Radical and palladium-mediated cyclizations of ortho-iodo benzyl enamines: A scope and limitation study

Article abstract of DOI:10.1055/s-1998-1900

Radical and palladium-mediated cyclizations of ortho-iodo benzyl enamines were compared to determine their potential application to the synthesis of libraries of compounds on solid support.

Full text of DOI:10.1055/s-1998-1900

November 1998
SYNLETT
1227
Radical and Palladium-Mediated Cyclizations of Ortho-Iodo Benzyl Enamines:
A Scope and Limitation Study.
Sabine Berteina, Alain De Mesmaeker*
Novartis Crop Protection AG, R-1060, CH-4002 Basel, Switzerland
Email: alain.de_mesmaeker@cp.novartis.com; Fax: +41-61-697-8529
Received 4 June 1998
Abstract: Radical and palladium-mediated cyclizations of ortho-iodo
benzyl enamines were compared to determine their potential application
to the synthesis of libraries of compounds on solid support.
0.5 eq. DMAP, CH2Cl2, RT, 15h, 96%. ix) 2 eq. nBu4NF, 2 eq. AcOH,
THF, RT, 15h, 90%. x) 2.5 eq. CBr4, 2.4 eq. Ph3P, dioxane, RT, 1.5h;
then 5 eq. nBuNH2, RT, 15h, 63%. xi) 1.1 eq. Me2C=C(NMe2)Cl
(chlorenamine), CH2Cl2, RT, 1.5h; then 1 eq. nBu4NBr, 1.5 eq. NEt3, 5
eq. nBuNH2, RT, 15h, 78%.
Cyclization reactions mediated by palladium (intramolecular Heck
reaction) or radical initiators would be very useful in combinatorial
chemistry. Both processes are complementary. They are feasible from
the same starting material (aromatic or vinylic iodides, bromides
carrying an unsaturated chain) leading to cyclized products with
different oxidation states. The palladium-mediated reaction would
deliver the unsaturated product. In contrast, radical cyclization mediated
by tributyltin hydride would lead to reduced product (Scheme 1). Both
processes are performed under mild conditions and tolerate many
functional groups. Intermolecular processes are also feasible starting
from the same precursors.
After reduction of the nitro group1, the iodination reaction2 vi was
performed directly on the crude compound 2 (95% determined by 1H
NMR of the crude reaction mixture). The iodo derivative was obtained
in 70% yield for the two steps v, vi.
Enamines having a β-electron withdrawing group (push-pull) were
selected as substrates for radical and palladium-mediated cyclizations,
mainly due to their synthetic accessibility by reaction of 4 with
acetylenes3 or β-alkoxy acrylates (Scheme 3).
However, radical cyclizations attempted on substrates 5-9, under
standard experimental conditions using tributyltin hydride or tris
(trimethylsilyl)silane with AIBN as initiator, led only to reduced
compounds 10-14 (Scheme 3). Slow-addition techniques were not
applied here, due to potential impractibility for automated multiple
parallel syntheses. To avoid reduction of the incipient radical before
cyclization, we used a catalytic amount of tributyltin chloride in
presence of sodium borohydride4 (0.1 eq. nBu3SnCl, 2 eq. NaBH4, 0.35
eq. AIBN, benzene/t-BuOH (5/1) 0.01M, reflux, 60h). The reaction was
performed in a benzene/t-BuOH mixture as solvent to avoid the
undesired transesterification reaction (cleavage of benzoate ester)
obtained with a benzene/EtOH (5/1) mixture. The expected 5-exo
cyclized compound 15 was obtained from 5 only in 15 % yield, the
major product being the reduced compound 10 (85%, (42%)).
Scheme 1
We report here an investigation of these reactions in solution, in order to
establish the feasibility of these processes and to identify standard
experimental conditions which could be translated to solid phase
synthesis.
The basic core structure chosen to carry out the cyclizations was
synthesized from commercially available 4-(bromomethyl)benzoic acid
1 (Scheme 2).
Scheme 3 i) 1.3 eq. methyl propiolate, CH2Cl2, RT, 15h, 95%; ii) 1.2 eq.
dimethyl acetylene dicarboxylate, CH2Cl2, RT, 3.5h, 97%; iii) 2 eq.
ethyl 2-cyano-3-ethoxyacrylate, t-BuOH/dioxane (1/1), 60°C, 15h,
97%. iv) 1.3 eq. ethynyl p-tolylsulfone, CH2Cl2, RT, 15h, 75%. v) 1.2
eq. 3-butyn-2-one, CH2Cl2, RT, 15h, 87%. vi) 1.1-1.5 eq. nBu3SnH, 0.1
eq. AIBN, benzene 0.01M, reflux, 24h.
Scheme 2 i) 90% HNO3, -10°C, 3h, 83%. ii) Na2CO3, acetone/H2O (1/
1), reflux, 12h, 91%. iii) 2 eq. Et2O.BF3, MeOH, reflux, 15h, 83%. iv)
1.2 eq. tBuPh2SiCl, 0.5 eq. DMAP, 1.5 eq. NEt3, CH2Cl2, RT, 2h, 95%.
v) 5 eq. Na2S2O4, EtOH/H2O : (2/1), 100°C, 30min, 75%. vi) 3 eq.
iAmONO, 0.5 eq. I2, benzene, 100°C, 80%. vii) 2 eq. DIBAH, toluene,
RT, 3h, 95%. viii) 1.1 eq. pMeOPhCOCl, 10 eq. pyridine, 1.1 eq. NEt3,
The reluctance of these systems to undergo the 5-exo radical cyclization
might arise from the strong delocalization (push-pull) of the lone pair of
the nitrogen atom through the C=C double bond carrying an electron

1228
LETTERS
SYNLETT
withdrawing group. In the cyclization transition state, this conjugation is
disrupted, having a negative impact on the reaction rate.5 It is
noteworthy that, even in the case of the vinylsulfone 8, no cyclized
product was detected.
In contrast, substituents disfavoring the push-pull electron
delocalization should increase the rate of cyclization.6 Therefore, the
enamines 16, 18 were synthesized (Scheme 4). The push-pull electron
delocalization through the C=C double bound should be reduced in 16,
compared to the enamines 5-9, due to the lower electron withdrawing
ability of an amide group. The 5-exo cyclized product 22 was formed in
36% yield (Table 1, entry 1).
The phenyl substituent on the nitrogen atom in 18 decreases also the
electron delocalization through the C=C double bound. In this case
(Table 1, entry 2), 15% of the 5-exo cyclized product 24 and 2 % of the
6-endo cyclized product 23 were isolated. Moreover cyclization on the
N-phenyl ring occurred (25, 27%), although this process requires the
disruption of the aromaticity in the transition state of cyclization. The
cyclic enamine 19 7 was synthesized due to its ability to undergo 5-exo
radical cyclization.8 Under our standard reaction conditions, the major
product was the tricyclic system 28 (Table 1, entry 3). However, 23% of
the reduced product 27 were detected in the crude reaction mixture. The
much higher propensity of the cyclic enamine 19 to undergo a 5-exo
radical cyclization compared to the acyclic ketone 9 probably arises
from the non planarity of the nitrogen atom in the 6-membered ring of
19. The pyramidalization of the nitrogen atom disfavours the
delocalization of the electron lone pair through the C=C double bound.
Finally, the vinyl ether 20, obtained from ortho iodobenzyl alcohol by
addition to methyl propiolate9, underwent a smooth 5-exo cyclization.10
a) 1.1-1.5 eq. Bu3SnH, 0.1 eq. AIBN, benzene (0.01M), reflux, 24h; b)
Yields were determined by 1H NMR of the crude reaction mixture. In
parentheses: yields obtained after purification by flash chromatography.
Scheme 4 i) 1.5 eq. 1-(1-oxo-2-propynyl)pyrrolidine11, toluene, reflux,
72h, 77%. ii) 1.5 eq. diisopropylethylamine, 5 eq. aniline, 1 eq nBu4NI,
dioxane, 100°C, 1.5h, 100%. iii) 3.7 eq. cis-3-chloroacrylic methyl
ester12, 1.2 eq. Schwesinger base13, toluene, 100°C, 120h, 32% (55% of
starting material recovered).
are in agreement with the ones observed14 for enamines without β-
electron accepting group on the C=C double bond, where small
substituents on nitrogen favoured the 5-exo cyclization mode.
In order to assess whether the low rate of cyclization of phenyl radicals
on push-pull enamines could arise from a too restricted conformational
freedom, we have synthesized the radical precursor 38.15 Although the
corresponding radical has a much higher degree of conformational
freedom it led only to the reduced product 39 without trace of a 5/6-
membered ring (Scheme 5). In summary, our results suggest that push-
pull enamines are not suitable substrates for 5-exo cyclization reactions
under standard reaction conditions (no high dilution technique).
In contrast, palladium-mediated cyclizations16 performed on 5 with
Pd(OAc)2, KOAc and nBu4NCl led to 6-endo cyclized product 40
which underwent a quantitative oxidation by air to give 41 (47%).
Practically, in order to increase the rate of conversion of 40 into 41, a
stream of oxygen was bubbled through the solution for 24h. The
cyclization reaction was further improved to 86% yield by the use of
Pd(OAc)2 in the presence of Ph3P, nBu4NCl and K2CO3 (Scheme 6).
The reluctance of β-amino acrylates to cyclize led us to investigate the
influence of the size of the N-substituent. An intramolecular hydrogen
atom transfer and/or the induction of a disfavourable conformation for
cyclization could take place with a N-butyl substituent compared to N-
methyl. Therefore, we performed the cyclization reactions of the
derivatives 30-32 with nBu3SnH and nBu3SnD (Table 2).
The radical reaction using Bu3SnD in benzene-d6 (Table 2, entry 4)
revealed that deuterium was incorporated in the aromatic ring (reduction
prior to cyclization). Only
a minor amount of deuterium was
incorporated in the n-butyl side chain. Therefore, intramolecular
hydrogen atom transfer cannot account for the large amount of reduction
product in this system. Only for the N-methyl substituent, a substantial
amount of cyclized product 33 was observed, suggesting that the system
could adopt a more favourable conformation for cyclization. Our results

November 1998
SYNLETT
1229
This process was applied to compounds 8, 9, 16, 30 and 31. The 6-endo
cyclized products were obtained in good yields (Table 3, entries1-5).
The enol ether 20 led to the 5-exo cyclized product 47.
a) Yields were determined by 1H NMR of the crude reaction mixture. In
parentheses: yields obtained after purification by flash chromatography.
Scheme 5 i) 1 eq. acryloyl chloride, 1.2 eq. n-butylamine, 1.2 eq.
diisopropylethylamine, CH2Cl2, 0°C, 1.5h, 100%. ii) 0.5 eq.
diphenyldiselenide, 1.5 eq. NaBH4, EtOH, RT, 70 h, reflux, 5h, 40%. iii)
3.5 eq. LiAlH4, diethyl ether, RT, 18h, 43%. iv) 1.2 eq. methyl
propiolate, CH2Cl2, RT, 15h, 85%. v) 1.1 eq. nBu3SnH, 0.05 eq. AIBN,
benzene (0.01M), reflux, 15h, 56%.
a) 0.1 eq. Pd(OAc)2, 0.2 eq. Ph3P, 1 eq. Bu4NCl, 2 eq. K2CO3, DMF
(0.05M), 100°C, 15h, O2, ClCH2CH2Cl, RT, 48h. b) The reactions led
exclusively to the expected compounds by 1H NMR of the crude
reaction mixture. Yields reported in parentheses are obtained after
purification by flash chromatography.
In conclusion, push-pull enamines (as 8, 9, 16, 30, 31) adopt probably a
disfavourable conformation for a 5-exo cyclization mode. However,
with Pd(OAc)2 a 6-endo cyclization can take place. This is not the case
under radical cyclization conditions with nBu3SnH where the reduction
of the initial radical occurs prior to the 6-endo cyclization. In contrast,
the corresponding push-pull enol ether 20 (Table 3, entry 6) adopts a
more favourable conformation and/or displays a higher intrinsic
reactivity for 5-exo cyclization modes (Pd(OAc)2 and nBu3SnH18
17
Scheme 6 i) 0.1 eq. Pd(OAc)2, 1 eq. nBu4NCl, 4 eq. KOAc, dioxane
(0.05M), 100°C, 15h, 47%. ii) 0.1 eq. Pd(OAc)2, 0.2 eq. Ph3P, 1 eq.
nBu4NCl, 2 eq. K2CO3, DMF (0.05M), 100°C, 15h, 86%. iii) O2,
ClCH2CH2Cl, RT, 24h.
mediated), leading exclusively to the 5-membered ring products 47 and
29, respectively.

1230
LETTERS
SYNLETT
References and Notes
11. 1-(1-Oxo-2-propynyl)pyrrolidine was synthesized from 3-
(trimethylsilyl)propynoic acid by reaction with chlorenamine
Me2C=C(NMe2)Cl, followed by treatment with pyrrolidine in
presence of diisopropylethylamine. The silyl protective group was
removed by reaction with AgNO3 and KCN in a mixture of H2O/
EtOH/THF : (0.5/1/1).
1.
Kitahara, Y.; Shimizu, M.; Kubo, A. Heterocycles 1990, 31, 2085.
Grove, J.F.; MacMillan, J.; Mulholland, T.P.C.; Zealley, J. J.
Chem. Soc. 1952, 3967.
2.
3.
4.
5.
Friedman, L.; Chlebowski, J.F. J. Org. Chem. 1968, 33, 1636.
Magnus, P. and Giles, M. Tetrahedron Lett. 1993, 34, 6355.
Corey, E.J.; Suggs, J.W. J. Org. Chem. 1975, 40, 2554.
12. Mal, D. Synthetic Commun. 1986, 16, 331.
13. 2-tert-Butylimino-2-diethylamino-1,3-dimethyl-perhydro-1,3,2-
Middleton, D.S.; Simpkins, N.S.; Terrett, N.K. Tetrahedron Lett.
1989, 30, 3865.
diazaphosphorine.
6.
For examples of radical cyclization reactions of enamines and
enamides, see: Beckwith, A.L.J.; Westwood, S.W. Tetrahedron
1989, 45, 5269. Ziegler, F.E.; Harran, P.G. J. Org. Chem. 1993, 58,
2768. Ziegler, F.E.; Harran, P.G. Tetrahedron Lett. 1993, 34, 4505.
Ziegler, F.E.; Belema, M. J. Org. Chem. 1994, 59, 7962. Clive,
D.L.J.; Bergstra, R.J. J. Org. Chem. 1991, 56, 4976. Ziegler, F.E.;
Jeroncic, L.O. J. Org. Chem. 1991, 56, 3479. Ueda, T.; Shuto, S.
Heterocycles 1982, 17, 95. Ueda, T.; Usui, H.; Shuto, S.; Inoue, H.
Chem. Pharm. Bull. 1984, 32, 3410. Yamagata, Y.; Fujii, S.;
Fujiwara, T.; Tomita, K.I.; Ueda, T. Biochimica et Biophysica
Acta. 1981, 654, 242. Sugawara, T.; Otter, B.A.; Ueda, T.
Tetrahedron Lett. 1988, 29, 75. Suzuki, Y.; Matsuda, A.; Ueda, T.
Chem. Pharm. Bull. 1987, 35, 1808. Schultz, A.G.; Guzzo, P.R.;
Nowak, D.M. J. Org. Chem. 1995, 60, 8044. Yuasa, Y.; Ando, J.;
Shibuya, S. J. Chem. Soc., Chem. Commun. 1994, 1383. Yuasa,
Y.; Ando, J.; Shibuya, S. J. Chem. Soc., Chem. Commun. 1994,
455. Takano, S.; Suzuki, M.; Ogasawara, K. Heterocycles 1990,
31, 1151. Takano, S.; Suzuki, M.; Kijima, A.; Ogasawara, K.
Tetrahedron Lett. 1990, 31, 2315. Glover, S.A.; Warkentin, J. J.
Org. Chem. 1993, 58, 2115. Cladingboel, D.E.; Parsons, P.J. J.
Chem. Soc., Chem. Commun. 1990, 1543.
14. Glover, S.A.; Warkentin, J. J. Org. Chem. 1993, 58, 2115.
15. Browman, W.R.; Stephenson, P.T.; Terrett, N.K.; Young, A.R.
Tetrahedron 1995, 51, 7959.
16. For examples of Pd-mediated cyclizations, see: Larock, R.C.;
Babu, S. T. Tetrahedron Lett. 1987, 28, 5291. Grigg, R.;
Sridharan, V.; Stevenson, P.; Sukirthalingam, S.; Worakun, T.
Tetrahedron 1990, 46, 4003. Grigg, R.; Sridharan, V.; Stevenson,
P.; Sukirthalingam, S. Tetrahedron 1989, 45, 3557. Grigg, R.;
Sridharan, V.; Stevenson, P.; Worakun, T. J. Chem. Soc., Chem.
Commun. 1986, 1697. Desarbre, E.; Merour, J.Y. Heterocycles
1995, 41, 1987.
17. Experimental procedure: 100 mg of 20 (0.31mmol) were
solubilized in DMA (6 ml), then 17 mg of Ph3P (0.2 eq.), 93 mg of
nBu4NCl (1 eq.) and 86 mg of K2CO3 (2 eq.) were added. The
solution was degassed for 15 min, then 7 mg (0.1 eq.) of
Pd(OAc)2 were added. The reaction mixture was stirred under
argon atmosphere, heated 15h at 100°C. DMA was removed by
bulb to bulb distillation. The residue was dissolved in CH2Cl2,
washed twice with an aqueous solution of NaH2PO4, then with
brine. Organic layers were combined, dried (Na2SO4) and
evaporated to lead to 47 in 90% yield by 1H NMR of the crude
reaction mixture. After purification by flash chromatography on
silica gel (eluent: toluene/AcOEt 85/15), 47 was obtained in 60%
yield.
7.
8.
Comins, D.L.; Brown, J.D. Tetrahedron Lett. 1986, 27, 4549.
Beckwith, A.L.J.; Joseph, S.P.; Mayadunne, R.T.A. J. Org. Chem.
1993, 58, 4198.
9.
Winterfeldt, E.; Preuss, H. Chem. Ber. 1966, 99, 450.
18. Experimental procedure: 100 mg of 20 (0.31mmol) were
solubilized in degassed benzene (32 ml) and stirred under argon
atmosphere, then 92 µl (1.1 eq.) of nBu3SnH, 5 mg (0.1 eq.) of
AIBN were added. The mixture was refluxed for 24h. After
removal of the solvent, the residue was dissolved in acetonitrile,
washed four times with hexane in order to eliminate the organotin
compounds. Acetonitrile was evaporated to lead to 29 in 90%
yield by 1H NMR of the crude reaction mixture. After purification
by flash chromatography on silica gel (eluent: toluene/AcOEt 9/
1), 47 was isolated in 66% yield.
10. For examples of radical cyclization reactions of enol ethers, see:
Middleton, D.S.; Simpkins, N.S.; Terrett, N.K. Tetrahedron 1990,
46, 545. Middleton, D.S.; Simpkins, N.S.; Terrett, N.K.
Tetrahedron Lett. 1988, 29, 1315. Moriya, O.; Urata, Y.; Ikeda, Y.;
Ueno, Y.; Endo, T. J. Org. Chem. 1986, 51, 4708. Zahouily, M.;
Journet, M.; Malacria, M. Synlett 1994, 366. Lee, E.; Tae, J.S.;
Lee, C.; Park, C.M. Tetrahedron Lett. 1993, 34, 4831. Evans, P.A.;
Roseman, J.D. Tetrahedron Lett. 1995, 36, 31. Lee, E.; Park, C.M.
J. Chem. Soc., Chem. Commun. 1994, 293. Munt, S.P.; Thomas,
E.J. J. Chem. Soc., Chem. Commun. 1989, 480.