1-(2,4,6-tri-tert-butylphenyl)-3-methylphosphole: A phosphole with a significantly flattened phosphorus pyramid having pronounced characteristics of aromaticity

Article abstract of DOI:10.1021/ja970463d

Single-crystal X-ray analysis of 1-(2,4,6-tri-tert-butylphenyl)-3- methylphosphole, synthesized to test the effect of the strongly sterically demanding P-substituent on the geometry of the molecule, revealed that the phosphorus pyramid was drastically fla

Full text of DOI:10.1021/ja970463d

J. Am. Chem. Soc. 1997, 119, 5095-5099
5095
1-(2,4,6-Tri-tert-butylphenyl)-3-methylphosphole: A Phosphole
with a Significantly Flattened Phosphorus Pyramid Having
Pronounced Characteristics of Aromaticity
Gyo1rgy Keglevich,*,† Zsolt Bo1cskei,‡ Gyo1rgy Miklo´s Keseru1,‡
Ka´lma´n UÄ jsza´szy,§ and Louis D. Quin*,|
Contribution from the Department of Organic Chemical Technology, Technical UniVersity of
Budapest, 1521 Budapest, Hungary, Chinoin Pharmaceuticals, 1045 Budapest, Hungary,
Department of General and Inorganic Chemistry, Lora´nd Eo¨tVo¨s UniVersity,
1518 Budapest, Hungary, and Department of Chemistry, UniVersity of Massachusetts,
Amherst, Massachusetts 01003
ReceiVed February 11, 1997X
Abstract: Single-crystal X-ray analysis of 1-(2,4,6-tri-tert-butylphenyl)-3-methylphosphole, synthesized to test the
effect of the strongly sterically demanding P-substituent on the geometry of the molecule, revealed that the phosphorus
pyramid was drastically flattened; the normal out of plane angle of 65° (formed between the P-substituent and the
ring C2-P-C5 plane) was reduced to 45.9°. Consistent with strong electron delocalization, the C3-C4 bond length
was dramatically shortened relative to that for other phospholes, and the Bird index of aromaticity was 56.5, almost
the equivalent of that in pyrrole (59). The phosphole ring, normally resisitant to electrophilic substitutions, underwent
reaction with acetyl chloride-aluminum chloride, consistent with considerable cyclic electron delocalization.
Introduction
P-C bonds. Another approach to planarization of phosphorus
has been taken by Schmidpeter that involves placement of
electron-withdrawing triphenylphosphonio substituents at the
two positions adjacent to phosphorus; this was effective in the
isophosphindole system6a and the 1,2-diphosphole system.6b
The first compound (1) prepared in this study contained the
2,4-di-tert-butyl-6-methylphenyl substituent,7 and it was readily
apparent from the photoelectron spectrum that increased electron
delocalization due to the partially planarized structure was
present; the ionization energy of the lone pair of 1 was the lowest
ever recorded for a phosphole and differed substantially from
that of the tetrahydro derivative.8 We use the angle of deflection
The family of phospholes constitutes an important place in
contemporary phosphorus chemistry, and detailed reviews of
this area1 and of aromaticity in general2 have recently been
published. As members of the group of heterocyclopentadienes,
phospholes might be expected to exhibit the cyclic electron
delocalization so characteristic of this structural type. In fact,
for the phospholes reported thus far, there is little delocalization,
and this is attributed to the pyramidal shape retained at
phosphorus that interferes with efficient orbital interaction.1a
However, theoretical calculations3 have suggested that if the
phosphorus atom could adopt planar geometry, then strong
delocalization, on the order of that of thiophene,1c could prevail.
We have embarked on an experimental program to test this idea.
Our premise is that placement of very large, space-demanding
substituents on phosphorus might induce a degree of flattening
of the phosphorus pyramid that would permit significant electron
delocalization which would be reflected in the physical and
chemical properties of the system. A recent report4 indeed
makes a connection between pyramidal character at P and the
degree of delocalization, but in the opposite sense: placing two
acetylenic substituents at the R-positions caused an increase in
pyramidal character relative to 1-benzylphosphole5 as seen in
an X-ray analysis, as well as lengthening of the C3-C4 and
of the P-substituent from the plane of C2-P-C5 in the
phosphole ring as a convenient descriptor of the flattening; in
an uncrowded phosphole such as 1-benzyl,5 this angle was
determined by X-ray diffraction analysis to be 66.9°. In
phosphole 1, calculations give an angle of 57.4° to account for
the lone pair ionization energy.8 The second phosphole (2)
prepared had the 2,4,6-triisopropylphenyl substituent, and X-ray
diffraction analysis suggested that partial flattening of the
phosphorus pyramid had occurred with an out of plane angle
of 58°.9
† Technical University of Budapest.
‡ Chinon Pharmaceuticals.
§ Lora´nd Eo¨tvo¨s University.
| University of Massachusetts.
X Abstract published in AdVance ACS Abstracts, May 15, 1997.
(1) (a) Mathey, F. Chem. ReV. 1988, 88, 429. (b) Hughes, A. N. In
Handbook of Organophosphorus Chemistry; Engel, R., Ed.; Marcel Dek-
ker: New York, 1992; Chapter 10. (c) Quin, L. D. In ComprehensiVe
Heterocyclic Chemistry; Katritzky, A., Rees, C. W., Scriven, E. F. V., Eds.;
Pergamon: Oxford, England, 1996; Vol. 2, pp 757-856.
(2) Minkin, V. I.; Glukhovtsev, M. N.; Simkin, B. Y. Aromaticity and
Antiaromaticity; Wiley: New York, 1994.
We have now succeeded in preparing phosphole 3 which has
the well-known highly space-demanding 2,4,6-tri-tert-butylphe-
(6) (a) Schmidpeter, A.; Thiele, M. Angew. Chem., Int. Ed. Engl. 1991,
30, 308. (b) Jochem, G.; No¨th, H.; Schmidpeter, A. Chem. Ber. 1996, 129,
1083.
(7) Quin, L. D.; Keglevich, Gy.; Ionkin, A. S.; Kalgutkar, R.; Szalontai,
G. J. Org. Chem. 1996, 61, 7801.
(3) Chesnut, D. B.; Quin, L. D. J. Am. Chem. Soc. 1994, 116, 9638.
(4) Holand, S.; Gandolfo, F.; Ricard, L.; Mathey, F. Bull. Soc. Chim.
Fr. 1996, 133, 33.
(8) Nyula´szi, L.; Keglevich, Gy.; Quin, L. D. J. Org. Chem. 1996, 61,
7808.
(9) Keglevich, Gy.; Quin, L. D.; Bo¨cskei, Zs.; Keseru¨, Gy. M.; Kalgutkar,
R.; Lahti, P. M. J. Organomet. Chem. In press.
(5) Coggon, P.; McPhail, A. T. J. Chem. Soc., Dalton Trans. 1973, 1888.
S0002-7863(97)00463-0 CCC: $14.00 © 1997 American Chemical Society

5096 J. Am. Chem. Soc., Vol. 119, No. 22, 1997
KegleVich et al.
Table 1. 13C NMR Data for Phosphole 3 (and the Phenyl
Scheme 1
a
Analogue13) in CDCl3
δ
JPC [Hz]
C2
C3
C4
C5
C3CH3
C7
125.8 (128.5)
138.5 (147.4)
131.0 (140.1)
132.0 (136.0)
19.0
11.1 (1.2)
23.4 (8.8)
20.5 (7.6)
7.7 (5.3)
3.8
121.8
18.2
C8
C9
151.9
119.6
2.2
C10
C11
C12
150.0
122.9
158.6
9.1
12.3
a Other data: 35.1 (C8CMe3), 35.2 (C10CMe3), 39.0 (J ) 2.4,
C12CMe3), 31.4 (C8C(CH3)3), 31.8 (C10C(CH3)3), 33.4 (J ) 4.1,
C12C(CH3)).
dehydrobromination with 2-picoline to form the desired phos-
phole 3 in a satisfactory yield of 62%.
Phosphole 3 could be purified by column chromatography
and crystallization and was characterized by mass and NMR
spectroscopic methods. The 31P NMR signal for the phosphole
appeared at δ -0.4, a value differing little from that (δ 1.8) of
the related phosphole 1.7 Calculations suggest that flattening
of the phosphorus pyramid and increased delocalization can be
expected to have a relatively small effect on the phosphorus
shift.3
In the 13C NMR spectrum of phosphole 3 (Table 1), the
effects on the phosphole ring carbons from the planarization
and increased electron delocalization can be clearly seen. These
effects are especially strong at the â-carbons, as is apparent when
a comparison is made with the spectrum of the uncrowded
1-phenyl analogue (Table 1).13 Thus, in 3, both â-carbons are
upfield-shifted by about 9 ppm. Nonbonded steric interactions
with the aryl substituent should be minimal at the â-carbons
and cannot account for this upfield shifting. Computations of
the carbon spectra for 1H-phosphole in both planar and
pyramidal form support this conclusion.14 At the R-carbons of
3, upfield shifting was also observed but was of smaller
magnitude (at C2, 2.7 ppm; at C5, 4.0 ppm). Also striking was
the substantial increase in phosphorus coupling to the ring
carbons of 3 compared to that in the 1-phenyl analogue13 (Table
1). It is possible that the coupling effect is related to electron
delocalization and the increased electron density at the ring
carbons, but it is also possible to develop an explanation based
upon relations established from many studies of other phos-
phorus compounds.15 Thus, two-bond P-C coupling is empiri-
cally related to the orientation of the phosphorus lone pair to
the coupled carbon. As the lone pair moves closer to a carbon,
the coupling constant increases. This is the case when a
phosphole is planarized. One-bond P-C coupling is sensitive
to the hybridization at phosphorus, increasing as the degree of
s-character increases. Increased planarization at phosphorus
would be accompanied by a change in this direction, as sp2
hybridization is approached in the fully planar form. The carbon
shifts and coupling constants for the phosphole ring carbons of
17 and 2,9 where the crowding is less than that in 3, show in
general the same trends but of intermediate magnitudes.
The 13C NMR signals for the aryl substituent of 3 clearly
show that the benzene and phosphole rings are orthogonal to
each other in the conformation adopted at room temperature,
nyl substituent. The synthesis and some properties are described
in this paper. It will be seen from both its physical and chemical
properties that the compound does indeed seem to have a quite
significant degree of electron delocalization, adequate to give
it a place in the family of “aromatic” heterocyclopentadienes.
We note that a phosphindole has been reported with the same
large P-substituent, but with no assessment of degree of
pyramidal character or delocalization.10
Results and Discussion
Synthesis and NMR Characterization of Phosphole 3. The
reactions used in our attempts to synthesize phosphole 3 are
summarized in Scheme 1. The starting material was 1-chloro-
3-phospholene (4). Steric factors prevented the formation of
the aryl Grignard reagent from the bromotri-tert-butylbenzene;
hence, we could not use our earlier method for the syntheses of
crowded phospholenes.7,9 The synthesis could, however, be
accomplished by reaction of the aryllithium reagent with the
chlorophospholene (4). The pathway followed earlier to phos-
pholes 1 and 2 involved the sequence shown as 4 f 5 f 6 f
7 f 8, in which the appropriate dibromophospholane 8 is the
immediate precursor of the phosphole on dehydrobromination.
The sequence as used in the present case was successful only
through the synthesis of the dibromophospholane oxide 7, since,
due to side reactions, this compound was very difficult to reduce
to the necessary phospholane stage (8). A second sequence,
developed early in our studies on phospholes11 proved to be
more successful. The 3-phospholene 5 was converted to the
bromophospholenium bromide 9 by the addition of 1 equiv of
bromine; 9 was then subjected to the Mathey procedure12 of
(13) Bundgaard, T.; Jakobsen, H. J. Tetrahedron Lett. 1972, 3358.
(14) Chesnut, D. B. Private communication.
(15) For a review, see: Quin, L. D. In Phosphorus-31 NMR Spectroscopy
in Stereochemical Analysis; Verkade, J. G., Quin, L. D., Eds.; VCH
Publishers: Deerfield Beach, FL, 1987; Chapter 12.
(10) Ma¨rkl, G.; Jin, G. Y.; Berr, K.-P. Tetrahedron Lett. 1993, 34, 3103.
(11) Quin, L. D.; Borleske, S. G.; Engel, J. F. J. Org. Chem. 1973, 38,
1858.
(12) Breque, A.; Mathey, F.; Savignac, P. Synthesis 1981, 983.

1-(2,4,6-Tri-tert-butylphenyl)-3-methylphosphole
J. Am. Chem. Soc., Vol. 119, No. 22, 1997 5097
Table 2. Selected Bond Lengths [Å] and Angles [deg] for 3
Table 3. Selected Bond Lengths [Å] and the Bird Index of
Aromaticity for Benzylphosphole (BP), Phospholes 2 and 3, and the
Phospholide Ion (PI)
P1-C5
P1-C2
P1-C7
C2-C3
C3-C4
C3-C6
C4-C5
C7-C12
C7-C8
C8-C9
1.741(8)
1.746(9)
1.812(8)
1.347(12)
1.402(12)
1.524(12)
1.352(12)
1.417(9)
1.449(9)
1.383(9)
C9-C10
1.379(10)
1.389(10)
1.395(10)
91.7(5)
118.3(4)
121.7(4)
109.9(7)
112.9(8)
115.2(8)
108.4(7)
C10-C11
BP5
29
PI17
3
C11-C12
C5-P1-C2
C5-P1-C7
C2-P1-C7
C3-C2-P1
C2-C3-C4
C5-C4-C3
C4-C5-P1
P-C2, P-C5
C3-C4
C2-C3, C4-C5
BI
1.786a
1.438
1.343a
35.5
1.780a
1.436
1.353a
40.4
1.751a
1.424
1.396a
1.743a
1.402
1.350a
56.5
a Average values.
angle of 90.7° with P out of the plane of C2-C3-C4-C5 by
0.21 Å; for phosphole 3, these values were 91.7° and 0.294 Å.
Of critical importance in the assessment of the extent of
electron delocalization in a cyclic compound is the determination
of bond lengths. In 5-membered heterocycles, electron release
from the heteroatom is well-known to cause shortening of the
bond to the sp2 R-carbon. Table 3 shows the average P-C
bond lengths measured for the series of 1-benzylphosphole,5
phospholes 29 and 3, and the phospholide ion.17 As can be seen
there is a relatively great reduction in the average P-C bond
length in 3, consistent with considerable electron delocalization.
Another indication of the extent of P-C bond shortening is
provided by comparison of the average internal sp2 C-to-P bond
length to the external sp2 C of the aryl ring; the latter one is
longer by 0.068 Å.
Also a strong indicator of delocalization is the shortening of
the sp2-sp2 bond at C3-C4 as the double-bond character
increases from the resonance interaction. The value for 3 is
again the smallest one in the series (Table 3) and is in fact in
the lower quartile of all those reported for the carbon-carbon
double bond in organic compounds in a recent survey of
structural data.18 It should in particular be compared to the
values19 for the related parent 5-membered heterocycles pyrrole
(1.417 Å), thiophene (1.423 Å), and furan (1.430 Å). The value
for 3 is therefore smaller than that in these parents and in each
case falls in the lower quartile (where q1 is the cutoff value)
established in a tabulation18 of bond lengths for known structures
containing each heterocyclic system (29 pyrroles, q1 ) 1.401
Å; 40 thiophenes, q1 ) 1.415 Å; 62 furans, q1 ) 1.412 Å).
Since the C3-C4 bond is relatively remote from the P-aryl
substituent, it is unlikely that nonbonded steric interactions are
involved in this substantial bond shortening, and we attribute it
to extensive electron delocalization in phosphole 3.
Figure 1. Perspective view of phosphole 3.
just as expected from studies on phosphole 1 which showed
the same effect.7 Thus, there are separate signals for the two
ortho carbons, as well as for the meta carbons, and also for the
tert-butyl groups attached to the ortho carbons. As a result of
the fixed conformation, one edge of the benzene ring is held in
closer proximity to the phosphorus lone pair orbital than is the
other as expressed in structure 3, and there are noticeable
differences in the coupling constants to carbons on the different
edges of the benzene ring (Table 1).
The proton NMR spectrum of phosphole 3 resembled that of
phospholes 1 and 2 and had no remarkable features that could
arise from the increased electron delocalization. We observed,
however, a surprisingly large four-bond P-coupling to the
3-methyl group; the value was 7.3 Hz, but was only 1.5 Hz in
phosphole 1.
Accompanying the shortening at the C3-C4 bond in 5-mem-
bered heterocycles should be some lengthening of the C2-C3
and C4-C5 bonds. The values for the above series are listed
in Table 3. Going from the benzylphosphole5 to 3, the effect
is not dramatic. It is not, however, inconsistent with values
for other heterocycles, as can be seen by comparison to the
values for the parent19 (P) and to the lowest quartile18 (q1) of
values for substituted heterocycles (58 pyrroles, P ) 1.382 Å,
q1 ) 1.361 Å; 60 thiophenes, P ) 1.370 Å, q1 ) 1.346 Å; 125
furans, P ) 1.361 Å, q1 ) 1.329 Å).
The effect of electron delocalization in aromatic systems on
bond parameters can be expressed by an index developed by
Bird.20 Here, a heterocyclic system is placed on a scale where
the fully delocalized benzene has a value of 100. The Bird
index for furan is 43, for pyrrole 59, and for thiophene 66. Bird
has also calculated a value of 35.5 for 1-benzylphosphole.21 We
Molecular Parameters for Phosphole 3 from X-ray Dif-
fraction Analysis. Molecular parameters of phosphole 3 are
supplied in Table 2. The conclusion drawn from the 13C NMR
spectrum that the phosphole and benzene rings were orthogonal
to each other was confirmed (see Figure 1). Another significant
observation on 3 was the very pronounced flattening of the
phosphorus pyramid. To describe this effect, we use the angle
(designated R) of deflection of the carbon of the P-substituent
from the plane of C2-P-C5 of the phosphole ring (the degree
of planarization could also be expressed by the π-orbital axis
vector method,16 as was done for phosphole 1). For the
unhindered 1-benzylphosphole,5 R is 66.9°. Replacement of
benzyl by 2,4,6-triisopropylphenyl9 caused measurable flattening
of the phosphorus pyramid (R 58.0°). In the more crowded
phosphole 3, R is reduced even further, as predicted, to a value
of 45.0°, and the molecule is of great significance in the
continuing study of the influence of planarization at phosphorus
on electron delocalization. The general shape of the phosphole
ring was not distorted by the steric crowding and resembled
that of 1-benzylphosphole.5 This phosphole had a C2-P-C5
(17) Douglas, T.; Theopold, K. H. Angew. Chem., Int. Ed. Engl. 1989,
28, 1367.
(18) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A.
G.; Taylor, R. J. Chem. Soc. 1987, S1.
(19) Katritzky, A. Handbook of Heterocyclic Chemistry; Pergamon:
Oxford, England, 1985; p 57.
(16) (a) Haddon, R. C. J. Am. Chem. Soc. 1976, 112, 3385. (b) Haddon,
R. C. Science 1993, 261, 1545.
(20) Bird, C. W. Tetrahedron 1985, 41, 1409.
(21) Bird, C. W. Tetrahedron 1990, 46, 5697.

5098 J. Am. Chem. Soc., Vol. 119, No. 22, 1997
KegleVich et al.
have calculated the value for our new phosphole 3 and find it
to be considerably higher (56.5), in close proximity to the value
for pyrrole and well above that for furan. On this basis,
phosphole 3 can be considered as a legitimate member of the
family of delocalized (“aromatic”) 5-membered heterocycles,
the first molecular phosphole to achieve this status.
The steric crowding in phosphole 3 is not without effect on
the aryl substituent. The phosphole substituent on the ipso
carbon causes considerable lengthening of the bond to the ortho
ring carbons (Table 2), an effect which lessens the crowding of
the tert-butyl substituents with the phosphole ring.
11 was also supported by the strongly deshielded signal for
C4-H at δ 7.65, which was split only by the phosphorus atom
(23.9 Hz).
Examining P(III) pyramids generally, Schmidpeter et al. found
that the sum of angles at phosphorus was informative; under
the influence of bulky substituents, this sum may increase up
to 344° from the normal value of 325°. On this scale, the sum
of 331.7° for 3 suggests a moderate change in the pyramidality.6b
The pronounced electron delocalization in phosphole 3 could
in principle reduce the reactivity of the phosphorus atom. The
phosphole indeed showed resistance to air oxidation, allowing
its handling in the atmosphere for several days. This stability
may also come from the steric hindrance due to the presence
of the ortho tert-butyl groups.
When the acylation was performed with approximately 2
equiv of acetyl chloride and AlCl3, another product (13) was
also formed in 44% along with 31% 10, 9% 11, and 16% 12.
Compound 13 with 31P NMR δ 11.7 could be separated from
the acetylphospholes by chromatography, but a sample of only
70% purity could be obtained due to decomposition. The mass
spectrum showed that the compound was a diacetyl derivative
(M+ ) 426). The 13C spectrum revealed that the second acetyl
group had entered the benzene ring, rather than the phosphole
ring; only one aryl carbon (δ 123.9, J ) 11.3 Hz, C11) possessed
a proton. In the phosphole ring, the carbonyl carbon had the
large constant associated with two-bond coupling (24.0 Hz),
indicating either a 2-acetyl or a 5-acetyl group. The proton
NMR spectrum contained a signal at δ 7.82 which showed only
splitting by 31P (22.9 Hz), and none by an adjacent C-H. This
eliminates the possibility of a 2-acetyl group, allowing the
assignment of 5,9-diacetyl substitution to 13.
Electrophilic Substitution of Phosphole 3. While not itself
a proof of aromaticity, it is a characteristic of aromatic
heterocyclic systems that they undergo electrophilic substitution
reactions. To the present, this is a reaction never observed with
a phosphole, but we suspected that phosphole 3, with its clear
indication of extensive delocalization from the X-ray data, might
well respond to such reactions. We chose to use Friedel-Crafts
acylation as a model reaction, since this has been successfully
applied to substitution on a phospholide ion moiety bonded in
complexes with Mn22 and Fe.23 Refluxing 3 in petroleum ether
with a small excess of acetyl chloride and aluminum chloride
did indeed result in nearly complete consumption of the
phosphole after 48 h. Analysis of the reaction mixture by 31P
NMR suggested that three new phospholes were formed (δ 8.3
(49%), δ 5.0 (22%), and δ 1.9 (22%)), with 7% of unreacted 3
present. The unreacted 3 was removed by silica gel chroma-
tography, but attempts so far to separate the mixture of acylation
isomers have not been successful. Spectral characterization
therefore has been performed on the best fraction from chro-
matography (ratio 51:33:16, respectively). That the phospholes
did indeed bear an acetyl group was demonstrated first by IR
(νCdO ) 1654 cm-1) and by mass spectral analysis of the three-
component mixture (M+ ) 384). The 1H NMR spectrum
contained readily recognized signals from the major component
with 31P NMR δ 8.3. The acetyl methyl was present as a
doublet (4JPH ) 8.1; note again the very large size for four-
bond coupling in this phosphole) at δ 2.48 and only two proton
signals from the phosphole ring were present that were coupled
by 6.9 Hz (as well as to 31P). Hence, they were adjacent,
allowing the assignment of structure 10 to this product. This
orientation is just that expected from analogy with pyrrole
chemistry; attack at the 2-position of 3-methylpyrrole is common
on electrophilic substitution.24 The 13C NMR spectrum was also
consistent with structure 10 (see Experimental Section). The
other two phospholes are assigned structures 11 and 12. The
5-acetyl isomer (11) with δP 5.0 was recognized from the large
31P coupling to the carbonyl carbon (δ 193.8, J ) 18.7), while
the 4-acetyl isomer (12) with δP 1.9 had a much smaller coupling
constant (δ 195.4, J ) 7.1 Hz). Structure of the 5-acetyl isomer
Experimental Section
General Considerations. The 31P, 1H, and 13C NMR spectra were
taken on a Bruker DRX-500 spectrometer operating at 202.4, 500, and
125.7 MHz, respectively. Mass spectra were obtained on an MS 25-
RFA instrument at 70 eV. All manipulations with air and moisture
sensitive compounds, such as chlorophosphine 4 and phosphines 3, 5,
and 10-13, as well as phospholium salt 9 were performed under an
atmosphere of nitrogen.
1-(2,4,6-Tri-tert-butylphenyl)-3-methyl-3-phospholene (5). A mix-
ture of 18.5 g (0.0569 mol) of 1-bromo-2,4,6-tri-tert-butylbenzene25
in 140 mL of dry THF was treated with 32.7 mL of a 2 M hexane
solution of butyllithium at -78 °C. After 2 h of stirring, the aryllithium
so obtained was slowly reacted with a solution of 7.65 g (0.0569 mol)
of chlorophospholene 426 in 30 mL of THF. After a 3 h stirring period
at -78 °C, the mixture was allowed to warm to room temperature,
and the contents of the flask were stirred overnight. Solvent was
evaporated, and the residue suspended in 250 mL of n-hexane.
Filtration of the suspension and evaporation of the filtrate afforded 19.1
(22) Mathey, F. Tetrahedron Lett. 1976, 4155.
(23) Mathey, F. J. Organomet. Chem. 1977, 139, 77.
(24) Newkome, G. R.; Paudler, W. W. Contemporary Heterocyclic
Chemistry; John Wiley: New York, 1982, p 111.
(25) Pearson, D. E.; Frazer, M. G.; Frazer, V. S.; Washburn, L. C.
Synthesis 1976, 621.
(26) Quin, L. D.; Szewczyk, J. Phosphorus Sulfur 1984, 21, 161.

1-(2,4,6-Tri-tert-butylphenyl)-3-methylphosphole
J. Am. Chem. Soc., Vol. 119, No. 22, 1997 5099
g (98%) of phosphine 5: 31P NMR (CDCl3) δ 8.8; MS, m/z (rel
intensity) 344 (M+, 60), 343 (100), 329 (10), 288 (21), 276 (35), 261
(29), 220 (47), 205 (15), 99 (17), 57 (45); HRMS calcd for C23H37P
344.2633, found 344.2701.
Monoacetylation of Phosphole 3. A mixture of 1.0 g (2.92 mmol)
of phosphole 3, 0.47 g (3.52 mmol) of aluminum chloride, and 0.35
mL (4.92 mmol) of acetyl chloride in 50 mL of 40-70 °C petroleum
ether was refluxed with stirring for 48 h. Volatile components were
evaporated, and the residue was taken up in 50 mL of chloroform. The
mixture was treated with 10 mL of water, and the organic phase was
dried (MgSO4). Evaporation of the solvent left an oil containing 93%
of acetylphospholes (10 (49%), 11 (22%), and 12 (22%)) and 7% of
unreacted 3 according to 31P NMR. Column chromatography (silica
gel, 2% methanol in chloroform) afforded a fraction (0.24 g) containing
51% of 10, 33% of 11, and 16% of 12: MS, m/z (rel intensity) 384
(M+, 49), 369 (M - Me, 40), 341 (M - Ac, 49), 231 (24), 57 (100);
HRMS calcd for C25H37OP 384.2582, found 384.2641.
1-(2,4,6-Tri-tert-butylphenyl)-3-methyl-3-phospholene 1-Oxide
(6). A solution of 15.0 g (0.0436 mol) of phosphine 5 in 100 mL of
chloroform was slowly treated with 6.0 mL (0.0528 mol) of 30%
hydrogen peroxide at 0 °C with intensive stirring. After a period of 1
h, the mixture was washed with 4 × 30 mL of water. The organic
phase was dried (MgSO4), and the solvent was evaporated to give crude
6. Column chromatography (silica gel, 3% methanol in chloroform)
furnished 9.1 g of oxide 6 in a purity of 95% (yield 55%): 31P NMR
1
(CDCl3) δ 58.6; H NMR (CDCl3) δ 1.29 (s, 9H, p-CMe3), 1.49 (s,
10: 31P NMR (CDCl3) δ 8.3; 1H NMR (CDCl3) δ 1.32 (s, p-CMe3),
18H, o-CMe3), 1.74 (s, 3H, C3CH3), 2.63-2.93 (m, 4H, CH2), 5.42 (d,
3JPH ) 30.8 Hz, 1H, CHd), 7.28 (m, 2H, ArH); MS, m/z (rel intensity)
360 (M+, 9), 359 (13), 345 (5), 303 (100), 57 (61); HRMS calcd for
C23H37OP 360.2582, found 360.2619.
1.37 (s, o-CMe3), 1.54 (d, 4JPH ) 2.0 Hz, C3CH3), 2.48 (d, 4JPH ) 8.1
3
3
Hz, C(O)CH3), 6.79 (dd, JPH ) 17.9 Hz, JHH ) 6.9, C4H), 7.26 (dd,
2JPH ) 30.1 Hz, 3JHH ) 6.9 Hz, C5H), 7.49 (s, ArH), 7.50 (s, ArH); 13
NMR (CDCl3) δ 18.8 (s, C3CH3), 30.7 (s, C(O)CH3), 119.4 (s, C9),
C
3,4-Dibromo-1-(2,4,6-tri-tert-butylphenyl)-3-methylphos-
pholane 1-Oxide (7). To 17.4 g (0.0482 mol) of phospholene oxide
6 in 120 mL of chloroform was added dropwise 2.48 mL (0.0482 mol)
of bromine in 20 mL of chloroform at 0 °C. After the addition, the
contents of the flask were stirred at room temperature for 1.5 h. The
crude product obtained after evaporation of the volatile components
was purified by column chromatography (silica gel, 3% methanol in
chloroform) to give 18.8 g (75%) of dibromide 7 as a mixture of two
diastereomers: 31P NMR (CDCl3) δ 49.8 for the major (90%) and 44.8
for the minor (10%) isomer; 1H NMR (CDCl3) δ 1.26 (s, 9H, p-CMe3),
1.49 (s, 18H, o-CMe3), 2.02 (s, 2.7H, C3CH3 (major)), 4.40 (m, 1H,
CH) 7.31 (s, 2H, ArH); 13C NMR (major isomer) δ 31.0 (C4′C(CH3)3),
32.0 (3JPC ) 8.0 Hz, C3CH3), 33.4 (C2′C(CH3)3), 34.4 (C4′CMe3), 40.3
(C2′CMe3), 46.2 (1JPC ) 70.0 Hz, C5), 50.1 (1JPC ) 70.2 Hz, C2), 57.6
(2JPC ) 4.8 Hz, C4), 67.0 (2JPC ) 7.5 Hz, C3), 123.9 (3JPC ) 13.2 Hz,
C3′), 126.5 (1JPC ) 108.2 Hz, C1′), 152.9 (C4′), 158.0 (2JPC ) 8.7 Hz,
C2′); CI-MS, 519 (M + H); MS, m/z (rel intensity) 461 (M - 57, 44),
381 (22), 303 (83), 231 (100), 57 (98).
3
2
123.3 (d, JPC ) 10.5 Hz, C11), 132.6 (d, JPC ) 15.1 Hz, C4),* 134.2
(d, 1JPC ) 12.3 Hz, C5),* 195.3 (d, 2JPC ) 25.6 Hz, CdO) (the asterisk
indicates may be reversed).
11: 31P NMR (CDCl3) δ 5.0; H NMR (CDCl3) δ 1.88 (C3CH3),
1
4
2
2.28 (d, JPH ) 6.3 Hz, C(O)CH3), 6.54 (d(b), JPH ) 35.4 Hz, C2H),
7.53 (s, ArH), 7.54 (s, ArH), 7.65 (dd, 3JPH ) 23.9 Hz, 4JHH ) 2.3 Hz,
C4H); 13C NMR (CDCl3) δ 19.9 (d, 3JPC ) 8.2, C3CH3), 193.8 (d, 2JPC
) 18.7, CdO).
12: 31P NMR (CDCl3) δ 1.9; H NMR (CDCl3) δ 1.54 (d, JPH
)
1
4
2
2.0 Hz, C3CH3), 2.47 (s, C(O)CH3), 6.93 (d, JPH ) 32.0 Hz, C2H),
C5H overlapped, 7.51 (s, ArH), 7.52 (s, ArH); 13C NMR (CDCl3) δ
3
3
18.6 (d, JPC ) 3.9 Hz, C3CH3), 195.4 (d, JPC ) 7.1 Hz, CdO).
Diacetylation of Phosphole 3. In the above reaction mixture, the
use of twice as much acetyl chloride (0.6 mL, 8.4 mmol) added in two
portions resulted in a mixture containing 31% of 10, 9% of 11, 16%
of 12, and 44% of 13. Column chromatography (as above) afforded
0.15 g (8%) of 13 in a purity of 70%: 31P NMR δ 11.7; H NMR
1
(CDCl3) δ 1.34 (s, p-CMe3), 1.39 (s, o-CMe3), 1.67 (d, 4JPH ) 1.7 Hz,
C3CH3), 2.55 (s, 9-C(O)CH3), 2.66 (d, 4JPH ) 8.6 Hz 5-C(O)CH3), ∼7.3
(d, partially overlapped, C2H), 7.53 (s, ArH), 7.82 (d, 3JPH ) 22.9 Hz,
C4H); 13C NMR (CDCl3) δ 17.2 (s, C3CH3), 30.9 (s, C5C(O)CH3),*
Attempted Deoxygenation of Dibromophospholane Oxide 7. The
reaction of dibromide 7 with 1.2 equiv of trichlorosilane and 3 equiv
of pyridine in boiling benzene was carried out as described in the
preparation of the triisopropylphenylphosphole 2.9 Due to extensive
decomposition of 7, resulting in, for example, the formation of tri-tert-
butylbenzene, none of the expected phospholane 8 was obtained. Traces
(ca. 5%) of phosphole 3 could be detected.
29.3 (s, C9C(O)CH3),* 123.7 (d, JPC ) 11.3 Hz, C11), 195.6 (d, 2JPC
)
4
24.0 Hz, C5C(O)CH3), 196.2 (d, JPC ) 6.7 Hz, C9C(O)CH3) (the
asterisk indicates may be reversed; GC-MS, m/z (rel intensity) 426 (M+,
29), 411 (M - Me, 19), 383 (M - Ac, 43), 231 (33), 57 (100); HRMS
calcd for C27H39O2P 426.2688, found: 426.2656.
1-(2,4,6-Tri-tert-butylphenyl)-3-methylphosphole (3). To 10.0 g
(0.0291 mol) of phospholene 5 in 400 mL of n-hexane was added 1.50
mL (0.0291 mol) of bromine in 25 mL of dichloromethane at 0 °C,
over a period of 1 h. After the addition, the mixture was stirred at
room temperature for 1.5 h. The 1-bromo-1-(2,4,6-tri-tert-butylphenyl)-
3-methylphospholium bromide (9) appeared as a yellow precipitate.
The suspension of the phospholium salt 9 so obtained was treated
with a solution of 6.3 mL (0.0636 mol) of 2-picoline in 25 mL of
dichloromethane at room temperature. After 1 h of stirring, the mixture
darkened. The stirring was continued for an additional 20 h. Then,
the volatile components were removed by vacuum distillation and the
solid material remaining in the bottom of the flask was extracted with
4 × 100 mL of n-hexane. The crude product obtained after removing
the solvent in Vacuo was purified by column chromatography (silica
gel, 3% methanol in chloroform) to afford 6.50 g of phosphole 3 in a
purity of 95% (yield 62%). Recrystallization from acetone yielded 3.5
g (35%) of 3: mp 102-104 °C; 31P NMR (CDCl3) δ -0.40; 1H NMR
X-ray Structure Analysis of C23H35P (3). Unit cell dimensions:
a ) 21.137(4) Å, b ) 22.327(5) Å, c ) 9.149(2) Å, V ) 4318(2) Å3,
Z ) 8, orthorombic, space group Pbca, d ) 1.054 g cm-3, R ) 0.0729.
Acknowledgments. We are grateful to Professor Donald B.
Chesnut (Duke University) for making available results on new
calculations before publication and for helpful discussions on
the electronic structure of phospholes. We also thank Professor
Andrew T. McPhail (Duke University) for aid in evaluating the
significance of certain aspects of molecular parameters. Gy.K.
thanks the OTKA support of this research (grant numbers T
014917 and U 21513).
Supporting Information Available: Crystallographic ex-
perimental details, coordinates, bond lengths and angles, and
anisotropic displacement coefficients, together with H atom
coordinates and isotropic displacement coefficients (8 pages).
See any current masthead page for ordering and Internet access
instructions.
4
(CDCl3) δ 1.34 (s, 9H, p-CMe3) 1.40 (s, 18H, o-CMe3), 2.24 (d, JPH
2
) 7.3 Hz, 3H, C3CH3), 6.53 (d(b), JPH ) 36.1 Hz, 1H, C2H), 6.69
(dd, 3JPH ) 19.0 Hz, 3JHH ) 6.8 Hz, 1H, C4H), 6.93 (ddd, 2JPH ) 32.9
Hz, 3JHH ) 6.8 Hz, 4JHH ) 2.0 Hz, 1H, C5H), 7.48 (s, 1H, ArH), 7.49
(s, 1H, ArH); 13C NMR, see Table 1; MS, m/z (rel intensity) 342 (M+,
100), 327 (83), 285 (24), 57 (33). Anal. Calcd for C23H35P: C, 80.70;
H, 10.23. Found: 80.49; H, 10.08.
JA970463D