Studies on the amino-heck reactions of unsaturated ketones o-phosphinyloximes for the preparation of substituted pyridines

Article abstract of DOI:10.3987/COM-08-11508

The amino-Heck cyclization process has been applied into a range of γ,δ-unsaturated ketone O-diethylphosphinyloximes 1 and δ,ε-unsaturated ketone O-diethylphosphinyloximes 7. Under the specific catalytic conditions developed by us, these substrates were found to preferentially undergo the 6-endo cyclization to give the formation of 2-substituted pyridines 3 and substituted methylpyridines 8, respectively, in moderate to good yields. Besides, several interesting aspects on the effects of phosphinyl groups, solvents, bases and molecular sieves on the regioselectivity of the cyclization of 1 have also been realized.

Full text of DOI:10.3987/COM-08-11508

HETEROCYCLES, Vol. 78, No. 2, 2009
369
HETEROCYCLES, Vol. 78, No. 2, 2009, pp. 369 - 387. © The Japan Institute of Heterocyclic Chemistry
Received, 29th July, 2008, Accepted, 19th September, 2008, Published online, 25th September, 2008.
DOI: 10.3987/COM-08-11508
STUDIES ON THE AMINO-HECK REACTIONS OF UNSATURATED
KETONES O-PHOSPHINYLOXIMES FOR THE PREPARATION OF
SUBSTITUTED PYRIDINES
Jia-Liang Zhu,a,* Yi-Lin Su,a Yu-Hui Chan,a I-Chia Chen,b and Chuan-Chen
Liaoa
a) Department of Chemistry, National Dong-Hwa University, Hualien 974,
Taiwan, R.O.C.; b) Department of Chemistry, National Tsing Hua University,
Hsinchu 300013, Taiwan, R.O.C.
E-mail: jlzhu@mail.ndhu.edu.tw
Abstract – The amino-Heck cyclization process has been applied into a range of
γ,δ-unsaturated ketone O-diethylphosphinyloximes 1 and δ,ε-unsaturated ketone
O-diethylphosphinyloximes 7. Under the specific catalytic conditions developed
by us, these substrates were found to preferentially undergo the 6-endo cyclization
to give the formation of 2-substituted pyridines 3 and substituted methylpyridines
8, respectively, in moderate to good yields. Besides, several interesting aspects on
the effects of phosphinyl groups, solvents, bases and molecular sieves on the
regioselectivity of the cyclization of 1 have also been realized.
INTRODUCTION
The palladium-catalyzed intramolecular Heck-type cyclization of unsaturated ketone oxime derivatives
provides a conceptually powerful means for the preparation of aza-heterocycles.1 According to literatures,
a great variety of heterocyclic compounds such as pyrroles,2-5 pyridines,6 isoquinolines,6b azaazulenes,7
and imidazoles,8 has been achieved via this approach. For instance, there are several types oxime
derivatives
including
γ,δ-unsaturated
ketone
O-pentafluorobenzoyloximes1,2
and
O-methylsulfonyloximes,2
3-butynyl
4-(methoxycarbonyl)-3-butynyl
phenyl
ketone
O-pentafluorobenzoyloximes5 and 3,3-difluoroallyl ketone O-pentafluorobenzoyloximes3 were reported
to undergo the cyclization under the catalysis of Pd(0) complexes to afford substituted pyrroles. As a
complement to these existing methods, we have recently developed a novel amino-Heck protocol for the
synthesis of substituted pyrroles by utilizing the readily prepared γ,δ-unsaturated ketone
O-diethylphosphinyloximes 1 as the substrates.9 In addition to their pronounced stability and the

370
HETEROCYCLES, Vol. 78, No. 2, 2009
efficiency for the cyclization, these precursors also exhibited the interesting regioselective tendency. For
example, during the initial model studies with phenyl-4-penten-1-one O-diethylphosphinyloxime (1a), we
discovered that under the reaction conditions of Pd(PPh3)4/Et3N/DMF, which were commonly employed
for other types oxime derivatives,1,2,8 the reaction occurred preferentially in a 6-endo fashion to give the
formation of 2-phenylpyridine (3a) in 46% yield. The desired product 2a arising from the 5-exo
cyclization was only formed in 16% yield (Scheme 1). After screening several reaction conditions, we
found that 2a could be achieved in almost quantitative yield (90%) by treating 1a with Pd(PPh3)4 and
1,8-diaza-bicyclo[5.4.0]undec-7-ene (DBU) in CH3CN. These results suggest that with this type of oxime
derivatives, the control of the regioselectivity of the cyclization can be attended to some extent with the
adjustment of the catalytic conditions. We then applied the reaction conditions favouring the 5-exo
cyclization [Pd(PPh3)4/DBU/CH3CN] into a number of γ,δ-unsaturated O-diethylphosphinyloximes, to
allow the generation of a series of 2-substituted 5-methylpyrroles and several indoles in good to high
yields with no trace of by-products resulting from Beckmann rearrangement, a competing side process
being frequently encountered by other types oxime derivatives.1a
Scheme 1. Pd(0)-catalyzed cyclization of 1a under the different reaction conditions
OPO(OEt)2
N
HN
Pd(PPh3)4 (0.2 eq), DBU (6 eq)
CH3CN, 80 oC, 3 h
Ph
1a
Ph
2a (90%)
(E/Z = 87:13)
Pd(PPh3)4 (0.1 eq), Et3N ( 5 eq)
DMF, 80 oC, 3 h
N
+
2a (16%)
Ph
3a (46%)
Similarly to other types of oxime derivatives, the cyclization of γ,δ-unsaturated ketone
O-phosphinyloximes 1 should be initiated by the oxidative addition of 1 to Pd(0) to generate an
alkylideneaminopalladium(II) intermediate A (Scheme 2). The subsequent 5-exo or 6-endo cyclization
affords the σ-complex species B or C, respectively. From which, the substituted pyrroles 2 or pyridines 3
can be released through β-elimination followed by isomerization or oxidative dehydrogenation.10
In the previous studies we did not further pursue the more favorable catalytic conditions for the 6-endo
cyclization. Nevertheless, it should be noted that the formation of 3a observed in our cases is quite
unusual, because all of the γ,δ-unsaturated ketone oxime derivatives previously documented, differing
from 1a only by the functional groups on the nitrogen atom, were revealed to undergo the 5-exo

HETEROCYCLES, Vol. 78, No. 2, 2009
371
Scheme 2. Proposed mechanism of the Pd(0)-catalyzed cyclization of γ,δ-unsaturated ketone
O-diethylphosphinyloximes
Pd(II)OPO(OR')
2
N
N
PdL / -H
N
2
2
R
R
R
C
3
H
Pd(II) OPO(OR')
2
6-endo
base
OPO(OR')
Pd(II)OPO(OR')
2
2
N
N
Pd(0)
R
R
A
1
base
5-exo
H
Pd(II) OPO(OR')
2
Pd(II)OPO(OR')
H
2
N
N
HN
R
R
R
2
B
cyclization exclusively under the same reaction conditions to afford 2a as the sole cyclization products.1,2
On the other hand, for the substrates including O-pentafluorobenzoyloximes6 and O-acetyloximes10,11
used for preparing pyridines, a β- acetoxy, a β-methoxy or an α,β-unsaturated functionality was shown to
be indispensable for the 6-endo cyclization in addition to the γ,δ-double bond. These substrates, as
outlined in Figure 1 (4-6) are obviously more difficult to be accessed compared with 1, and for some of
them, the yields of the cyclization reactions are extremely low. For these reasons and due to the broad
synthetic and biological utilities of the pyridinyl compounds,12 we envisaged that it would be valuable if
we could turn 1 into the useful precursors for the generation of pyridines. To this end, an extensive
investigation was carried out, which eventually allowed us to find a catalytic system to further promote
the 6-endo cyclization of 1 on the basis of what we had already accomplished in the previous studies.
With its application, the synthesis of a range of 2-substituted pyridines was achieved mostly in good
yields. In addition, we also applies the hitherto unexplored δ,ε-unsaturated ketone
O-diethylphosphinyloximes to the cyclization process for the preparation of substituted methyl pyridines.
Herein, we wish to report the details of these studies.
Figure 1. Oxime derivatives previously reported to undergo 6-endo cyclization
OAc
OAc
OCOC6F5
OMe
OCOC6F5
N
N
N
R
R
R
4
5
6

372
HETEROCYCLES, Vol. 78, No. 2, 2009
RESULTS AND DISCUSSION
We speculated that the unusual regioselectivity observed for 1a under the conditions of
Pd(PPh3)4/Et3N/DMF (Scheme 1) might be attributed to the unique coordination nature and/or the steric
effect of the phosphinyl group. With true reason(s) being obscure for us, we began our investigation by
surveying the substitution effect of phosphinyl groups on the regioselectivity. Accordingly, substrates 1b
and 1c, with a bulkier diphenylphosphinyl and a smaller dimethylphosphinyl groups to replace the
diethylphosphinyl moeity of 1a, were prepared following the procedure previously developed for the
preparation of 1a.9 As shown in Scheme 3, treatment of the O-hydroxyoxime intermediate derived from
benzaldehyde in three synthetic steps with diphenyl chlorophosphate [ClPO(OPh)2] and dimethyl
chlorophosphate [ClPO(OMe)2], respectively, in the presence of NaH readily led to the formation of 1b
and 1c in high yields. Since the stereogeometry of 1a had been proven to have no influence on the
regioselectivity of the cyclization,9 both 1b (E/Z = 78:22) and 1c (E/Z = 90:10) were used as the
unseparated isomeric mixtures13 for the subsequent investigation. The results of the cyclization reactions
of 1a-c under the same catalytic conditions are given in Table 1.
Scheme 3. Preparation of phenyl-4-penten-1-one O-phosphinylphinyloximes 1a-c
OH
O
N
3 steps
ClPO(OR)2 (2 eq), NaH (1.2 eq)
THF, 0 oC to rt, 12 h
Ph
H
Ph
OPO(OR)2
N
Ph
1a (R = Et, 83%, E/Z = 87:13)
1b (R = Ph, 79%, E/Z = 78:22)
1c (R = Me, 85%, E/Z = 90:10)
Table 1. Substitution effect of the phosphinyl groups on the Pd(0)-catalyzed cyclization
OPO(OR)2
Pd(PPh3)4 (0.1eq), Et3N (5 eq)
DMF, heat, 24 h
N
3a
+
2a
Ph
1a
yield (%)b
ratio of
3a:2a
entrya
substrate
3a
2a
1
2
3
1a (R = Et)
1b (R = Ph)
1c (R = Me)
46
30
50
16
53
18
74:26
36:64
74:26
a ) All reactions were performed using one equivalent of substrate in a 0.02 M DMF
solution. b) Isolated yields.
Among three substrates, 1b was shown to be mostly in favour of the 5-exo cyclization to give the highest
ratio between 2a and 3a (64:36) (Table 1, entry 2). This result can presumably be rationalized in term of a

HETEROCYCLES, Vol. 78, No. 2, 2009
373
steric effect, i.e. a bulky diphenylphosphinyl group being prone to adopt the exo position in a cyclic
transition state (Scheme 2, intermediate B) in order to minimize the conformational strain. However, what
was not clear from these experiments was the same 3a/2a ratios obtained from 1a and 1c (74:26) (Table 1,
entries 1 and 3). The assumption that the steric difference between the diethylphosphinyl and the
dimethylphosphinyl groups was not sufficiently large enough to make the difference in regioselectivity
may somewhat account for this observation. Given the similar regioselectivity, reactivity and the
preparative manner, the unsaturated O-diethylphosphinyl oximes should be more appropriate to be used
as substrates than the corresponding O-dimethylphosphinyl oximes with regard to the much higher cost of
ClPO(OMe)2 than that of ClPO(OEt)2. Consequently, we used O-diethylphosphinyl oximes as the
substrates for the following experiments.
After the aforementioned investigation, we then turned our attention to identifying reaction conditions
that would enhance the desired 6-endo cyclization. The previous experiments had already demonstrated
that when accompanied with Pd(PPh3)4 and Et3N, DMF was superior to CH3CN as a solvent in respect of
offering higher yield of 3a.9 On the basis of these results, several other solvent systems including
tetrahydrofurane (THF), 1,4-dioxane, dimethyl sulfoxide (DMSO) and N,N-dimethyl acetamide (DMA)
were investigated with 1a. As shown in Table 2, the reactions carried out with THF and 1,4-dioxane
proceeded sluggishly to afford only trace amounts of 3a (5% and 6%) and 2a (8%), along with recovered
1a in 40% and 52% yields even after prolonged reaction times (Table 2, entries 1 and 2). Moreover, the
use of DMSO also turned out to be unsuccessful in giving no detectable formation of 3a and 2a, and 80%
of intact 1a (entry 3). The reaction in DMA gave a comparable 3a/2a ratio (70:30) with that obtained
from DMF (Table 1, entry 1), but a much lower yield of 3a (Table 2, entry 4). These disappointing results
led us to look for other variants to modify reaction conditions.
Table 2. Effect of solvents on the palladium(0)-catalyzed cyclization of 1a
OPO(OEt)2
Pd(PPh3)4 (0.1eq), Et3N (5 eq)
solvent, heat, 24 h
N
3a
+
2a
Ph
1a
yield (%)d
ratio of
3a:2a
entrya
solvent
3a
2a
1b
2c
3c
4c
THF
1,4-dioxane
DMSO
5
6
-
8
8
-
38:62
43:57
-
DMA
28
12
70:30
a) All reactions were performed using 1 equiv of 1a in a 0.02 M solution. b) The reaction was
performed in refluxing THF. c) The reactions were carried out at 80 oC. d) Isolated yields.

374
HETEROCYCLES, Vol. 78, No. 2, 2009
Having proved DMF to be not displaceable for the 6-endo cyclization, we next evaluated a range of bases
by using 1a as the substrate and Pd(PPh3)4 as the catalyst. As illustrated in Table 3, when potassium
carbonate and acetate were employed, the reactions displayed some preference for the 6-endo cyclization
to give 3a and 2a in the ratios of 55:45 and 64:36, respectively, but the yields of 3a were only modest
(Table 3, entries 1, 2). Furthermore, we observed that the reaction performed with potassium t-butoxide
only afforded a trace amount of 3a (5%) as the only detectable cyclization product (entry 3), and the
starting material easily decomposed within a short period of time (＜10 min) to yield a complex mixture
under such strong basic conditions. The use of DBU produced only 9% of 3a as the minor regioisomeric
product together with 27% of 2a (entry 4). In the case of 1,4-diazabicylo[2,2,2]octane (Dabco) (entry 5),
the desired 3a was also formed in low yield (19%) but as the major cyclization product (3a:2a 59:41).
The use of pyridine resulted in the relative higher 3a/2a ratio in comparison with that given in entry 5, but
the yield of 3a was still only moderate (29%) (entry 6). Finally, we were gratified to discover that with
the employment of diisopropylethyl amine (DIPEA) as the base, the yield of 3a could be significantly
improved to 71% (entry 7), higher than those obtained from the other types of oxime derivatives.6,11
Meanwhile, the competing 5-exo cyclization was also suppressed to a large extend as reflected by the
ratio between 3a and 2a (83:17). Thus, the results given in entry 7 represented the best ones obtained
from 1a in terms of both yield of 3a and 3a/2a ratio.
Table 3. Effect of bases on the Pd(0)-catalyzed cyclization of 1a
OPO(OEt)2
N
Pd(PPh3)4 (0.1eq), base (5 eq)
DMF, 80 oC
3a
+
2a
Ph
1a
yield (%)b
ratio of
3a:2a
entrya
base
time (h)
3a
2a
1
2
3
4
5
6
7
K2CO3
KOAc
12
12
3
12
12
12
12
27
27
5
22
15
-
27
13
14
15
55:45
64:36
100:0
25:75
59:41
67:33
83:17
t-BuO-K+
DBU
9
Dabco
pyridine
DIPEA
19
29
71
a) All reactions were performed using one equivalent of 1a in a 0.02 M DMF solution.
b) Isolated yields.
In most cases listed in Table 3, we also noted the formation of 1-phenyl-pent-4-en-1-one (10-47% and
10% for entry 7) arising from the hydrolysis of 1a. According to the literature reports,9,14 such type of
side reactions can be minimized in the presence of molecular sieves. To our surprise, the addition of

HETEROCYCLES, Vol. 78, No. 2, 2009
375
molecular sieves (MS) 4 Å to the reaction under the conditions of Pd(PPh3)4/DIPEA/DMF caused no
apparent reduction of the ketone but a dramatic change in regioselectivity to give 2a in 29% yield and a
trace amount of 3a (~ 3%) (3a:2a 9:91). To verify if this 5-exo promoting effect of MS was common, we
combined a few above-described catalytic conditions [Pd(PPh3)4/Et3N/DMF (Table 1, entry 1),
Pd(PPh3)4/K2CO3/DMF (Table 3, entry 1) and Pd(PPh3)4/DBU/DMF (Table 3, entry 4) ] with the use of
MS and applied to 1a. In contrast to the examples given in Table 1 and 3, these reactions were all
overwhelmingly dominated by the 5-exo cyclization with MS, resulting in the drastically decreased ratios
between 3a and 2a (Table 4, entries 2-4). The underlying causes of the above-mentioned base and MS
effects on the regiochemistry are still not fully understood for us and will be subjected to our further
exploration.
Table 4. Effect of molecular sieves 4 Å on the Pd(0)-catalyzed cyclization of 1a
OPO(OEt)2
N
Pd(PPh3)4 (0.1eq), base (5 eq)
4 Å MS, DMF, 80 oC, 12 h
3a
+
2a
Ph
1a
yield (%)c
ratio of
3a:2a
entrya
catalytic conditionb
3a
2a
1
2
3
4
Pd(PPh3)4/DIPEA
Pd(PPh3)4/Et3N
Pd(PPh3)4/K2CO3
Pd(PPh3)4/DBU
3
16
13
3
29
56
54
42
9:91
22:78
19:81
7:93
a) All reactions were performed using one equivalent of 1a in a 0.02 M DMF solution.
b) 100 mg of MS 4 Å was used for 1 mmol of 1a. c) Isolated yields.
The reaction conditions indicated in entry 7 of Table 3 were subsequently applied into a range of
γ,δ-unsaturated ketone O-diethylphosphinyloximes with various substitutions (1d-i), which were all
readily prepared from the corresponding aldehydes following the procedure given in Scheme 3. Among
which, 1d-f and 1h-i were previously used for the preparation of the pyrroles,9 while 1g were synthesized
for the first time.13 As shown in Table 5, under the reaction conditions, the substrates possessing the
aromatic (1d, 1f-g) and heteroaromatic (1e) substituents cyclized preferentially in a 6-endo pathway to
give rise to the 2-substituted pyridines 3d-g as the major products in good yields (50-61%) (entries 1-4).
These compounds could be easily separated from pyrrole byproducts (5-16%) by flash chromatography
on basic aluminium oxide. We additionally observed that the electronic natures of the 4-substituents on
the phenyl moiety seemed to have no effect on the 6-endo cyclization as illustrated by the almost equal
yields of 3f and 3g (entries 3 and 4). However, the current catalytic system appeared to be less applicable
to the cycloalkyl or alkyl substituted substrates. The cyclization reactions of 1h and 1i did not show

376
HETEROCYCLES, Vol. 78, No. 2, 2009
strong regioselective bias to furnish pyridines 3h and 3i in 26% and 20% yield, respectively (entries 5 and
6), together with the corresponding pyrroles in almost equal yields (23% and 18%, respectively).
To further extend the utility of the cyclization process, we also investigated the possibility of using the
homologated δ,ε-unsaturated ketone O-diethylphosphinyloxime precursors for the generation of
substituted methylpyridines. It was expected that by using these hitherto unexplored substrates, the
complete control of the regioselectivity of the cyclization could be achieved. At first, several
δ,ε-unsaturated ketone O-diethylphosphinyloximes 7a-e were prepared in the analogous manner as for
synthesizing 1, except using pent-4-enylmagnesium bromide for the first addition step (Scheme 4).13
Table 5. Preparation of 2-substituted pyridines 3d-i via the Pd(0)-catalyzed cyclization of 1d-i
OPO(OEt)2
N
N
Pd(PPh3)4 (0.1eq), DIPEA (5 eq)
DMF, 80 oC, 12 h
R
R
1d-i
3d-i
entrya
starting material
product
yield (%)b
1
2
3
4
5
6
1d (R = 2-naphthalyl, E/Z = 93:7)
1e (R = 4-pyridinyl, E/Z = 83:17)
1f (R = 4-methoxyphenyl, E/Z = 70:30)
1g (R = 4-fluorophenyl, E/Z = 84:16)
1h (R = cyclohexyl, E/Z = 68:32)
1i (R = n-heptyl, E/Z = 50:50)
3d
3e
3f
3g
3h
3i
60
61
51
50
26
20
a) All reactions were performed using one equivalent of substrate in a 0.02 M DMF
solution. b) Isolated yields.
Scheme 4. Synthesis of δ,ε-unsaturated ketone O-diethylphosphinyloxime precursors 7a-e
O
OH
CH2=CHCH2CH2CH2MgBr (1 eq), THF
-20 oC, 10 min, then rt, 12 h, 49-83%
H
R
R
1) NH2OH.HCl (1.1 eq)
NaOAc (1.1 eq)
CH3OH, 20 oC, 6 h
O
PDC (1.5 eq), CH2Cl2
rt 12 h, 79-95%
R
2) ClPO(OEt)2 (2 eq)
NaH (1.2 eq), THF, 0 oC
to rt, 12 h
OPO(OEt)2
N
R
7a (R = Ph, 80%, E/Z = 86:14)
7b (R = 4-pyridinyl, 79%, E/Z = 75:25)
7c (R = 4-methoxyphenyl, 81%, E/Z = 88:12)
7d(R = 4-fluorophenyl, 73%, E/Z = 85:15)
7e (R = n-heptyl, 63%, E/Z = 50:50)
We then explored the suitable cyclization conditions by using 7a as the substrate. However, with the
extension of one carbon unit, 7a was found to be more difficult to cyclize than 1a under most of the

HETEROCYCLES, Vol. 78, No. 2, 2009
377
reaction conditions given in Table 3 (entries 1 and 2, 4-7) and Table 1, to provide only small amounts of
2-methyl 6-phenylpyridine (8a) (< 10%) plus 60-86% of 1-phenyl-hex-5-en-1-one resulting from the
hydrolysis. After several unsuccessful attempts, we discovered that the addition of 5 equivalents of
tetrabutylammonium chloride [(n-Bu)4NCl] to the reaction under the catalytic system of Pd(PPh3)4 (0.2
eq)/Et3N (5 eq)/MS 4 Å/DMF could result in the generation of 8a in 50 % yield. Based on this, it was
further found that the yield of 8a could be improved to 63% when (n-Bu)4NCl was replaced by
tetrabutylammonium bromide [(n-Bu)4NBr]. The reaction conditions of Pd(PPh3)4 (0.2 eq)/Et3N (5
eq)/MS 4 Å/DMF/(n-Bu)4NBr (5eq) were then applied to 7b-e and the results are compiled in Table 6.
For 7b-d bearing the aryl and heteroaryl substituents, the reactions offered the desired pyridines 8b-d in
modest to good yields (36-50%), as well as the ketone byproducts in 26-54% yields (entries 2-4). The
relatively low-yielding formation of the pyridine was observed for 7e substituted with an alkyl group. The
reaction of which afforded 8e in 21% yield, along with a large amount of the ketone (74%) (entry 5).
The cyclization products 3a, 3d-i and 8a-e were all purified by flash chromatography on basic aluminium
13
oxide and satisfactory spectra (1H, C-NMR, IR and HRMS) were obtained for them. For those known
compounds (3a, 3d-h, 8a-c), the NMR data were found to agree well with those reported in the
literature.15
Table 6. Preparation of substituted methylpyridines 8a-e via the Pd(0)-catalyzed cyclization of 7a-e
CH3
OPO(OEt)2
Pd(PPh3)4 (0.2eq), Et3N (5 eq)
N
N
4 Å MS, (n-Bu)4NBr (5eq), DMF
80 oC, 12 h
R
R
7a-e
8a-e
entrya,b
substrate
yield (%)c
product
1
2
3
4
5
7a
7b
7c
7d
7e
8a (R = Ph)
63
36
50
42
21
8b (R = 4-pyridinyl)
8c (R = 4-methoxyphenyl)
8d (R = 4-fluorophenyl)
8e (R = n-heptyl)
a) All reactions were performed using one equivalent of substrate in a 0.02 M
DMF solution. b) 100 mg of MS 4 Å was used for 1 mmol of oxime substrate. c)
Isolated yields.
In conclusion, we have demonstrated that the regioselectivity of the amino-Heck reactions of
γ,δ-unsaturated ketone O-diethylphosphinyloximes, particularly for those bearing aromatic or
heteroaromatic substituents, could be deliberately controlled to a large extend with the adjustment of
catalytic conditions. Thus, these compounds can serve as the versatile and useful precursors for the
preparation of either substituted methylpyrroles or 2-substituted pyridines. Under the catalytic conditions

378
HETEROCYCLES, Vol. 78, No. 2, 2009
of Pd(PPh3)4/DIPEA/DMF, they were shown to preferentially undergo the 6-endo cyclization to afford
2-substituted pyridines in moderate to good yields. During the course of the studies, several interesting
aspects regarding the effects of phosphinyl groups, solvents, bases and molecular sieves on the
regioselectivity have also been observed, and the further investigation to disclose the true reasons behind
these observations should be necessary. The synthetic scope of the protocol has been further expanded to
the synthesis of substituted methylpyridines form δ,ε-unsaturated ketone O-diethylphosphinyloximes,
thereby providing a new entry into this type of heterocyclic compounds.
EXPERIMENTAL
All of starting materials were obtained from commercial suppliers and used without further purification.
Reactions were performed under an atmosphere of Nitrogen. Tetrahydrofuran was distilled from
sodium-benzophenone, and dichloromethane, triethylamine, dimethylformamide, 1,4-dioxane, dimethyl
sulfoxide and N,N-dimethylacetamide were distilled from calcium hydride before use. TLC analysis was
carried out on Merck 25 DC-Alufolien Kieselgel 60F254 aluminum-backed plates visualised by using UV
light, or by means of ethanolic solution of vanillin (5%) with sulphuric acid (5%). All of the compounds
were purified by flash chromatography on Merck Art.9385 Kiesegel 60 silica gel (230-400 mesh) or
Brockmann I basic aluminum oxide (~150 mesh). NMR spectra (1H, 13C) were recorded on a Bruker 400
spectrometer using deuteriochloroform (CDCl3) as solvent. Chemical shifts measurements are reported in
delta (δ) units. Splitting patterns are described as singlet (s), doublet (d), triplet (t), quartet (q) or multiplet
(m). Coupling constants (J) are reported in Hertz (Hz). Infrared (IR) spectra were recorded on an IR-FT
JASCO 410 spectrophotometer (neat) and resonances are reported in wave numbers (cm-1). High
resolution mass spectra (HRMS) were determined by using a A. E. I. model MS-50 mass spectrometer in
fast atom bombardment (FAB) or electron impact (EI) mode. Elemental analyses were performed on
Perkin Elmer 240C apparatus.
General Procedure for the Preparation of γ,δ- and δ,ε-Unsaturated Phosphinyl Oxime Substrates
from the Corresponding Ketones; 1-Phenyl-4-penten-1-one O-diethylphosphinyloxime (1a): To a
solution of 1-phenyl-4-penten-1-one (317 mg, 1.98 mmol) in MeOH (8 mL), NaOAc (179 mg, 2.18
mmol) and NH2OH·HCl (152 mg, 2.18 mmol) were successively added. The reaction mixture was stirred
at 20 °C for 6 h, then diluted with CH2Cl2 (30 mL) and washed sequentially with H2O (2 x 10 mL) and
saturated NaCl aq soln (10 mL). After dried over anhydrous Na2SO4 and concentrated at ambient
temperature, the crude hydroxyoxime was dissolved in anhyd THF (13 mL). The solution was cooled to
0 °C and added with NaH (60%, 95 mg, 2.38 mmol). Stirring was continued at 0 °C for another 20 min
under N2, before (EtO)2POCl (0.61 ml, 3.96mmol) was added dropwise in 3 min. The reaction mixture
was allowed to warm to rt and continued to stir for 12 h. After concentration, the crude residue was

HETEROCYCLES, Vol. 78, No. 2, 2009
379
subjected to chromatographic purification on silica gel (hexane-EtOAc 5:1, 1:1) to afford 1a as a mixture
of two isomers (E/Z = 87:13) (513 mg, 83%).
IR (neat): 2983 , 1276 , 1034 , 919 cm-1; 1H NMR (400 MHz, CDCl3): Z-isomer : δ 1.30 (td, J1 = 7.1 Hz,
J2 = 0.7 Hz , 6H), 2.24 (dt, J1 = 7.4 Hz, J2 = 7.2 Hz , 2H), 2.74 (broad t, J = 7.6 Hz, 2H), 4.10-4.19 (m,
4H), 4.96-5.09 (m, 2H), 5.70-5.83 (m, 1H), 7.31-7.37 (m, 2H), 7.37-7.48 (m, 3H) ppm; E-isomer: δ 1.38
(td, J1 = 7.1 Hz, J2 = 0.7 Hz , 6H), 2.32 (dt, J1 = 7.8 Hz, J2 = 7.3 Hz , 2H), 2.95 (broad t, J = 7.9 Hz, 2H),
4.21-4.34 (m, 4H), 5.00 (broad d, J = 11.5 Hz, 1H), 5.04 (broad d, J = 17.1 Hz, 1H), 5.82 (dm, J = 17.1
Hz, 1H), 7.37-7.45 (m, 3H), 7.65 (broad d, J = 7.8 Hz, 2H) ppm; 13C NMR (100 MHz, CDCl3): Z-isomer:
δ 16.1 (d, J = 6.1 Hz), 30.2, 34.7, 64.5 (d, J = 5.5 Hz), 115.8, 127.5, 128.3, 129.6, 132.1, 136.5, 166.9 (d, J
= 13.0 Hz) ppm; E-isomer: δ 16.2 (d, J = 6.3 Hz), 27.0, 30.5, 64.6 (d, J = 6.0 Hz), 115.8, 127.1, 128.7,
130.5, 133.6, 136.6, 166.7 (d, J = 13.4 Hz) ppm; HRMS-FAB: m/z [M + H]+ calcd. for C15H23O4NP:
312.1365; found: 312.1357.
1-Phenyl-4-penten-1-one O-diphenylphosphinyloxime (1b):The typical procedure for the preparation
of 1a was followed by using ClPO(OPh)2 instead of ClPO(OEt)2. From 210 mg (1.31 mmol) of
1-phenyl-4-penten-1-one, 1b was obtained as a pale yellow oil after flash chromatography on silica gel
(hexane-EtOAc 3:1, 1:1, 1:2) (422 mg, E/Z = 78:22, 79%).
1
IR (neat): 3074, 1687, 1591, 1489, 1442, 1298, 1188, 766, 688 cm-1; H NMR (400 MHz, CDCl3):
E-isomer: δ 2.21 (td, J1 = 7.8 Hz, J2 = 7.3 Hz, 2H), 2.90 (broad t, J = 7.9 Hz, 2H), 4.96 (broad d, J = 9.0
Hz, 1H), 4.99 (broad d, J = 17.3 Hz, 1H), 5.72 (dm, J = 17.3 Hz, 1H) 7.17-7.24 (m, 3H), 7.27-7.53 (m,
10H), 7.63 (broad d, J = 8.1 Hz, 2H) ppm; Z-isomer: δ 2.18-2.26 (m, 2H), 2.76 (broad t, J = 7.5 Hz, 2H),
5.01-5.13 (m, 2H), 5.79-5.93 (m, 1H), 7.15-7.22 (m, 3H), 7.27-7.53 (m, 10H), 7.52 (broad d, J = 8.0Hz,
2H) ppm; 13C NMR (100 MHz, CDCl3): E-isomer: δ 27.3, 30.4, 115.9, 120.4 (d, J = 4.7 Hz), 125.5, 127.3,
128.7, 129.8, 130.8, 133.0, 136.3, 150.6 (d, J = 7.3 Hz), 168.0 (d, J = 13.6 Hz) ppm; Z-isomer: δ 30.0,
34.7, 115.7, 119.8 (d, J = 4.1 Hz), 125.4, 127.3, 128.9, 129.6, 129.7, 131.9, 137.0, 150.5 (d, J = 7.3 Hz),
168.2 (d, J = 12.5 Hz) ppm; HRMS-FAB: m/z [M + H]+ calcd. for C23H23O4NP: 408.1365; found:
408.1367.
1-Phenyl-4-penten-1-one O-dimethylphosphinyloxime (1c): The typical procedure for the preparation
of 1a was followed by using ClPO(OMe)2 instead of ClPO(OEt)2. From 157 mg (0.98 mmol) of
1-phenyl-4-penten-1-one, 1c was obtained as a yellow oil after flash chromatography on silica gel
(hexane-EtOAc 4:1, 2:1) (236 mg, E/Z = 90:10, 85%).
IR (neat): 2954, 1641, 1572, 1444, 1281, 1043, 764, 694 cm-1; 1H NMR (400 MHz, CDCl3): E-isomer: δ
2.30 (td, J1 = 7.7 Hz, J2 = 7.2 Hz, 2H), 2.95 (broad t, J = 7.9 Hz, 2H), 3.88 (d, J = 11.1 Hz, 6H), 4.99
(broad d, J = 9.8 Hz, 1H), 5.02 (broad d, J = 17.0 Hz, 1H), 5.80 (dm, J = 17.0 Hz, 1H), 7.35-7.46 (m, 3H),
7.59-7.68 (m, 2H) ppm; Z-isomer: δ 2.21 (td, J1 = 7.6 Hz, J2 = 6.8 Hz, 2H), 2.73 (broad t, J = 7.6 Hz, 2H),

380
HETEROCYCLES, Vol. 78, No. 2, 2009
3.78 (d, J = 11.2 Hz, 6H), 4.95-5.06 (m, 2H), 5.70-5.81 (m, 1H), 7.30-7.34 (m, 2H) 7.35-7.46 (m, 3H)
13
ppm; C NMR (100 MHz, CDCl3): E-isomer: δ 27.0, 30.5, 55.1 (d, J = 6.1 Hz), 115.9, 127.2, 128.7,
130.6, 133.4, 136.5, 167.2 (d, J = 13.2 Hz) ppm; Z-isomer: δ = 30.2, 34.7, 54.9 (d, 6.0 Hz), 127.5, 128.3,
129.7, 131.9, 136.4, 167.3 (d, 12.5 Hz) ppm; HRMS-EI: m/z [M]+ calcd. for C13H18O4NP: 283.0973;
found: 283.0979.
1-(2-Naphthalyl)-4-penten-1-one O-diethylphosphinyloxime (1d):The typical procedure for the
preparation of 1a was followed; 1-(2-naphthalyl)-4-penten-1-one (120 mg, 0.57 mmol) was used as
starting material. Flash chromatography on silica gel (hexane-EtOAc 5:1, 1:1) gave 1d as a yellow oil.
Yield: 200 mg (97%) (E/Z = 93:7).
IR (neat): 2982, 1443, 1275, 1165, 1033, 917 cm-1; 1H NMR (400 MHz, CDCl3): E-isomer: δ 1.41 (td, J1
= 7.1 Hz, J2 = 0.9 Hz, 6H), 2.39 (td, J1 = 7.8 Hz, J2 = 7.8 Hz, 2H), 3.08 (broad t, J = 7.9 Hz, 2H),
4.23-4.38 (m, 4H), 5.02 (broad d, J = 10.1 Hz, 1H), 5.07 (broad d, J = 17.1 Hz, 1H), 5.87 (dm, J = 17.1
Hz, 1H), 7.50-7.58 (m, 2H), 7.82-7.92 (m, 4H), 8.08 (s, 1H) ppm; Z-isomer: δ 1.30 (td, J1 = 7.0 Hz, J2 =
0.7 Hz, 6H), 2.27 (td, J1 = 7.7 Hz, J2 = 7.5 Hz, 2H), 2.85 (broad t, J = 7.6 Hz, 2H), 4.08-4.24 (m, 4H),
4.98-5.11 (m, 2H), 5.69-5.82 (m, 1H), 7.42-7.52 (m, 2H), 7.79-7.89 (m, 4H), 8.08 (s, 1H) ppm; 13C NMR
(100 MHz, CDCl3): E-isomer: δ 16.2 (d, J = 6.4 Hz), 26.8, 30.7, 64.7 (d, J = 5.8 Hz), 115.8, 123.9, 126.6,
127.3, 127.5, 127.7, 128.4, 128.7, 131.0, 132.9, 134.3, 136.6, 166.6 (d, J = 13.5 Hz) ppm; HRMS-EI: m/z
[M]+ calcd. for C19H24O4NP: 361.1443; found: 361.1446.
1-(4-Pyridinyl)-4-penten-1-one O-diethyl phosphinyloxime (1e): The typical procedure for the
preparation of 1a was followed; 1-(4-pyridinyl)-4-penten-1-one (171 mg, 1.06 mmol) was used as starting
material. Flash chromatography on silica gel (hexane-EtOAc 1:1, EtOAc) gave 1e as a viscous oil. Yield:
262 mg (79%) (E/Z = 83:17).
IR (neat): 2984, 1595, 1544, 1410, 1274, 1032, 916 cm-1; 1H NMR (400 MHz, CDCl3): E-isomer: δ 1.38
(t, J = 7.1 Hz, 6H), 2.31 (td, J1 = 7.8 Hz, J2 = 7.0 Hz, 2H), 2.94 (broad t, J = 7.8 Hz, 2H), 4.19-4.35 (m,
4H), 5.00 (broad d, J = 9.2 Hz, 1H), 5.03 (broad d, J = 17.3 Hz, 1H), 5.78 (dm, J = 17.3 Hz, 1H), 7.53 (d,
J = 6.1 Hz, 2H), 8.67 (broad d, J = 6.1 Hz, 2H) ppm; Z-isomer: δ 1.29 (t, J = 7.1 Hz, 6H), 2.18-2.29 (m,
2H), 2.73 (broad t, J = 7.5 Hz, 2H), 4.08-4.19 (m, 4H), 4.92-5.08 (m, 2H), 5.70-5.90 (m, 1H), 7.21 (d, J =
6.1 Hz, 2H), 8.62-8.72 (m, 2H) ppm; 13C NMR (100 MHz, CDCl3): E-isomer: δ 16.2 (d, J = 6.3 Hz), 26.4,
30.3, 64.9 (d, J = 5.9 Hz), 116.3, 121.2, 136.0, 141.3, 150.3, 164.9 (d, J = 13.9 Hz) ppm; Z-isomer: δ 16.1
(d, J = 6.0 Hz), 29.8, 34.2, 64.8 (d, J = 6.0 Hz), 116.2, 121.8, 135.9, 140.3, 150.0, 164.8 (d, J = 13.5 Hz)
ppm; HRMS-FAB: m/z [M + H]+ calcd. for C14H22O4N2P: 313.1317; found: 313.1318.
1-(4-Methoxyphenyl)-4-penten-1-one O-diethylphosphinyloxime (1f): The typical procedure for the
preparation of 1a was followed; 1-(4-methoxyphenyl)-4-penten-1-one (177 mg, 0.93 mmol) was used as
starting material. Flash chromatography on silica gel (hexane-EtOAc 5:1, 1:1) gave 1f as a pale yellow

HETEROCYCLES, Vol. 78, No. 2, 2009
381
oil. Yield: 257 mg (81%) (E/Z = 70:30).
IR (neat): 2974, 2929, 1606, 1514, 1255, 1032, 922 cm-1; 1H NMR (400 MHz, CDCl3): E-isomer: δ 1.37
(td, J1 = 7.1 Hz, J2 = 0.7 Hz, 6H), 2.31 (td, J1 = 7.4 Hz, J2 = 7.2 Hz, 2H), 2.91 (broad t, J = 7.9 Hz, 2H),
3.83 (s, 3H), 4.20-4.31 (m, 4H), 4.99 (broad d, J = 10.0 Hz, 1H), 5.03 (broad d, J = 17.2 Hz, 1H), 5.82
(dm, J = 17.2 Hz, 1H), 6.90 (d, J = 8.9 Hz, 2H), 7.62 (d, J = 8.9 Hz, 2H) ppm; Z-isomer: δ 1.31 (td, J1 =
7.0 Hz, J2 = 0.6 Hz, 6H), 2.22 (td, J1 = 7.6 Hz, J2 = 7.4 Hz, 2H), 2.73 (broad t, J = 7.6 Hz, 2H), 3.83 (s,
3H), 4.11-4.20 (m, 4H), 4.95-5.05 (m, 2H), 5.72-5.89 (m, 1H), 6.91 (d, J = 8.9 Hz, 2H), 7.38 (d, J = 8.9
Hz, 2H) ppm; 13C NMR (100 MHz, CDCl3): E-isomer: δ 16.2 (d, J = 6.5 Hz), 26.8, 30.7, 55.3, 64.5 (d, J
= 5.9 Hz), 114.0, 115.7, 125.9, 128.6, 136.7, 161.5, 166.1 (d, J = 13.2 Hz) ppm; Z-isomer: δ 16.1 (d, J =
6.4 Hz), 30.5, 34.5, 55.3, 64.4 (d, J = 6.0 Hz), 113.6, 124.0, 129.7, 136.6, 160.5, 165.8 (d, J = 12.1 Hz)
ppm; HRMS-EI: m/z [M + H]+ calcd. for C16H25O5NP: 342.1470; found: 342.1473.
1-(4-Fluorophenyl)-4-penten-1-one O-diethylphosphinyloxime (1g): The typical procedure for the
preparation of 1a was followed; 1-(4-fluorophenyl)-4-penten-1-one (160 mg, 0.90 mmol) was used as
starting material. Flash chromatography on silica gel (hexane-EtOAc 6:1, EtOAc) gave 1g as a yellow oil.
1
Yield: 216 mg (73%) (E/Z = 84:16); IR (neat): 3078, 2983, 1641, 1603, 1510, 1277, 1034, 843 cm-1; H
NMR (400 MHz, CDCl3): E-isomer: δ 1.37 (t, J = 7.1 Hz, 6H), 2.28 (td, J1 = 7.8 Hz, J2 = 7.6 Hz, 2H),
2.91 (broad t, J = 7.9 Hz, 2H), 4.18-4.31 (m, 4H), 4.97 (broad d, J = 9.7 Hz, 1H), 5.01 (broad d, J = 17.4
Hz, 1H), 5.78 (dm, J = 17.4 Hz, 1H), 7.06 (tm, J = 8.5 Hz, 2H), 7.64 (dd, J1 = 8.8 Hz, J2 = 5.4 Hz, 2H)
ppm; Z-isomer: δ 1.29 (t, J = 7.1 Hz, 6H), 2.19 (td, J1 = 7.6 Hz, J2 = 7.4 Hz, 2H), 2.71 (broad t, J = 7.6 Hz,
2H), 4.08-4.18 (m, 4H), 4.93-5.02 (m, 2H), 5.69-5.81 (m, 1H), 7.04-7.11 (m, 2H), 7.35 (dm, J = 8.7 Hz,
2H) ppm; 13C NMR (100 MHz, CDCl3): E-isomer: δ 16.2 (d, J = 6.4 Hz), 26.9, 30.5, 64.7 (d, J = 6.0 Hz),
115.8 (d, J = 21.8 Hz), 116.0, 129.2 (d, J = 8.5 Hz), 129.6 (d, J = 3.1 Hz), 136.4, 164.1 (d, J = 251.0 Hz),
165.8 (d, J = 13.4 Hz) ppm; Z-isomer: δ 16.1 (d, J = 6.5 Hz), 30.2, 34.6, 64.6 (d, J = 6.0 Hz), 115.5 (d, J =
21.9 Hz), 115.9, 127.9 (d, J = 3.4 Hz), 129.8 (d, J = 8.5 Hz), 136.3, 163.1 (d, J = 250.1 Hz), 165.7 (d, J =
12.8 Hz) ppm; HRMS-FAB: m/z [M]+ calcd. for C15H21FO4NP: 329.1192; found: 329.1189.
1-Cyclohexyl-4-penten-1-one O-diethylphosphinyloxime (1h): The typical procedure for the
preparation of 1a was followed; 1-cyclohexyl-4-penten-1-one (97 mg, 0.58 mmol) was used as starting
material. Flash chromatography on silica gel (hexane-EtOAc 10:1, 3:1) gave 1h as a colourless oil. Yield:
128 mg (69%) (E/Z = 68:32).
IR (neat): 2981, 2931, 1666, 1641, 1450, 1275 cm-1; 1H NMR (400 MHz, CDCl3): E-isomer: δ 1.16-1.35
(m, 5H), 1.32 (td, J1 = 7.0 Hz, J2 = 0.8 Hz, 6H), 1.63-1.82 (m, 5H), 2.18-2.43 (m, 5H), 4.09-4.25 (m, 4H),
4.97 (broad d, J = 10.2 Hz, 1H), 5.02 (broad d, J = 17.0 Hz, 1H), 5.78 (dm, J = 17.0 Hz, 1H) ppm;
Z-isomer: δ 1.16-1.35 (m, 5H), 1.32 (td, J1 = 7.0 Hz, J2 = 0.8 Hz, 6H), 1.63-1.82 (m, 5H), 2.18-2.43 (m,
5H), 4.09-4.25 (m, 4H), 4.93-5.06 (m, 2H), 5.73-5.83 (m, 1H) ppm; 13C NMR (100 MHz, CDCl3):

382
HETEROCYCLES, Vol. 78, No. 2, 2009
E-isomer: δ 16.2 (d, J = 6.3 Hz), 25.7, 25.8, 25.9, 27.3, 30.0, 43.8, 64.4 (d, J = 5.9 Hz), 115.4, 137.0,
172.8 (d, J = 12.2 Hz) ppm; Z-isomer: δ 16.2 (d, J = 6.3 Hz), 28.9, 29.8, 30.2, 30.4, 38.2, 64.4 (d, J = 5.9
Hz), 115.3, 137.4, 172.0 (d, J = 12.5 Hz) ppm; HRMS-FAB: m/z [M + H]+ calcd. for C15H29O4NP:
318.1834; found: 318.1829.
Dodec-1-en-5-one O-diethylphosphinyloxime (1i): The typical procedure for the preparation of 1a was
followed; dodec-1-en-5-one (136 mg, 0.75 mmol) was used as starting material. Flash chromatography on
silica gel (hexane-EtOAc 10:1, 3:1) gave 1i as a viscous oil. Yield: 162 mg (65%) (E/Z = 50:50).
IR (neat): 2929, 1277, 1036, 918 cm-1; 1H NMR of isomeric mixture (400 MHz, CDCl3): δ 0.88 (t, J = 6.4
Hz, 3H), 1.21-1.32 (m, 8H), 1.35 (t, J = 7.1 Hz, 6H), 1.45-1.58 (m, 2H), 2.25-2.35 (m, 3H), 2.35-2.42 (m,
2H), 2.48 (broad t, J = 7.8 Hz, 1H) 4.12-4.29 (m, 4H), 5.00 (broad d, J = 10.2 Hz, 1H), 5.05 (broad d, J =
17.1 Hz, 1H), 5.80 (dm, J = 17.1 Hz, 1H) ppm; 13C NMR of isomeric mixture (100 MHz, CDCl3): δ 14.0,
16.1 (d, J = 6.3 Hz), 22.6, 25.8, 26.0, 28.1, 28.9, 29.2, 29.6, 29.8, 30.0, 31.6, 31.7, 33.2, 33.9, 64.3 (d, J =
5.8 Hz), 115.6, 115.7, 136.7, 136.9, 169.7 (d, J = 12.6 Hz), 169.8 (d, J = 11.9 Hz) ppm; HRMS-EI: m/z
[M - CH3]+ calcd. for C15H29O4NP: 318.1834; found: 318.1837.
1-Phenyl-5-hexen-1-one O-diethylphosphinyloxime (7a): The typical procedure for the preparation of
1a was followed; 1-phenyl-5-hexen-1-one (160 mg, 0.92 mmol) was used as the starting material. Flash
chromatography on silica gel (hexane-EtOAc 8:1, 3:1) gave 7a as a viscous oil. Yield: 239 mg (80%)
(E/Z = 86:14).
1
IR (neat): 2981, 1641, 1444, 1275, 916 cm-1; H NMR (400 MHz, CDCl3): E-isomer: δ 1.38 (td, J1 = 7.0
Hz, J2 = 0.8Hz, 6H), 1.63 (m, 2H), 2.14 (td, J1 = 7.2 Hz, J2 = 7.0 Hz, 2H ), 2.86 (broad t, J = 8.0 Hz, 2H),
4.21-4.38 (m, 4H), 4.99 (broad d, J = 10.1 Hz, 1H), 5.02 (broad d, J = 17.1 Hz, 1H), 5.79 (dm, J = 17.1
Hz, 1H), 7.37-7.47 (m, 3H), 7.66 (dd, J1 = 7.8 Hz, J2 = 1.2 Hz, 2H) ppm; Z-isomer: δ 1.30 (td, J1 = 7.0 Hz,
J2 = 0.9 Hz, 6H ), 1.54 (m, 2H), 2.05-2.13 (m, 2H), 2.65 (broad t, J = 7.7 Hz, 2H), 4.19-4.21 (m, 4H),
13
4.93-5.04 (m, 2H), 5.66-5.78 (m, 1H), 7.32-7.37 (m, 2H), 7.37-7.47 (m, 3H) ppm; C NMR (100 MHz,
CDCl3): E-isomer: δ 16.2 (d, J = 6.3 Hz), 25.7, 27.0, 33.6, 64.6 (d, J = 5.9 Hz), 115.5, 127.1, 128.6, 130.5,
133.6, 137.5, 167.3 (d, J = 13.2 Hz) ppm; Z-isomer: δ 16.1 (d, J = 6.3 Hz), 25.4, 33.0, 34.7, 64.4 (d, J =
5.8 Hz), 115.4, 127.4, 128.3, 129.5, 132.3, 137.6, 167.4 (d, J = 13.4 Hz) ppm; HRMS-FAB: m/z [M + H]+
calcd. for C16H25O4NP: 326.1521; found: 326.1514.
1-(4-Pyridinyl)-5-hexen-1-one O-diethylphosphinyloxime (7b): The typical procedure for the
preparation of 1a was followed; 1-(4-pyridinyl)-5-hexen-1-one (173 mg, 0.99 mmol) was used as the
starting material. Flash chromatography on silica gel (hexane-EtOAc 5:1, EtOAc) gave 7b as a yellow
oil. Yield: 254 mg (79%) (E/Z = 75:25).
IR (neat): 3078, 2983, 1641, 1595, 1545, 1275, 1032, 916 cm-1; 1H NMR (400 MHz, CDCl3): E-isomer: δ
1.38 (t, J = 7.1 Hz, 6H), 1.65 (m, 2H), 2.13 (td, J1 = 7.2 Hz, J2 = 7.1 Hz, 2H), 2.83 (broad t, J = 8.0 Hz,

HETEROCYCLES, Vol. 78, No. 2, 2009
383
2H), 4.20-4.34 (m, 4H), 5.00 (broad d, J = 10.1 Hz, 1H), 5.02 (broad d, J = 17.5 Hz, 1H), 5.77 (dm, J =
17.5 Hz, 1H), 7.57 (d, J = 6.2 Hz, 2H), 8.67 (broad d, J = 6.2 Hz, 2H) ppm; Z-isomer: δ 1.29 (t, J = 7.0 Hz,
6H), 1.54-1.62 (m, 2H), 2.06-2.14 (m, 2H), 2.62 (broad t, J = 7.7 Hz, 2H), 4.09-4.20 (m, 4H), 4.95-5.01
(m, 2H), 5.66-5.76 (m, 1H), 7.21 (broad d, J = 6.2 Hz, 2H), 8.67-8.70 (m, 2H) ppm; 13C NMR (100 MHz,
CDCl3): E-isomer: δ 16.2 (d, J = 6.3 Hz), 25.5, 26.4, 33.5, 64.9 (d, J = 5.9 Hz), 115.9, 121.1, 137.1, 141.2,
150.4, 165.5 (d, J = 13.8 Hz) ppm; Z-isomer: δ 16.1 (d, J = 6.3 Hz), 25.1, 32.9, 34.3, 64.8 (d, J = 5.9 Hz),
115.8, 121.7, 137.2, 140.4, 150.1, 165.4 (d, J = 13.4 Hz) ppm; HRMS-FAB: m/z [M + H]+ calcd. for
C15H24O4N2P: 327.1483; found: 327.1474.
1-(4-Methoxyphenyl)-5-hexen-1-one O-diethylphosphinyloxime (7c): The typical procedure for the
preparation of 1a was followed; 1-(4-methoxyphenyl)-5-hexen-1-one (210 mg, 1.03 mmol) was used as
the starting material. Flash chromatography on silica gel (hexane-EtOAc 10:1, 2:1) gave 7c as a yellow
oil. Yield: 296 mg (81%) (E/Z = 88:12).
IR (neat): 2981, 1608, 1514, 1255, 1032, 918, 837 cm-1; 1H NMR (400 MHz, CDCl3): E-isomer: δ 1.36 (t,
J = 7.1 Hz, 6H), 1.65 (m, 2H), 2.12 (td, J1 = 7.1 Hz, J2 = 7.1 Hz, 2H), 2.81 (broad t, J = 8.0 Hz, 2H), 3.81
(s, 3H), 4.19-4.31 (m, 4H), 4.97 (broad d, J = 10.2 Hz, 1H), 5.01 (broad d, J = 17.1 Hz, 1H), 5.77 (dm, J =
17.1 Hz, 1H), 6.89 (d, J = 8.8 Hz, 2H), 7.61 (d, J = 8.8 Hz, 2H) ppm; Z-isomer: δ 1.30 (t, J = 7.1 Hz, 6H),
1.55 (m, 2H), 2.03-2.10 (m ,2H), 2.63 (broad t, J = 7.6 Hz,, 2H), 3.81 (s, 3H), 4.09-4.21 (m, 4H),
5.01-5.10 (m, 2H), 5.66-5.79 (m, 1H), 6.86-6.93 (m, 2H), 7.37 (d, J = 8.7 Hz, 2H) ppm; 13C NMR (100
MHz, CDCl3): E-isomer: δ 16.2 (d, J = 6.4 Hz), 25.9, 26.8, 33.6, 55.3, 64.5 (d, J = 5.9 Hz), 114.0, 115.5,
125.9, 128.6, 137.6, 161.5, 166.6 (d, J = 13.3 Hz) ppm; Z-isomer: δ 16.1 (d, J = 6.4 Hz), 25.8, 33.0, 34.5,
55.2, 64.4 (d, J = 6.1 Hz), 113.6, 115.3, 124.1, 129.7, 137.7, 160.5, 166.3 (d , J = 13.3 Hz) ppm;
HRMS-FAB: m/z [M]+ calcd. for C17H26O5NP: 355.1549; found: 355.1539.
1-(4-Fluorophenyl)-5-hexen-1-one O-diethylphosphinyloxime (7d): The typical procedure for the
preparation of 1a was followed; 1-(4-fluorophenyl)-5-hexen-1-one (101 mg, 0.53 mmol) was used as the
starting material. Flash chromatography on silica gel (hexane-EtOAc 6:1, 1:1) gave 7d as a yellow oil.
Yield: 132 mg (73%) (E/Z = 85:15).
IR (neat): 3076, 2983, 1641, 1602, 1512, 1275, 1063, 918, 842 cm-1; 1H NMR (400 MHz, CDCl3):
E-isomer: δ 1.37 (td, J1 = 7.1 Hz, J2 = 0.7 Hz, 6H), 1.64 (m, 2H), 2.12 (td, J1 = 7.2 Hz, J2 = 7.0 Hz, 2H),
2.82 (broad t, J = 8.0 Hz, 2H), 4.19-4.31 (m, 4H), 4.98 (broad d, J = 10.1 Hz, 1H), 5.01 (broad d, J = 17.1
Hz, 1H), 5.76 (dm, J = 17.1 Hz, 1H), 7.07 (t, J = 8.8 Hz, 2H), 7.65 (dm, J = 8.8 Hz, 2H) ppm; Z-isomer: δ
1.30 (td, J1 = 7.0 Hz, J2 = 0.7 Hz, 6H), 1.55 (m, 2H), 2.03-2.10 (m, 2H), 2.63 (broad t, J = 7.7 Hz, 2H),
4.08-4.19 (m, 4H), 4.91-5.01 (m, 2H), 5.65-5.78 (m, 1H), 7.05-7.12 (m, 2H), 7.36 (dm, J = 8.8 Hz, 2H)
ppm; 13C NMR (100 MHz, CDCl3): E-isomer: δ 16.2 (d, J = 6.4 Hz), 25.7, 26.9, 33.5, 64.6 (d, J = 6.0 Hz),
115.6, 115.7 (d, J = 22.4 Hz), 129.1 (d, J = 8.5 Hz), 129.7 (d, J = 3.5 Hz), 137.4, 164.2 (d, J = 251.0 Hz),

384
HETEROCYCLES, Vol. 78, No. 2, 2009
166.3 (d, J = 13.5 Hz) ppm; Z-isomer: δ 16.1 (d, J = 5.9 Hz), 25.5, 32.9, 34.7, 64.5 (d, J = 6.2 Hz), 115.5,
115.5 (d, J = 22.1 Hz), 128.1 (d, J = 3.8 Hz), 129.8 (d, J = 8.4 Hz), 137.5, 163.1 (d, J = 250.1 Hz), 166.2
(d, J = 13.0 Hz) ppm; HRMS-FAB: m/z [M + H]+ calcd. for C16H24FO4NP: 344.1427; found: 344.1433.
Tridec-1-en-6-one O-diethylphosphinyloxime (7e): The typical procedure for the preparation of 1a was
followed; tridec-1-en-6-one (87 mg, 0.44 mmol) was used as the starting material. Flash chromatography
on silica gel (hexane-EtOAc 10:1, 5:1) gave 7e as a colourless oil. Yield: 97 mg (63%) (E/Z = 50:50).
IR (neat):2979, 1641, 1277, 1066, 920 cm-1; 1H NMR of the isomeric mixture (400 MHz, CDCl3): δ 0.85
(t, J = 6.5 Hz, 3H), 1.22-1.30 (m, 8H), 1.33 (td, J1 =7.1 Hz, J2 = 0.8 Hz, 6H), 1.44-1.53 (m, 2H), 1.55-1.65
(m, 2H), 2.03-2.10 (m, 2H), 2.22-2.29 (m, 2H), 2.32-2.39 (m, 2H), 4.12-4.23 (m, 4H), 4.96 (broad d, J =
11.6 Hz, 1H), 4.99 (broad d, J = 17.2 Hz, 1H), 5.75 (dm, J = 17.2 Hz, 1H) ppm; 13C NMR of the isomeric
mixture (100 MHz, CDCl3): δ = 14.0, 16.2 (d, J = 6.4 Hz), 22.6, 25.0, 25.2, 25.8, 26.1, 28.2, 28.8, 28.9,
29.2, 29.7, 31.6, 31.7, 33.1, 33.2, 33.6, 33.8, 64.3 (d, J = 5.7 Hz), 115.3, 115.4, 137.5, 137.7, 170.3 (d, J =
12.6 Hz), 170.4 (d, J = 12.4 Hz) ppm; HRMS-FAB: m/z [M + H]+ calcd. for C17H35O4NP: 348.2304;
found: 348.2297.
Typical Experimental Procedure for the Amino-Heck Reactions of γ,δ-Unsaturated
O-Diethylphosphinyloximes for the Generation of 2-Substituted Pyridines; 2-Phenylpyridine (3a):
Under a nitrogen atmosphere, Pd(PPh3)4 (37.5 mg, 0.032 mmol) and DIPEA (210 mg, 1.62 mmol) were
sequentially added to a solution of 1a (101 mg, 0.32 mmol) in anhyd DMF (16.2 mL, 0.02 M) at r.t. The
resulting yellow suspension was quickly warmed to 80 °C and continued to stir for 12 h. The reaction
mixture was then cooled to r.t., filtrated through a Celite pad and washed with EtOAc (80 mL). The
filtrate was successively washed with H2O (2 × 10 mL) and saturated NaCl aq soln (10 mL). After
concentration, the crude products were subjected to chromatographic purification on aluminum oxide
(hexane-EtOAc 5:1, 3:1) to afford 3a in 71% yield (35.7 mg).
1
IR (neat): 1735, 1266, 739 cm-1; H NMR (400 MHz, CDCl3): δ 7.19-7.26 (m, 1H), 7.40-7.52 (m, 3H),
7.70-7.79 (m, 2H), 8.00 (d, J = 7.3 Hz, 2H), 8.70 (d, J = 4.7 Hz, 1H) ppm; 13C NMR (100 MHz, CDCl3):
δ 120.6, 122.1, 126.9, 128.8, 129.0, 136.7, 139.4, 149.7, 157.5 ppm; HRMS-FAB: m/z [M]+ calcd. for
C11H9N: 155.0735; found: 155.0733.
2-Heptylpyridine (3i): The typical procedure for the preparation of 3a was followed; 1i (158 mg, 0.47
mmol) was used as the starting material. Flash chromatography on aluminum oxide (hexane-EtOAc 15:1,
7:1) gave 3i in 20% yield (16.8 mg).
IR (neat): 2954, 2925, 1591, 1570, 1473, 1435, 758 cm-1; 1H NMR (400 MHz, CDCl3): δ 0.87 (t, J = 6.9
Hz, 3H), 1.22-1.40 (m, 7H), 1.64-1.76 (m, 3H), 2.77 (t, J = 7.8 Hz, 2H), 7.09 (ddd, J1 = 7.6 Hz, J2 = 4.9
Hz, J3 = 0.6 Hz, 1H), 7.14 (d, J = 7.6 Hz, 1H), 7.58 (td, J1 = 7.6 Hz, J2 = 1.8 Hz, 1H ), 8.52 (broad d, J =
13
4.9 Hz, 1H) ppm; C NMR (100 MHz, CDCl3): δ 14.1, 22.7, 29.2, 29.4, 30.0, 31.8, 38.5, 120.8, 122.7,

HETEROCYCLES, Vol. 78, No. 2, 2009
385
136.2, 149.2, 162.6 ppm; HRMS-EI: m/z [M – Et]+ calcd. for C10H14N: 148.1126; found: 148.1127.
Typical Experimental Procedure for the Amino-Heck Reactions of δ,ε-unsaturated ketone
O-diethylphosphinyloximes for the Generation of Substituted Methylpyridines;
2-Methyl-6-phenylpyridine (8a): Under a nitrogen atmosphere, Pd(PPh3)4 (79.6 mg, 0.069 mmol) and
molecular sieves (MS) 4 Å (34 mg) were added to a solution of 7a (112 mg, 0.34 mmol) in anhyd DMF
(17.2 mL, 0.02 M) at rt. The reaction mixture was stirred at r.t. for 10 min, then Et3N (174 mg, 1.72 mmol)
and (n-Bu)4NBr (555 mg, 1.72 mmol) were successively added. The mixture was quickly warmed to
80 °C and continued to stir for 12 h. The reaction mixture was then cooled to rt, filtrated through a Celite
pad and washed with EtOAc (100 mL). The filtrate was successively washed with H2O (2 × 15 mL) and
saturated NaCl aq soln (15 mL). After concentration, the crude products were subjected to
chromatographic purification on aluminum oxide (hexane-EtOAc 5:1, 3:1) to afford 8a in 63% yield
(36.7 mg).
IR (neat): 2954, 1589, 1448, 1444, 758, 694 cm-1; 1H NMR (400 MHz, CDCl3): δ 2.63 (s, 3H), 7.10 (d, J
= 7.5 Hz, 1H), 7.39 (t, J = 7.2 Hz, 1H), 7.46 (t, J = 7.4 Hz, 2H), 7.52 (d, J = 7.7 Hz, 1H), 7.64 (t, J = 7.7
Hz, 1H), 7.97 (d, J = 7.6 Hz, 2H) ppm; 13C NMR (100 MHz, CDCl3): δ 24.8, 117.6, 121.6, 127.0, 128.6,
128.7, 136.9, 139.8, 157.0, 158.4 ppm; HRMS-EI: m/z [M]+ calcd. for C12H11N: 169.0891; found:
169.0889.
2-(4-Fluorophenyl)-6-methylpyridine (8d):
The typical procedure for the preparation of 8a was followed; 7d (97 mg, 0.28 mmol) was used as the
starting material. Flash chromatography on aluminum oxide (hexane-EtOAc 10:1, 1:1) gave 8d in 42%
yield (22.2 mg).
IR (neat): 3062, 2925, 1601, 1456, 845 cm-1; 1H NMR (400 MHz, CDCl3): δ 2.62 (s, 3H), 7.09 (d, J = 7.6
Hz, 1H), 7.14 (tm, J = 8.8 Hz, 2H), 7.46 (d, J = 7.8 Hz, 1H), 7.63 (t, J = 7.7 Hz, 1H), 7.96 (dm, J = 8.8 Hz,
2H) ppm; 13C NMR (100 MHz, CDCl3): δ 24.7, 115.5 (d, J = 21.5 Hz), 117.3, 121.5, 128.8 (d, J = 8.2 Hz),
135.9 (d, J = 2.9 Hz), 136.9, 155.9, 158.4, 163.4 (d, J = 247.8 Hz) ppm; HRMS-EI: m/z [M]+ calcd. for
C12H10FN: 187.0797; found: 187.0797; Anal. Calcd for C12H10FN: C, 76.99; H, 5.38; N, 7.48. Found: C,
76.63; H, 5.34; N, 7.51.
2-Heptyl-6-methylpyridine (8e): The typical procedure for the preparation of 8a was followed; 7e (95
mg, 0.27 mmol) was used as the starting material. Flash chromatography on aluminum oxide
(hexane-EtOAc 15:1, 10:1) gave 8e in 20% yield (10.5 mg).
IR (neat):2954, 1593, 1456, 789 cm-1; 1H NMR (400 MHz, CDCl3): δ 0.87 (t, J = 6.8 Hz, 3H), 1.23-1.29
(m, 4H), 1.31-1.37 (m, 3H), 1.66-1.72 (m, 3H), 2.52 (s, 3H), 2.73 (broad t, J = 7.9 Hz, 2H), 6.94 (d, J =
7.6 Hz, 1H), 6.95 (d, J = 7.6 Hz, 1H), 7.46 (t, J = 7.6 Hz, 1H) ppm; 13C NMR (100 MHz, CDCl3): δ 14.1,
22.7, 24.6, 29.2, 29.5, 30.3, 31.8, 38.6, 119.4, 120.3, 136.4, 157.6, 162.0 ppm; HRMS-FAB: m/z [M +

386
HETEROCYCLES, Vol. 78, No. 2, 2009
H]+ calcd. for C13H22N: 192.1752; found: 192.1757; Anal. Calcd for C13H21N: C, 81.61; H, 11.06; N, 7.32.
Found: C, 81.64; H, 11.06; N, 7.30.
ACKNOWLEDGEMENTS
We are grateful to the National Science Council of Republic of China (Taiwan) and National Dong-Hwa
University for financial support.
REFERENCES AND NOTES
1. For recent reviews, see: a) M. Kitamura and K. Narasaka, Chem. Record., 2002, 268; b) K. Narasaka,
Pure Appl. Chem., 2003, 75, 19.
2. H. Tsutsui and K. Narasaka, Chem. Lett., 1999, 45.
3. K. Sakoda, J. Mihara, and J. Ichikawa, Chem. Commun., 2005, 4684.
4. M. Kitamura, H. Yanagisawa, and M. Yamane, Heterocycles, 2005, 65, 273.
5. S. Zaman, M. Kitamura, and A. D. Abell, Aust. J. Chem., 2007, 60, 624.
6. T. Hosokawa, N. Shimo, K. Maeda, A. Sonoda, and S. Murahashi, Tetrahedron Lett., 1976, 383; b) H.
Tsutsui and K. Narasaka, Chem. Lett., 2001, 526.
7. S. Chiba, M. Kitamura, O. Saku, and K. Narasaka, Bull. Chem. Soc. Jpn., 2004, 77, 785.
8. S. Zaman, K. Mitsuru, and A. D. Abell, Org. Lett., 2005, 7, 609.
9. J. L. Zhu and Y. H. Chan, Synlett, 2008, 1250.
10. R. A. Sheldon and J. K. Kochi, Metal-Catalyzed Oxidations of Organic Compound, Academic Press:
New York, 1981; b) R. I. Rudakova, E. S. Rudakov, and A. K. Sheinkman, Chem. Heterocycl.
Compd., 1976, 713.
11. M. Kitamura, D. Kudo, and K. Narasaka, ARKIVOC, 2006 (iii), 148.
12. J. A. Joule and K. Mills, Heterocyclic Chemstry, Blackwell Science Ltd, 2000, pp. 71–86.
13. The configurations the E, Z-isomers were assigned based on the chemical shifts of the α-methylene
protons on the 1H NMR spectrum. The cis-relationship between α-methylene protons and the
phosphinyloxime group causes their signal to appear at higher frequency.
14. M. Kitamura, S. Zaman, and K. Narasaka, Synlett, 2001, 974.
15. For the characterization of 3a, see: a) H. Ohta, M. Tokunaga, Y. Obora, T. Iwai, T. Iwasawa, T.

HETEROCYCLES, Vol. 78, No. 2, 2009
387
Fujihara, and Y. Tsuji, Org. Lett., 2007, 9, 89; b) for 3d, see: A. Núñez, A. Sánchez, C. Burgos, and
J. Alvarez-Builla, Tetrahedron, 2004, 60, 6217; c) for 3e, see: G. Lunn, J. Org. Chem., 1992, 57,
6317; d) for 3f, see: M. Parmentier, P. Gros, and Y. Fort, Tetrahedron, 2005, 61, 3261; e) for 3g, see:
O. P. Shkurko, S. G. Baram, and V. P. Mamaev, Chem. Heterocycl. Compd., 1983, 60; f) for 3h, see:
X. L. Cui and R. S. Brown, J. Org. Chem., 2000, 65, 5653; g) for 5a, see: H. Anderson, F. Almqvist,
and R. Olsson, Org. Lett., 2007, 9, 1335; h) for 5b, see: I. Minoru, O. Tsukasa, and T. Masanao,
Chem. Pharm. Bull., 1985, 33, 4755; i) for 5c, see: S. A. Petrich, F. A. Hicks, D. R. Wilkinson, J. G.
Tarrant, S. M. Bruno, M. Vargas, K. N. Hosein, J. T. Gupton, and J. A. Sikorski, Tetrahedron, 1995,
51, 1575.