Unexpected products from the Fp2-catalyzed reductive cyclization of nitroaromatics bearing pendant unsaturation

Article abstract of DOI:10.1016/S0040-4020(02)01627-7

A representative o-nitroenone (Z=O) was cyclized by reduction with CO and [CpFe(CO)₂]₂ (Fp₂) as the catalyst to give the corresponding 4-quinolone. In contrast, Baylis-Hillman adducts derived from o-nitrobenzaldehydes were

Full text of DOI:10.1016/S0040-4020(02)01627-7

Tetrahedron 59 (2003) 747–754
Unexpected products from the Fp2-catalyzed reductive cyclization
of nitroaromatics bearing pendant unsaturation
*
David K. O’Dell and Kenneth M. Nicholas
Department of Chemistry and Biochemistry, University of Oklahoma, 620 Parrington Oval, Norman, OK 73019, USA
Received 28 August 2002; revised 12 December 2002; accepted 12 December 2002
Abstract—A representative o-nitroenone (Z¼O) was cyclized by reduction with CO and [CpFe(CO)2]2 (Fp2) as the catalyst to give the
corresponding 4-quinolone. In contrast, Baylis–Hillman adducts derived from o-nitrobenzaldehydes were cyclized to N-formylindolines and
indoles under the same conditions. q 2003 Published by Elsevier Science Ltd.
1. Introduction
to catalyze the intermolecular reductive reaction of
nitroarenes with alkenes to give allylamines (Scheme 1).10
We havenow soughttoexamine the efficacy ofthisprotocolin
anintramolecularreaction, toprepare4-quinolonederivatives,
an important class of chemotherapeutic agents.11
The transition metal-catalyzed cyclization of anilines
bearing pendant unsaturation is a well established route to
important heterocycles.1,2 Several groups have also focused
upon the reduction of the corresponding o-nitro compounds
with concomitant cyclization. These aromatic nitro group
deoxygenations, pioneered by Cadogan and Sundberg’s
phosphine and phosphite chemistry,3,4 have also been
effected by carbon monoxide, catalyzed by Pd com-
plexes,5–7 Se,8 and Group 8 transition metal carbonyls,9
affording a variety of heterocycles, including indoles,
quinolones and quinolines.
We also wanted to investigate the transition metal catalyzed
carbonylative cyclization of the Baylis–Hillman (B–H)
adducts of 2-nitrobenzaldehydes and methylacrylate, again
in an effort to prepare 4-quinolone derivatives. The Baylis–
Hillman reaction is a powerful carbon–carbon bond
forming reaction between an aryl aldehyde and a Michael-
acceptor, catalyzed by a non-nucleophilic base, giving a
2-functionalized allylic alcohol.12–14 The synthesis and
subsequent elaboration of such adducts has thus received
increasing attention. Several groups have targeted the B–H
In an ongoing effort to discover new amination reactions, we
have used the cyclopentadienyliron dicarbonyl dimer (Fp2)
Scheme 1.
Scheme 2.
Keywords: cyclization; hydrogenation; indoles.
*
Corresponding author. Tel.: þ1-4053253696; fax: þ1-4053254811; e-mail: knicholas@ou.edu
0040–4020/03/$ - see front matter q 2003 Published by Elsevier Science Ltd.
PII: S0040-4020(02)01627-7

748
D. K. O’Dell, K. M. Nicholas / Tetrahedron 59 (2003) 747–754
adducts of 2-nitrobenzaldehyde as precursors to nitrogen
heterocycles (Scheme 2). Recently Basavaiah has reported
the synthesis of 2-quinolones by reduction of these
compounds with Fe/AcOH.15 Kim and coworkers were
able to synthesize the 4-quinolones and 4-quinolone-N-
oxides through irradiation of the B–H adducts, as well as
synthesis of the later by treatment with TFA.16,17 Kaye et.
al. have used catalytic hydrogenation of B–H adducts to
produce 2-quinolones.18
styrene 2. It has been found previously that aroyl halides can
be used in place of aryl halides in Pd-catalyzed coupling
reactions, readily decarbonylating to give the intermediate
aryl-Pd(II) species.20
A pressure of 300 psi was required to nearly completely
suppress the formation of 2 and give 1 in 35% yield.
2-Nitrostyrenes have been reductively cyclized to give
indoles under a variety of conditions.3–9 Indeed, when 2
was heated at 1508C with 10 mol% Fp2 at 750 psi CO,
3-methylindole (skatole) was formed as the primary product
(40% GC yield). Given the relatively large number of such
indole-forming reactions and the instability of the styrene
precursors, a more extensive study of these reactions was
not carried out. Instead, we focused our attention upon
the nitroenone 1 that could produce the corresponding
4-quinolone. To our delight when 1 was heated in dioxane
with 5 mol% Fp2 under CO (800 psi), 3-methylquinolin-
4(1H)-one 3 was obtained in 76% yield (Scheme 4).
2. Results and discussion
Our initial studies focused on the reaction of the nitroenone
1, which could be obtained by Stille coupling19 of 2-nitro-
benzoyl chloride with 2-propenyltributyltin under CO
pressure using a catalytic amount of Pd(CH2Ph)Cl(PPh3)2
(Scheme 3). When CO was bubbled through the reaction
mixture the major product was the (decarbonylated) 2-nitro-
Scheme 3.
Next we turned our attention to the easily prepared Baylis–
Hillman adducts, hoping to produce, by analogy, dihydro-4-
quinolones. Compounds 4–6 were prepared in excellent
yield by reaction of the appropriate aldehyde with DABCO
in methyl acrylate (Scheme 5).
The Baylis–Hillman adduct 4 was heated (1508C) in
Scheme 4.
Scheme 5.
Scheme 6.

D. K. O’Dell, K. M. Nicholas / Tetrahedron 59 (2003) 747–754
749
Figure 1. Temperature-dependent 1H NMR spectra of 10.
dioxane with 10 mol% of Fp2 under 750 psi CO (Scheme 6).
Chromatography of the product mixture gave two major
products, one of apparent mass 175 and another with m/e
205, a few minor products, and some tar, which accounts for
a poor mass balance. As Smith noted in his review of related
deoxygenations, ‘cyclizations of o-nitrostyrenes that take
place in low yields are usually accompanied by extensive
tar formation’.21 The m/e 175 product (17% yield),
remarkably, is the result of the loss of one carbon (as CO,
CH2O?) from the precursor 4. It was identified as methyl
3-indolecarboxylate 7 and was spectroscopically identical
to commercial material. Also detected was the known
2-quinolone 8 (4%), arising from apparent ester aminolysis,
reduction and dehydration.18
the N-formylindoline 10. To verify this assignment, we
applied standard reaction protocols to synthesize 10 from
the commercially available indole 7 (Scheme 7). Thus 7 was
protected as the N-Boc derivative 11,24 followed by
catalytic hydrogenation to give N-Boc-indoline 12.25
A
one pot deprotection/formylation26 using formic acid,
followed by treatment with formic acetic anhydride, gave
the N-formylindoline 10 which was spectroscopically
identical to that formed through our cyclization.
Regardless of the efficiency of this transformation, for-
mation of the indole 7 and the N-formylindoline 10 is
remarkable. In a control experiment reaction of 4 with Fp2
(35 mol%) in the absence of CO also gave the indole 7,
albeit in 7% yield, as the sole identifiable product.
Additionally, when either the indole 7 or the indoline 10
were subjected to the catalytic cyclization conditions, they
were recovered unchanged after several days. This result
indicates that both 7 and 10 are primary products of the
cyclization reaction. It is also worth noting that heating the
Baylis–Hillman adduct 4 to 1508C under 750 psi CO
The other major product (20%, m/e 205) corresponded
formally to the removal of two oxygen atoms, consistent
with the expected dihydroquinolone 9. The 1H NMR
spectrum of this compound in d6-DMSO was tempera-
ture-dependent, the low field portion of which is shown in
Figure 1. The two singlets at 8.45 and 9.0 ppm (approxi-
mately 3:1 ratio at 208C) and two doublets at 7.5 and
7.9 ppm coalesced at 1008C; the original spectrum was
restored upon cooling to room temperature. The NMR
results suggested that two isomers of the compound were in
equilibrium. Although this could be explained by the
existence of a keto-enol tautomeric mixture for 9, the
sharpness of the low field resonances, the differing melting
point of the compound compared to that reported for 9,22
and other data (vide infra) pointed to an alternative structure
for the m/e 205 product.
When this compound was subjected to oxidation with
MnO2, it gave a product with m/e 203 (loss of 2H) and a
small amount of the indole 7. We thus turned our attention
to prospective structures that could be oxidized to indoles.
Examination of the literature of N-formylindolines,23
1
revealed H NMR spectra strikingly similar to those we
obtained, with temperature-dependent spectra arising from
the slowly interconverting N-formyl rotamers. Ultimately,
we assigned the structure of the m/e 205 compound to be
Scheme 7.

750
D. K. O’Dell, K. M. Nicholas / Tetrahedron 59 (2003) 747–754
Scheme 8.
without Fp2 for several days did not produce any of the
products mentioned above.
the indole 17 (5%). Each of the N-formyl indolines
apparently exists as a rotomeric mixture, judging from
the appearance of two formyl proton resonances in their
1H NMR spectra as well as the presence of multiple pairs of
resonances in their 13C NMR spectra.
We also compared these results to corresponding reactions
using other CO-based reduction methods, specifically the
PdCl2(PPh3)/SnCl2/CO system of Watanabe6 and the Se/
NEt3/CO combination employed by Sonoda.8 In both cases
the reduction of 4 was incomplete after prolonged reaction
times and the product mixtures were extremely complex.
In neither case was the N-formylindoline 10 detected;
however, the indole 7 was formed in trace amounts.
An admittedly speculative, but precedented, mechanistic
outline is suggested in Scheme 9. Initial Fp2-promoted
reduction of the nitro group of 4 to the hydroxylamine 18
would be followed by Michael addition to the terminal
olefinic carbon.27 Intermediate 19 could then undergo
reversible retro-aldol reaction to generate aldehyde 20.
Deformylation of 20 could be effected by seventeen electron
CpFe(CO)2,28 either by a radical29,30 or an organometallic
pathway,31 to generate an aryl-iron or aryl radical
intermediate, which then could cyclize to indoline 21 by
addition to the ester enol. The N-hydroxyindoline 21 could
then dehydrate to give the indole 7 or be further reduced
to the indoline 22, which could be re-carbonylated to
The cyclization of the other two B–H adducts, 5 and 6, gave
similar results, as shown in Scheme 8. The 5-chloro-2-
nitrobenzaldehyde adduct 5 provided a 13% yield of
N-formylindoline 13 and 8% of the methyl 5-chloro-
indole-3-carboxylate 14; a small amount of the 6-chloro-
quinoline 15 was also isolated. Similarly, the adduct of the
2-nitropiperonal 6 gave the N-formylindoline 16 (8%) and
Scheme 9.

D. K. O’Dell, K. M. Nicholas / Tetrahedron 59 (2003) 747–754
751
produce the N-formylindoline 10. The N-carbonylation of
amines to formamides by group 8 transition metal
complexes and carbon monoxide is also established.32
Anal. calcd: C 62.82, H 4.74, N 7.33; found: C 62.52, H
4.79, N 7.27.
4.1.2. 3-Methylquinolin-4(1H)-one (3).35 In a glass-lined
15 mL Parr reaction vessel the nitroenone 1 (75 mg,
0.39 mmol) and Fp2 (7 mg, 0.02 mmol) were combined in
dioxane (10 mL). The vessel was purged thrice with CO
(fume hood) and charged to 800 psi. The vessel was then
stirred at 2008C for 8.5 h. After cooling and venting of the
CO (fume hood), the reaction mixture was filtered to remove
insoluble material and concentrated. Preparative TLC,
eluting with EtOAc/hexanes (1:1), gave the quinolone 3,
Rf¼0.1 (48 mg, 76%). 1H NMR (300 MHz; d6-DMSO) 1.95
(s, 3H), 7.23 (td, J¼7.5, 1.2 Hz, 1H), 7.46 (dd, J¼6.6,
0.6 Hz, 1H), 7.55 (td, J¼7.7, 1.5 Hz, 1H), 7.85 (s, 1H), 8.07
(dd, J¼8.1, 1.2 Hz, 1H), 8.29 (s, 1H); 13C NMR (d6-DMSO)
13.6, 116.2, 117.8, 122.3, 123.8, 124.7, 130.8, 136.5, 139.5,
176.4; IR (KBr) 3460, 2865, 1630 cm21; MS (EI) 159 (83),
130 (100), 104 (10), 92 (5), 77 (18), 63 (8), 51(8); HRMS
(ESI) calcd for C10H9NONa (MþNa) 182.0582; found:
182.0591.
3. Conclusions
Entry to the 4-quinolone ring system has been demonstrated
by the Fp2-catalyzed reductive cyclization of an o-acryloyl
nitroarene derivative. In contrast to other reductions of
Baylis–Hillman adducts the Fp2–CO system gives mark-
edly different products, the formation of indole and
N-formylindolines being quite unusual. Further studies are
underway to improve product selectivity, to expand the
range of substrates, and to elucidate the mechanistic details.
4. Experimental
4.1. General
Unless specified otherwise all reactions were carried out
under an atmosphere of nitrogen with magnetic stirring,
using dried, freshly distilled solvents and oven dried
glassware. High pressure reactions were carried out in
Parr stainless steel reaction vessels. All reagents and
[CpFe(CO)2]2 were purchased from US suppliers. TLC
was performed using ANALTECH HPTLC with fluorescent
indicator. Flash chromatography was performed using silica
gel 230–400 mesh (Merck). 1H and 13C NMR spectra were
recorded on a 300 MHz Varian spectrometer; TMS was
used as an internal standard for all spectra taken in acetone.
Melting points are uncorrected.
4.1.3. Methyl 2-[hydroxy(2-nitrophenyl)methyl]acrylate
(4).36 2-Nitrobenzaldehyde (2.00 g, 13.2 mmol) was dis-
solved in methyl acrylate (20 mL), DABCO (1.48 g,
13.2 mmol) was added slowly over 5 min upon which the
mixture became a dark orange-red. The mixture was stirred
at room temperature for 5 days. The methyl acrylate was
removed in vacuo, and the residue was dissolved in Et2O,
washed with 1 M HCl (3£20 mL), and then with water
(2£10 mL). The ether extract was concentrated on a rotary
evaporator and then chromatographed over silica gel (Et2O/
hexanes, 1:1) to give 4 (Rf¼0.28) as a straw colored oil
(3.1 g, 99%), which crystallized from cold ether. Mp 28–
1
298C; IR (film) 3445, 2946, 1699, 1537 cm21; H NMR
4.1.1. 2-Methyl-1-(2-nitrophenyl)prop-2-en-1-one (1).
(PPh3)2Pd(CH2Ph)Cl33 (64 mg, 0.090 mmol) was sealed in
a glass vial which was scored to ensure breakage. This vial
along with 2-nitrobenzoyl chloride (1.67 g, 9.00 mmol),
isopropenyltributyltin34 (3.40 g, 10.3 mmol), and HMPA
(4 mL) were added to a 120 mL glass-lined Parr reaction
vessel equipped with a large magnetic stirrer. The vessel
was flushed three times with CO (fume hood) and finally
charged to 300 psi. The vessel was manually shaken until
the vial containing the catalyst was broken, after which the
reaction was stirred at 708C for 10 h. After cooling and
venting of CO (fume hood), the reaction mixture was diluted
with ether and stirred with saturated aqueous KF (30 mL)
for 1 day. The ether layer was separated and filtered to
remove a white precipitate (Bu3SnCl), and washed with
brine (3£20 mL). Solvent was removed in vacuo and the
residue was purified by flash chromatography (Et2O/
hexanes, 1:4) to give the product slightly contaminated
with Bu3SnCl. Recrystallization from Et2O/hexanes gave
pale yellow prisms of the nitroenone 1 (0. 60 g, 35%); mp
61.5–638C; Rf¼0.5 (1:1, Et2O/hexanes); IR (KBr) 3321,
3097, 3027, 2957, 1660, 1521, 1436, 1343, 1189 cm21. 1H
NMR (300 MHz; d6-DMSO) 1.96 (dd, J¼1.2, 0.3 Hz, 3H),
5.35 (d, J¼0.9 Hz, 1H), 5.99 (q, J¼1.2 Hz, 1H), 7.57 (ddd,
J¼7.12, 1.2, 0.3 Hz 1H), 7.77 (tdd, J¼8.1, 1.2, 0.6 Hz, 1H),
7.88 (td, J¼7.5, 1.2 Hz, 1H), 8.20 (dd, J¼8.25, 1.2 Hz, 1H);
13C NMR (75.45 MHz; d6-DMSO) 17.8, 125.1, 129.5,
129.6, 131.9, 135.4, 134.8, 143.7, 146.1, 194.5; MS (ESI)
192 (MþH, 30), 214 (MþNa, 100), 405 (M2þNa, 11.4);
(300 MHz; d6-acetone) 3.64 (s, 3H), 5.1 (br s, 1H), 5.81 (d,
J¼1.2 Hz, 1H), 6.26 (m, 2H), 7.52 (td, J¼8.1, 1.8 Hz, 1H),
7.69 (td, J¼7.8, 1.2 Hz, 1H), 7.76 (dd, J¼7.8, 1.8 Hz, 1H),
7.9 (dd, J¼8.1, 1.2 Hz, 1H); 13C NMR (75.45 MHz;
d6-acetone) 51.7, 66.4, 124.4, 124.6, 128.9, 129.2, 133.2,
137.3, 143.1, 148.8, 165.8; MS (EI) 220 (Mþ –OH, 10),
191 (Mþ –NO2, 100), 160 (40), 132 (146), 117 (38), 117
(38), 104 (40), 77 (42), 51 (28); HRMS (ESI) calcd for
C11H11NO5Na (MþNa) 260.0535; found: 260.0463.
4.1.4. Methyl 2-[(5-chloro-2-nitrophenyl)(hydroxy)-
methyl]acrylate (5). 5-Chloro-2-nitrobenzaldehyde (1.00 g,
5.3 mmol), sold as a mixture from Aldrich of at least 77%
purity, was dissolved in 15 mL of methyl acrylate. DABCO
(0.60 g, 5.4 mmol) was then added over 5 min after which
the mixture darkened. The mixture was stirred at room
temperature for 4 days, followed by the workup described
for compound 4. Chromatography on silica gel with
Et2O/hexanes (gradient elution: 1:5 to 1:1) gave 1.0 g
(93% yield based upon the purity of the starting material)
of the Baylis–Hillman adduct 5 as a yellow solid, mp 63–
658C, Rf¼0.40 (1:1, Et2O/hexanes). IR (KBr) 3475, 1722,
1
1529, 1367, 1297 cm21; H NMR (300 MHz; d6-acetone)
3.69 (s, 3H), 5.38 (d, J¼5.1 Hz, 1H), 5.81 (d, J¼0.9 Hz,
1H), 6.29 (m, 2H), 7.59 (dd, J¼8.7, 2.4 Hz, 1H), 7.79 (d, J¼
2.4 Hz, 1H), 8.0 (d, J¼8.7 Hz, 1H); 13C NMR (75.45 MHz;
d6-acetone) 52.2, 66.8, 125.7, 127.0, 129.3, 129.6, 139.5,
140.5, 143.2, 147.5, 166.1; MS (EI) 254 (Mþ –OH, 1), 227

752
D. K. O’Dell, K. M. Nicholas / Tetrahedron 59 (2003) 747–754
(30), 225 (Mþ –NO2, 100), 194 (20), 166 (15), 138 (6);
HRMS (ESI) calcd for C11H10NO5Na (MþNa) 294.0145;
found: 294.0103.
(35 mg, 0.10 mmol) were placed in a glass-lined, stainless
steel 15 mL Parr reaction vessel with benzene (10 mL). The
vessel was flushed thrice with CO (fume hood) and charged
to 800 psi. The mixture was heated to 1508C while stirring
for 18 h. After cooling and venting the CO (fume hood), the
resulting dark brown solution was filtered to remove
insoluble material, the filtrate concentrated, and then
chromatographed on silica gel with a gradient elution of
Et2O/hexanes (1:5–9:1) giving the following compounds:
4.1.5. Methyl 2-[hydroxy(6-nitro-1,3-benzodioxol-5-
2-Nitropiperonal
yl)methyl]acrylate
(6).
(3.00 g,
15.4 mmol) was dissolved in methyl acrylate (20 mL),
DABCO (1.72 g, 15.4 mmol) was added slowly over 5 min
upon which the mixture became a dark orange-red. The
mixture was stirred at room temperature for 7 days; after
one day a yellow precipitate formed. After the above
described workup, chromatography on silica gel using a
gradient elution with Et2O/hexanes (20–50% Et2O) gave
4.10 g (95%) of 6, Rf¼0.2 (1:1, Et2O/hexanes); mp 122–
123.58C; IR (KBr) 3494, 1712, 1637, 1531, 1482, 1343,
4.3.1. Methyl 5-chloro-1-formylindoline-3-carboxylate
(13).37 30 mg (13%); Rf¼0.13 (Et2O/hexanes, 3:1); IR
(KBr) 3452, 2975, 1738, 1684, 1491, 1367, 1321,
;
819 cm21 1H NMR (300 MHz; CD3CN) 3.72 (s, 3H),
4.05–4.15 (m, 1H), 4.24–4.47 (m, 2H), 7.26 (m, 2H), 7.41
(m, 0.7H), 7.93 (d, J¼8.7 Hz, 0.3H, minor isomer), 8.42 (s,
0.3H, minor), 8.86 (s, 0.7H, major); 13C NMR (75.45 MHz;
CD3CN) 45.1, 45.6, 48.1, 50.2, 53.2, 111.8, 117.6, 126.0,
126.3, 127.1, 128.8, 129.3, 131.7, 158.5, 160.5, 171.7; MS
(EI) 241 (Mþ, 15), 239 (50), 211 (16), 180 (31), 154 (32),
152 (100), 117 (78), 89 (14); Anal. calcd for C11H10ClNO3:
C, 55.13; H, 4.21; N, 5.84. Found: C, 54.91; H, 4.33; N,
5.54.
1
1256, 1030, 879, 822 cm21; H NMR (300 MHz; CD3CN)
3.71 (s, 3H), 4.02 (br s, 1H), 5.67 (d, J¼0.9 Hz, 1H), 6.15
(m, 3H), 7.17 (s, 1H), 7.49 (s, 1H); 13C NMR (75.45 MHz;
CD3CN) 52.6, 67.3, 104.5, 105.7, 108.2, 125.5, 135.2,
142.2, 143.4, 148.2, 153.0, 166.8; MS (EI) 235 (Mþ –NO2,
20), 221 (32), 204 (76), 188 (64), 176 (100), 160 (26), 148
(18), 135 (9), 120 (7), 104 (4), 77 (3), 53 (8); HRMS (ESI)
calcd for C12H11NO7Na (MþNa) 304.0433; found:
304.0412; Anal. calc: C 51.25, H 3.94, N 4.98; found: C
51.25, H 3.99, N 4.86.
4.3.2. Methyl 5-chloro-1H-indole-3-carboxylate (14).
23 mg (11%); mp 110–1128C; Rf¼0.18 (Et2O/hexanes,
1:1); IR (KBr) 3250, 1680, 1524, 1443; 1H NMR (300 MHz;
CD3CN) 3.87 (s, 3H), 7.23 (dd, J¼8.8, 2.1 Hz, 1H), 7.51 (d,
J¼8.7 Hz, 1H), 7.97 (s, 1H), 8.07 (d, J¼2.1 Hz, 1H), 10.06
(br s, 1H); 13C NMR (75.45 MHz; CD3CN) 51.4, 114.3,
120.8, 123.6, 133.8, 165.4; MS (EI) 211 (15), 209 (50),
178 (100), 150 (24), 123 (8); HRMS (ESI) calcd for
C10H8ClNO2Na (MþNa) 232.0141; found: 232.0080.
4.2. Cyclization of Baylis–Hillman adduct 4
Compound 4 (0.20 g, 0.84 mmol) and Fp2 (30 mg,
0.085 mmol) were placed in a glass-lined, stainless steel
Parr reaction vessel with dioxane (10 mL). The vessel was
flushed three times with CO (fume hood) and charged to
800 psi. The mixture was the heated to 1508C with stirring
for 14 h. After cooling and venting of CO (fume hood), the
resulting brown solution was filtered to remove insoluble
material, concentrated and then chromatographed on silica
gel with a gradient elution of Et2O/hexanes (1:5–9:1) to
give the following compounds:
4.3.3. Methyl 6-chloroquinoline-3-carboxylate (15). 8 mg
(4%); Rf¼0.2 (Et2O/hexanes, 1:3), mp 170–1728C; IR
(KBr) 3066, 2950, 1722, 1491, 1436, 1336, 1267, 1097,
1
819 cm21. H NMR (300 MHz; d6-acetone) 4.00 (s, 3H),
7.89 (dd, J¼9.0, 2.4 Hz, 1H), 8.12 (d, J¼9.0 Hz, 1H), 8.24
(d, J¼2.1 Hz, 1H), 8.93 (d, J¼1.5 Hz, 1H), 9.35 (d, J¼
2.1 Hz, 1H); 13C NMR (75.45 MHz; CDCl3) 52.7, 132.7,
127.5, 130.9, 132.7, 133.3, 137.7, 148.0, 150.0, 165.3; MS
(EI) 221 (Mþ, 100), 190 (95), 162 (70), 127 (27), 99 (16), 74
(5); HRMS (ESI) calcd for C11H8ClNO2Na (MþNa)
244.0141; found: 244.0092.
4.2.1. Methyl 1-formylindoline-3-carboxylate (10). 34 mg
(20%); Rf¼0.20 (Et2O/hexanes, 3:1); mp 84–85.58C; IR
(KBr) 1732, 1665, 1594, 1497, 1367, 1283, 1220,
1
1174 cm21; H NMR (300 MHz; d6-DMSO) 3.69 (s, 3H),
4.0–4.5 (m, 3H), 7.09 (m, 1H), 7.26 (m, 1H), 7.37 (d, J¼
7.8 Hz, 1H), 7.48 (d, J¼8.1 Hz, 0.7H, major isomer), 7.91
(d, J¼7.8 Hz, 0.3H, minor), 8.50 (s, 0.3H, minor), 9.04 (s,
0.7H, major); 13C NMR (75.45 MHz; d6-DMSO) 44.0, 44.5,
46.9, 48.9, 52.5, 110.2, 115.6, 123.7, 124.1, 125.3, 125.9,
128.5, 128.6, 128.8, 129.4, 140.8, 158.3, 160.0, 171.2,
171.3; MS (EI) 205 (Mþ, 60), 177 (8),145 (35), 118 (100),
91 (17); Anal. calcd for C11H11NO3: C, 64.38; H, 5.40; N,
6.83. Found: C, 64.45; H, 5.47; N, 6.74.
4.4. Cyclization of Baylis–Hillman adduct 6
The Baylis–Hillman adduct 6 (200 mg, 0.71 mmol) and Fp2
(27 mg, 0.08 mmol) were placed in a glass-lined, 15 mL
stainless steel Parr reaction vessel with dioxane (10 mL).
The vessel was flushed thrice with CO (fume hood) and
charged to 800 psi. The mixture was heated to 1508C while
stirring for 24 h. After cooling and venting of CO (fume
hood) the resulting dark brown solution was filtered to
remove insoluble material, concentrated, and then chro-
matographed on silica gel with a gradient elution of
Et2O/hexanes (1:5 to 9:1) giving the following compounds:
4.2.2. Methyl 1H-indole-3-carboxylate (7). 25 mg (17%);
Rf¼0.25 (Et2O/hexanes, 1:1); spectroscopically identical to
material purchased from Aldrich.
4.2.3. 3-Methylquinolin-2(1H)-one (8). 6 mg (4%); Rf¼
0.26 (Et2O/hexanes, 3:1); identical to literature compound.37
4.4.1. Methyl 5-formyl-6,7-dihydro-5H-[1,3]dioxolo[4,5-f]-
indole-7-carboxylate (16). 14 mg (8%); Rf¼0.25 (Et2O/
hexanes, 3:1); IR (KBr) 2973, 2919, 1730, 1669, 1483,
4.3. Cyclization of Baylis–Hillman adduct 5
1
The Baylis–Hillman adduct 5 (0.24 g, 0.89 mmol) and Fp2
1297, 1220, 1035 cm21; H NMR (300 MHz; CDCl3) 3.76

D. K. O’Dell, K. M. Nicholas / Tetrahedron 59 (2003) 747–754
753
(s, 3H), 4.10–4.24 (m, 2H), 4.40–4.58 (m, 1H), 5.95 (m,
2H), 6.69 (s, 0.7H, major isomer), 6.86 (s, 0.3H, minor
isomer), 6.88 (s, 0.7H, major), 7.69 (s, 0.3H), 8.41 (s, 0.3H,
minor), 8.74 (s, 0.7H, major); 13C NMR (75.45 MHz;
CDCl3) 44.5, 44.9, 47.9, 49.7, 52.8, 92.2, 99.3, 101.6, 105.4,
106.4, 120.5, 130.7, 156.5, 158.2, 171.1; MS (EI) 249 (Mþ,
65), 190 (100), 160 (40), 132 (60), 104 (40), 77 (10); HRMS
(ESI) calcd for C12H11ClNO5Na (MþNa) 272.0535; found:
272.0551.
(100), 118 (58), 89 (8); HRMS (ESI) calcd for
C15H19NO4Na (MþNa) 300.1212; found: 300.1252.
4.5. One pot deprotection and formylation of N-Boc
indoline 12
The N-Boc indoline 12 (0.730 g, 2.64 mmol) was stirred in
formic acid (8 mL) for 1 h at rt. Acetic formic anhydride,
preformed by the addition of acetic acid (0.5 mL) to formic
acid (1.0 mL) with stirring for 1 h, was then added. The
mixture was stirred at room temperature for 4 h. Formic acid
was removed in vacuo, the residue was dissolved in ether
(20 mL) and then extracted with aqueous saturated NaHCO3
(3£15 mL). The ether extract was dried with MgSO4,
concentrated, and chromatographed using Et2O/hexanes
(gradient elution 1:1–9:1) to give 0.295 g (55%) of the
previously described N-formyl-indoline 10 and an undeter-
mined amount of starting material.
4.4.2. Methyl 5H-[1,3]dioxolo[4,5-f]indole-7-carboxylate
(17). 8 mg (5%); Rf¼0.20 (Et2O/hexanes, 1:1); IR (KBr)
3351, 2950, 1676, 1545, 1467, 1297, 1197, 1081, 1027, 942,
1
819. H NMR (300 MHz; d6-acetone) 3.7 (s, 3H), 5.96 (s,
2H), 6.98 (s, 3H), 7.45 (s, 1H), 7.85 (s, 1H), 10.81 (br s, 1H);
13C NMR (75.45 MHz; d6-acetone) 50.3, 92.8, 99.6, 101.1,
108.0, 120.7, 130.1, 131.6, 144.5, 145.5, 165.0; MS (EI) 219
(100), 188 (80), 160 (27), 130 (5), 103 (6), 74 (10); HRMS
(ESI) calcd for C11H9NO4Na (MþNa) 242.0429; found:
242.0483.
Acknowledgements
4.4.3. Methyl 1-(2,2-dimethylpropanoyl)-1H-indole-3-
carboxylate (11). The indole 7 (3.00 g, 17.1 mmol) was
placed in a 50 mL round bottom flask with NaH (0.54 g,
22.3 mmol) and cooled with an ice bath. THF (20 mL) was
added with stirring, and gas evolution was observed. After
5 min Boc2O (4.86 g, 22.2 mmol) was added and a solid
precipitated. The mixture was stirred overnight. The
solution was then quenched with sat. NH4Cl (20 mL),
diluted with ether and washed with water (3£20 mL). The
ethereal solution was then dried with MgSO4 concentrated
under vacuum, and chromatographed with Et2O/hexanes
(1:10) to give a 4.5g (95%) of a white solid. Mp 125–1278C;
Rf¼0.25 (1:20, Et2O/hexanes); IR (KBr) 3166, 2981, 1745,
We are grateful for financial support provided by the
National Science Foundation and for an O.U. Graduate
Alumni Fellowship. We also thank Professor S. Blechert of
T.U. Berlin for stimulating discussions.
References
1. Li, J. J.; Gribble, G. W. E. Palladium in Heterocyclic
Chemistry: A Guide for the Synthetic Chemist; Pergamon:
New York, 2000.
1714, 1560, 1451, 1374, 1150, 726 cm21
;
1H NMR
2. Hegedus, L. S. Angew. Chem. 1988, 100, 1147–1161.
3. Cadogan, J. I. G.; Cameron-Wood, M.; Mackie, R. K.; Searle,
R. J. G. J. Chem. Soc. 1965, 4831–4837.
4. Sundberg, R. J. J. Org. Chem. 1965, 30, 3604–3610.
5. Soderberg, B. C.; Shriver, J. A. J. Org. Chem. 1997, 62,
5838–5845.
(300 MHz; d6-acetone) 1.71 (s, 9H), 3.89 (s, 3H), 7.32–
7.44 (m, 2H), 8.12–8.21 (m, 2H), 8.23 (s, 1H); 13C NMR
(75.45 MHz; d6-acetone) 28.1, 51.5, 85.8, 112.4, 115.8,
122.0, 124.4, 125.7, 128.1, 132.3, 136.1, 149.3, 164.3; MS
(EI) 275 (Mþ, 11), 175 (100), 144 (63), 116 (10); HRMS
(ESI) calcd for C15H17NO4Na (MþNa) 298.1055; found:
298.1144.
6. Akazome, M.; Kondo, T.; Watanabe, Y. J. Org. Chem. 1994,
59, 3375–3380.
7. Aoyagi, Y.; Mizusaki, T.; Ohta, A. Tetrahedron. Lett. 1996,
37, 9203–9206.
4.4.4. Methyl 1-(2,2-dimethylpropanoyl)indoline-3-car-
boxylate (12). The N-Boc indole 11 (2.25 g, 8.20 mmol)
was placed in Fischer–Porter bottle with 5% Pd/C (0.50 g),
MeOH (10 mL) and EtOAc (30 mL). The bottle was purged
thrice with H2 (fume hood) and charged to 110 psi. After
stirring for 3 days at room temperature the vessel was
recharged with H2 to 110 psi and an additional 300 mg of
5% Pd/C was added. After another 2 days of stirring the
reaction was stopped and the solution was filtered through
Celite and concentrated. Chromatography on silica gel with
Et2O/hexanes (gradient elution 1:20–1:10) gave 1.54 g
(68%) of a white solid with Rf¼0.1 (1:20, Et2O/hexanes)
and an undetermined amount of recovered starting material.
Mp 53–548C. IR (KBr, cm21) 2981, 1745, 1699, 1491,
8. Nishiyama, Y.; Maema, R.; Ohno, K.; Hirose, M.; Sonoda, N.
Tetrahedon Lett. 1999, 40, 5717–5720.
9. Tollari, S.; Penoni, A.; Cenini, S. J. Mol. Catal. A: Chem.
2000, 152, 47–54.
10. Srivastava, R. S.; Nicholas, K. M. J. Chem. Soc., Chem.
Commun. 1998, 24, 2705–2706.
11. Andriole, V. T. E. The Quinolones; 2nd ed, Academic: San
Diego, 1998.
12. Ciganek, E. Org. React. 1997, 51, 201–350.
13. Basavaiah, D.; Rao, P. D.; Hyma, R. S. Tetrahedron 1996, 52,
8001–8062.
14. Drewes, S. E.; Roos, G. H. P. Tetrahedron 1988, 44,
4653–4670.
1
1398, 1004, 741; H NMR (300 MHz; d6-acetone) 1.58 (s,
15. Basavaiah, D.; Reddy, R. M.; Kumaragurubaran, N.; Sharada,
D. S. Tetrahedron 2002, 58, 3693–3697.
9H), 3.75 (s, 3H), 4.12 (m, 3H), 4.31 (m, 2H), 6.96 (td,
J¼7.4, 1.2 Hz, 1H), 7.23 (tdd, J¼7.7, 1.5, 0.6 Hz, 1H), (ddd,
J¼8.0, 1.5, 0.6 Hz), 7.84 (br s, 1H); 13C NMR (75.45 MHz;
d6-acetone) 29.5, 50.5, 52.6, 65.9, 115.1, 122.6, 125.5,
129.1, 152.1, 172.0; MS (EI) 277 (Mþ, 20), 221 (80), 177
16. Chung, Y. M.; Lee, H. J.; Hwang, S. S.; Kim, J. N. Bull. Kor.
Chem. Soc. 2001, 22, 799–800.
17. Kim, J. N.; Lee, K. Y.; Kim, H. S.; Kim, T. Y. Org. Lett. 2000,
2, 343–345.

754
D. K. O’Dell, K. M. Nicholas / Tetrahedron 59 (2003) 747–754
18. Familoni, O. B.; Kaye, P. T.; Klaas, P. J. J. Chem. Soc., Chem.
Commun. 1998, 24, 2563–2564.
29. Giese, B.; Thoma, G. Helv. Chim. Acta 1991, 74, 1143–1155.
(b) Giese, B.; Thoma, G. Helv. Chim. Acta 1991, 74,
1135–1142. (c) Thoma, G.; Giese, B. Tetrahedron Lett.
1989, 30, 2907–2910.
19. Milstein, D.; Stille, J. K. J. Am. Chem. Soc. 1978, 100,
3636–3638.
20. Obora, Y.; Tsuji, Y.; Kawamura, T. J. Am. Chem. Soc. 1993,
115, 10414–10415.
30. Belani, R. M.; James, B. R.; Dolphin, D.; Rettig, S. J. Can.
J. Chem. 1988, 66, 2072. (b) Applequist, D. E.; Kaplan, L.
J. Am. Chem. Soc. 1965, 87, 2194.
21. Smith, P. A. S. In Nitrenes: Reactive Intermediates in Organic
Chemistry; Lwowski, W. E., Ed.; Interscience: New York,
1970; Chapter 4.
31. Kampmeier, J. A.; Harris, S. H.; Wedegaertner, D. K. J. Org.
Chem. 1980, 45, 315. (b) Beck, C. M.; Rathmill, S. E.; Park,
Y. J.; Crabtree, R. H.; Liable-Sands, L. M.; Reingold, A. L.
Organometallics 1999, 18, 5311.
22. Proctor, G. R.; Ross, W. I.; Tapia, A. J. Chem. Soc., Perkin
Trans. 1 1972, 14, 1803–1808.
23. Nagarajan, K.; Nair, M. D. Tetrahedron 1967, 23, 4493–4497.
24. Macor, J. E.; Ryan, K.; Newman, M. E. J. Org. Chem. 1989,
54, 4785–4795.
32. Jenner, G.; Bitsi, G.; Schleiffer, E. J. Mol. Catal. 1987, 39,
233–236.
33. Fitton, P.; McKeon, J. E.; Ream, B. C. J. Chem. Soc. D 1969,
7, 370–371.
25. Collot, V.; Schmidt, M.; Marwah, P.; Bourguignon, J.
Heterocycles 1999, 51, 2823–2847.
34. Seyferth, D.; Stone, F. G. A. J. Am. Chem. Soc. 1957, 79,
515–517.
26. Marino, J. P.; Bogdan, S.; Kimura, K. J. Am. Chem. Soc. 1992,
114, 5566–5572.
35. Barker, A. J.; Paterson, T. M.; Smalley, R. K.; Suschitzky, H.
J. Chem. Soc., Perkin Trans. 1 1979, 9, 2203–2208.
36. Mason, P. H.; Emslie, N. D. Tetrahedron 1994, 50,
12001–12008.
27. Baldwin, J. E.; Harwood, L. M.; Lombard, M. J. Tetrahedron
1984, 40, 4363–4370.
28. Cutler, A. R.; Rosenblum, M. J. Organomet. Chem. 1976, 120,
87–96. Tenhaeff, S. C.; Covert, K. J.; Castellani, M. P.;
Grunkemeier, J.; Kunz, C.; Weakley, T. J. R.; Koenig, T.;
Tyler, D. R. Organometallics 1993, 12, 5000–5004.
37. Gonzalez, R.; Ramos, M. T.; De la Cuesta, E.; Avendano, C.
Heterocycles 1993, 36, 315–322.