A study on the ability of 2,5-dihydro- and 2,3,4,5-tetrahydro-1H-phosphole oxides, as well as 7-phosphanorbornene 7-oxide derivatives to undergo UV light-mediated fragmentation-related phosohinylation of methanol

Article abstract of DOI:10.1002/jhet.5570430329

Reactivity of the title P-heterocycles (1-14) in the photoinduced fragmentation-related phosphinylation of methanol was found to be influenced by the extent of ring strain and the UV absorption at 254 nm. The 7-phosphanorbomene oxides (7-14) are universal

Full text of DOI:10.1002/jhet.5570430329

May-Jun 2006
723
A Study on the Ability of 2,5-Dihydro- and 2,3,4,5-Tetrahydro-1H-
phosphole oxides, as well as 7-Phosphanorbornene 7-Oxide
Derivatives to Undergo UV Light-mediated Fragmentation-related
Phosphinylation of Methanol
György Keglevich,1* János Kovács,1 Helga Szelke2 and Tamás Körtvélyesi3
1Department of Organic Chemical Technology, Budapest University of Technology and Economics, 1521
Budapest, Hungary. E-mail: keglevich@oct.mail.bme.hu
2Research Group of the Hungarian Academy of Sciences at the Department of Organic Chemical Technology,
Budapest University of Technology and Economics, 1521 Budapest, Hungary
3Department of Physical Chemistry, University of Szeged, 6701 Szeged, Hungary
Received October 14, 2005
Y
O
Me
....
P
hν
MeOH
hν
MeOH
O
P
Y
H
H
MeCN
MeCN
P
Me
OMe
O
Y
H
Y = Ph, Et, Bn, C6H11
Reactivity of the title P-heterocycles (1-14) in the photoinduced fragmentation-related phosphinylation
of methanol was found to be influenced by the extent of ring strain and the UV absorption at 254 nm. The
7-phosphanorbornene oxides (7-14) are universal precursors due to their ring strain, no matter if they are
UV-active or not at 254 nm. The easily available 2,5-dihydro-1H-phosphole oxides can be applied only in
case of 1-phenyl substitution that enhances the absorption at 254 nm. The ring strain of representative P-
heterocycles (5-8) was evaluated by HF/6-31G* and B3LYP/6-31+G* calculations. UV spectra of
compounds 5-8 were interpreted by ZINDO/S and MNDO-d calculations. The new precursors (11-14)
made possible the extension of the phosphinylations.
J. Heterocyclic Chem., 43, 723 (2006).
Introduction.
In the present paper, the factors governing the reactivity
of P-heterocycles in fragmentation-related phosphinyl-
ation are explored to make possible to chose the
appropriate precursor suitable for the preparation of the
target H-phosphinates (HP(O)(OMe)Y, Y = aryl, alkyl,
cycloalkyl). On the other hand, new precursors of
phosphinylation are introduced.
It has been known for long that the UV light-
mediated reaction of 1-phenyl-2,5-dihydro-1H-phos-
phole oxides with methanol results in the formation of
methyl phenyl-H-phosphinate [1,2]. Tomioka and co-
workers proposed that irradiation of the dihydrophos-
phole oxide provides an oxophosphine (or phosphin-
idene oxide, PhP(O)) that reacts fast with the methanol
present in the mixture [3,4]. It was Quin and Jankowski
who pointed out later on that the real intermediate is
rather a pentacoordinated adduct formed by the
addition of methanol on the P=O group of the starting
P-cycle [5]. A variety of 7-phosphanorbornene deriv-
atives including P-sulfides were also utilized in the
above type fragmentation-related phosphinylation of
simple alcohols [6-10]. A study of 7-phosphan-
orbornene 7-oxides with sterically demanding 2,4,6-
trialkylphenyl substituents on the phosphorus atom
confirmed the addition–elimination reaction path [8,9].
Recently, the reaction was extended to the preparation
of P-aryl, P-alkyl- and P-cycloalkyl H-phosphinates and
it was evaluated in which cases the easily available 2,5-
dihydrophosphole oxides can be used and in which
instances the application of 7-phosphanorbornene 7-
oxides is inevitable [11].
Results and Discussion.
Figure 1 shows the variety of P-heterocycles (1-14)
studied in fragmentation-related phosphinylations. As can
be seen, the precursors include 2,5-dihydro-1H-phosphole
oxides (1-5), a 2,3,4,5-tetrahydrophosphole oxide (6) and
7-phosphanorbornene oxides, such as the dimers of 1H-
phosphole oxides (7, 9, 11 and 13), as well as the cyclo-
adducts of 1H-phosphole oxides with N-phenylmaleimide
(8, 10, 12 and 14). Bridged P-heterocycles 7-14 were
prepared by well-established general methods [8]. A part
of the 7-phosphanorbornene 7-oxides (11-14) are new
whose synthesis is shown in Scheme I. It can be seen that
the 2,5-dihydro-1H-phosphole oxides (3 or 4) were
converted to dibromo-tetrahydrophosphole oxides 15 and
16 giving phosphole oxides 17 and 18 by double
dehydrobromination. Intermediates 17 and 18 were
dimerised spontaneously to afford phosphanorbornenes 11
and 13, or trapped by NFMI to yield species 12 or 14. The

724
G. Keglevich, J. Kovács, H. Szelke and T. Körtvélyesi
Vol 43
dibromo-tetrahydrophosphole oxides (15 and 16)
consisted of diastereomeric pairs (A and B).
dihydrophosphole oxides that are only of medium ring
strain (cf. with 3-cyclopentenone) are useful only in the
case of a strong UV absorption at 254 nm. This is fulfilled
only for P-heterocycles 1 and 5 due to the UV activity of
the phenyl ring attached to the P=O group (Figure
2/curves 1 and 5). The less strained 1-phenyl-tetrahydro
derivative 6 with a high absorbance at 254 nm (Figure
2/curve 6) displays only a lower reactivity with a reaction
time of 14 h. The P-cyclohexyl, P-ethyl and P-benzyl
dihydrophosphole oxides (2, 3 and 4, respectively) can be
regarded to be useless in the reaction under discussion due
to the low absorbance at 254 nm (Figure 2/curves 2-4).
Me
Me
Me
P
P
P
O
Y
O
Ph
O
Ph
5
6
Y
1
Ph
Ch
Et
2
3
4
Bn
Ph
O
Ph
O
P
P
Me
Me
H
H
O
Table 1
H
H
The use of P-heterocycles in fragmentation-related phosphorylations.
N
P
Me
Ph
O
Ph
O
7
8
254 nm
20 °C
MeOH
O
Y
O
Y
O
P
P
Y
P H
P
MeCN
O
Y
OMe
Me
Me
H
O
H
H
Y
1-14
19
Me
H
N
Starting
P-cycle
Y
Time of
irradiation
[h]
Conversion
[%]
Yield
of 15
[%]
Ref.
P
Ph
O
O
Y
Y
9
Ch
10 Ch
12 Et
14 Bn
1
Ph
Ch
Et
1
6
6
1
1
14
1
1
1
1
100
12
0
[a]
95
[11]
[11]
[11]
11 Et
13 Bn
2 [14]
3 [15]
4
5 [16]
6 [16]
7 [14]
8 [14]
9 [14]
10 [14]
11
0
0
95
[b]
~93
~93
~93
~93
77
Bn
Ph
Ph
Ph
Ph
Ch
Ch
Et
Figure 1
100
100
100
100
100
100
100
100
100
100
[11]
[11]
[14]
[14]
[14]
[14]
Table 1 summarises the results of the photoinduced
fragmentation-related phosphinylations. As can be seen,
from among the 2,5-dihydro-1H-phosphole oxides (1-5),
only 1-phenyl derivatives 1 and 5 could be used well in
the preparation of methyl phenyl-H-phosphinate (19, Y =
Ph). Reaction of the 1-cyclohexyl-dihydrophosphole
oxide (2) proceeded reluctantly (12% conversion after a 6
h’s irradiation), while the 1-ethyl derivative (3) failed to
enter into reaction under the conditions applied. The P-
benzyl derivative (4) participated only in side-reactions.
Phenyl-tetrahydrophosphole oxide 6 could be involved in
the reaction discussed, that was, however, much slower
(13% conversion after a 2 h’s irradiation), as compared to
that of dihydro derivative 1. The results can be interpreted
by considering the measure of the ring strain and the
extent of the UV absorption of the individual P-
heterocycles at 254 nm (Table 2). The 2,5-
1
1
1
1
12
13
14
Et
Bn
Bn
75
80
83
In all experiments 0.10 g of the P-cycle was irradiated in the mixture of
40 ml of acetonitrile and 4 ml of methanol; [a] Only side-reactions could
be observed; [b] No yield was provided.
Regarding the 7-phosphanorbornene oxides, all precursors
(7-14) could be used well in the fragmentation-related
phosphinylation of methanol, practically no matter what
P-substituent the bridged P-heterocycle bears. The
enhanced reactivity is obviously the consequence of the
high ring strain of the 7-phosphanorbornene framework
Scheme I
R
O
P
Me
Br
Me
Br
Me
17/18
or
NEt3
Br2
Me
H
CHCl3
PhMe
NFMI
P
P
P
H
O
R
O
R
O
R
R = Et
3
4
15
16
17
18
11 or 12
13 or 14
R = Bn

May-Jun 2006
UV light-mediated fragmentation-related phosphinylation
725
Table 2
The effect of parameters on the fragmentation properties of P-heterocycles 1-11.
P-cycle
Ring strain
moderate
UV absorption at
254 nm
Overall ability of the P-cycle to undergo
fragmentation-related phosphorylation
medium
high
low
strong
+
none
moderate
high
+
1
2
3
4
5
6
7
8
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
9
+
+
+
10
11
12
13
14
+
+
+
[8]. The P-phenyl derivatives (7 and 8) were found to be
UV active at 254 nm (Figure 3/curve 7, Figure 4/curve 8).
It was observed that the phenyl group attached to the
imide moiety (as in 10, 12 and 14), may also enhance the
UV absorption at 254 nm (Figure 4/curves 10, 12 and 14)
and hence the reaction under discussion. Worth noting is
that the cyclohexyl-, ethyl- and benzyl-1H-phosphole
oxide dimers (9, 11 and 13, respectively) can be used in
phosphinylation, although they are practically UV
inactive at 254 nm (Figure 3/9, 11 and 13).
1,6
1,4
1,2
1,0
0,8
0,6
0,4
0,2
0,0
1
2
3
4
5
6
1,6
190
210
230
250
270
290
310
330
350
8
Wavelegth (nm)
1,4
1,2
1,0
0,8
0,6
0,4
0,2
0,0
10
12
14
Figure 2. UV spectra of dihydro- and tetrahydro-1H-phosphole oxides
obtained in acetonitrile.
1,6
7
1,4
1,2
1,0
0,8
0,6
0,4
0,2
0,0
9
11
13
190
210
230
250
270
290
310
330
350
Wavelength (nm)
Figure 4. UV spectra of 7-Phosphanorbornene derivatives (II) obtained
in acetonitrile.
The stereostructure and ring strain of P-heterocycles 5-8
was evaluated by HF/6-31G* and B3LYP/6-31+G*
calculations. Stereostructures of 2,5-dihydro-1H-phos-
phole oxide 5 and 7-phosphanorbornene oxide 8 are
shown in Figures 5 and 6, respectively. In the former case,
the more stable diastereomer is represented.
190
210
230
250
270
290
310
330
350
Wavelength (nm)
Figure 3. UV spectra of 7-phosphanorbornene derivatives (I) obtained
in acetonitrile.

726
G. Keglevich, J. Kovács, H. Szelke and T. Körtvélyesi
Vol 43
the 2,3,4,5-tetrahydro-1H-phosphole ring is less strained
than the 2,5-dihydro heterocycle. Comparing the results of
HF/6-31G* and B3LYP/6-31+G* calculations, position of
the plane of the P-phenyl ring was found to be slightly
different in phosphanorbornenes 7 and 8.
The possible mechanism for the fragmentation-related
phosphinylations governed mainly by the ring strain
discussed above is that on irradiation the P-heterocycle (1-
14) gets to an excited state to afford species 20 that reacts
fast with the nucleophile. The pentacoordinated
intermediate (21) so formed is then stabilized to yield H-
phosphinate 19 (Scheme II).
Figure 5. Perspective view of the more stable diastereomer of 2,5-
dihydro-1H-phosphole oxide 5 obtained by HF/6-31G* calculation
[P(1)-C(2): 1.840, C(2)-C(3): 1.510, C(3)-C(4): 1.322, C(4)-C(5): 1.519,
C(5)-P(1): 1.852 (Å); C(5)-P(1)-C(1'): 107.17, C(2)-P(1)-O: 116.54, O-
P(1)-C(1'): 112.19, O-P(1)-C(2)-CH3: 7.22 (°)].
The UV spectra of P-heterocycles 5-8 that are of
diagnostic value from the point of view photolysis were
subjected to a detailed study. The results of ZINDO/S
calculations for compounds 5-8 support the experimental
spectra. The S0ꢂS1 transition of molecules 5-7 was
found at 262.6–262.7 nm, with a similar oscillator
strength (0.0005–0.0006) that may be due to the O=P–Ph
chromophore. For compound 8, maximum of the S0ꢂS1
transition is located at 320.7 nm with an oscillator
strength of 0.0049 that is probably due to the presence of
the (–CO)2N–Ph chromophore. In its tendency, similar
results were obtained by MNDO-d C.I.=6 calculations
(with 400 microstates) for the ground- and the first
excited singlet electronic state. With the exception of
compound 8, there is no significant absorbance above 320
nm. The S0ꢂS1 excitation energy for compounds 5-7
was found to be 3.928 eV, 3.930 eV and 3.928 eV,
respectively, supporting similar spectra in the lower wave
number range of the UV spectra. The torsion angle of the
phenyl groups were changed in the different excited
states. It is probable that in all cases (5-8), the ꢀ*ꢃ ꢀ
transition of the phenyl group is involved that is affected
by the presence of the P=O or the N(CO–)2 moiety. For
compound 8, the band at 320 nm must be due to the
ꢀ*ꢃ n transition of the C=O group. The S0ꢂS2, and the
S0ꢂS3 transitions were found at lower wave lengths with
greater oscillator strengths.
Figure 6. Perspective view of 7-phosphanorbornene 8 obtained by HF/6-
31G* calculation [C(1)-C(2): 1.559, C(2)-C(3): 1.545, C(3)-C(4): 1.569,
C(4)-C(5): 1.516, C(5)-C(6): 1.327, C(6)-C(1): 1.512, C(1)-P(7): 1.850,
C(4)-P(7): 1.861 (Å); C(1)-P(7)-O: 116.42, C(4)-P(7)-O: 115.28, C(1)-
P(7)-C(1'): 114.97, C(4)-P(7)-C(1'): 114.55, O-P(7)-C(1'): 110.51 (°).
The bond angles in the hetero rings of 5-8 are listed in
Table 3. The sum of the absolute values of the deviations
of the actual angles from the ideal ones (109.5° or 120°)
may be regarded to be an indicator of the ring strain. This
number (ꢀꢁ1) was found to be 28.94, 27.49 96.56 and
106.94 for P-cycles 5, 6, 7 and 8, respectively, confirming
that the 7-phosphanorbornene precursors are indeed of
significant ring strain as compared to the 5-ring
heterocycles (5 and 6). The C–P–C angle of the bridging
unit in 7 and 8 was found to be 82.79 and 82.77°,
respectively, that is in accord with the literature data of
around 82° obtained for analogous derivatives [12,13].
The strain of phosphole rings 5 and 6 can be compared by
the sum of the absolute values of the deviations of the
actual angles from 105° that is the standard angle in 5-
membered rings. This number (ꢀꢁ2) was found to be
37.50 and 17.19 for 5 and 6, respectively, suggesting that
It can be concluded that the reactivity of the five-ring P-
heterocycles and 7-phosphanorbornenes in fragmentation-
related phosphinylations is governed by the UV activity
and the ring strain of the precursor. An intensive UV
absorption at 254 nm or a highly strained ring is the sine
qua non of the reaction under discussion. This means that
simple 2,5-dihydro-1H-phosphole oxides can be used only
Scheme II
fast
slow
fast
OH
O
P
*
hν
fragmentation
MeOH
.
.
Y
P
Y
H
P
OH
OMe
21
P
P
Y
OMe
OMe
O
Y
O
Y
1-14
20
19'
19

May-Jun 2006
UV light-mediated fragmentation-related phosphinylation
727
Table 3
Bond angles (ꢁ) in typical precursors obtained by HF/6-31G* calculation
(ꢂꢃ1 is the sum of the absolute values of the differences from the ideal angle (109.5° or 120°),
while ꢂꢃ2 is the sum of the absolute values of the differences from the standard angle (105°) for 5-ring compounds).
O
Ph
P
7
4
3
4
3
2
5
2
5
1
1
Me
Me
Me
P
P
4
5
6
3
1
O
Ph
O
Ph
2
5
6
7
8
P1–C2–C3
C2–C3–C4
C3–C4–C5
C4–C5–P1
C5–P1–C2
103.40
118.38
117.65
104.21
95.92
103.41
107.29
107.48
106.33
95.50
C1–C2–C3
C2–C3–C4
C3–C4–C5
C4–C5–C6
C5–C6–C1
C6–C1–C2
C3–C4–P7
C2–C1–P7
C1–P7–C4
C5–C4–P7
C6–C1–P7
106.55
107.24
106.68
102.19
111.50
107.86
96.47
96.94
82.79
99.82
100.52
106.25
106.40
107.05
112.11
111.42
108.39
97.12
97.17
82.77
99.80
100.46
ꢂꢃ1 = 106.94
ꢂꢃ1 = 28.94
ꢂꢃ2 = 37.50
ꢂꢃ1 = 27.49
ꢂꢃ2 = 17.19
ꢂꢃ1 = 96.56
General Procedure for the Preparation of Dibromo-tetrahydro-
phosphole oxides 16 and 15.
in case of phenyl- (and obviously aryl) substitution. If the
target molecule is an alkyl- or cycloalkyl H-phosphinate,
the corresponding 7-phosphanorbornene should be the
starting material of choice. Results of the theoretical
calculations are in accord with the experimental data.
Bromine (1.5 ml, 28.6 mmol) of in 15 ml of chloroform was
added to the 50 ml chloroform solution of 26.0 mmol of
dihydrophosphole oxide 4 and 3 at 0 °C. After 0.5 h, the mixture
was allowed to warm up to room temperature and the stirring
was continued for 4 h. Volatile components were then removed.
Purification of the crude product by column chromatography
(silica gel, 3% methanol in chloroform) afforded the
corresponding dibromo-tetrahydrophosphole oxides (16 and 17,
respectively) as a mixture of two isomers (A and B).
EXPERIMENTAL
1
The 31P, 13C and H NMR spectra were obtained on a Bruker
DRX-500 spectrometer operating at 202.4, 125.7 and 500 MHz,
respectively. Chemical shifts are downfield relative to 85%
H3PO4 or TMS. The couplings are given in Hz. Mass spectro-
metry was performed on a ZAB-2SEQ instrument. As the new
compounds were dense oils, their purity (that was ca. 98%) was
1-Benzyl-3,4-dibromo-3-methyl-2,3,4,5-tetrahydro-1H-phos-
phole 1-oxide (16).
16: Yield, 68%, dense oil as a 66:34 mixture of diastereomers (A and
B). HR-MS, (M+H)+found = 364.9279, C12H16POBr2 requires 364.9305.
16A (66%): 31P NMR (CDCl3) ꢀ 60.1; 13C NMR (CDCl3) ꢀ 31.9
(J = 8.3, C3–Me), 37.1 (J = 60.7, CH2–Ph),a 39.6 (J = 60.9, C2),a
41.3 (J = 61.0, C5),a 57.1 (J = 4.6, C4), 68.0 (J = 7.7, C3), 127.0 (J
= 3.3, C4'), 128.6 (J = 2.8, C3'),b 129.2 (J = 5.6, C2'),b 130.7 (J =
8.5, C1'); a,b may be reversed; 1H NMR (CDCl3) ꢀ 2.16 (s, 3H, C3–
CH3), 4.61–4.68 (m, 1H, C4–H), 7.24–7.38 (m, Ar–H).
1
established by H NMR and mass spectroscopy, as well as thin
layer chomatography. UV spectra were obtained using ca. 10–5
mol·l–1 acetonitrile solutions.
The 3-methyl-2,5-dihydro- and 2,3,4,5-tetrahydro-1H-phos-
phole oxides (2 [14], 3 [15], 5 [16] and 6 [16]), as well as the 7-
phosphanorbornene 7-oxides (7-10) [14] were prepared as
described earlier. P-cycles 5 and 6 were used as ca. 1:1 mixtures
of diasteromers.
1-Benzyl-2,5-dihydro-1H-phosphole oxide 4 was synthesized
by the method applied for the preparation of 2 [14]. The 13C
NMR spectral data of 4 were identical with those described in
the literature [17]. 13C NMR (CDCl3) ꢀ 20.4 (J = 10.5, C3–Me),
31.6 (J = 63.8, C5),a 34.6 (J = 66.8, C2),a 37.5 (J = 57.4, CH2Ph),a
120.7 (J = 8.3, C4), 127.1 (J = 3.0, C4'), 128.9 (J = 2.6, C3'),b
129.6 (J = 5.1, C2'),b 132.2 (J = 7.8, C1'), 137.0 (J = 13.0, C3);
MS, (M+H)+ = 207; a,bmay be reversed; ꢀ (lit [17]) 20.2 (J =
11.3, C3–Me), 31.4 (J = 64.1, C5), 34.4 (J = 67.1, C2), 37.3 (J =
56.9, CH2Ph), 120.6 (J = 8.0, C4), 126.9 (C4'), 128.8 (C3'), 129.4
(J = 4.0, C2'), 132.0 (J = 8.8, C1'), 136.7 (J = 12.2, C3).
16B (34%): 31P NMR (CDCl3) ꢀ 61.8; 13C NMR
(CDCl3) ꢀ 32.5 (J = 10.4, C3–Me), 37.7 (J = 60.9, CH2–
Ph),a 40.6 (J = 61.0, C2),a 40.8 (J = 64.1, C5),a 58.3 (J =
5.0, C4), 67.4 (J = 5.7, C3), 126.9 (J = 3.2, C4'), 128.5 (J =
3.7, C3'),b 129.3 (J = 5.5, C2'),b 130.8 (J = 8.1, C1'); a,bmay
1
be reversed; H NMR (CDCl3) ꢀ 2.08 (s, 3H, C3–CH3),
4.89–5.03 (m, 1H, C4–H), 7.24–7.38 (m, Ar–H).
3,4-Dibromo-1-ethyl-3-methyl-2,3,4,5-tetrahydro-1H-
phosphole 1-oxide (15).
15: Yield, 80%, dense oil as a 56:44 mixture of
diastereomers (A and B). MS, (M+H)+
= 302.9132,
Synthesis of the new 7-phosphanorbornene 7-oxides (11-14)
and that of their intermediates (15 and 16) is described below.
found
C7H14OPBr2 requires 302.9149.

728
G. Keglevich, J. Kovács, H. Szelke and T. Körtvélyesi
Vol 43
15A (56%): 31P NMR (CDCl3) ꢀ 71.3; 13C NMR (CDCl3) ꢀ
6.2 (J = 5.9, CH2CH3), 25.8 (J = 66.6, CH2CH3), 32.8 (J = 9.4,
C3–Me), 38.0 (J = 61.4, C5),* 42.2 (J = 61.9, C2),* 57.5 (J = 2.8,
C4), 68.7 (J = 7.3, C3); *may be reversed.
15B (44%): 31P NMR (CDCl3) ꢀ 72.3; 13C NMR (CDCl3) ꢀ
6.3 (J = 4.8, CH2CH3), 26.8 (J = 66.6, CH2CH3), 33.2 (J = 10.8,
C3–Me), 38.9 (J = 61.2, C5),* 41.2 (J = 64.6, C2),* 58.5 (J = 4.5,
C4), 67.7 (J = 4.9, C3); *may be reversed.
3,10-Dibenzyl-5,8-dimethyl-3,10-diphosphatricyclo[5.2.1.02,6]-
deca-4,8-diene 3,10-dioxide (13).
13: Yield, 65%, light yellow oil. 31P NMR (CDCl3) ꢀ 63.0
(P1), 88.6 (P8), J = 34.3; 13C NMR (CDCl3) ꢀ 19.0 (J 1 = 16.5,
C3–Me), 19.3 (J 2 = 3.9, C5–Me), 28.6 (J 2 = 56.2, P8–CH2Ph),
36.7 (J 1 = 74.5, J 2 = 11.0, C7a), 39.8 (J 1 = 63.8, P1–CH2Ph),
41.4 (J 1 = 2.5, J 2 = 57.6, C7), 47.7 (J 2 = 61.8, C4), 52.0 (J 1 = J 2
= 11.8, C3a), 123.3 (J 1 = 95.5, J 2 = 2.3, C2), 124.3 (J 1 = 5.0, J 2
= 8.9, C6), 126.5 (J 1 = 3.0, C4'),a 126.8 (J 2 = 2.6, C4"),a 128.4 (J 1
= 2.5, C3'),b 128.6 (J 2 = 2.0, C3"),b 128.9 (J 1 = 5.4, C2'),c 129.5
(J 2 = 5.2, C2"),c 131.6 (J 1 = 7.2, C1'),d 132.3 (J 2 = 8.6, C1"),d
136.2 (J 2 = 11.8, C5), 159.3 (J 1 = 22.4, J 2 = 8.7, C3); a-dmay be
reversed; J 1: coupled by P1, J 2: coupled by P8; 1H NMR
(CDCl3) ꢀ 1.70 (s, 3H, CH3), 1.79 (s, 3H, CH3), 2.77-2.81 (m,
1H, CH), 2.97–3.04 (m, 1H, CH), 3.06–3.09 (m, 1H, CH), 3.17–
3.27 (m, 4H, 2ꢀPCH2), 3.47–3.53 (m, 1H, CH), 5.71–5.78 (m,
1H, =CH), 6.10–6.16 (m, 1H, =CH), 7.14–7.33 (m, 10H, ArH);
HR-MS, (M+H)+found = 409.1456, C24H27P2O2 requires 409.1486.
General Procedure for the Preparation of 7-Phosphanorbornene
7-oxides 14, 12, 13 and 11.
A) Trapping of 3-Methyl-1H-phosphole oxides with NPMI.
Triethylamine (2.7 ml, 19.6 mmol) was added to the 50 ml
toluene solution of 8.9 mmol of dibromo-tetrahydrophosphole
oxide (16 and 15) and 1.9 g (11.1 mmol) of NFMI. After 6 days
of stirring at room temperature, the mixture was stirred at the
boiling point for 4 h. After filtration, the filtrate was evaporated
and the crude product so obtained was purified by column
chromatography (silica gel, 3% methanol in chloroform) to
afford the corresponding trapped products (14 and 12,
respectively).
3,10-Diethyl-5,8-dimethyl-3,10-diphosphatricyclo[5.2.1.02,6]-
deca-4,8-diene 3,10-dioxide (11).
11: Yield, 50%, colourless oil. 31P NMR (CDCl3) ꢀ 67.6 (P1),
93.0 (P8), J = 33.5; 13C NMR (CDCl3) ꢀ 7.6 (P1–CH2CH3), 7.4
(P8–CH2CH3), 13.1 (J 2 = 63.1, P8–CH2), 19.1 (C5–Me), 19.0 (J 1
10-Benzyl-8-methyl-4-phenyl-4-aza-10-phosphatricyclo[5.2.1.02,6]-
dec-8-ene-3,5-dione 10-oxide 14.
14: Yield, 71%, dense oil; 31P NMR (CDCl3) ꢀ 92.4; 13C NMR
(CDCl3) ꢀ 19.3 (J = 3.5, C2–Me), 29.0 (J = 57.9, CH2–Ph), 43.5 (J
= 11.6, C5),a 43.6 (J = 59.6, C4),b 44.7 (J = 12.2, C6),a 46.6 (J =
59.5, C1),b 122.2 (J = 9.1, C3), 126.5 (C2"),c 127.5 (J = 3.0, C4'),
129.0 (C4"), 129.1 (J = 3.2, C2'), 129.2 (C3"),c 129.3 (C3'),c 131.7
(C1"), 132.0 (J = 8.6, C1'), 141.2 (J = 11.1, C2), 175.1 (J = 13.0,
C10),d 175.5 (J = 13.4, C8)d; a-dmay be reversed; 1H NMR (CDCl3)
ꢀ 1.89 (s, 3H, C2–CH3), 3.22–3.27 (m, 1H, CH), 3.32–3.38 (m,
2H, CH2–Ph), 3.40-3.45 (m, 1H, CH), 3.96–4.01 (m, 2H, 2ꢀCH),
5.93–5.98 (m, 1H, =CH), 7.24–7.54 (m, 10H, ArH); HR-MS,
(M+H)+found = 378.1228, C22H21PO3N requires 378.1259.
= 4.8, C3–Me), 24.4 (J 2 = 70.9, P8–CH2), 36.6 (J 1 = 73.5, J 2
=
9.2, C7a), 40.3 (J 2 = 57.7, C7), 47.3 (J 2 = 61.1, C4), 51.9 (J 1 = J 2
= 11.5, C3a), 122.9 (J 1 = 94.7, C2), 123.8 (broad signal, C6),
135.9 (J 1 = 11.6, C5), 158.9 (J 1 = 21.8, J 2 = 8.7, C3); J 1:
coupled by P1, J 2: coupled by P8; 1H NMR (CDCl3) ꢀ 0.96–1.19
(m, 6H, 2ꢀCH2–CH3), 1.67 (s, 3H, skeletal–CH3), 1.84 (s, 3H,
skeletal–CH3), 2.88–2.92 (m, 1H, CH), 2.96–3.03 (m, 1H, CH),
3.05–3.11 (m, 1H, CH), 3.70–3.79 (m, CH), 5.65–5.78 (m, 1H,
=CH), 6.00–6.08 (m, 1H, =CH); HR–MS, (M+H)+
285.1148, C14H23O2P2 requires 285.1173.
=
found
10-Ethyl-8-methyl-4-phenyl-4-aza-10-phosphatricyclo[5.2.1.02,6]-
dec-8-ene-3,5-dione 10-oxide (12).
General Procedure for the Preparation of H-Phosphinates 19
using 7-Phosphanorbornene 7-oxides 11-14.
A solution of 0.52 mmol of phosphanorbornene oxide 11-14
in 45 ml of acetonitrile and 4.0 ml of methanol was irradiated by
a mercury lamp (125 W) in a quartz reactor for 1 hour. Solvent
was evaporated and the crude product so obtained purified by
flash column chromatography (silica gel, 3% methanol in
chloroform) to afford the corresponding H-phosphinate (19) as
shown in Table 1.
12: Yield, 62%, light yellow crystals, mp.: 176–177 °C (10%
dichloromethane in hexane); 31P NMR (CDCl3) ꢀ 96.7; 13C NMR
(CDCl3) ꢀ 7.7 (J = 5.1, CH2CH3), 13.6 (J = 63.7, CH2CH3), 19.3 (J
= 3.3, C2–CH3), 43.2 (J = 50.1, C4),a 43.7 (J = 1.3, C5),b 44.8 (J =
12.3, C ),b 46.6 (J = 58.6, C ),a 121.8 (J = 9.0, C ), 126.5 (C ),c
'
2
6
1
3
128.9 (C4'),c 129.2 (C3'),c 131.6 (C1'), 141.1 (J = 10.9, C2), 175.3 (J
1
= 12.7, C10),d 175.6 (J = 13.1, C8)d; a-dmay be reversed; H NMR
(CDCl3) ꢀ 1.27 (dt, 3JPH = 18.0, 3JHH = 7.8, 3H, CH2CH3), 1.92 (s,
3H, C2–CH3), 1.88–1.96 (m, 2H, CH2CH3), 3.30–3.40 (m, 1H,
CH), 3.45–3.55 (m, 1H, CH), 3.96–4.05 (m, 2H, 2ꢀCH), 5.86–
5.95 (m, 1H, =CH), 7.11–7.14 and 7.37–7.48 (m, 5H, ArH); MS,
(M+H)+found = 316.1115, C17H19O3PN requires 316.1103.
The use of precursors 13 or 14 led to product 19, Y = Bn in a
yield of 80 and 83%, respectively. 31P NMR (CDCl3) ꢀ 39.2, JPH
= 546.5; 13C NMR (CDCl3) ꢀ 37.0 (J = 88.4, CH2Ph), 53.2 (J =
7.1, CH3O), 127.5 (J = 3.6, C4'), 129.2 (J = 3.3, C3'),* 129.9 (J =
6.3, C2')*; *may be reversed; MS, (M+H)+
C8H12O2P requires 171.0575.
= 171.0569,
found
B) Dimerisation of 3-Methyl-1H-phosphole oxides.
Applying phosphanorbornenes 11 and 12, H-phosphinate 19,
Y = Et was obtained in 77 and 75% yield, respectively. 31P NMR
(CDCl3) ꢀ 43.8, ꢀ (lit [11]) 44.3.
Triethylamine (3.4 ml, 24.4 mmol) was added to the 35 ml of
toluene solution of 11.1 mmol of dibromo-tetrahydrophosphole
oxide 16 and 15. The mixture was stirred at 110 °C for a day and
at 26 °C for 2 other days. After filtration, the filtrate was
evaporated. Purification of the crude product by column
chromatography (silica gel, 2% methanol in chloroform) led to a
light yellow oil, containing the corresponding dimers (13 and 11,
respectively).
The other P-heterocycles (1, 5-10) were used similarly.
Theoretical Calculations.
Structures 5-8 were calculated by ab initio and DFT quantum
chemical methods with a basis set of HF/6-31G* and B3LYP/6-
31+G*, respectively, implemented in Gaussian '03 [18]. The

May-Jun 2006
UV light-mediated fragmentation-related phosphinylation
729
[8] L. D. Quin, Rev. Heteroatom Chem., 3, 39 (1990).
[9] Gy. Keglevich, K. Ludányi and L. D. Quin, Heteroatom
Chem., 8, 135 (1997).
[10] Gy. Keglevich, M. Trecska, Z. Nagy and L. Tꢀke,
Heteroatom Chem., 12, 6 (2001).
[11] H. Szelke, J. Kovács and Gy. Keglevich, Synth. Commun,
2005, 35, 2927.
[12] L. D. Quin, K. C. Caster, J. C. Kisalus and K. Mesch, J. Am.
Chem. Soc., 106, 7021 (1984).
[13] Gy. Keglevich, L. Tꢀke, Zs. Böcskei and V. Harmat,
Heteroatom Chem., 8, 527 (1997).
[14] J. Kovács, N. Balázsdi Szabó, Z. Nagy, K. Ludányi and Gy.
Keglevich, Heteroatom Chem., 16, 320 (2005).
[15] Gy. Keglevich, L. Tꢀke, Cs. Lovász, K. Újszászy and G.
Szalontai Heteroatom Chem., 5, 395 (1994).
[16] J. Kovács, Gy. Keglevich, A. Kerényi, T. Imre, K. Ludányi
and L. Tꢀke Heterocyclic Commun., 10, 283 (2004).
force matrix of the fully optimized structures were found to be
positive definite supporting that a minimum was found.
In order to support the change in the spectra at molecules 5-8,
the ground and the first excited singlet electronic states were
calculated by MNDO-d with C.I.=6 method implemented in CS
Chem3d Ultra Version 9.0.1 [19]. The structures were fully
optimized in gas phase with a tolerance in gradient norm of 0.1.
The energies of the S0ꢀS1 transition for molecules 5-8 were
compared with the experimental spectra. The energies of the
excited states of 5-8 were also calculated by ZINDO/S
implemented in Arguslab [20]. 10 Electronic states were
considered in the C.I. method. The dielectric constant was 37.4
(acetonitrile) in the SCRF. The geometry, we applied was the
HF/6-31G* ab initio optimized structure. The results are only
informative, because the phosphorus d-orbital is not
parameterized.
[17] G. W. Buchanan and V. L. Webb, Org. Magn. Reson., 21,
436 (1983).
Acknowledgements.
[18] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K.
N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V.
Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G.
A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S.
Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A.
D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A.
G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C.
Gonzalez and J. A. Pople, Gaussian 03, Revision B.01, Gaussian,
Inc., Pittsburgh PA, 2003.
The above project was supported by the Hungarian Scientific
and Research Fund (OTKA T042479 (G.K. and T.K.) and
T032190 (T.K.)). T.K. is grateful for the support of computation
in the Hungarian Supercomputer Centre (www.niif.hu). The
authors thank János Deme for his assistance in the synthetic work.
REFERENCES
[1] H. Tomioka, Y. Hirano and Y. Izawa, Tetrahedron Lett.,
1865, (1974).
[2] H. Tomioka, S. Nakamura, T. Ohi and Y. Izawa, Bull. Chem.
Soc. Japan, 49, 3707 (1976).
[3] H. Tomioka and Y. Izawa, J. Org. Chem., 42, 582 (1977).
[4] H. Tomioka, S. Takata, Y. Kato and Y. Izawa, J. Chem. Soc.,
Perkin Trans II, 1017 (1980).
[5] L. D. Quin, S. Jankowski, J. Rudzinski, A. G. Sommese and
X.-P. Wu, J. Org. Chem., 58, 6212 (1993).
[19] Chem3d Ultra Version 9.0.1, Cambridgesoft, Cambridge,
MA, USA, 2005.
[20] ArgusLab 4.0.1, Mark A. Thompson, Planaria Software LLC,
Seattle, WA.; http://www.arguslab.com.
[6] H. Tomioka, Y. Hirano and Y. Izawa, Tetrahedron Lett.,
4477 (1974).
[7] S. Holand and F. Mathey, J. Org. Chem., 46, 4386 (1981).