Methylenecyclopropane Rearrangement as a Probe for Free Radical Substituent Effects. σ·Values for Potent Radical-Stabilizing Nitrogen-Containing Substituents

Article abstract of DOI:10.1021/jo990732d

A series of nitrogen-containing 2-aryl-3,3-dimethylmethylenecyclopropanes have been prepared and rearrangement rates to the corresponding 2-arylisopropylidenecyclopropanes have been measured. These rates are dependent on the nature of the nitrogen-containing group in the para-position of the aryl group. Rearrangement rates have been used to calculate σ· values, which are a measure of the radical stabilizing ability of the substituent. Groups such as p-N=N-Bu-t, p-CH=N-Bu-t, p-NH₂, p-CH=N-OH, and p-CH=N-OCH₃, are "good" radical stabilizers. We have also classified groups such as p-CH=N-NMe₂, p-N=N-Ph, p-N=N(O)-Bu-t, p-CH=N(O)-Bu-t, and p-CH=N-O^- M⁺, which have an extraordinarily large radical stabilizing effect, as "Super Stabilizers". These substituents stabilize the transition state of the methylenecyclopropane rearrangement by extensive spin delocalization. In the case of the latter three substituents, nitroxyl type stabilization is proposed. Density functional calculations (B3LYP/6-31G*) have been carried out on a series of nitrogen-containing substituted benzylic radicals. Rates of the methylenecyclopropane rearrangement correlate with radical stabilization energies (ΔE) determined from an isodesmic reaction of substituted benzylic radicals with toluene. These calculations confirm substantial spin delocalization onto the nitrogen-containing substituents on the para-position of the benzylic radical.

Full text of DOI:10.1021/jo990732d

5634
J . Org. Chem. 1999, 64, 5634-5643
Meth ylen ecyclop r op a n e Rea r r a n gem en t a s a P r obe for F r ee
Ra d ica l Su bstitu en t Effects. σ • Va lu es for P oten t
Ra d ica l-Sta bilizin g Nitr ogen -Con ta in in g Su bstitu en ts
Xavier Creary,*,† Paul S. Engel,*,‡ Natasha Kavaluskas,† Li Pan,‡ and Allison Wolf†
Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, Indiana 46556, and
Department of Chemistry, Rice University, P.O. Box 1892, Houston, Texas 77251
Received May 3, 1999
A series of nitrogen-containing 2-aryl-3,3-dimethylmethylenecyclopropanes have been prepared and
rearrangement rates to the corresponding 2-arylisopropylidenecyclopropanes have been measured.
These rates are dependent on the nature of the nitrogen-containing group in the para-position of
the aryl group. Rearrangement rates have been used to calculate σ • values, which are a measure
of the radical stabilizing ability of the substituent. Groups such as p-NdN-Bu-t, p-CHdN-Bu-t,
p-NH2, p-CHdN-OH, and p-CHdN-OCH3, are “good” radical stabilizers. We have also classified
groups such as p-CHdN-NMe2, p-NdN-Ph, p-NdN(O)-Bu-t, p-CHdN(O)-Bu-t, and p-CHdN-
O- M+, which have an extraordinarily large radical stabilizing effect, as “Super Stabilizers”. These
substituents stabilize the transition state of the methylenecyclopropane rearrangement by extensive
spin delocalization. In the case of the latter three substituents, nitroxyl type stabilization is proposed.
Density functional calculations (B3LYP/6-31G*) have been carried out on a series of nitrogen-
containing substituted benzylic radicals. Rates of the methylenecyclopropane rearrangement
correlate with radical stabilization energies (∆E) determined from an isodesmic reaction of
substituted benzylic radicals with toluene. These calculations confirm substantial spin delocalization
onto the nitrogen-containing substituents on the para-position of the benzylic radical.
A number of scales have been developed over the last
splitting parameter D of cyclopentane-1,3-diyl biradicals
5.8 One of the present authors proposed a σ • scale based
on the rearrangement rate of substituted methylenecy-
clopropanes 6 to their isomers 8,9 while another has
published thermolysis rates of several R-substituted
azoalkanes 9.10-12 The azoalkane-based scale was pio-
neered by Timberlake13 and others,14 but more recent
results suggest the intervention of polar effects.15 Theo-
retical calculations of increasing sophistication have also
yielded reasonable radical-stabilization energies.8,16-20
20 years to quantify the ability of substituents to stabilize
a carbon-centered free radical. Some examples are J ack-
son’s scale based on pyrolysis rates of dibenzylmercurials
1,1 Fisher’s scale based on the free radical bromination
of 2,2 and J iang’s cyclodimerization rates of 3.3 Neumann
has evaluated radical-stabilizing effects from the dis-
sociation equilibrium constant of substituted triphenyl-
methyl dimers,4 while Bordwell has deduced bond dis-
sociation energies from pKa’s of substituted fluorenes and
the oxidation potential of their anions.5
(4) Neumann, W. P.; Penenory, A.; Stewen, V.; Lehnig, M. J . Am.
Chem. Soc. 1989, 111, 5845-51.
(5) Zhang, S.; Zhang, X.-M.; Bordwell, F. G. J . Am. Chem. Soc. 1995,
117, 602-606.
(6) Wayner, D. D. M.; Arnold, D. R. Can. J . Chem. 1985, 63, 2378-
83.
(7) J ackson, R. A.; Sharifi, M. J . Chem. Soc., Perkin Trans. 2 1996,
775-778.
(8) Adam, W.; Harrer, H. M.; Kita, F.; Korth, H.-G.; Nau, W. M. J .
Org. Chem. 1997, 62, 1419-1426.
(9) Creary, X.; Mehrsheikh-Mohammadi, M. E.; McDonald, S. J . Org.
Chem. 1987, 52, 3254-3263.
(10) Engel, P. S.; Bishop, D. J . J . Am. Chem. Soc. 1972, 94, 2148-
2149.
(11) Engel, P. S.; Bishop, D. J . J . Am. Chem. Soc. 1975, 97, 6754-
6762.
(12) Engel, P. S.; Wu, W.-X. J . Org. Chem. 1990, 55, 2720-2725.
(13) Luedtke, A.; Meng, K.; Timberlake, J . W. Tetrahedron Lett.
1987, 28, 4255-4258.
(14) For references to other azoalkane-based scales, see ref 15.
(15) Nau, W.; Harrer, H. M.; Adam, W. J . Am. Chem. Soc. 1994,
116, 10972-10982.
Two of the published scales do not involve kinetics or
equilibria, but instead they rely on the ESR spin density
of substituted benzylic radicals 46,7 and on the zero-field
(16) Pasto, D. J .; Krasnansky, R.; Zercher, C. J . Org. Chem. 1987,
52, 3062-3072.
(17) J ursic, B. S.; Timberlake, J . W.; Engel, P. S. Tetrahedron Lett.
1996, 37, 6473-6474.
(18) Lehd, M.; J ensen, F. J . Org. Chem. 1991, 56, 884-885.
(19) Leroy, G.; Peeters, D.; Sana, M.; Wilante, C. In Substituent
Effets in Radical Reactions; Reidel: Dordrecht, 1986; p 1.
(20) Wu, Y.-D.; Wong, C.-L.; Chan, K. W. K.; J i, G.-Z.; J iang, X.-K.
J . Org. Chem. 1996, 61, 746-750.
† University of Notre Dame.
‡ Rice University.
(1) J ackson, R. A. J . Organomet. Chem. 1992, 437, 77-83.
(2) Fisher, T. H.; Meierhoefer, A. W. J . Org. Chem. 1978, 43, 224-
228.
(3) J iang, X. K.; J i, G. Z. J . Org. Chem. 1992, 57, 6051-6056.
10.1021/jo990732d CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/02/1999

Probe for Free-Radical Substituent Effects
J . Org. Chem., Vol. 64, No. 15, 1999 5635
Comparison of the various scales is hampered in many
cases by an insufficient number of substituents in com-
mon.
route gave the requisite azoaldehyde 25 in low yield.
p-Formylbenzoic acid was converted30 to p-formylphenyl-
isocyanate 23,31 which was then reacted with tert-
butylamine to form Schiff base urea 24.32 Hypochlorite
oxidation33 and hydrolysis afforded p-formylphenylazo-
tert-butane 25, the precursor to 10, while treatment of
25 with m-chloroperbenzoic acid34 gave azoxy compound
26, the precursor to 11.
The Creary scale9 is appealing because polar effects
are minimal in breaking the strained cyclopropane
bond.3,7 The arylmethylenecyclopropanes present some
synthetic challenges since the cyclopropane must be
generated in the presence of the aryl substituent or the
substituent must be introduced in the presence of the
strained ring without heating above ∼50 °C. Neverthe-
less, a large number of substrates of type 6 have been
prepared and studied in the past. Our recent interest in
the radical-stabilizing ability of the azo21-23 and azoxy
groups24 has now led us to determine the rearrangement
rates of compounds 10-12. For comparison with 10-12,
we have also studied the nitrone derivative 13, the Schiff
base 14, the amine 15, the azide 16, the hydrazone 17,
and the phenylazo compound 18. Finally, we studied
three oxime derivatives, 19-21. With the exception of
NH2 and Ph-NdN, these substituents are not included
in any of the existing radical stabilization scales, yet some
of them had the potential to be extraordinary radical
stabilizing groups. Reported here are the results of these
studies.
The transformation of p-nitrosobenzaldehyde 31 to 32
by Kovacic’s method35 was carried out in 78% yield.
However, in our hands, the reported36,37 synthesis of
p-nitrosobenzaldehyde, 31, from p-nitrobenzaldehyde, 27,
was not reproducible, probably because the intermediate
hydroxylamine from reduction of 27 can condense with
an aldehyde group. We therefore protected 27 as the
ethylene acetal before converting nitro to nitroso.38,39 In
this manner, the synthesis of 32 could be achieved in 48%
overall yield from 27.
Resu lts
Syn th eses of 10-12. Attempts to formylate phenylazo-
tert-butane25-27 using POCl3/DMF28 or SnCl4/CHCl2OMe29
led to recovery of starting material, but the following
(21) Engel, P. S.; Wang, C.; Chen, Y.; Ru¨chardt, C.; Beckhaus, H.-
D. J . Am. Chem. Soc. 1993, 115, 65-74.
(22) Engel, P. S.; Chen, Y.; Wang, C. J . Am. Chem. Soc. 1991, 113,
4355-4356.
(23) Engel, P. S.; Wang, C.; Chen, Y.; He, S.-L.; Andrews, B. K.;
Weisman, R. B. J . Org. Chem. 1994, 59, 6257-6261.
(24) Engel, P. S.; He, S.-L.; Wang, C.; Duan, S.; Smith, W. B. J . Am.
Chem. Soc. 1999, submitted.
(25) Curtin, D. Y.; Ursprung, J . A. J . Org. Chem. 1956, 21, 1221-
1225.
(26) Kisch, H.; Reisser, P.; Knoch, F. Chem. Ber. 1991, 124, 1143-
1148.
(27) Garst, M. E.; Lukton, D. Synth. Commun. 1980, 10, 155-160.
(28) Campaigne, E.; Archer, W. L. In Organic Syntheses; Wiley: New
York, 1963; Collect. Vol. 4, pp 331-333.
The desired methylenecyclopropanes were prepared
from 25, 26, and 32, in the usual manner: conversion of
aldehyde to tosylhydrazone, reaction of tosylhydrazone

5636 J . Org. Chem., Vol. 64, No. 15, 1999
Creary et al.
with base, pyrolysis of the tosylhydrazone salt, and Cu-
(OTf)2-catalyzed decomposition of the resulting diazo
compound in the presence of 1,1-dimethylallene.9
Syn th eses of 13-21. The p-bromophenylmethylenecy-
clopropane 3640 provided a convenient starting material
for the preparation of a variety of substrates. Lithiation
of 36 with n-butyllithium followed by reaction with
dimethylformamide gave the key aldehyde 37. This was
converted to the nitrone 13 by reaction with N-tert-
butylhydroxylamine. Likewise, the imine 14 and the
hydrazone 17 could be prepared by the toluenesulfonic
acid-catalyzed condensation of 37 with tert-butylamine
and 1,1-dimethylhydrazine, respectively. The O-methy-
lated oxime derivative 19 and the simple oxime 20 were
also prepared from 37 using O-methylhydroxylamine
hydrochloride and hydroxylamine hydrochloride, respec-
tively. Deprotonation of the oxime derivative 20 with
alkali metal methoxides in DMSO gave the anions 21.
Amine 15 and azide 16 were prepared by lithiation of
36 followed by reaction with trimethylsilylmethyl azide41
and toluenesulfonyl azide,42,43 respectively. Base-cata-
lyzed condensation of 15 with nitrosobenzene afforded
the phenylazo compound 18.44
Kin etic Stu dies. Methylenecyclopropanes 10-19 were
thermally rearranged to the corresponding isopropy-
lidenecyclopropanes 8 in C6D6, while 20 and the anionic
substrates 21a -d were thermally rearranged in DMSO-
d6. Rate data and the corresponding σ • values are
summarized in Tables 1 and 2. To eliminate errors due
to potential temperature differences in kinetic studies
carried out in two different laboratories, rates of the
parent compound 6 (X ) H) were determined in both
laboratories. Relative rate data from each laboratory
were used in determining σ • values for compounds
studied in that laboratory. All of the substrates 10-21
rearrange faster than the unsubstituted derivative 6 (X
) H). The observed rate enhancements indicate that the
nitrogen-containing substituents stabilize the developing
benzylic radical intermediate 7 to varying degrees.
(29) Meier, H.; Kretzschmann, H.; Kolshorn, H. J . Org. Chem. 1992,
57, 6847-6852.
(30) Yamada, S.; Ninomiya, K.; Shioiri, T. Tetrahedron Lett. 1973,
2343-2346.
(31) Kuppe, A.; Mrsny, R. J .; Shimizu, M.; Firsan, S. J .; Keana, J .
F. W.; Griffith, O. H. Biochemistry 1987, 26, 7693-7701.
(32) Fowler, J . S. J . Org. Chem. 1972, 37, 510-511.
(33) Porter, N. A.; Marnett, L. J . J . Am. Chem. Soc. 1973, 95, 4361-
4367.
Discu ssion
Consider first the magnitude of the rate enhancements
that have previously been observed in the methylenecy-
clopropane rearrangement. Since the developing radical
center is insulated from the substituent by the aromatic
ring, substituent effects are attenuated. Some rate
enhancements are “small”, i.e., in the range of 1.3-1.5
for groups such as p-CH3 and p-Cl.45 In contrast, unsat-
urated conjugating groups such as cyano, nitro, and vinyl
are “moderate” to “good” radical stabilizers, showing rate
enhancements (krel’s) of 2.87, 3.76, and 4.67, respectively.9
On the basis of krel’s of 4.85 and 3.97 in 10 and 14, the
unsaturated groups p-NdN-Bu-t and p-CHdN-Bu-t are
classified as “good” radical stabilizers. Spin delocalization
in the biradical intermediate, as in 38 and 39, accounts
(34) Sullivan, F. R.; Luck, E.; Kovacic, P. J . Org. Chem. 1974, 39,
2967-2970.
(35) Nelson, V.; Serianz, A.; Kovacic, P. J . Org. Chem. 1976, 41,
1751-1754.
(36) Riad, Y.; Assad, A. N.; Gohar, G.-A. S.; Abdullah, A. A. Z.
Naturforsch. 1984, 39, 893-898.
(37) Alway, F. J . Chem. Ber. 1903, 36, 2303-2311.
(38) Wood, W. W.; Wilkin, J . A.; Kent, S. Synth. Commun. 1992,
22, 1683-1686.
(39) Entwistle, I. D.; Gilkerson, T.; J ohnstone, R. A. W.; Telford, R.
P. Tetrahedron 1978, 34, 213-215.
(40) Creary, X. J . Org. Chem. 1978, 43, 1777-1783.
(41) Nishiyama, K.; Tanaka, N. J . Chem. Soc., Chem. Commun.
1983, 1322-1323.
(42) Smith, P. A. S.; Rowe, C. D.; Bruner, L. B. J . Org. Chem. 1969,
34, 3430-3433.
(43) Spagnolo, P.; Zanirato, P.; Gronowitz, S. J . Org. Chem. 1982,
47, 7, 3177-3180.
(44) Brown, E. V.; Kipp, W. H. J . Org. Chem. 1971, 36, 170-173.
(45) Creary, X. J . Org. Chem. 1980, 45, 280-284.

Probe for Free-Radical Substituent Effects
J . Org. Chem., Vol. 64, No. 15, 1999 5637
Ta ble 1. Rea r r a n gem en t Ra tes a n d σ • Va lu es of
compound 6 (X ) p-CHdCH2). Thus, substitution of
carbon by nitrogen in a double bond has only a minor
effect on radical stabilization. In previous studies of R-azo
radicals, we were only able to state that an aliphatic azo
group was less than 3.1 kcal/mol more stabilizing than
vinyl.21,23
Meth ylen ecyclop r op a n es in C6D6
∆H‡
∆S‡
substrate
6 (p-H)
T, °C
k (s-1
)
(kcal/mol) (eu) krel
σ •
80.0 5.57 × 10-5
70.0 1.72 × 10-5
60.0 5.05 × 10-6
80.1 5.86 × 10-5
80.0b 5.78 × 10-5
70.1 1.79 × 10-5
60.2 5.23 × 10-6
27.4
-0.8 1.00 0.00
6 (p-H)a
27.7
26.3
0.2 1.00 0.00
12 (p-N(O)dNBu-t) 80.0 1.11 × 10-4
70.0 3.60 × 10-5
16 (p-N3)a
-2.4 2.00 0.30
60.0 1.10 × 10-5
80.1 1.41 × 10-4
80.0b 1.40 × 10-4
70.1 4.55 × 10-5
60.2 1.41 × 10-5
80.0 2.14 × 10-4
70.0 7.22 × 10-5
60.0 2.04 × 10-5
80.0 2.70 × 10-4
70.0 8.83 × 10-5
60.0 2.66 × 10-5
80.1 2.85 × 10-4
80.0b 2.82 × 10-4
70.0 9.46 × 10-5
60.2 2.96 × 10-5
26.4
26.8
26.4
-1.7 2.41 0.38
0.4 3.97 0.60
-0.4 4.85 0.69
14 (p-CHdNBu-t)
10 (p-NdNBu-t)
15 (p-NH2)a
Unusually large krel’s of 9.45 and 13.5 are seen for the
azoxy derivative 11 and the isoelectronic nitrone 13. Prior
to this paper, there were no single substituents with krel
above 7.88, the value for dimethylamino.9 Transition-
state stabilization by nitroxyl forms such as 40 and 41
accounts for these rapid rearrangement rates. The high
thermodynamic stability of nitroxyl radicals is apparent
from the >30 kcal/mol lower O-H bond dissociation
energy (BDE) of N,N-dialkylhydroxylamines relative to
ordinary alcohols.46,47
In contrast to the large krel seen in the azoxy-
substituted methylenecyclopropane 11, the isomeric azoxy
derivative 12 rearranges only 2.0 times faster than the
unsubstituted system. The observation that p-N(O)dN-
Bu-t is a weaker stabilizer than the isoelectronic p-NO2
substituent (krel ) 3.76) is attributable to the fact that
the resonance structures of an R-nitro radical are identi-
cal but those of the R-azoxy radical are not. In support
of this rationale, a computational study (vide infra) shows
significant spin delocalization onto the nitrogen atom of
a p-N(O)dN-Bu-t substituted benzylic radical, but mini-
mal delocalization onto the oxygen atom.
25.9
25.7
25.7
25.8
-1.8 4.88 0.69
-2.3 5.13 0.71
-1.3 8.27 0.92
-0.8 9.45 0.98
19 (p-CHdNOCH3) 80.0 2.86 × 10-4
70.0 9.31 × 10-5
17 (p-CHdNNMe2) 80.0 4.61 × 10-4
70.0 1.50 × 10-4
11 (p-NdN(O)Bu-t) 80.0 5.26 × 10-4
70.0 1.75 × 10-4
18 (p-NdNPh)a
60.0 2.99 × 10-5
60.0 4.80 × 10-5
60.0 5.45 × 10-5
80.1 7.12 × 10-4
80.0b 7.00 × 10-4
70.0 2.38 × 10-4
60.1 7.76 × 10-5
25.2
25.0
-1.8 12.1 1.08
-2.4 13.5 1.13
13 (p-CHdN(O)Bu-t) 80.0b 7.53 × 10-4
70.0 2.60 × 10-4
60.0 8.31 × 10-5
50.0 2.53 × 10-5
a
Data from Rice University. Other data are from University of
b
Notre Dame. See the Experimental Section. Extrapolated from
data at other temperatures.
Ta ble 2. Rea r r a n gem en t Ra tes a n d σ • Va lu es of
Meth ylen ecyclop r op a n es in DMSO-d 6
According to Table 1, the amino group is a good radical
stabilizer, comparable to vinyl (krel ) 4.67) but it is not
as strong as dimethylamino (krel ) 7.88). An early mass
spectroscopic study suggested that more alkyl substitu-
ents on nitrogen greatly increased the stabilization of an
R-amino radical.48 However, more recent computa-
tional49,50 and experimental8,51,52 work showed no such
effect. The C-H BDE of 9-dimethylaminofluorene is 7
kcal/mol greater than that of the amino analogue,52 the
opposite of our krel values. While steric inhibition of
resonance can rationalize the fluorene results, it is in the
wrong direction to explain ours. A powerful electron
donor like Me2N may result in some ground-state desta-
∆H‡
∆S‡
substrate
6 (p-H)
T, °C
k (s-1
)
(kcal/mol) (eu) krel σ •
80.0 7.90 × 10-5
70.0 2.45 × 10-5
60.0 7.44 × 10-6
80.0 4.01 × 10-4
70.0 1.35 × 10-4
60.0 4.23 × 10-5
26.9
25.6
25.9
-1.5 1.00 0.00
-1.9 5.08 0.71
20 (p-CHdNOH)
21a (p-CHdNO- Li+) 80.0a 1.83 × 10-3
60.0 1.90 × 10-4
1.8 23
-2.2 28
0.5 42
1.2 45
1.36
1.45
1.62
1.65
50.0 5.36 × 10-5
40.0 1.47 × 10-5
21b (p-CHdNO- Na+) 80.0a 2.23 × 10-3
24.3
25.0
25.2
60.0 2.57 × 10-4
50.0 8.47 × 10-5
40.0 2.32 × 10-5
bilization of 6 (X ) p-NMe2), raising krel
.
21c (p-CHdNO- K+) 80.0a 3.35 × 10-3
60.0 3.72 × 10-4
50.0 1.13 × 10-4
(46) Mahoney, L. R.; Mendenhall, G. D.; Ingold, K. U. J . Am. Chem.
Soc. 1973, 95, 8610-8614.
40.0 3.15 × 10-5
21d (p-CHdNO- Cs+) 80.0a 3.55 × 10-3
(47) Bordwell, F. G.; Liu, W.-Z. J . Am. Chem. Soc. 1996, 118, 8777-
60.0 3.89 × 10-4
8781.
(48) Griller, D.; Lossing, F. P. J . Am. Chem. Soc. 1981, 103, 1586-
1587.
50.0 1.16 × 10-4
40.0 3.22 × 10-5
(49) Kysel, O.; Mach, P. THEOCHEM 1986, 139, 333-337.
(50) Goddard, J . D. Can. J . Chem. 1982, 60, 1250-1255.
(51) Wayner, D. D. M.; Clark, K. B.; Rauk, A.; Yu, D.; Armstrong,
D. A. J . Am. Chem. Soc. 1997, 119, 8925-8932.
(52) Bordwell, F. G.; Zhang, X.-M.; Alnajjar, M. S. J . Am. Chem.
Soc. 1992, 114, 7623-7629.
a
Extrapolated from data at other temperatures.
for this radical-stabilizing ability of the azo and imino
groups, completely analogous to that seen in the vinyl

5638 J . Org. Chem., Vol. 64, No. 15, 1999
Creary et al.
The azido group is an effective carbocation-stabilizing
substituent,53,54 being comparable to methoxy. We there-
fore were interested in determining whether p-N3 would
be a good radical-stabilizing group. However, data in
Table 1 show that p-N3 is only a “moderate” radical
stabilizer (krel ) 2.41), somewhat below cyano (krel ) 2.87)
but slightly better than methoxy (krel ) 1.97).
nitrogen and thermolyzed 11 times faster at 150 °C than
43.22 The resonance stabilization of R-azo radical 45 was
estimated to be at least 6 kcal/mol greater than that of
its tert-butyl analogue.21
The potent carbocation-stabilizing properties of the
O-methyloxime group55,56 prompted us to examine the
radical-stabilizing ability of p-CHdNOCH3. On the basis
of the krel values of 5.13 and 5.08 for 19 and 20, we
classify the p-CHdNOCH3 and p-CHdNOH groups as
“good” radical stabilizers. These groups enhance the
rearrangement rate to approximately the same extent as
p-CHdCH2, p-NdN-Bu-t, and p-CHdN-Bu-t, indicating
a similar spin delocalization mechanism.
Introduction of the dimethylamino group onto the
nitrogen in the form of the hydrazone 17 (p-CHdN-
NMe2) gives a further rate enhancement relative to the
oxime 19 (p-CHdNOCH3). The rearrangement rate of 17
now approaches that of the azoxy and nitrone derivatives
11 and 13. The extraordinary radical-stabilizing ability
of p-CHdN-NMe2 is attributed to further spin delocal-
ization as in 42a and 42b , which is recognizable as a
hydrazyl radical. These radicals are highly stabilized by
three-electron bonding, as evidenced by the 26.5 kcal/
mol difference between the N-H BDE of ammonia (107.3
kcal/mol)57 and the most recent value for hydrazine (80.8
kcal/mol).58 The fact that 17 rearranges faster than 19
suggests that the Me2N group stabilizes the nitrogen-
centered radical more effectively than does the MeO
group of 19. Indeed, the same trend is found in comparing
the N-H BDE of phenylhydroxylamine (77.5 kcal/mol)52
with that of 1,2-diphenylhydrazine (73.1 kcal/mol).59
These differences are easily rationalized by the better
electron-donating properties of nitrogen than oxygen.
The best of the radical stabilizing groups in this paper
is the anionic p-CHdN-O- M+ group of 21a -21d . The
rearrangement rates of these anionic substrates show a
dependence on the counterion, with moderate increases
as the metal cation becomes larger. This trend suggests
that the “free anion” p-CHdNO- is the best radical
stabilizer. The krel of 45 exhibited by the cesium salt 21d
is exceeded by no other simple substituent on a phenyl
ring; however, such large factors have been observed
when the attenuating benzene ring is omitted. The σ•
value for 21d of 1.65 is comparable to the γ• values of
the 2-thienyl (γ• ) 1.70) and 2-furanyl (γ• ) 1.64) groups
and exceeded only by the most potent radical stabilizing
group that we have reported to date, the 4-pyridyl-N-
oxide group (γ• ) 1.88).60,61 We attribute this “super”
radical stabilizing effect of p-CHdN-O- to spin delocal-
ization as in form 46a and additional delocalization
utilizing the readily available nonbonding electrons on
oxygen as in 46b.
Tables 1 and 2 include activation parameters for the
rearrangement of 16 arylmethylenecyclopropanes. We
attribute the 4.2 eu range of ∆S‡ to random experimental
error and conclude that the krel values are governed by
∆H‡. This outcome is as expected if krel measures stabi-
lization of biradical 7 by the benzene ring substituents.
The average ∆S‡ for all entries is -0.9 eu, a figure that
should be quite reliable and is consistent with ∆S‡ for
related methylenecyclopropane rearrangements.62,63
Com p u ta tion a l Stu d ies. To gain further insight into
the radical-stabilizing nature of the nitrogen-containing
substituents, density functional molecular orbital calcu-
lations were carried out on a series of benzylic radicals
and the corresponding substituted toluenes. Similar
calculations have been reported20 for 15 benzylic radicals
The krel of 12.1 for the phenylazo derivative 18 is nearly
as large as that of nitrone 13 and greatly exceeds that of
alkylazo compound 10 (krel ) 4.88). Extended conjugation
in biradical 7 is surely responsible for this rate enhance-
ment. Only one previous radical-stability scale has
included the Ph-NdN substituent and surprisingly placed
it between NO2 and CN.2 The present result is in accord
with thermolysis studies of bisazoalkanes 43 and 44.
Whereas 43 afforded nitrogen quantitatively, 44 gave no
(53) Hoz, S.; Wolk, J . L. Tetrahedron Lett. 1990, 31, 4085-4088.
(54) Amyes, T. L.; Richard, J . P. J . Am. Chem. Soc. 1991, 113, 1867-
1869.
(55) Creary, X.; Wang, Y.-X.; J iang, Z. J . Am. Chem. Soc. 1995, 117,
3044-3053.
(60) Creary, X.; Mehrsheikh-Mohammadi, M. E. Tetrahedron Lett.
1988, 29, 749-752.
(56) Creary, X.; J iang, Z. J . Org. Chem. 1996, 61, 3482-3489.
(57) Lias, S. G.; Bartmess, J . E.; Liebman, J . F.; Holmes, J . L.; Levin,
R. D.; Mallard, W. G. J . Phys. Chem. Ref. Data 1988, 17 (Suppl. 1),
1-872.
(61) Creary, X.; Mehrsheikh-Mohammadi, M. E.; McDonald, S. J .
Org. Chem. 1989, 54, 2904-2910.
(62) LeFevre, G. N.; Crawford, R. J . J . Org. Chem. 1986, 51, 747-
749.
(63) Roth, W. R.; Winzer, M.; Lennartz, H.-W.; Boese, R. Chem. Ber.
1993, 126, 2717-2715.
(58) Ruscic, B.; Berkowitz, J . J . Chem. Phys. 1991, 95, 4378-4384.
(59) Ingemann, S.; Fokkens, R. H.; Nibbering, N. M. M. J . Org.
Chem. 1991, 56, 607-612.

Probe for Free-Radical Substituent Effects
J . Org. Chem., Vol. 64, No. 15, 1999 5639
Ta ble 3. B3LYP /6-31G* En er gies of Ra d ica ls 48 a n d
Tolu en es 50 a n d ∆E Va lu es for th e Isod esm ic Rea ction of
Su bstitu ted Ben zylic Ra d ica ls 48 w ith Tolu en e
substituted radicals 48 by spin delocalization. The bond
from the aromatic ring to the benzylic radical center gets
progressively shorter as radical stabilization (measured
by ∆E values) increases. For example, the C-C bond
length of 1.407 Å for the unsubstituted benzyl radical
49 exceeds the 1.395 Å for the “stabilized” p-NdN-CH3-
substituted radical 51, which in turn exceeds the 1.389
Å for the “super stabilized” p-NdN(O)CH3-substituted
radical 52. In valence bond terms, these bond length
changes indicate progressively increasing importance of
forms with spin delocalized onto the para substituent.
B3LYP
energy (au)
ZPE
(au)
∆E
(kcal/mol)
substituent
48 (p-H)
48 (p-CH3)
48 (p-N(O)dNCH3)
48 (p-N3)
48 (p-CHdNCH3)
48 (p-NH2)
48 (p-NMe2)
48 (p-CHdCH2)
48 (p-NdNCH3)
48 (p-CHdNOCH3)
48 (p-CHdNNMe2)
48 (p-NdNPh)
48 (p-NdN(O)CH3)
48 (p-CHdN(O)CH3)
48 (p-CHdNO-)
4-pyridyl-CH2
3-pyridyl-CH2
4-pyridyl-N-oxide-CH2 -362.123719 0.107943
50 (p-H)
50 (p-CH3)
50 (p-N(O)dNCH3)
50 (p-N3)
50 (p-CHdNCH3)
50 (p-NH2)
50 (p-NMe2)
50 (p-CHdCH2)
50 (p-NdNCH3)
50 (p-CHdNOCH3)
50 (p-CHdNNMe2)
50 (p-NdNPh)
50 (p-NdN(O)CH3)
50 (p-CHdN(O)CH3)
50 (p-CHdNO-)
4-pyridyl-CH3
3-pyridyl-CH3
-270.915143 0.114935
-310.233556 0.142461
-494.873123 0.157968
-434.506512 0.118004
-403.677819 0.164847
-326.269835 0.131645
-404.883745 0.188564
-348.317735 0.147986
-419.687565 0.152339
-478.852187 0.168931
-498.316384 0.210117
-611.433669 0.205541
-494.877912 0.158037
-478.851831 0.170159
-438.971717 0.127345
-286.951160 0.103251
-286.951463 0.103099
0.00
0.34
0.60
1.09
1.39
1.46
1.72
1.73
1.80
1.89
2.34
2.89
2.96
3.48
11.0
-1.09
-0.43
6.00
-271.566648 0.128296
-310.884526 0.155818
-495.523671 0.171280
-435.156276 0.131334
-404.327110 0.178093
-326.919009 0.144898
-405.532506 0.201748
-348.966485 0.161228
-420.336197 0.165622
-479.500686 0.182089
-498.964157 0.223281
-612.080569 0.218711
-495.524706 0.171461
-479.497792 0.183509
-439.605655 0.139936
-287.604404 0.116599
-287.603654 0.116626
The spin density at the benzylic carbon also mirrors
these findings. There is a progressive drop in spin density
at the benzylic position with increasing radical stabiliza-
tion. Substantial spin delocalization onto certain atoms
of the para substituent, which is in accord with predic-
tions based on valence bond considerations, is also
indicated by these computational studies.
Interestingly, the calculated structure of the azoxy-
substituted benzylic radical 52 shows a shorter C-N
bond and a longer NdN bond than the corresponding
substituted toluene 54. Analogous trends are seen in the
nitrone-substituted radical 53, and both radicals 52 and
53 show high calculated spin densities on nitrogen and
oxygen. In valence bond terms, these bond length changes
4-pyridyl-N-oxide-CH3 -362.765666 0.120791
different from those included here. Table 3 summarizes
our B3LYP/6-31G* energies and the corresponding values
of ∆E for the isodesmic reaction of toluene with substi-
tuted benzylic radicals 48. The degree of stabilization
and spin density trends are consistent with substantial
contributions from nitroxyl radical forms. These compu-
tational results are in line with the observed large rate
enhancements seen in the azoxy and nitrone derivatives
11 and 13.
The calculated structure of the radical anion 55 also
merits further discussion. A high degree of spin delocal-
ization is indicated by a large calculated spin density on
oxygen as well as smaller spin density on nitrogen
relative to the other entries in Table 4. Also of interest
are calculated charge densities in 55, where the benzylic
carbon has considerable negative charge. In valence bond
terms, nitroxyl form 55a has major importance. The
value of ∆E for the reaction of 55 with toluene (11 kcal/
mol)64 is the largest of the substituents studied and
indicative of major stabilization in the radical 55. The
experimental observation that the anionic systems 21 are
parallels the rearrangement rates of the substituted
methylenecyclopropanes. Figure 1 shows a linear free
energy relationship between krel and the isodesmic reac-
tion energy. The correlation is quite good (r ) 0.983) and
further demonstrates that density functional calculations
can be useful in determining radical-stabilizing effects.
The slope of the plot is 0.7 and suggests that ap-
proximately 70% of the radical-stabilizing effect is mani-
fested in the transition state of the methylenecyclopro-
pane rearrangement. The points for p-NH2 and p-NMe2
(open circles) show the greatest deviations from the line
and the larger than expected krel values may indicate
ground-state destabilization of the corresponding meth-
ylenecyclopropanes.
Calculated bond length and spin density data (Table
4) are in accord with extensive stabilization of the

5640 J . Org. Chem., Vol. 64, No. 15, 1999
Creary et al.
F igu r e 1. Plot of ln krel for rearrangement of substituted methylenecyclopropanes 6 vs calculated (B3LYP/6-31G*) ∆E values for
the isodesmic reaction of toluene with substituted benzylic radicals 48.
Ta ble 4. B3LYP /6-31G* Bon d Len gth s (C1-Ar yl Bon d )
a n d Sp in Den sities of Ra d ica ls 48
bond spin density at
length C1 (benzylic
spin density on X
(N, O, or C atoms)
substituent
4-pyridyl-CH2
48 (p-H)
3-pyridyl-CH2
48 (p-CH3)
48 (p-N(O)dNCH3)
48 (p-NH2)
48 (p-NMe2)
(Å)
carbon)
1.4086
1.4067
1.4047
1.4052
1.4011
1.4028
1.4021
1.4002
1.3981
1.3966
1.3956
1.3952
1.3951
1.3944
1.3888
1.3887
0.814
0.792
0.795
0.783
0.754
0.757
0.750
0.746
0.732
0.721
0.714
0.710
0.709
0.699
0.653
0.656
0.637
0.495
0.571
N, 0.108; O, -0.021
N, 0.064
N, 0.075
48 (p-N3)
N(2), 0.009; N(3), 0.098
N, 0.162
C, 0.223
N, 0.178; O, 0.036
N, 0.178; O, 0.038
N, 0.198
N, 0.148; N, 0.074
N, 0.223
N, 0.144; O, 0.205
N, 0.166; O, 0.234
N, 0.066; O, 0.300
N, 0.190; O, 0.371
48 (p-CHdNCH3)
48 (p-CHdCH2)
48 (p-CHdNOH)
48 (p-CHdNOCH3)
48 (p-NdNCH3)
48 (p-CHdNNMe2)
48 (p-NdNPh)
48 (p-NdN(O)CH3)
tion state for the rearrangement accounts for these large
rate enhancements. Computational studies (B3LYP/6-
31G*) on substituted benzylic radicals reveal an excellent
linear free energy relationship between rate of the
methylenecyclopropane rearrangement and radical-
stabilization energy. These calculations confirm extensive
spin delocalization in many nitrogen-substituted benzylic
radicals. Carbocation studies would be lacking without
inclusion of substituents, such as p-methoxy, which have
special cation-stabilizing features. By the same token, we
suggest that correlations of radical reactions should
include substituents that have special radical-stabilizing
properties. Some of the nitrogen-containing substituents
described in this paper provide excellent examples of
potent radical stabilizers that could be employed as
criteria for evaluation of radical processes.
48 (p-CHdN(O)CH3) 1.3867
48 (p-CHdNO-)
1.3852
4-pyridyl-N-oxide-CH2 1.3776
the fastest rearranging methylenecyclopropanes in the
present study confirms this suggestion.
Con clu sion s. The radical-stabilizing ability of a va-
riety of nitrogen containing groups has been evaluated
by measuring their effect on the rearrangement rate of
2-aryl-3,3-dimethylmethylenecyclopropanes. The fastest
rearrangement occurred when the aryl substituent was
the anionic p-CHdNO- group. The azoxy, nitrone, and
phenylazo groups, (p-NdN(O)-Bu-t, p-CHdN(O)-Bu-t,
and p-NdN-Ph) were also outstanding rate enhancers.
We classify these groups as “super radical stabilizers”.
Radical stabilization by spin delocalization in the transi-
(64) The anionic p-CHdNO- substituent is not included in the
correlation in Figure 1. The comparison of calculated energies of
charged molecules in the gas phase with neutral molecules is unwar-
ranted.

Probe for Free-Radical Substituent Effects
J . Org. Chem., Vol. 64, No. 15, 1999 5641
method39 followed by PCC oxidation of the hydroxylamine.38
A mixture of 2.5 g of p-nitrobenzaldehyde ethylene acetal (28)65
(0.0128 mol) in 25 mL of THF containing 40 mg of 5% Rh/C
was stirred as 15 mL of 65% aqueous hydrazine solution was
added dropwise. Gas evolution was immediate. After 1 h, TLC
showed no remaining starting material. The mixture was
poured into 80 mL of distilled water and quickly extracted with
3 × 50 mL of ether. The combined organic extracts were dried
over Na2SO4 under nitrogen. The ether was rapidly removed
by rotary evaporation to give crude p-hydroxylaminobenzal-
dehyde ethylene acetal (29). To a solution of this crude product
in 25 mL of freshly distilled, dry THF was added 2.98 g of
pyridimium chlorochromate (1.05 equiv). After 10 min, the
dark brown slurry was filtered through a dry pad of silica gel,
and the silica gel was washed with THF. The solvent was
removed, and the crude p-nitrosobenzaldehyde ethylene acetal
(30) was further purified by column chromatography on silica
gel with 30% ethyl acetate in hexane as eluent. Compound 30
(60% yield) was a green solid that turned yellow on standing
due to dimerization: 1H NMR (CDCl3) δ 4.10 (m, 4 H), 5.89
(s, 1 H), 7.73 (d, 2 H, J ) 8.5 Hz), 7.91 (d, 2 H, J ) 8.5 Hz);
13C NMR (CDCl3) δ 65.45, 102.57, 120.92, 127.39, 145.12,
165.56.
Compound 30 (1.4 g) was hydrolyzed by stirring with 100
mL of 4 M HCl under nitrogen for 50 min. The bright yellow
solid (dimer of 31) was collected by filtration, washed five times
with distilled water, and recrystallized from THF at -5 °C.
The yield of 31 (as the dimer) was 90%, mp 135-136 °C. This
compound was found to be sensitive to dry potassium carbon-
ate. The dimer of 31 slowly dissolved in CDCl3 to give the
monomer in solution: 1H NMR of 31 (CDCl3) δ 8.03 (d, 2 H, J
) 8.4 Hz), 8.16 (d, 2 H, J ) 8.4 Hz), 10.19 (s, 1 H); 13C NMR
(CDCl3) δ 121.05, 131.04, 139.42, 163.71, 191.26.
Exp er im en ta l Section
p-F or m ylp h en yl isocya n a te (23) was prepared using a
modification of Yamada’s method.30 To 4-carboxybenzaldehyde
(3 g, 20 mmol) suspended in dry methylene chloride (50 mL)
was added triethylamine (2.78 mL, 20 mmol). When all of the
solid had dissolved, diphenylphosphoryl azide (4.3 mL, 20
mmol) was added dropwise. The resulting solution was re-
fluxed for 6 h, and the solvent was removed by rotary
evaporation. The resulting light yellow solid was dissolved in
50 mL of ether and washed with 2 × 50 mL portions of aqueous
NaHCO3 (pH 9). The ether layer was dried over K2CO3, and
the solvent was evaporated to give 23 as a white solid (3 g, 20
mmol, yield: 100%): 1H NMR (CDCl3) δ 8.00 (d, 2 H, J ) 8.2
Hz), 8.27 (d, 2 H, J ) 8.2 Hz), 10.12 (s, 1 H); 13C NMR (CDCl3)
δ 129.65, 130.0, 135.27, 139.96, 171.68, 191.32; IR 3096, 2918,
2855, 2173, 2140, 1676, 1250 cm-1
.
N-ter t-Bu tyl-N′-(p-for m ylp h en yl)u r ea , ter t-bu tylim in e
(24) was prepared according to a method modified from
Fowler.32 To p-formylphenyl isocyanate (3 g, 20 mmol) in 100
mL of benzene was added dry tert-butylamine (2 equiv, 2.93
g, 4.20 mL), and the solution was heated to reflux for 5.5 h.
The solvent was removed by rotary evaporation, and the
residual white solid was recrystallized from ether. Purification
was difficult due to partial Schiff base hydrolysis. The yield
of 24 was 65%: 1H NMR (CDCl3) δ 1.29 (s, 9 H), 1.37 (s, 9 H),
7.33 (d, 2 H, J ) 8.6 Hz), 7.66 (δ, 2 H, J ) 8.6 Hz), 8.18 (s, 1
H). For N-tert-butyl, N′-(p-formylphenyl) urea: 1H NMR
(CDCl3) δ 1.39 (s, 9 H), 5.08 (s, 1 H), 7.09 (s, 1 H), 7.49 (d, 2 H,
J ) 8.5 Hz), 7.77 (d, 2 H, J ) 8.5 Hz), 9.85 (s, 1 H); 13C NMR
δ 29.20, 51.13, 118.03, 130.61, 131.41, 145.49, 191.27.
p-(ter t-Bu tyla zo)ben za ld eh yd e (25) was prepared by
Porter’s method.33 Potassium tert-butoxide (49 mg, 0.44 mmol,
1.2 equiv) was dissolved in 2 mL of tert-butyl alcohol, and 0.1
g (0.364 mmol) of the urea 24 in 2 mL of tert-butyl alcohol
was then added. After the slurry was stirred for 35 min at
room temperature, a solution of 69 mL (1.6 equiv) of t-BuOCl
in 1 mL of tert-butyl alcohol was added over 10 min. Stirring
was continued for another 30 min. Ice (5 g) was added, the
solution was stirred for 5 min, and then ether (25 mL) was
added. The organic layer was separated and washed with 3 ×
50 mL of ice-water. The water layer was extracted with
hexane, and the ether and hexane solutions were combined
and treated with aqueous HCl (15 mL). After being stirred
for 1 h at room temperature, the organic layer was separated
and dried over K2CO3. The solvent was removed by rotary
evaporation, and the crude product 25 was purified by column
chromatography on silica gel with 20% ethyl acetate in hexane
p-(ter t-Bu tyla zoxy-NNO)ben za ld eh yd e (32) was pre-
pared by a modification of Kovacic’s method.35 p-Nitrosoben-
zaldehyde, 31 (0.5 g, 3.7 mmol), was dissolved in 20 mL of
acetonitrile with stirring and heat (monomer formation). Solid
potassium iodide (0.615 g) was then added followed by the
dropwise addition of 0.52 g (3.7 mmol) of freshly prepared tert-
butyl N,N-dichloroamine66 at room temperature. The solution
turned brown immediately. The mixture was stirred overnight
at room temperature under nitrogen. The solvent was removed
by rotary evaporation, and the aldehyde 32 was purified by
silica gel chromatography, eluting with 10% ethyl acetate in
hexane. The yield of 32 (brown oil) was 78%: 1H NMR (CDCl3)
δ 1.45 (s, 9 H), 7.95 (d, 2 H, J ) 8.7 Hz), 8.26 (d, 2 H, J ) 8.7
Hz), 10.09 (s, 1 H); 13C NMR (CDCl3) δ 25.61, 59.45, 123.01,
129.97, 137.92, 152.22, 190.93; HRMS (CI) (M+ + H) calcd for
1
as eluent. The yield of 25 was 30%: mp 43-45 °C; H NMR
C
11H15N2O2 207.1133, found 207.1142.
(CDCl3) δ 1.37 (s, 9 H), 7.75 (d, 2 H, J ) 8.4 Hz), 7.98 (d, 2 H,
J ) 8.4 Hz), 10.08 (s, 1 H); 13C NMR (CDCl3) δ 26.85, 68.83,
122.40, 130.612, 136.89, 155.85, 191.69; HRMS (70 eV, EI)
calcd for C11H14N2O 190.1106, found 190.1105. Anal. Calcd for
Ar yld ia zom eth a n es (33-35) were prepared by the previ-
ously described tosylhydrazone salt solution pyrolysis method.67
Approximately 0.78 mmol of tosylhydrazine was suspended in
3 mL of methanol containing about 5 mg of pyridine, and 0.74
mmol of the appropriate aldehyde 25, 26, or 32 was added to
the stirred solution. After 8 h at room temperature, 0.80 mmol
of NaOCH3 in methanol was added. The methanol was
removed using a rotary evaporator, and 8 mL of ethylene glycol
was added. The mixture was stirred until the tosylhydrazone
salt dissolved, and then the solution was heated in an oil bath
at 80-90 °C with periodic extraction with ether. When no more
diazo compound formed, the combined ether extracts were
washed with water and dried over MgSO4. Solvent removal
left the corresponding diazo compounds 33, 34, and 35, which
were used without further purification. The thermally labile,
light-sensitive diazo compounds were stored in the dark at -20
°C.
C
11H14N2O: C, 69.44; H 7.42; N, 14.72. Found: C, 69.36; H,
7.42; N, 14.76.
p-(ter t-Bu tyl-ONN-a zoxy)ben za ld eh yd e (26) was pre-
pared by a modification of the method of Kovacic.34 A solution
of 1 g of p-(tert-butylazo)benzaldehyde, 25, in 30 mL of
methylene chloride was cooled to 0 °C, and 1.8 g of 50%
m-chloroperbenzoic acid (1.0 equiv) was added. The mixture
was stirred for 3 h in an ice bath, and the solid m-chlorobenzoic
acid was removed by filtration. Sodium hydroxide (3 M, 15
mL) was added dropwise to the filtrate cooled in an ice bath,
and the solution was stirred for 15 min. The methylene
chloride was removed by rotary evaporation, and the crude
product was purified by column chromatography on silica gel
with 20% ethyl acetate in hexane as eluent. The aldehyde 26
was then recrystallized from pentane at low temperature. The
In a typical procedure, reaction of 153 mg of aldehyde 32
with 145 mg of tosylhydrazine gave 146 mg (90% yield) of diazo
1
yield of 26 was 40%: mp 47-48 °C; H NMR (CDCl3) δ 1.68
(s, 9 H), 7.87 (d, 2 H, J ) 8.6 Hz), 7.94 (d, 2 H, J ) 8.6 Hz),
10.01 (s, l H); 13C NMR (CDCl3) δ 28.38, 78.59, 124.68, 130.15,
135.33, 149.11, 191.28. Anal. Calcd for C11H14N2O2: C, 64.06;
H, 6.84; N, 13.58. Found: C, 64.21; H, 6.93; N, 13.47.
p-Nitr osoben za ld eh yd e (31) was made by reduction of the
nitro group to the hydroxylamine according to Entwistle’s
(65) Walton, R.; Lahti, P. M. Synth. Commun. 1998, 28(6), 1087-
1092.
(66) Kling, T. A.; White, R. E.; Kovacic, P. J . Am. Chem. Soc. 1972,
94, 7416-7421.
(67) Creary, X. Organic Syntheses; Wiley: New York, 1990; Collect.
Vol VII, p 438.

5642 J . Org. Chem., Vol. 64, No. 15, 1999
Ta ble 5. 1H NMR Sp ectr a l Da ta (δ) for Meth ylen ecyclop r op a n es 10-21 in CDCl3
Creary et al.
other
compd
CH3
CH
CH2
aromatic
10 p-NdNBu-t
11 p-NdN(O)Bu-t
12 p-N(O)dNBu-t
13 p-CHdN(O)Bu-t
14 p-CHdNBu-t
15 p-NH2
16 p-N3
17 p-CHdNNMe2
18 p-NdNPh
0.853, 1.357
0.850, 1.348
0.835, 1.359
0.836, 1.348
0.835, 1.349
0.831, 1.306
0.830, 1.334
0.838, 1.337
0.897, 1.382
0.846, 1.348
0.856, 1.353
0.793, 1.283
2.499
2.481
2.501
2.487
2.484
2.369
2.432
2.451
2.541
2.464
2.472
2.436
5.56, 5.50
5.55, 5.58
5.56, 5.60
5.55, 5.59
5.55, 5.59
5.51, 5.54
5.53, 5.58
5.54, 5.57
5.58, 5.62
5.54, 5.59
5.55, 5.59
5.51, 5.54
7.25, 7.56
7.21, 7.87
7.22, 7.99
7.22, 8.19
7.21, 7.64
6.61, 6.98
6.93, 7.17
7.14, 7.47
7.32, 7.49
7.18, 7.47
7.20, 7.47
6.94, 7.28
1.327
1.648
1.463
1.606, 7.506
1.287, 8.249
3.552
2.947, 7.257
7.82-7.92
3.959, 8.037
8.116
19 p-CHdNOCH3
20 p-CHdNOH
21 p-CHdNO- K+ a
7.811
a
In DMSO-d6.
Ta ble 6. 13C NMR Sp ectr a l Da ta (δ) for Meth ylen ecyclop r op a n es 10-21 in CDCl3
CH3 CH CH2 aromatic
compd
other
10 p-NdNBu-t
11 p-NdN(O)Bu-t
12 p-N(O)dNBu-t
18.43, 26.16 32.13 103.62 121.58, 129.39, 145.45, 150.65
18.40, 26.15 32.26 103.58 124.54, 128.90, 142.35, 145.42
18.35, 26.11 31.86 103.95 121.75, 128.91, 142.32, 144.90
24.32, 27.05, 67.42, 140.76
24.38, 28.44, 77.38, 139.49
24.60, 25.85, 58.91
13 p-CHdN(O)Bu-t 18.28, 26.11 32.39 103.61 128.46, 128.87, 141.02, 145.18
24.47, 28.31, 70.41, 128.70, 129.81
24.24, 29.80, 57.13, 134.99, 155.14
22.96, 146.35
23.78, 145.37
23.90, 42.95, 133.35, 134.58
14 p-CHdNBu-t
15 p-NH2
16 p-N3
17 p-CHdNNMe2
18 p-NdNPh
19 p-CHdNOCH3
20 p-CHdNOH
18.39, 26.17 32.26 103.55 127.56, 129.10, 140.93, 145.46
18.44, 25.94 31.44 102.90 114.88, 128.24, 129.79, 144.31
18.37, 25.97 31.53 103.54 118.57, 130.23, 135.25, 137.52
18.40, 26.14 32.17 103.31 125.28, 129.10, 137.78, 145.79
18.37, 26.18 32.35 103.71 129.04, 129.53, 130.68, 145.32, 151.02, 152.80 24.77, 142.24
18.38, 26.17 32.25 103.64 126.68, 129.28, 140.68, 145.30
18.38, 26.15 32.22 103.67 126.71, 129.35, 140.98, 145.23
24.37, 61.97, 129.79, 148.65
24.43, 129.53, 150.31
22.72, 141.50, 145.70
21 p-CHdNO- K+ a 18.29, 25.65 31.50 103.26 122.78, 128.33, 133.04, 137.31
a
In DMSO-d6.
compound 35: 1H NMR of 35 (CDCl3) δ 1.461 (s, 9 H), 5.034
(s, 1 H), 6.91 and 8.04 (AA′BB′ quartet, 4 H); 13C NMR of 35
(CDCl3) δ 25.91, 48.08, 58.88, 120.69, 123.06, 134.09; 1H NMR
of 33 (CDCl3) δ 1.322 (s, 9 H), 5.014 (s, 1 H), 7.63 and 6.96
(AA′BB′ quartet, 4 H); 13C NMR of 33 (CDCl3) δ 27.10, 48.01,
matographed on 22 g of silica gel, and the column was eluted
with increasing amounts of ether in hexanes. A small amount
of debrominated product (6, X ) H) eluted with pure hexane.
The aldehyde 37 (containing increasing amounts of the
isomeric aldehyde 8 (X ) p-CHO) eluted with 4% ether in
hexanes. The purest fraction contained 95% of 37 along with
5% of the isomeric aldehyde 8 (X ) p-CHO). The total yield of
products was 433 mg (78% yield): 1H NMR of 37 (CDCl3) δ
0.869 (s, 3 H), 1.376 (s, 3 H), 2.526 (m, 1 H), 5.56 (m, 1 H),
5.62 (m, 1 H), 7.78 and 7.33 (AA′BB′ quartet, 4 H), 9.961 (s, 1
H); 13C NMR of 37 (CDCl3) δ 18.25, 25.32, 26.18, 32.55, 104.04,
129.39, 129.47, 134.37, 144.64, 146.28, 191.93. Anal. Calcd for
1
67.20, 121.42, 123.15, 132.17; H NMR of 34 (CDCl3) δ 1.641
(s, 9 H), 5.011 (s, 1 H), 8.00 and 6.92 (AA′BB′ quartet, 4 H);
13C NMR of 34 (CDCl3) δ 28.43, 48.43, 77.21, 120.92, 126.26,
130.76.
Cu (OTf)2-Ca t a lyzed R ea ct ion of Ar yld ia zom et h a n es
33-35 w ith 1,1-Dim eth yla llen e. Gen er a l P r oced u r e. Ap-
proximately 15 mg of Cu(OTf)2 was stirred in 2.5 mL of 1,1-
dimethylallene as a solution of approximately 80-100 mg of
aryldiazomethanes 33-35 in 13 mL of 1,1-dimethylallene was
added dropwise over a 1 h period. The color of the diazo
compound disappeared during the addition. The excess allene
was removed at 15 mm pressure (and recovered), and the
residue was chromatographed on approximately 7 g of silica
gel. The product 10 eluted with hexanes, while the products
11 and 12 eluted with 5% ether in hexanes. Solvent removal
gave 10, 11, or 12, which contain varying amounts of the
“rearranged” methylenecyclopropane 8 (formed by carbene
addition to the less substituted double bond of the allene).
Satisfactory high-resolution mass spectral data (EI) were
obtained for 10, 11, and 12. 1H and 13C NMR data are
summarized in Tables 5 and 6.
2-(p-F or m ylp h en yl)-3,3-d im et h ylm et h ylen ecyclop r o-
p a n e 37. A solution of 795 mg of the 2-(p-bromophenyl)-3,3-
dimethylmethylenecyclopropane 36 (which contained 10% of
the isomeric methylenecyclopropane)40 in 4 mL of dry THF was
cooled to -78 °C, and 3.2 mL of 1.6 M n-BuLi in hexane was
added dropwise to the stirred solution. The mixture was
warmed to -50 °C for 5 min and then recooled to -78 °C. After
30 min at -78 °C, a solution of 360 mg of dimethylformamide
in 3 mL of ether was added dropwise. The mixture was allowed
to slowly warm to 0 °C, and water was then added. The organic
phase was separated, and the aqueous phase was extracted
with an additional portion of ether. The combined organic
extracts were washed with saturated NaCl solution and then
dried over MgSO4. After filtration, the organic solvents were
removed using a rotary evaporator. The residue was chro-
C
13H14O: C, 83.83; H, 7.58. Found: C, 83.61; H, 7.42.
P r ep a r a tion of Ald eh yd e Der iva tives 13, 19, a n d 20.
Gen er a l P r oced u r e. The methylenecyclopropanes 13, 19, and
20 were prepared by the reaction of aldehyde 37 with either
N-tert-butylhydroxylamine hydrochloride, O-methylhydroxyl-
amine hydrochloride, or hydroxylamine hydrochloride in pryi-
dine. A mixture of 0.30 mmol of aldehyde 37, and 0.60 mmol
of the appropriate hydroxylamine hydrochloride in 1 mL of
pyridine was stirred at room temperature for 24 h. The
mixture was then taken up into ether, washed with water,
dilute HCl solution, and saturated NaCl solution, and dried
over MgSO4. After filtration, the ether solvent was removed
using a rotary evaporator, and the residue was chromato-
graphed on silica gel and eluted with increasing amounts of
ether in hexanes. Solvent removal gave 13, 19, or 20 in 92,
96, and 96% yields, respectively. Satisfactory high-resolution
1
mass spectral data (EI) were obtained for 13, 19, and 20. H
and 13C NMR data are summarized in Tables 5 and 6.
2-(p-F or m ylp h en yl)-3,3-d im et h ylm et h ylen ecyclop r o-
p a n e, t-Bu tyl Im in e. A solution of 98 mg of aldehyde 37 and
80 mg of tert-butylamine in 1.0 mL of CCl4 was stirred at room
temperature as 13 mg of p-toluenesulfonic acid was added.
Anhydrous MgSO4 (356 mg) was then added, and the mixture
was stirred at room temperature for 140 h. The mixture was
then diluted with 2 mL of hexane and centrifuged. The solvent
was carefully decanted and removed by rotary evaporator to
give 113 mg (89% yield) of imine 14, which was used without
1
further purification for kinetic studies. H and 13C NMR data

Probe for Free-Radical Substituent Effects
J . Org. Chem., Vol. 64, No. 15, 1999 5643
are given in Tables 5 and 6: HRMS (EI+) calcd for C17H23
241.1831, found 241.1859.
N
the mixture was stirred for another 5 h, during which time a
precipitate formed. The ether was removed by rotary evapora-
tion, and the solid was washed with 3 × 3 mL dry hexane and
redissolved in 3 mL of ether. Sodium pyrophosphate (0.24 g,
1 equiv) in 2 mL of H2O was added at 0 °C, and stirring was
continued overnight. The organic layer was separated, and the
aqueous solution was extracted with 3 × 2 mL of ether. The
pure azide 16 was obtained in 50% yield by silica gel chroma-
tography, eluting with hexane. The NMR data can be found
in Tables 5 and 6: IR 3064, 3038, 2953 (s), 2862, 2431 (w),
2260 (w), 2129 (s), 2089 (s), 1514, 1454 (m), 1376 (w), 1291(s),
1115(m); HRMS (CI+) calcd for (C12H13N3 + 1H) 200.11877,
found 200.11926.
2-(p -P h e n y la zo )-3,3-d im e t h y lm e t h y le n e c y c lo p r o -
p a n e (18) was prepared according to Brown’s method.44 2-(p-
Aminophenyl)-3,3-dimethylmethylenecyclopropane (15) (60
mg) was dissolved in a mixture of 4.5 mL of pyridine and 1.1
mL of distilled water in an ice-water bath. Then 2 equiv of
nitrosobenzene and 4 equiv of tetramethylammonium hydrox-
ide pentahydrate were added into the mixture. The ice-water
bath was removed, and the solution was allowed to stand at
room temperature for 48 h. After 1 mL of distilled water was
added, the solution was extracted with 3 × 3 mL toluene. The
toluene was removed by rotary evaporation. Purification by
silica gel chromatography using 5% EtOAc in hexane afforded
18 as a thick, orange-red liquid in 38% isolated yield. The NMR
data can be found in Tables 5 and 6: UV λmax 440 nm; HRMS
(EI+) calcd for C18H18N2 262.1470, found 262.1469.
Kin etics P r oced u r es. Kinetics procedures for the thermal
rearrangements of 10-21 were analogous to those previously
described.9,61 A solution of approximately 3.0 mg of the
appropriate methylenecyclopropane and 1.5 mg of dimethyl
maleate in 0.75 mL of C6D6 was sealed under nitrogen in an
NMR tube. The tube was immersed in a constant-temperature
bath (allowing 30 s to reach the bath temperature) for a given
amount of time. The tube was withdrawn from the bath,
immediately quenched in cold water, and then analyzed for
unreacted methylenecyclopropane by 300 MHz NMR at 22 °C
(where the rate is negligible). The olefinic singlet of the
dimethyl maleate at δ 5.66 served as the internal standard,
and the relative area of the vinyl region of the unreacted
methylenecyclopropane at δ 5.56-5.35 was determined as a
function of time. In determining peak areas, care was taken
to use 90° pulses, long relaxation delays, as well as well-phased
spectra. First-order rate constants were determined by stan-
dard least-squares methods. Correlation coefficients were
greater than 0.9995. The kinetics of 15, 16, and 18 were carried
out similarly at Rice University but without the 30 s equilibra-
tion period, which has no effect on relative rate constants. To
ensure the accuracy of our krel values, the kinetics of 6 (X )
H) were repeated at Rice. Both values for 6 (X ) H) are shown
in Table 1, and they differ by 3.8% at 80 °C. In an alternative
kinetics procedure, the relative area of the methyl signals of
unreacted methylenecyclopropane (δ 1.1-1.3) and the area of
the methyl signal of the rearranged product (δ 1.8-1.9) was
determined as a function of time. Rate constants were deter-
mined by standard methods. Rates in DMSO-d6 were deter-
mined using this method.
2-(p-F or m ylp h en yl)-3,3-d im et h ylm et h ylen ecyclop r o-
p a n e, N,N-Dim eth yl Hyd r a zon e 17. A solution of 34.6 mg
of aldehyde 37 and 31 mg of 1,1-dimethylhydrazine in 1.0 mL
of ether was stirred at room temperature, and 3.3 mg of
p-toluenesulfonic acid was added. After 6 h, about 20 mg of
Na2CO3 was added followed by a small amount of anhydrous
MgSO4. The mixture was then filtered, and the solvent was
removed using a rotary evaporator leaving 41.2 mg of hydra-
zone 17 (97% yield), which was used without further purifica-
tion for kinetic studies. 1H and 13C NMR data are given in
Tables 5 and 6: HRMS (EI+) calcd for C15H20N2 228.1626,
found 228.1625.
Oxim e An ion s 21a -d were prepared by reaction of 1.2
equiv of the corresponding metal methoxide (prepared by
dissolving the appropriate alkali metal in absolute methanol)
with 1.0 equiv of oxime 20. In a typical procedure, 82 µL of
1.03 M NaOCH3 in methanol was placed a 5 mL flask under
nitrogen, and the solvent was carefully removed under aspira-
tor vacuum. The dry NaOCH3 (under nitrogen) was dissolved
with stirring in 1.5 mL of DMSO-d6, and a solution of 14.2 mg
of oxime 20 in 0.6 mL of DMSO-d6 was added. The DMSO-d6
solution was kept under nitrogen at all times and carefully
transferred to three NMR tubes. The tubes were sealed under
nitrogen and used directly for kinetic studies. 1H and 13C NMR
data for 21c in DMSO-d6 are given in Tables 5 and 6.
2-(p -Am in op h en yl)-3,3-d im et h ylm et h ylen ecyclop r o-
p a n e (15) was synthesized by a modification of Nishiyama’s
method.41 tert-Butyllithium (2.0 eq of 1.7 M in pentane) was
slowly added into a solution of 0.22 g of 2-(p-bromophenyl)-
3,3-dimethylmethylenecyclopropane 36 in 2 mL of dry THF
at -78 °C under nitrogen. The solution was stirred at -78 °C
for 40 min, and then the temperature was allowed to rise and
the solution was stirred for another 5 min at 0 °C. After 1.2
equiv of trimethylsilylmethyl azide in 3 mL of dry THF was
added dropwise at -78 °C, the mixture was allowed to warm
slowly to room temperature; then it was stirred for another 3
h. Water (2 mL) and then ether (2 mL) were added dropwise
at 0 °C, and the aqueous solution was drained off and then
washed with 3 × 3 mL of Et2O. The organic phases were
combined and dried sequentially over Na2SO4 and MgSO4. A
0.5 g portion of silica gel was added to the organic phase, and
the mixture was stirred at room temperature until the
bubbling subsided (2-3 h). The silica gel was removed by
filtration and was washed with 3 × 5 mL of EtOAc. Rotary
evaporation of the combined organic phases yielded the crude
15, which was purified by silica gel chromatography using 20%
EtOAc in hexane as eluent. The yield of light brown oil was
49%. The NMR data can be found in Tables 5 and 6: HRMS
(EI+) calcd for C12H15N 173.12045, found 173.1205.
2-(p -Azid op h e n yl)-3,3-d im e t h ylm e t h yle n e cyclop r o-
p a n e (16) was prepared by a modification of the method of
Smith.42 A 0.21 g portion of 2-(p-bromophenyl)-3,3-dimethyl-
methylenecyclopropane 36 was dissolved in 2.5 mL of dry
ether. Under argon, 2.0 equiv of tert-butyllithium (1.7 M in
pentane) was added dropwise with stirring at -78 °C. After
40 min, the bath temperature was raised to 0 °C, and the
mixture was stirred for 5 min. A solution of p-toluenesulfo-
nyl azide (1.1 equiv in 1.5 mL of Et2O) was added dropwise at
-78 °C, and the mixture was stirred for 1 h at -78 °C. The
bath was allowed to warm slowly to room temperature, and
Com p u ta tion a l Stu d ies. Molecular orbital calculations
were performed using the Gaussian 94 series of programs.68
Structures were characterized as energy minima via frequency
calculations that showed no negative frequencies.
(68) Frisch, M, J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Peng, C. Y.; Ayala, P.
Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
94, Revision B.3, Gaussian, Inc., Pittsburgh, PA, 1995.
Ack n ow led gm en t. We thank the National Science
Foundation and the Robert A. Welch Foundation for
financial support. Helpful suggestions from Professor
Marco Ciufolini are gratefully acknowledged.
J O990732D