforma-
mide and N-methylacetamide hardly changes with
solvent polarity.7 On the other hand, protonation of a
formamide carbonyl is known to increase the formamide
cis/trans ratio8 and the amount of cis formamide rotamer
is concentration dependent in CDCl3 due to hydrogen-
bonded homodimerization.9 Cis secondary amides10–12
and secondary carbamates13,14 have been observed
The fraction of cis amide bonds in proteins is much
lower than the 1.5% cis reported for N-methylacetamide,
a compound that has often been viewed as a model for the
peptide bond.1,5 A recent study showed that the fraction
*Correspondence to: B. D. Smith, Department of Chemistry and
Biochemistry, University of Notre Dame, Notre Dame, IN 46556-
5670, USA.
Email: smith.115@nd.edu
Contract/grant sponsor: National Science Foundation.
Contract/grant sponsor: University of Notre Dame.
Contract/grant sponsor: Pharmacia.
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 463–467

464
M. J. DEETZ, J. E. FAHEY AND B. D. SMITH
Table 1. Percentage cis rotamer in various solvent systems at 298 Ka
% cis [DG(kcal molÀ1)]b
Entry
Solvent
1
2
3
4
5
6
1
2
3
4
5
6
CDCl3
13 [1.12]
17 [0.94]
34 [0.39]
19 [0.86]
8 [1.44]
3 [2.05]
52 [À0.05]
55 [À0.12]
75 [À0.65]
50 [0]
22 [0.75]
29 [0.53]
27 [0.59]
27 [0.59]
46 [0.09]
26 [0.62]
20 [0.82]
20 [0.82]
49 [0.01]
15 [1.02]
9:1 CDCl3/CF3CH2OH
9:1 CDCl3/CH3COOH
CD3COOD
–
–
–
14 [1.07]
5 [1.74]
70 [À0.50]
39 [0.26]
23 [0.72]
36 [0.33]
CDCl3, 25 mM TBAAa
CDCl3, 25 mM TBACa
–
–
–
–
–
–
10 [1.30]
a
[amide] = 25 mM, TBAA = tetrabutylammonium acetate, TBAC = tetrabutylammonium chloride.
b DG for trans to cis equilibrium. Error Æ 5% of the stated value.
within supramolecular assemblies, but these systems
have not been examined systematically.
solution. Since the isomerization rate is slow on the NMR
time scale, the cis/trans ratios can be determined at room
temperature by direct integration. The percentages of cis
rotamer found for solutions of 1–6 in various solvent
systems are listed in Table 1.
The amount of cis rotamer in 9:1 CDCl3/CF3CH2OH is
slightly higher than that observed in CDCl3 alone
(compare entries 1 and 2 in Table 1). This is attributed
to the increased steric hindrance arising in the solvated
complex A shown in Scheme 1. A more significant
increase in amide cis/trans ratio is observed when the
solvent is 9:1 CDCl3/CH3COOH. In this case, the trans to
cis equilibrium for the six amides is moved to the right by
an average of 0.7 kcal molÀ1, presumably because the cis
rotamer forms the chelated hydrogen bonded complex B.
The aim of this study is determine the effect that
different hydrogen bonding motifs have on the structure
of a secondary amide bond. The study is in two related
parts. First, we report that the cis/trans ratios for six
model secondary amides increase slightly upon addition
of trifluorethanol, a monotopic hydrogen bond donor, and
increase quite significantly in the presence of acetic acid,
a donor–acceptor diad. Conversely, addition of anionic
hydrogen bond acceptors decreases amide cis/trans
ratios. Second, we provide 15N NMR evidence suggest-
ing that hydrogen bonding to trans secondary amides
increases the barrier to amide C—N rotation. The paper
concludes with a short discussion on the relationship of
these results to protein structure.
RESULTS AND DISCUSSION
Amide cis/trans rotamer ratios
Secondary amides 1–6 were chosen for study because
they have a measurable fraction of cis rotamer in CDCl3
Scheme 1
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 463–467

NMR STUDIES OF HYDROGEN BONDING INTERACTIONS
465
Scheme 2
Interestingly, lower amounts of cis rotamer are observed
when the solvent is 100% CD3COOD (Table 1, entry 4).
In this case, the concentration of acetic acid is so high that
polysolvated trans amide complexes such as C are the
predominant supramolecular structure(s).
Figure 1. Down®eld change in 15N chemical shift +15N Dꢀ) in
DMSO-d6 at 298 K for host 8 +67 mM) as a function of
increasing amounts of tetrabutylammonium dihydrogen-
phosphate +‡) and host 9 +100 mM) with tetrabutylam-
monium acetate +*) and tetrabutylammonium chloride +*)
We determined the effect of anionic hydrogen bond
acceptors on secondary amide conformation by measur-
ing the percentage of cis rotamer for amides 1, 3 and 4 in
the presence of tetrabutylammonium acetate and tetra-
butylammonium chloride (Table 1, entries 5 and 6). In
each case, the anions decreased the percentage of cis,
with acetate inducing the largest decrease. A likely
explanation is that the anion forms a hydrogen bond with
the amide N—H residue, which destabilizes the cis
rotamer (see D in Scheme 1). In the case of formamides 1
and 3, the anion may also stabilize the trans rotamer by
chelating with the aldehyde C—H as shown in E.15
Since anionic hydrogen bond acceptors decrease cis/
trans amide ratios (Table 1), it was of particular interest
to determine how anions affect the structure of a trans
amide. Thus, a solution of 7 in CDCl3 was treated with
one molar equivalent of tetrabutylammonium chloride.
This resulted in a 2.8 ppm downfield movement in 15N
chemical shift, which again suggests an increase in C—N
bond order, this time due to the zwitterionic complex G
in Scheme 2. A factor that complicates the interpretation
of this NMR result is the likelihood that 7 and chloride do
not form a discrete 1:1 hydrogen-bonded complex in
CDCl3. Therefore, we decided to confirm this effect by
using some amide-derived host compounds that are
known to form 1:1 hydrogen-bonded complexes with
anions.
Hydrogen bonding and trans amide structure
15N NMR spectroscopy was used to gain additional
insight into the effect of hydrogen bond donors and
acceptors on amide bond structure. Although 15N NMR is
commonly employed in biomolecule structure determi-
nation,16,17 it is rarely employed in organic supramol-
ecular chemistry.18,19 The inherent insensitivity of the
15N nucleus is a disadvantage, and often 15N-enriched
samples are required; however, this means that it is easy
to monitor an isotopically enriched nitrogen compound in
a mixture with other non-enriched compounds.
Two 15N-enriched compounds were prepared, the
organometallic bis(anilide) 823,24 and the urea 9.25 Both
compounds are known to form chelated 1:1 host/guest
complexes with anions, and they were used as hosts in
15N NMR titration experiments. Addition of tetrabutyl-
Specifically, we wished to learn how the structure of a
trans secondary amide is affected by hydrogen bonding.
As noted in the Introduction section, a possible reason for
the scarcity of cis amides in proteins is that the trans
amide is preferentially stabilized by non-covalent inter-
actions. Initially, we examined 15N-enriched acetanilide,
(7) as a model amide that adopts a predominantly trans
conformation. The 15N chemical shift for 7 in 9:1 CDCl3/
CF3CH2OH occurs 1.8 ppm downfield of the signal in
CDCl3. This agrees with previous reports on the effect of
hydrogen bond donors on amides and indicates that the
C—N bond order increases as shown in zwitterionic
complex F (Scheme 2).20 The downfield movement in
chemical shift is due to increased delocalization of the
nitrogen lone pair, which results in increased anisotropic
deshielding by the partial C=
N double bond.18,20–22
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 463–467

466
M. J. DEETZ, J. E. FAHEY AND B. D. SMITH
ammonium dihydrogenphosphate to 8 in DMSO-d6
results in a downfield movement of the amide 15N
resonance (Fig. 1). As expected, the titration isotherm
fitted nicelÀy1 to a 1:1 binding model and a value of
methods.25 A downfield movement in 15N chemical shift
was also observed when urea derivative 9 was titrated
with tetrabutylammonium acetate and tetrabutylammo-
nium chloride in DMSO-d6 (Fig.À11). The binding
constants were calculated to be 18 M for chloride and
340 MÀ1 for acetate.
folding step for proteins containing cis amides. It is
intriguing that a cis/trans isomerase for non-prolyl
peptides has yet to be identified.27
Ka = 110 M
was extracted by iterative curve-fitting
EXPERIMENTAL
All NMR samples were 25 mM unless stated otherwise.
Compounds 1–4 and 6 were purchased from Aldrich, and
compound 7 was made by a literature method.22
As in the case of 7, the approximate 3 ppm downfield
movement of the amide (and urea) 15N chemical shifts
upon saturation of 8 and 9 with anions indicates that the
C—N bond order increases as shown in complex G.
Martin et al. have established that there is a linear
correlation between C—N rotational activation energy
and 15N chemical shift.26 According to their relationship,
a 3 ppm downfield shift corresponds to a 0.7 kcal molÀ1
increase in activation energy for C—N rotation. In other
words, forming a hydrogen bond with the NH residue of a
trans secondary amide rigidifies the amide C—N bond.
2,6-Dimethylformanilide 25)
2,6-Dimethylaniline (165.1 ml, 0.67 mmol) was dissolved
in benzene in a Dean–Stark apparatus. An 88% solution
of formic acid (3 ml, excess) was added and the mixture
was stirred under nitrogen for 30 min. The mixture was
then heated to reflux for 15 h and the excess water and
formic acid driven off as an azeotrope. The benzene was
removed, leaving a white solid that was shown to be pure
by NMR.30 1H NMR (CDCl3) ꢀ 2.28 (s, 3H), 2.32 (s, 3H),
6.80/6.92 (bs, cis/trans NH signal, 1H), 7.12–7.15 (m,
3H), 8.11 (d, trans CHO signal, J = 12 Hz), 8.43 (d, cis
CHO signal, J = 1.5 Hz) ppm.
CONCLUSIONS
For a protein to contain a backbone cis amide bond it
must overcome an approximate 2.8 kcal molÀ1 energy
difference in favor of the trans rotamer.3,27 Our
investigation of six model secondary amides in chloro-
form solution consistently shows that the trans to cis
equilibrium is only slightly affected by the hydrogen
bond donor, trifluorethanol, but it is moved to the right by
an average of 0.7 kcal molÀ1 when acetic acid (in
chloroform solvent) is used as an intermolecular donor–
acceptor template. Conversely, amide interaction with
anionic hydrogen bond acceptors decreases the percen-
tage of cis rotamer. It appears that if a protein is to induce
one of its amide groups to adopt a cis conformation then
the protein must fold in a way so as to stabilize the cis
amide with at least two intramolecular hydrogen bonds in
a relatively non-polar environment.
[4,4'-Bis-2phenyl-15N-carbamoyl)-2,2'-bipyridine]-
bis22,2'-bipyridine)ruthenium2II) bis2hexa¯uoro-
phosphate) 28)
4,4'-Dicarboxy-2,2'-bipyridine (0.318 g, 1.3 mmol) was
suspended in 15 ml of SOCl2 and the mixture refluxed for
24 h under an atmosphere of nitrogen. The solvent was
removed in vacuo and the yellow residue suspended in
dry THF. 15N-Aniline (263 ml, 2.86 mmol) was added via
dropping funnel as a solution in THF over 5 min. The
mixture was stirred at room temperature overnight under
nitrogen. The mixture was filtered and the residue
dissolved in dimethyl sulfoxide and precipitated upon
addition of water. The precipitate was filtered and rinsed
twice with water, leaving 4,4'-bis(phenyl-15N-carba-
moyl)-2,2'-bipyridine as an off-white solid (0.488 g,
15N NMR is an attractive method for studying the
supramolecular chemistry of amide and urea compounds.
Not only is it a useful probe for host/guest titration
experiments (Fig. 1), but the direction and the magnitude
of the 15N complex-induced-shift provides insight into
the change in C—N bond order and the corresponding
C—N rotational barrier. The results of our 15N NMR
experiments indicate that both hydrogen bond donors and
acceptors can raise amide rotational barriers by increas-
ing the C—N bond order, as shown in structures F and G
in Scheme 2.20,28 Thus, hydrogen bonding with a
protein’s backbone amides is a way of rigidifying and
polarizing the backbone structure.29 The barrier for
conversion of trans to cis amide is around 20 kcal molÀ15
which means that the process can be the rate-limiting
1
1.23 mmol) in 95% yield. H NMR (DMSO-d6) ꢀ 7.15
(t, 2H, J = 7.5 Hz), 7.39 (t, 4H, J = 8 Hz), 7.81 (d, 4H,
J = 8.5 Hz), 7.98 (dd, 2H, J = 5, 1.5 Hz), 8.91 (s, 2H),
8.96 (d, 2H, J = 5 Hz), 10.71 (d, 2H, JN— H = 90.5 Hz).
13C NMR (DMSO-d6) ꢀ 163.8 (JN— C = 15.5 Hz), 155.4,
150.0, 143.4 (JN— C = 9.5 Hz), 138.5 (JN— C = 14.5 Hz),
128.5, 124.1, 122.2, 120.5, 118.5. 15N NMR (DMSO-d6)
(referenced relative to CH3NO2 at 0 ppm) À246.9. MS
‡
‡
(FAB , NBA) [M ‡ H] at m/z 397.1449 (calc.
397.1450).
Complex 8 was formed by dissolving the above solid
(0.227 g, 0.57 mmol) in a mixture of 25 ml of water,
25 ml of ethanol, and 12.5 ml of glacial acetic acid. This
mixture was heated to 50°C for 20 min under an
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 463–467

NMR STUDIES OF HYDROGEN BONDING INTERACTIONS
467
atmosphere of nitrogen. cis-Ru(bpy)2Cl2ÁnH2O (0.328 g,
0.63 mmol) was added and the mixture was heated at
reflux for 3 days under nitrogen. The mixture was cooled
to room temperature and filtered through a Celite plug,
and upon removal of the solvent in vacuo gave a dark
brown–orange glass. Column chromatography on alumi-
na (10% methanol in CHCl3 → 100% methanol)
produced 8 as the dichloride salt (0.44 g, 0.54 mmol) in
95% yield. The bis-PF6 salt was formed by precipitation
Foundation, the University of Notre Dame (George M.
Wolf and Reilly fellowships to M.J.D.) and Pharmacia
(Summer Undergraduate Fellowship to J.E.F). This paper
is dedicated to Professor Hans-Jo¨rg Schneider, a leading
teacher/scholar in the field of supramolecular chemistry.
REFERENCES
1
after addition of NH4PF6(aq). H NMR (DMSO-d6) ꢀ
1. Weiss MS, Jabs A, Hilgenfeld R. Nat. Struct. Biol. 1998; 5: 676.
2. Herzberg O, Moult J. Protein: Struct. Funct. Genet. 1991; 11: 223–
229.
3. Banerjee S, Shiematsu N, Pannell LK, Ruvinov S, Orban J, Scharz
F, Herberg O. Biochemistry 1997; 36: 10857–10866.
4. Bouckaert J, Poortmans F, Wyns L, Loris R. J. Biol. Chem. 1996;
271: 16144–16150.
5. Jabs A, Weiss MS, Hilgenfeld R. J. Mol. Biol. 1999; 286: 291–304.
6. Scherer G, Kramer ML, Schutkowski M, Reimer U, Fischer G. J.
Am. Chem. Soc. 1998; 120: 5568–5574.
7. Radzicka A, Pederson L, Wolfenden R. Biochemistry 1988; 27:
4538–4541.
8. Stewart WE, Siddall TH. Chem. Rev. 1970; 70: 517–551.
9. Manea VP, Wilson KJ, Cable JR. J. Am. Chem. Soc. 1997; 119:
2033–2039.
10. Pern´ı GJ, Kilburn JD, Essex JW, Mortshire-Smith RJ, Rowley M.
J. Am. Chem. Soc. 1996; 118: 10 220–10 227.
11. Bairaktari E, Mierke DF, Mammi S, Peggion E. J. Am. Chem. Soc.
1990; 112: 5383.
7.18 (t, 2H, J = 7.5 Hz), 7.41 (t, 4H, J = 8 Hz), 7.54–7.59
(m, 4H), 7.75–7.77 (m, 6H), 7.81 (d, 2H, J = 5.5 Hz), 7.94
(d, 2H, J = 6 Hz), 7.99 (d, 2H, J = 6 Hz), 8.20 (d, 2H,
J = 6.5 Hz), 8.23 (d, 2H, J = 8 Hz), 8.87 (d, 4H, J = 8 Hz),
9.39 (s, 2H), 10.77 (d, 2H, JN—C = 90 Hz). 13C NMR
(CD3CN) ꢀ 163.0 (JN—C = 17 Hz), 158.6, 157.9 (JN—C
= 18 Hz), 153.6, 152.8 (JN—C = 17 Hz), 144.0 (JN—C
= 9.5 Hz), 139.3 (JN—C = 4.5 Hz), 139.0 (JN—C
=
14 Hz), 130.0, 128.8 (JN—C = 6.5 Hz), 126.5, 126.1,
125.5, 123.4, 121.8. 15N NMR (CD3CN) (referenced
relative to CH3NO2 at 0 ppm) À246.6 (JN—H = 90.5 Hz).
‡
‡
MS (FAB , NBA) [M À PF6] at m/z 955.1447 (calc.
955.1431).
12. Beijer FH, Sijbesma RP, Vekemans JAJM, Meijer EW, Kooijman
H, Spek AL. J. Org. Chem. 1996; 61: 6371–6380.
13. Marcovici-Mizrahi D, Gottlieb HE, Marks V, Nudelman A. J. Org.
Chem. 1996; 61: 8402–8406.
15N-Phenyl-N'-octyl urea 29)
14. Moraczewski AL, Banaszynski LA, From AM, White CA, Smith
BD. J. Org. Chem. 1998; 63: 7258–7262.
15. Corey EJ, Barnes Seeman D, Lee TW. Tetrahedron Lett. 1997; 38:
1699–1702.
16. Von Philipson W, Mu¨ller R. Angew. Chem. Int. Ed. Engl. 1986; 25:
383–413.
17. Cordier F, Crzesiek S. J. Am. Chem. Soc. 1999; 121: 1601–1602.
18. Haushalter KA, Lau J, Roberts JD. J. Am. Chem. Soc. 1996; 118:
8891–8896.
19. Dega-Szafran Z, Szafran M, Sitkowski J, Stefaniak L. J. Phys. Org.
Chem. 1996; 9: 746–750.
20. Saito H, Tanake Y, Nukada K. J. Am. Chem. Soc. 1971; 93: 1077–
1081.
21. Lina´s M, Horsley WJ, Klein MP. J. Am. Chem. Soc. 1976; 98:
7554–7558.
22. Lumsden MD, Wasylishen RE, Eichele K, Schindler M, Penner
GH, Power WP, Cutirs RD. J. Am. Chem. Soc. 1994; 116: 1403–
1411.
15N-Aniline (100 ml, 1.14 mmol) was dissolved in
CH3CN (7 ml). Octyl isocyanate (242 ml, 1.37 mmol)
was added and the mixture heated to reflux for 24 h under
an atmosphere of nitrogen. The mixture was cooled to
room temperature and placed in a freezer for 20 min. A
white solid was collected by suction filtration and rinsed
with cold CH3CN, leaving pure 9 as a white solid
(0.251 g, 1.01 mmol) in 88% yield. 1H NMR (DMSO-d6)
ꢀ 0.85 (t, 3H, J = 7 Hz), 1.23–1.42 (m, 12H), 3.04 (q, 2H,
J = 7 Hz), 6.08 (bt, 1H, J = 5.5 Hz), 6.85 (t, 1H, J = 7 Hz),
7.19
(t, 2H, J = 8 Hz), 7.36 (d, 2H, J = 8.5 Hz), 8.35 (d, 1H,
JN—H = 88.5 Hz). 13C NMR (DMSO-d6) ꢀ 155.1 (JN—C
= 20 Hz), 140.6 (JN—C = 16 Hz), 128.6, 120.8, 117.5,
39.0, 31.2, 29.7, 28.7, 28.7, 26.4, 22.1, 13.9. 15N NMR
(DMSO-d6) (referenced relative to CH3NO2 at 0 ppm)
23. Beer PD, Szemes F, Balzani V, Sala` CM, Drew MGB, Dent SW,
Maestri M. J. Am. Chem. Soc. 1997; 119: 11864–11875.
24. Deetz MJ, Smith BD. Tetrahedron Lett. 1998; 39: 6841–6844.
25. Hughes MP, Smith BD. J. Org. Chem. 1997; 62: 4492–4501.
26. Martin GJ, Gouesnard JP, Dorie J, Rabiller C, Martin ML. J. Am.
Chem. Soc. 1977; 99: 1381–1384.
‡
‡
À270.9 (JN—H = 88.0 Hz). MS (FAB , NBA) [M ‡ H]
at m/z 250.1945 (calc. 250.1937).
27. Scholz C, Scherer G, Mayr L, Schindler T, Fischer G, Schmid FX.
Biol. Chem. 1998; 379: 361–365.
28. Suezawa H, Tsuchiya K, Tahara E, Hirota M. Bull. Chem. Soc. Jpn.
1988; 61: 5047–4062.
29. Scheiner S, Kern CW. J. Am. Chem. Soc. 1977; 99: 7042–7050.
30. Deslongchamps P, Gerval P, Cheriyan UO, Guida A, Tailefer RJ.
Nouv. J. Chim. 1978; 2: 631–636.
Acknowledgements
This work was supported by the National Science
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 463–467