The Direct Alkylation of π-Rich, Acid-Sensitive Heterocyclic Compounds via Essentially Free Carbocations

Article abstract of DOI:10.1021/jo971081t

Alkylation of π-rich heteroaromatics such as pyrroles and furans by the standard Friedel-Crafts approach is impractical because the acid catalysts employed (Bronsted or Lewis) induce polyalkylation, ring opening, and polymerization. The present study describes the facile benzylation of π-excessive heteroaromatics using the nitrosoamide approach, which generates nitrogen-separated carbocation-counterion ion-pairs as the alkylating agent with no catalyst being required. N-Nitrosoamides are favorable sources of carbocations because of the following variables: mildness of the conditions required to generate cations, high reactivity of the unsolvated carbocations formed, solubility of the precursors in a wide range of solvents, homogeneity of the reactions, wide range of decomposition temperatures possible, straightforward chemistry, and excellent product balance. The majority of the cations that are generated in pyrrole (80%) are intercepted by the solvent, and only 20% are intercepted by the counterion; this result provides support for the intermediacy of nitrogen-separated ion-pairs in deamination. A nucleophilicity scale is presented for reactions of selected nucleophiles with essentially free carbocations.

Full text of DOI:10.1021/jo971081t

J . Org. Chem. 1997, 62, 8091-8094
8091
Th e Dir ect Alk yla tion of π-Rich , Acid -Sen sitive Heter ocyclic
Com p ou n d s via Essen tia lly F r ee Ca r boca tion s1
Ron W. Darbeau* and Emil H. White
Department of Chemistry, McNeese State University, Lake Charles, Louisiana 70609 and Department of
Chemistry, The J ohns Hopkins University, Baltimore, Maryland 21218
Received J une 16, 1997X
Alkylation of π-rich heteroaromatics such as pyrroles and furans by the standard Friedel-Crafts
approach is impractical because the acid catalysts employed (Brønsted or Lewis) induce polyalkyl-
ation, ring opening, and polymerization. The present study describes the facile benzylation of
π-excessive heteroaromatics using the nitrosoamide approach, which generates nitrogen-separated
carbocation-counterion ion-pairs as the alkylating agent with no catalyst being required.
N-Nitrosoamides are favorable sources of carbocations because of the following variables: mildness
of the conditions required to generate cations, high reactivity of the unsolvated carbocations formed,
solubility of the precursors in a wide range of solvents, homogeneity of the reactions, wide range
of decomposition temperatures possible, straightforward chemistry, and excellent product balance.
The majority of the cations that are generated in pyrrole (80%) are intercepted by the solvent, and
only 20% are intercepted by the counterion; this result provides support for the intermediacy of
nitrogen-separated ion-pairs in deamination. A nucleophilicity scale is presented for reactions of
selected nucleophiles with essentially free carbocations.
In tr od u ction
counterion (after diffusion of the nitrogen molecule from
3
its blocking position) (eq 1). Neutral conditions in
Carbocations are reaction intermediates in the thermal
decomposition of N-alkyl-N-nitrosoamides (eq 1); they
nonpolar solvents can be used,2 and since catalysts are
not used, the reaction constitutes a particularly useful
way of investigating the intrinsic reactions of essentially
,4b
2
are generated in the form of nitrogen-separated ion-pairs.
Carbocations formed in this way are little solvated and
6
free carbocations introduced by a unimolecular process
2
thus highly reactive; they are probably the “freest” type
into essentially any medium.3b The nitrosoamide ap-
proach has been used in alkylation of compounds ranging
of carbocations that can be readily prepared in liquid
solution. The nitrogen molecule, through its screening
4
from hydrocarbons (e.g. benzene, toluene) to carboxylic
acids.6b Several related reactions lead to analogous
7
intermediates, and through their use, a wide range of
temperatures can be employed for the generation and
4
study of carbocations.
Heterocyclic compounds such as furan and pyrrole are
“π-rich” or “π-excessive,” because resonance involving
lone-pair electrons on the heteroatom result in an
enhanced electron density on the carbon atoms of the
ring.8 Consequently, these heteroaromatics are as reac-
tive as anilines and phenols in undergoing electrophilic
substitutions.8 Mono C-alkylation of pyrrole cannot be
achieved through direct reaction with simple alkyl ha-
lides, either alone or with Lewis-acid catalysts, because
(
3) (a) The decomposition products (∼98%) in a reactive solvent are
4b
the corresponding ester and the solvent-derived products (SDP). (b)
The yields of solvent-derived products (SDP) increase
4
b,5
with the
2
c,4b
reactivity of the cation,
inversely with the nucleophilicity of the counterion.
4) (a) White, E. H.; Woodcock, D. J . In The Chemistry of the Amino
with the nucleophilicity of the solvent, and
2
b,4b
action with respect to the counterion, provides the
carbocation with time to interact with molecules compris-
ing the solvent cage. Thus, the overall reaction is one in
which the carbocations are scavenged in competing
processes: by nucleophiles in the solvent cage and by the
(
Group; Patai, S., Ed.; J ohn Wiley and Sons, Inc.: New York, 1968;
Chapter 8. (b) From the Ph.D. Thesis of Ron W. Darbeau, The J ohns
Hopkins University, Baltimore, Maryland, 1996.
(
5) The yield of SDP also increases, but to a lesser extent, as the
4b
inert molecule is varied from N
reactions at lower temperatures.
2
to N
2
O
and by performing the
4
b
X
Abstract published in Advance ACS Abstracts, October 15, 1997.
(6) (a) Huisgen, R u¨ chardt, C. J ustus Liebigs Ann. Chem. 1956, 601,
1. (b) White, E. H.; Dolak, L. A. J . Am. Chem. Soc. 1966, 88, 3790.
(7) (a) White, E. H.; Ribi, M. A.; Cho, L. K.; Egger, N.; Dzadzic, P.
M.; Todd, M. D. J . Org. Chem. 1984, 49, 4866. (b) White, E. H. J . Am.
Chem. Soc. 1955, 77, 6014. (c) White, E. H.; Maskill, H.; Woodcock, D.
J .; Schroeder, M. A. Tetrahedron Lett. 1969, 1713. (d) White, E. H.;
Lewis, C. P.; Ribi, M. A.; Ryan, T. J . J . Org. Chem. 1981, 46, 552. (e)
M u¨ ller, H.; Haiss, H. Chem. Ber. 1963, 96, 570. (f) Olah, G. A.;
Friedman, N.; Bollinger, J . M.; Lukas, J . J . Am. Chem. Soc. 1966, 88,
5328. (g) Schmid, H.; Sami, A. F. Monatsh. Chem. 1955, 86, 904. (h)
Soll, H. Methoden der Organischen Chemie; Muller, E., Ed., (Houben-
Weyl), Georg Thieme Verlag: Stuttgart, 1958; Vol. 11, Part 2, p 133.
(
1) Publication no. 56 in a series on alkane diazonium ion-pairs and
deamination. Preceding publication: White, E. H. et al. J . Org. Chem.
996, 61, 7986.
2) (a) White, E. H.; Field, K. W.; Hendrickson, W. H.; Dzadzic, P.;
Roswell, D. F.; Paik, S.; Mullen, P. W. J . Am. Chem. Soc. 1992, 114,
023. (b) White, E. H.; DePinto, J . T.; Polito, A. J .; Bauer, I.; Roswell,
1
(
8
D. F. J . Am. Chem. Soc. 1988, 110, 3708. (c) White, E. H.; McGirk, R.
H.; Aufdermarsh, C. A.; J r.; Tiwari, H. P.; Todd, M. J . J . Am. Chem.
Soc. 1973, 95, 8107. (d) White, E. H.; Tiwari, H. P.; Todd, M. J . J . Am.
Chem. Soc. 1968, 90, 4734. (e) White, E. H.; Field, K. W. J . Am. Chem.
Soc. 1975, 97, 2148.
S0022-3263(97)01081-5 CCC: $14.00 © 1997 American Chemical Society

8
092 J . Org. Chem., Vol. 62, No. 23, 1997
Darbeau and White
Ta ble 1. Ben zyla tion of Heter ocyclic Com p ou n d s w ith
benzylpyrroles was 59% (vide infra). Deuteriomethyl-
benzene was not observed in decompositions of 1a in 10:
N-Ben zyl-N-n itr osoben za m id e (1a )a
%
yields
1, 1:1, or 1:10 mixtures (vol/vol) of either pyrrole/CDCl
3
isomer
4
b
distribution (%)
or of hexane/CDCl
abstract hydrogen from chloroform, it appears that free-
radical pathways are absent in these reactions.
;
since alkyl radicals are known to
alkylated
3
1
0
temp ester heteroaromatics 2-isomer 3-isomer
solvent
pyrrole
(°C)
3
4 + 5
4
5
40
20
41
80
59
62
64
38
36
Ben zyla tion of N-Meth ylp yr r ole via th e Nitr oso-
a m id e Ap p r oa ch . Compound 1a was decomposed in
N-methylpyrrole at 40 °C (the half-life of the decomposi-
equimolar pyrrole/ 40
CDCl
3
N-methylpyrrole
furan
40
25
32
77
68
23
49
88
51
12
9
a,b
tion was also ∼7 h at 40 °C as above ). In the present
case, the yield of solvent-derived products (SDP) (4 + 5)
was 68% (Table 1), and the isomer distribution (4:5) was
a
[
Nitrosoamide] ≈ 0.05 M. The data are averages of two or more
runs; standard deviation for yields and for isomer distributions
8
a
e0.4.
roughly 1:1 (Table 1). N-Methylpyrrole (pK
) -2.9)
) -3.8),8 but from the
larger yield of SDP in the latter case (80% vs 68%) (Table
) it would appear that pyrrole competes better for the
a
a
is more basic than pyrrole (pK
a
pyrroles are polymerized by strong acids.8 Similarly,
traditional Friedel-Crafts alkylation is not practical with
furan because of catalyst-promoted polymerization and
polyalkylation.8
The present study demonstrates that mono C-benzyl-
ated, π-rich heteroaromatics can be readiliy prepared via
the nitrosoamide route.
1
carbocation than does N-methylpyrrole. Further, the
relative yield of 2-benzylpyrrole, 4a (62%) (Table 1), is
larger than that of 2-benzyl-1-methylpyrrole (49%) (Table
1
). The steric effect of the N-methyl group may be
responsible for these observations by decreasing the rate
of alkylation at the adjacent 2-position of the pyrrole
nucleus.
Resu lts a n d Discu ssion
The results shown in Table 1 indicate that the ni-
trosoamide approach allows for direct syntheses of mono
C-benzylpyrroles in good to excellent yields under mild
conditions.
Dir ect Ben zyla t ion of P yr r ole Usin g t h e Ni-
tr osoa m id e Ap p r oa ch . N-Benzyl-N-nitrosobenzamide
1a ) was decomposed in pyrrole (eq 2) at 25 °C and at 40
(
°
C; the half-lives of these decompositions were ∼2 days
9a,b
The high yield of benzylpyrroles formed (80%) (Table
at 25 °C and ∼7 h at 40 °C.
1
) indicates that even in the case of the relatively
unreactive, resonance-stablized benzyl cation, the major-
ity of the carbocations of the inert-molecule-separated
ion-pair are interceptible by solvent molecules if the latter
are sufficiently reactive. This fact provides further
evidence for the initial presence of a nitrogen molecule
between the cation and its counterion, since in its absence
one would have expected that a facile carbocation-anion
reaction would have produced principally the ester (3).
The 80% yield of SDP is similar to values of 75-80%
observed for the yields of acetate ester from decomposi-
2
c
From the reaction in neat pyrrole, the yield of C-
tion in acetic acid, and for methyl alkyl ethers from
decompositions in methanol.4b In general, it would
appear that ∼80% is the maximum amount of solvent
interception that is possible. This limit is set presumably
by the high diffusion rate of nitrogen into the medium,
which would allow a finite amount of ester formation
through ion-pair collapse involving the cation and its
counterion.
benzylpyrroles (4a + 5a ) was 80% (Table 1) with a
2
-benzylpyrrole/3-benzylpyrrole ratio (4a /5a ) of 62/38.
The latter ratio is consistent with the larger nucleophi-
8
a,b
licity at the C
2
center vs that at C
3
.
Interestingly, no
N-benzylpyrrole was observed which may appear a little
surprising given the high reactivity of the deaminatively
formed benzyl cation.2
,4,7
It has been proposed, however,
that five-membered heteroaromatics do not react with
electrophiles (alkyl groups, protons, etc.) at the
heteroatom8 because reaction at this position would
lead to a substantial loss of resonance stabilization.8
In the presence of equimolar CDCl
3
(an inert diluent
the yield of SDP from the
benzylation of pyrrole remains high (59%; Table 1). The
fact that equimolar CDCl reduces the solvent-capture
for the benzyl cation),4
b,9d
a,b
c
3
In equimolar pyrrole/CDCl
tribution was virtually unchanged (Table 1), the yield of
3
, although the isomer dis-
yield only from 80% to 59% and not to 40%, the value
expected on statistical grounds, suggests that not all
abortive collisions of the benzyl cation with inert mol-
ecules will lead to ion-pair collapse (ester formation).
Evidently carbocations which have undergone such col-
lisions still have the ability to alkylate pyrrole.
Ben zyla tion of F u r a n . Furan was benzylated using
nitrosoamide 1a at 25 °C; a 23% yield of benzylfurans
was observed (Table 1). The lower yield of SDP is
(8) (a) J oule, J . A.; Mills, K.; Smith, G. F. Heterocyclic Chemistry,
3
rd ed., Chapman and Hall: London, 1995; pp 231-232. (b) Schofield,
K. Hetero-Aromatic Nitrogen Compounds: Pyrroles and Pyridines,
Plenum Press: New York, 1967; p 3. (c) N-Alkylpyrroles, however, are
readily prepared by reaction of pyrryl salts with iodoalkanes (Candy,
C. F.; J ones, A. R. J . Org. Chem. 1971, 36, 3990 and Hobbs, C. F. et
al. J . Am. Chem. Soc., 1962, 84, 43).
(9) (a) Half-lives were determined by measuring the changes in the
integral of the benzyl signal as a function of time at a constant
temperature in the NMR probe. (b) With bulkier nitrosoamides such
as N-benzyl-N-nitrosopivalamide (which has a half-life of ∼30 min in
consistent with furan being less nucleophilic than the
8a,b
pyrroles.
The 2-isomer/3-isomer ratio was 88/12 (Table
4
b
1). A second approach to the benzylation was pursued
∼
0.1 M solutions in benzene/toluene at 25 °C) benzylation would be
faster. (c) No alkylpyridines were observed presumably because of the
low concentration of the relatively unreactive pyridine. (d) No reaction
between the benzyl cations and chloroform has been observed in that
in which 1H NMR spectra were recorded at -80 °C to
1
neither toluene nor hexachloroethane were detected by H NMR or
(10) Fossey, J .; Lefort, D.; Sorba, J . Free Radicals in Organic
Chemistry; J ohn Wiley and Sons: New York, 1955; pp 202-203.
GC and by GC, respectively (see ref 2e for comparison).

Alkylation of π-Rich, Acid-Sensitive Heterocyclics
J . Org. Chem., Vol. 62, No. 23, 1997 8093
determine whether O-benzylfuranonium ion, 9, was a
reaction intermediate. To this end, furan in 30% furan/
≡
CDCl
3
(vol/vol) was benzylated using nitrosation of
N-benzyl-O-benzoylhydroxylamine (6) at -80 °C as a
source of benzyl cations (eq 3).7 With the exception of
benzyl benzoate (3) and benzyl nitrate (8) (eq 3), no new
a
of solvent-derived products (SDP) (when different sol-
vents, but a common cation/counterion pair are used) can
be considered as a measure of the relative nucleophilici-
ties of the solvents.
The data show (Table 2) that the nucleophilicities
decrease in the order: pyrrole > methanol > N-meth-
ylpyrrole > acetone > furan ≈ acetonitrile > toluene >
1
4a,b
benzene > chloroform ≈ pentane.
The present scale
(
Table 2) is more compressed than scales that are based
5a
on S
as trityl with nucleophiles.
N
2 reactions1 or on reactions of stable cations such
1
5b
This observation is con-
sistent with the high reactivity of deaminatively gener-
ated carbocations.
Ta ble 2. Yield sa of Solven t-Der ived P r od u cts (SDP )
fr om th e Rea ction of Ben zyl Ca tion s Gen er a ted fr om
N-Ben zyl-N-n itr osoben za m id es or fr om
P h en yld ia zom eth a n e
1
11,12
signals ( H NMR) were observed between 5-7,
at -80
°
C or on warming to 25 °C. The signals for the C-
yield of SDP
benzylpyrroles were already present in the first spectrum
run at -80 °C (time from mixing ≈ 45 s). This observa-
tion indicates that either the O-benzylfuranonium ion (9)
did not form or that it is too labile for detection at -80
yield of
in solvent
X yield of SDP
pKb SDP (%) yield of ester in benzene
yield of
SDP
solvent
pyrrole
methanol
17.8b
80c,d
4.0
3.2
2.1
0.7
0.3
0.3
0.2
0.08
0.0
0.0
11.4
10.9
9.7
5.9
3.3
3.3
2.6
1.0
0.0
0.0
°
C. The yield of benzylfurans in this latter experiment
e
f-h
16
76
68
41
23
b
c,d
N-methylpyrrole 16.9
i
g,j
acetone
furan
21.2
b
c,g
27
24
e
g,k
acetonitrile
toluene
benzene
chloroform
pentane
23
c,d
-
-
-
-
18
7
0
0
c,d
c,d
c,d
was ∼5%, consistent with the low mole fraction of furan
a Yields independent of nitrosoamide or diazoalkane concentra-
tion over the range 0.05-0.5 M; all reactions were performed in
in the medium.4 The isomer distribution [90% (2-isomer)/
9
c
b
c
the presence of 2 equiv of pyridine.
References 8a,b. Com-
1
0% (3-isomer)] was similar to that obtained in the
d
e
pound 1a was used. Run temperature ) 40 °C. Reference 16a.
nitrosoamide aproach.
f
17 g
Compound 10 and benzoic acid were used.
Run temperature
A Nu cleop h ilicity Sca le for Su bstr a tes in Rea c-
tion s w ith High ly Rea ctive Ca r boca tion s. Ther-
molyses of N-benzyl-N-nitrosoamides (eq 1) and acidifi-
cation of phenyldiazomethane (10) (eq 4) lead to the
h
i
j
)
25 °C. Reference 15c. Reference 16b. Compound 10 and
acetic acid were used.17 k N-Benzyl-N-nitrosopivalamide was used.
Exp er im en ta l Section
formation of equivalent suites of particles: nitrogen-
Ma ter ia ls a n d Meth od s. Dinitrogen tetraoxide was ob-
tained from The Matheson Gas Co. The pyrroles and furan
were distilled prior to use; all other commercial reagents were
separated ion-pairs.4
b,17
These ion-pairs consist of es-
sentially free benzyl carbocations that react with either
4
the counterion or with solvent molecules. Thus the yield
(
15) (a) McManus, S. P.; Pittman, C. U., J r. Organic Reactive
(
11) The chemical shifts (in CDCl
3
) of the benzyl protons of the
Intermediates; McManus, S. P.; Ed., Academic Press: New York, 1973;
Vol. 26, p 299. (b) McClelland, R. A.; Banait, N.; Steenken, S. J . Am.
Chem. Soc. 1986, 108, 7023. (c) Interestingly, no methoxytoluenes were
observed.
(16) (a) Gordon, A. J .; Ford, R. A. The Chemists’ Companion:
Handbook of Practical Data, Techniques and References; Wiley-
Interscience: New York, 1972; p 6063. (b) Streitweiser, A. J r.;
Heathcock, C. H. Introduction to Organic Chemistry, 3rd ed.; Mac-
millan Publishing Co.: New York, 1981; p 1155.
(17) Product distributions (% ester, % SDP) from thermolysis of
compound 1a in equimolar benzene/toluene are the same within
experimental error as that from acidification of phenyldiazomethane
(compound 10) with benzoic acid in the same solvent. Thus it would
appear that phenyldiazomethane on general acid protonation leads
directly to the corresponding intimate ion-pair and eventually into the
nitrogen-separated ion-pair (eqs 1 and 4) which is the active alkylating
agent in these deaminations. Ion-pair formation is likely to be
facilitated in more polar solvents such as acetone and methanol.
N-benzylacetonitrilium ion and of N-benzylpyridinium ion are, respec-
tively, δ 5.38b and 6.4 (Friedrich, E. C.; Vartanian, P. F. J . Organomet.
Chem. 1976, 110, 159). Since oxygen is more electronegative than
nitrogen it would be expected to deshield the benzylic protons to a
greater extent than would nitrogen (Silverstein, R. M.; Bassler, G. C.;
Morrill, T. C. Spectrometric Identification of Organic Compounds, 4th
ed.; J ohn Wiley and Sons: New York, 1981; p 220).
4
b
A
(
12) The absence of a signal between δ 5-7 which rose then fell
with time also indicates that the N-benzyl-N-nitroso-O-benzoylhy-
4
a,13
droxylamine is too labile to be detected at -80 °C.
(
13) White, E. H.; Lim, H. M. J . Org. Chem. 1987, 52, 2162.
4
b
(14) (a) Ideal data would be obtained by decomposing the same
precursor (e.g., 1a ) at the same temperature in a medium comprised
of the solvent under study dissolved in a large excess (∼90%) of an
inert diluent. (b) Interestingly, with the more reactive 1-norbornyl
cation, the yields of ester in ethanol and chloroform were 67% and
4
9%, respectively.2c

8
094 J . Org. Chem., Vol. 62, No. 23, 1997
Darbeau and White
conditions); (3) elution of the band at R ) 0.4 from a
used without further purification. NMR spectra were recorded
on a 300 MHz, FT instrument; FT-IR data are reported. TLC
analyses were performed on UV-fluorescent silica gel plates.
N-Ben zylben za m id e. This compound was prepared by the
procedure of Heyns and von Bebenburg. mp 103-104°C (lit.
1
f
preparative TLC plate [using the same conditions as in (2)]
followed by evaporation of the solvent and redissolution of the
1
residue in CDCl
3
gave a H NMR spectrum which was identical
1
8
18
with that of commercial benzyl benzoate.
-
1
1
04-105 °C); IR (Nujol) 3289, 1638, 1491 cm
;
H NMR
Decom p osition s of N-Ben zyl-N-n itr osoben za m id e in
P yr r ole a n d N-Meth ylp yr r ole. N-Benzyl-N-nitrosobenza-
mide (5 mg, 20.8 mmol) and pyridine (3.3 mg, 41.6 mmol) were
dissolved in 500 mL of pyrrole or N-methylpyrrole in an NMR
(
(
CDCl ) δ 4.65 (d, 2H, J ) 5.7 Hz), 6.55 (bs, 1H), 7.29-7.50
3
m, 8H), 7.80 (d, 2H, J ) 6.9 Hz).
N-Ben zyl-N-n itr osoben za m id e. This compound was pre-
1
9
pared by the method of White et al. IR (Nujol) 1719, 1609,
tube containing ∼20 mg of Na
evacuated, and sealed and was incubated at 40 °C. After 4
days, the solvent was removed by a gentle stream of N , and
H NMR spectra revealed
2 4
SO . The tube was cooled,
-
1 1
1
(
502, 1375, cm ; H NMR (CDCl
m, 10H).
Ha n d lin g a n d Stor a ge. N-Nitrosoamides are thermo-
3
) δ 5.14 (s, 2H), 7.36-7.60
2
1
3
the residue was dissolved in CDCl .
labile and are also decomposed by acids, bases, and moisture.
Dry, neutral N-benzyl-N-nitrosobenzamide (1) has been stored
at -25 °C for ∼1 month before significant decomposition into
ester was detected. Its half-life at 40 °C in ∼0.1 M solutions
in benzene is ∼5 h. N-Benzyl-O-benzoylhydroxylamine (6) has
been stored for months (dry and neutral) at -25 °C. These
reagents and their reaction mixtures were handled in the dark.
Ca u tion ! Nitrosoamides should be handled with care because
of their possible mutagenicity and carcinogenicity (local
and systemic). Efficient fume hoods and appropriate personal
protection (chemical-resistant gloves, safety glasses, lab coat,
etc.) are recommended when handling these compounds.
Decom p osit ion of N-Ben zyl-N-n it r osob en za m id e in
P yr r ole/CDCl . N-Benzyl-N-nitrosobenzamide (2.1 mg, 8.75
3
mmol) and pyridine (1.4 mL, 17.5 mmol, 2 equiv) were
dissolved in equimolar pyrrole (freshly distilled, bp760 ) 125-
1
NMR tube. The tube was cooled in liquid nitrogen and
evacuated at oil pump pressure; it was sealed and incubated
at 40 °C. Decomposition was complete in 3 days; the relative
yields of products were 38% 2-benzylpyrrole (δ 3.74), 21%
3
The solution was divided into two parts, and one part was
concentrated by evaporation of the solvent using a gentle
2 3
stream of N gas. Equimolar CDCl /pyrrole was added, and
NMR integrals were retaken to ensure that this mode of
evaporation did not affect the product distribution. The ratio
of the 2- to 3-isomer was unchanged by the evaporations. The
solvent was removed as before, and this time it was replaced
by 9:1 CCl
The chemical shifts thus obtained agreed very well with the
literature values for the 2- and 3-benzylpyrroles in CCl
that the reactions were complete. The observed product
distributions were as follows: In pyrrole: 20% benzyl benzoate
2
1a
(δ 5.23), 50% 2-benzylpyrrole (δ 3.74),
and 30% 3-benzylpyr-
role (δ 3.71)2 (Table 1). In N-methylpyrrole: 32% benzyl
benzoate, 33% 2-benzyl-N-methylpyrrole (δ 3.88), and 35%
1a
2
1b
-benzyl-N-methylpyrrole (δ 3.80).2
1c
3
Decom p osit ion of N-Ben zyl-N-n it r osob en za m id e in
F u r a n . N-Benzyl-N-nitrosobenzamide (5 mg, 20.8 mmol) and
pyridine (3.3 mg, 41.6 mmol) were dissolved in 500 mL of furan
2
0a
20b
in an NMR tube with ∼20 mg of Na
and set aside at 25 °C in the dark. After 14 days, the solvent
and the residue was
; H NMR analysis revealed 70% decomposi-
2 4
SO . The tube was sealed
was removed by a gentle stream of N
2,
1
dissolved in CDCl
3
tion had occurred to form 76% benzyl benzoate (δ 5.37), 21%
2
1
-benzylfuran (δ 3.97),23 and 3% 3-benzylfuran (δ 3.61) (Table
24
). The chemical shifts of the products agreed with the
2
3,24
literature values.
3
28 °C) (210 mL, 3 mmol) and CDCl (240 mL, 3 mmol) in an
Nitr osa tion of N-Ben zyl-O-ben zoylh yd r oxyla m in e (6)7a
in 30% F u r a n /CDCl
benzoylhydroxylamine (5 mg, 22 mmol) and pyridine (3.6 mL,
3
(vol/vol) a t -80 °C. N-Benzyl-O-
2
1a
44 mmol) were dissolved in a mixture of furan (150 mL) and
-benzylpyrrole (δ 3.71),21a and 41% benzyl benzoate (δ 5.22).
CDCl (350 mL) in an NMR tube. The solution was cooled to
3
-
80 °C in the NMR probe. Using a jet of dry N
2
, the tube
was raised to the top of the probe into a plastic bag filled with
3
dry N
2
. N
2
O
4
(0.27 cm , 11 mmol) was injected quickly ∼0.5
cm above the liquid surface through a serum stopper the inner
surface of which was lined with Teflon tape. The tube was
shaken quickly and was allowed to descend back into the
cooled probe. The total time out of the cooled probe was ∼5 s.
1
1
H NMR spectra were taken immediately and after 5 min and
4
/CDCl
3
(vol/vol), and H NMR spectra were retaken.
1
0 min. The temperature was raised in 10° intervals (∼5 min
per interval), and spectra were recorded at the end of each
interval. All the spectra had approximately the same appear-
ance. The product distribution: 92% benzyl benzoate (δ 5.41),
4
: δobs
2
1a
for 2-benzylpyrrole ) 3.77 (δlit. ) 3.78);
δobs for 3-benzylpyr-
2
1a
role ) 3.73 (δlit. ) 3.70). No N-benzylpyrrole was observed
2
3
2
2
3% benzyl nitrate (δ 5.25), 4.5% 2-benzylfuran (δ 4.01), and
.5% 3-benzylfuran (δ 3.68)24 did not change with time or
temperature. There was no evidence of the intermediacy of
(
absence of signal at δ 4.87).
The presence of benzyl benzoate was confirmed by (1)
spiking the product solution with commercial benzyl benzoate;
2) a TLC of the product mixture on silica gel using 10% ether/
hexane (vol/vol) gave a spot of R ) 0.4 (the same value
observed for commercial benzyl benzoate under the same
0
1
1
O-benzylfuranonium ions (absence of signal at δ ∼6).
(
f
Ack n ow led gm en t is made to the donors of the
Petroleum Research Fund, administered by the Ameri-
can Chemical Society, and to the D. Mead J ohnson
Foundation for support of this research. The authors
also thank Mr. Fenhong Song for the results in acetone
(Table 2).
(
18) Heyns, K.; v. Bebenburg, W. Chem. Ber. 1953, 86, 278.
(19) White, E. H.; Aufdermarsh, C. A., J r. J . Am. Chem. Soc. 1961,
8
3, 1174.
20) (a) Lee, K.; Gold, B.; Mirvish, S. Mutat. Res. 1977, 48, 131. (b)
(
Preussman, R.; Stewart, B. W. Chemical Carcinogenesis; Searle, C.,
Ed., ACS Monograph No. 182, American Chemical Society, Washing-
ton, DC, 1984; pp 643-828.
J O971081T
(21) (a) Patterson, J . M.; Burka, L. T. Tetrahedron Lett. 1969, 27,
(22) Katritzky, A. R.; Lang, H.; Lan, X. Tetrahedron 1993, 49, 2829.
(23) Pelter, A.; Rowlands, M.; Clements, G. Synthesis 1987, (1), 51.
(24) Kotake, H.; Inomata, K.; Aoyama, S.; Kinoshita, H. Chem. Lett.
1977, (1), 73.
2
1
4
215. (b) Logan, N. J .; Davies, C. S. J . Organomet. Chem. 1972, 39,
29 (c) Groves, J . K.; Anderson, H. J .; Nagy, H. Can. J . Chem. 1971,
9, 2427.