Unprecedented ring transformation of an αα′- monosubstituted 2,4,5-triphenylpyrylium salt with η3-phosphines: Efficient synthesis of aryl- and alkylphosphonium triphenylcyclopentadienylides

Article abstract of DOI:10.1021/om100872f

The ring transformation of the α-reactive 2,4,5-triphenylpyrylium salt 1 with η³-phosphines to the aryl- and alkylphosphonium triphenylcyclopentadienylides 5 is reported. The mechanism of this unprecedented transformation might involve tandem c

Full text of DOI:10.1021/om100872f

6744 Organometallics 2010, 29, 6744–6748
DOI: 10.1021/om100872f
Unprecedented Ring Transformation of an r,r0-Monosubstituted
2,4,5-Triphenylpyrylium Salt with η3-Phosphines: Efficient Synthesis
of Aryl- and Alkylphosphonium Triphenylcyclopentadienylides
Tao Xu, Wei-tao Gong,* Jun-wei Ye, Yuan Lin, and Gui-ling Ning*
State Key Laboratory of Fine Chemicals, School of Chemical Engineering, Dalian University of Technology,
158 Zhongshan Road, Dalian 116012, People’s Republic of China
Received September 8, 2010
The ring transformation of the R-reactive 2,4,5-triphenylpyrylium salt 1 with η3-phosphines to the aryl-
and alkylphosphonium triphenylcyclopentadienylides 5 is reported. The mechanism of this unprece-
dented transformation might involve tandem conjugate addition and intramolecular Wittig cyclization.
Phosphonium cyclopentadienylide (PCPY) ligands have
attracted increasing attention recently, owing to their poten-
tial applications in organometallic chemistry and catalytic
fields.1 However, the difficulties in the synthetic method and
in characterization have induced very few PCPYs except for
C5H4PPh3, thus limiting the study of their chemistry to some
extent. For example, the reactivities of most of these com-
pounds remain largely unexplored and the effects of ligand
substitution on metal complex structures and reactivities
have been very little examined. On the other hand, it is pro-
posed that the metal coordinating and subsequent catalytic
behavior of the PCPY ligands could be partially dependent
on the substituents on phosphorus.2 As a result, the devel-
opment of a novel synthetic route to PCPY ligands with differ-
ent substituents on phosphorus is significant and necessary.
Ring transformation of pyrylium salts with various nucleo-
philes to synthesize a wide range of cyclic as well as hetero-
cyclic compounds is of currently great interest to organic
chemists. Up to now, various functional compounds start-
ing from pyrylium salts have been prepared which are
utilized in many fields, such as biology,3 pharmaceutics,4
nonlinear materials,5 polymer science,6 etc. Surprisingly,
ring transformation of R-active (i.e., R.R0-nonsubstituted
and monosubstituted) pyrylium salts has been much less
investigated. In fact, owing to the “naked” positive charge at
the R-position, such pyrylium salts should have much higher
reactivity than their R-nonreactive counterparts.7
In our previous work, we reported an efficient synthetic
method for pyrylium salts, especially for R-reactive pyrylium
salts, by aerobic oxidation of polysubstituted cyclopentadienes.8
Encouraged by this discovery, we investigate herein the
reaction of the R-reactive 2,4,5-triphenylpyrylium salt 1 with
nucleophilic η3-phosphines. As a result, an unprecedented
ring transformation of 1 to new PCPY ligands with different
substituents on phosphorus was discovered. The correspond-
ing results of our investigation are reported in this paper.
Experimental Section
All syntheses were carried out under a dry, deoxygenated
argon atmosphere using standard Schlenk line techniques.
NMR spectra were recorded on a Varian INOVA-400 spectrom-
eter. Mass spectra were recorded on a HP 1100 MS spectrom-
eter. Elemental analyses were carried out on a Shimadzu 6800
instrument.
All solvents used in this work such as CH2Cl2, CH3CN,
ethanol, toluene, and hexane were purified by the methods
descried by Perrin et al.9 All deuterated solvents were purchased
*To whom correspondence should be addressed. E-mail: wtgong@
dlut.edu.cn (W.G.); ninggl@dlut.edu.cn (G.N.).
(1) (a) Ramirez, F.; Levy, S. J. Org. Chem. 1956, 21, 488. (b) Ramirez,
F.; Levy, S. J. Am. Chem. Soc. 1957, 79, 67. (c) Holy, N. L.; Baenziger,
N. C.; Flynn, R. M. Angew. Chem., Int. Ed. Engl. 1978, 17, 686. (d) Shin,
J. H.; Bridgewater, B. M.; Parkin, G. Organometallics 2000, 19, 5155.
(e) Brownie, J. H.; Baird, M. C. Coord. Chem. Rev. 2008, 252, 1734.
(f) Higham, L. J.; Muldoon, J.; Kelly, P. G.; Corr, D. M.; Mller-Bunz, H.;
Gilleany, D. G. J. Org. Chem. 2007, 72, 8780. (g) Ortega, E.; Kleigrewe, N.;
(6) (a) Bhowmik, P. K.; Kamatam, S.; Han, H.; Nedeltchev, A. K.
Polymer 2008, 49, 1748. (b) Hoffmann, K.; Mix, R.; Resch-Genger, U.;
Friedrich, J. Langmuir 2007, 23, 8411. (c) Garcia, F.; Garcia, J. M.; Garcia-
Acosta, B.; Martinez-Manez, R.; Sancenon, F.; Soto, J. Chem. Commun.
2005, 22, 2790. (d) Garcia-Acosta, B.; Garcia, F.; Garcia, J. M.; Martinez-
Manez, R.; Sancenon, F.; San-Jose, N.; Soto, J. Org. Lett. 2007, 9, 2429.
(e) Mahler, C.; Mueller, U.; Mueller, W. M.; Enkelmann, V.; Moon, C.;
Brunklaus, G.; Zimmermann, H.; Hoeger, S. Chem. Commun. 2008, 39,
4816.
(7) Balaban, A. T.; Ficher, G. W.; Dincylescu, A.; Koblik, A. V.;
Dorofeenko, G. N.; Mezheritskii, V. V.; Schroth, W. Pyrylium salts:
synthesis, reactions and physical properties. In Advances in Heterocyclic
Chemistry; Katritzky, A. R., Ed.; Academic Press: New York, 1982;
Supplement 2.
€
Kehr, G.; Erker, G.; Frohlich, R. Eur. J. Inorg. Chem. 2006, 3769. (h) Petrov,
A. R.; Rufanov, K. A.; Ziemer, B.; Neubauer, P.; Kotov, V. V.; Sundermeyer,
J. Dalton Trans. 2008, 909.
(2) Tolman, C. A. Chem. Rev. 1977, 77, 313.
(3) (a) Wetzl, B. K.; Yarmoluk, S. M.; Craig, D. B.; Wolfbeis, O. S.
Angew. Chem., Int. Ed. Engl. 2004, 43, 5400. (b) Ilies, M. A.; Seitz, W. A.;
Ghiviriga, I.; Johnson, B. H.; Miller, A.; Thompson, E. B.; Balaban, A. T.
J. Med. Chem. 2004, 47, 3744. (c) de Freitas, V.; Sousa, C.; Silva, A. M. S.;
Santos-Buelga, C.; Mateus, N. Tetrahedron Lett. 2004, 45, 9349.
(4) (a) Ilies, M. A.; Seitz, W. A.; Johnson, B. H.; Ezell, E. L.; Miller,
A. L.; Thompson, E. B.; Balaban, A. T. J. Med. Chem. 2006, 49, 3872.
(b) Kovacic, P. Curr. Med. Chem. 2005, 5, 501.
(5) (a) Nakayama, H.; Sugihara, O.; Okamoto, N.; Saito, H.; Mizuno,
A.; Matsushima, R. J. Opt. Soc. Am. B 1998, 15, 477. (b) Li, F. Y.; Zheng, J.;
Huang, C. H.; Jin, L. P.; Zhuang, J. Y.; Guo, J. Q.; Zhao, X. S.; Liu, T. T. J. Phys.
Chem. B 2000, 104, 5090.
(8) Gong, W. T.; Ning, G. L.; Li, X. C.; Wang, L.; Lin, Y. J. Org.
Chem. 2005, 70, 5768.
(9) Perrin, D. D.; Armarego, W. L.; Perrin, D. R. Purification of
Laboratory Chemicals; Pergamon Press: New York, 1980.
r
pubs.acs.org/Organometallics
Published on Web 11/19/2010
2010 American Chemical Society

Article
Organometallics, Vol. 29, No. 24, 2010 6745
from Cambridge Isotope Laboratories. Most chemicals were
obtained from Aldrich and were used as received.
130.8, 132.6, 135.5, 135.9, 136.5, 136.7, 144.1. 31P NMR (aceto-
nitrile-d3, 100 MHz): δ 31.26. MS (API-ES): m/z 369.1(Mþ).
Anal. Calcd for (C26H26Pþ)(ClO4-): C, 66.67; H, 5.56; P, 6.62.
Found: C, 66.77; H, 5.53; P, 6.55.
X-ray crystal structure determinations were performed on a
Bruker SMART APEX CCD diffractometer with graphite-
˚
monochromated Mo KR radiation (λ = 0.710 73 A) using the
Triethyl(2,3,5-triphenylcyclopentadienyl)phosphonium Perchlorate
(4b). A procedure analogous to that for 4a gave 4b (0.312 g,
83%) as gray crystals. 1H NMR (acetonitrile-d3, 0.400 MHz): δ
7.13-7.66 (m, 16H), 5.75 (d, 1H, 2JH-P = 20.8 Hz), 1.51-1.94
(m, 6H), 1.03 (m, 9H). 13C NMR (acetonitrile-d3, 100 MHz): δ
6.5, 12.2, 49.1, 48.7, 128.5, 129.3, 129.5, 129.8, 130.2, 130.6,
SMART and SAINT programs. The structures were solved by
direct methods and refined on F2 by full-matrix least-squares
methods with SHELXTL version 5.1. All non-hydrogen atoms
were refined anisotropically.
Trimethyl(2,3,5-triphenyl-1,3-pentadien-5-one)phosphonium
Perchlorate (3a). A solution of 2,4,5-triphenylpyrylium perchlo-
rate (1; 0.3 g, 0.735 mmol) in 20 mL of CH3CN was treated
dropwise with 0.056 g of PMe3 (0.750 mmol) in 0.5 mL of toluene.
The reaction mixture was stirred at room temperature for 1 h
under an Ar atmosphere. When TLC showed the complete con-
sumption of 1, the solvent was removed in vacuo to yield an
organic solid. Recrystallization from a mixture of CH2Cl2 and
hexane gave product 3a (0.310 g, 87%) as light yellow crystals.
1H NMR (acetonitrile-d3, 400 MHz): δ 8.07-7.38 (m, 16H), 6.64
131.0, 133.7, 135.4, 135.8, 137.5, 138.9, 139.5, 144.8, 147.7. 31
P
NMR (acetonitrile-d3, 100 MHz): δ 42.67. MS (API-ES): m/z
411.3 (Mþ). Anal. Calcd for (C29H32Pþ)(ClO4-): C, 68.24; H,
6.27; P, 6.08. Found: C, 68.43; H, 6.26; P, 6.07.
Dimethylphenyl(2,3,5-triphenylcyclopentadienyl)-phosphonium
Perchlorate (4c). A procedure analogous to that for 4a gave 4c
(0.318 g, 82%) as gray crystals. H NMR (acetonitrile-d3, 400
1
2
MHz): δ 6.99-7.66 (m, 21H), 5.97 (d, 1H, JH-P = 22.0 Hz),
2
1.67 (t, 6H, JH-P = 13.6 Hz). 13C NMR (acetonitrile-d3, 100
2
2
(d, 1H, JH-P = 16.4 Hz), 1.73 (d, 9H, JH-P = 14.4 Hz). 13C
NMR (acetonitrile-d3, 100 MHz): δ 10.2, 108.1, 124.4, 127.4,
128.1, 128.7, 129.1, 129.7, 131.1, 131.5, 133.9, 136.7, 137.3, 137.5,
137.8, 149.3, 162.2, 189.2. 31P NMR (acetonitrile-d3, 100 MHz):
δ 15.4. MS (API-ES): m/z 385.1 (Mþ). Anal. Calcd for
(C26H26OPþ)(ClO4-): C, 64.46; H, 5.37; P, 6.40. Found: C, 64.53;
H, 5.34; P, 6.36.
MHz): δ 7.7, 51.5, 51.9, 128.3, 129.1, 129.4, 129.5, 129.7, 129.9,
130.0, 130.4, 130.6, 132.6, 132.7, 134.6, 134.9, 135.5, 136.9,
137.0, 138.9, 143.8, 147.4. 31P NMR (acetonitrile-d3, 100 MHz):
δ 28.92. MS (API-ES): m/z431.1 (Mþ). Anal. Calcd for (C31H28Pþ)-
(ClO4-): C, 70.19; H, 5.28; P, 5.85. Found: C, 70.01; H, 5.29;
P, 5.83.
Diphenylmethyl(2,3,5-triphenylcyclopentadienyl) phosphonium
Perchlorate (4d). A procedure analogous to that for 4a gave 4d
(0.344 g, 79%) as gray crystals. H NMR (acetonitrile-d3, 400
Triethyl(2,3,5-triphenyl-1,3-pentadien-5-one)phosphonium
Perchlorate (3b). A procedure analogous to that for 3a gave 3b
(0.351 g, 89%) as yellow crystals. 1H NMR (acetonitrile-d3, 400
1
2
MHz): δ 6.86-8.42 (m, 26H), 6.33 (d, 1H, JH-P = 22.0 Hz),
2
1.12 (t, 3H, JH-P = 7.2 Hz). 13C NMR (acetonitrile-d3, 100
2
MHz): δ 7.36-8.03 (m, 16H), 6.50 (d, 1H, JH-P = 14.0 Hz),
2.11-1.82 (m, 6H), 1.05-0.96 (m, 9H). 13C NMR (acetonitrile-
d3, 100 MHz): δ 5.9, 14.1, 103.4, 118.3, 125.0, 128.4, 129.0, 129.5,
129.9, 130.5, 132.0, 134.7, 138.2, 138.7, 138.9, 150.9, 166.6, 189.8.
31P NMR (acetonitrile-d3, 100 MHz): δ 31.8. MS (API-ES): m/z
427.5 (Mþ). Anal. Calcd for (C29H32OPþ)(ClO4-): C, 66.03; H,
5.10; P, 4.94. Found: C, 65.97; H, 5.06; P, 4.91.
MHz): δ 5.9, 52.1, 116.6, 118.3, 128.3, 128.4, 128.9, 129.1, 129.4,
129.7, 129.9 130.3, 130.4, 130.5, 130.9, 131.0, 132.8, 133.5, 134.5,
134.7, 135.3, 135.5, 136.3, 139.2, 143.6, 147.3. 31P NMR
(acetonitrile-d3, 100 MHz): δ 23.8. MS (API-ES): m/z 493.2
(Mþ). Anal. Calcd for (C36H30Pþ)(ClO4-): C, 70.19; H, 5.28; P,
5.85. Found: C, 70.01; H, 5.29; P, 5.83.
Trimethylphosphonium 2,3,5-Triphenylcyclopentadienylide (5a).
A suspension of 4a (0.468 g, 0.001 mol) in ethanol (20 mL) was
added to an aqueous solution of NaOH (2 N, 2 mL). The mixture
was stirred for 10 min and then left for 6 h at room temperature.
The deep red precipitate was filtered off and recrystallized from
acetonitrile to give the ylide 5a (0.323 g, 88%). 1H NMR
(chloroform-d, 400 MHz): δ 6.91-7.49 (m, 15H), 6.41 (d, 1H,
Dimethylphenyl(2,3,5-triphenyl-1,3-pentadien-5-one)phosphonium
Perchlorate (3c). A procedure analogous to that for 3a gave 3c
(0.352 g, 86%) as yellow crystals. 1H NMR (acetonitrile-d3, 400
2
MHz): δ 6.80-7.84 (m, 21H), 5.93 (d, 1H, JH-P = 21.6 Hz),
2
1.67 (t, 6H, JH-P = 12.8 Hz). 13C NMR (acetonitrile-d3, 100
MHz): δ 11.6, 106.3, 107.2, 118.2, 124.3, 128.3, 128.9, 129.3,
129.5, 129.7, 130.0, 130.1, 130.4, 130.7, 132.1, 134.5, 137.2, 1385,
139.0, 144.0, 150.2, 165.4, 189.4. 31P NMR (acetonitrile-d3, 100
MHz): δ 14.1. MS (API-ES): m/z 447.5 (Mþ). Anal. Calcd for
(C31H28OPþ)(ClO4-): C, 68.01; H, 5.12; P, 5.67. Found: C, 67.89;
H, 5.02; P, 5.66.
2
4JH-P = 6.4 Hz), 1.37 (d, 9H, JH-P = 13.2 Hz). 13C NMR
(chloroform-d, 100 MHz): δ 16.5, 82.7, 83.9, 116.0, 123.2, 125.3,
125.9, 126.9, 127.0, 127.4, 128.0, 129.6, 129.9, 131.6, 132.8, 139.8,
141.5, 141.9. 31P NMR (chloroform-d, 100 MHz): δ 2.87. Anal.
Calcd for C26H25P: C, 84.78; H, 6.79; P, 8.43. Found: C, 84.72;
H, 6.83; P, 8.45.
Triethylphosphonium 2,3,5-Triphenylcyclopentadienylide (5b).
A procedure analogous to that for 5a gave 5b (0.365 g, 89%). 1H
NMR (chloroform-d, 400 MHz): δ 6.89-7.82 (m, 15H), 6.56
(d,1H,4JH-P=6.8Hz),1.92(m,6H),1.43(d,9H,2JH-P=13.2 Hz).
13C NMR (chloroform-d, 100 MHz): δ 14.9, 79.0, 80.0, 116.7,
123,2, 125.0, 125.7, 127.4, 127.5, 129.5, 130.8, 131.0, 131.9, 132.3,
133.7, 133.8, 139.9, 141.1, 141.6. 31P NMR (chloroform-d, 100 MHz):
δ 4.83. Anal. Calcd for C29H31P: C, 84.88; H, 7.56; P, 7.56.
Found: C, 84.84; H, 7.57; P, 7.59.
Diphenylmethyl(2,3,5-triphenyl-1,3-pentadien-5-one)phosphonium
Perchlorate (3d). A procedure analogous to that for 3a gave 3d
(0.351 g, 77%) as yellow crystals. 1H NMR (acetonitrile-d3, 400
MHz): δ 7.04-7.84 (m, 26H), 6.25 (d, 1H, 2JH-P=22.0 Hz), 1.10
2
(t, 3H, JH-P = 7.2 Hz). 13C NMR (acetonitrile-d3, 100 MHz):
δ 10.6, 105.6, 118.3, 121.1, 124.3, 128.3, 128.6, 129.0, 129.6, 130.2,
130.5, 130.8, 132.9, 133.5, 133.7, 133.9, 134.5, 135.5, 135.6, 137.0,
138.4, 138.6, 139.4, 149.9, 167.1, 189.6. 31P NMR (acetonitrile-d3,
100 MHz): δ 13.6. MS (API-ES): m/z 509.6 (Mþ). Anal. Calcd
for (C36H30OPþ)(ClO4-): C, 70.94; H, 4.93; P, 5.09. Found: C,
70.91; H, 4.91; P, 5.08.
Trimethyl(2,3,5-triphenylcyclopentadienyl)phosphonium Perchlo-
rate (4a). A solution of 2,4,5-triphenylpyrylium perchlorate
(1; 0.3 g, 0.735 mmol) in 20 mL of CH3CN was treated dropwise
with 0.112 g of PMe3 (1.500 mmol) in 1 mL of toluene. The
reaction mixture was refluxed for 6 h under an Ar atmosphere.
When TLC monitoring showed the complete consumption of 1,
the solvent was removed in vacuo to yield an pale solid.
Recrystallization from a mixture of CH2Cl2 and hexane gave
product 4a (0.29 g, 85%) as gray crystals. 1H NMR (acetonitrile-
d3, 400 MHz):δ 7.16-7.69 (m, 16H), 5.81 (d, 1H, 2JH-P=18.8 Hz),
1.30(d,9H,2JH-P=14.0Hz).13C NMR (acetonitrile-d3,100MHz):
δ 7.6, 55.7, 79.3, 128.9, 129.5, 129.6, 129.9, 130.3, 130.5, 130.6, 130.7,
Dimethylphenylphosphonium 2,3,5-Triphenylcyclopentadien-
ylide (5c). A procedure analogous to that for 5a gave 5c (0.353 g,
1
82%). H NMR (chloroform-d, 400 MHz): δ 6.89-7.80 (m,
20H), 6.56 (d, 1H, 4JH-P=6.8 Hz), 1.40 (d, 6H, 2JH-P=13.2 Hz).
13C NMR (chloroform-d, 100 MHz): δ 14.9, 79.0, 80.0, 116.6,
123.2, 125.0, 125.7, 127.6, 127.9, 128.0, 129.3, 129.5, 130.9, 131.0,
131.1, 131.7, 131.9, 132.3, 133.8, 139.9, 141.2, 141.6. 31P NMR
(chloroform-d, 100 MHz): δ 4.88. Anal. Calcd for C31H27P: C,
86.51; H, 6.28; P, 7.21. Found: C, 86.47; H, 6.30; P, 7.23.
Diphenylmethylphosphonium 2,3,5-Triphenylcyclopentadien-
ylide (5d). A procedure analogous to that for 5a gave 5d (0.395 g,
80%). 1H NMR (chloroform-d, 400 MHz): δ 6.84-7.54 (m, 25H),

6746 Organometallics, Vol. 29, No. 24, 2010
Scheme 1. Ring Transformation of 1 with Various Nucleophilic η3-Phosphines
Xu et al.
6.60 (d, 1H, 4JH-P =6.4 Hz), 1.66 (d, 3H, 2JH-P =12.8 Hz). 13
C
NMR (chloroform-d, 100 MHz): δ 31.0, 79.4, 116.9, 123.2, 124.4,
125.2, 127.5, 127.7, 127.9, 128.1, 128.4, 128.9, 129.6, 130.6,
130.7, 131.7, 131.9, 132.2, 132.9, 134.8, 140.0, 141.3. 31P NMR
(chloroform-d, 100 MHz): δ 9.04. Anal. Calcd for C36H29P: C,
87.63; H, 5.88; P, 6.49. Found: C, 87.58; H, 5.90; P, 6.52.
Trimethyl(1,3,5-triphenyl-4-trimethylphosphoranylidene-2-en-
5-one)phosphornium Perchlorate (8). A procedure analogous to
that for the synthesis of compounds 4 gave 8 (0.21 g, 51%) as
gray crystals. 1H NMR (acetonitrile-d3, 400 MHz): δ 7.44-7.85
(m, 16H), 1.81 (d, 1H, 2JH-P = 13.6 Hz), 1.58 (d, 9H, 2JH-P
=
14.0 Hz), 1.34 (d, 9H, 2JH-P=14.0 Hz). 13C NMR (acetonitrile-
d3, 100 MHz): δ 5.8, 11.2, 28.5, 28.7, 28.9, 29.1, 29.3, 41.4, 41.7,
54.2, 122.4, 127.5, 128.4, 128.7, 129.5, 129.8, 129.9, 130.4, 141.5.
31P NMR (acetonitrile-d3, 100 MHz): δ 32.61, 13.46. MS (API-
ES): m/z 461.1 (Mþ). Anal. Calcd for (C29H35OP2þ)(ClO4-): C,
62.03; H, 6.24; P, 11.05. Found: C, 62.09; H, 6.19; P, 11.04.
Figure 1. Crystal structure of compound 4a. The anion has been
omitted for clarity.
Table 1. Access to PCPY Ligands 5 from Ring Contraction of 1
Results and Discussion
The pyrylium salt 1 was synthesized on the basis of our
previously reported work.8 The key distinguishing structural
feature of 1 is that there is no substituent at the 6-position of
the pyrylium ring, which causes it to be more easily attacked
by nucleophiles. In this case, we tried the reaction of 1 with
nucleophilic η3-phosphines with the aim of obtaining some
new phosphorus-containing functional compounds. As shown
in Scheme 1, the ring transformation of 1 exhibited two differ-
ent pathways, depending on the properties of the η3-phosphines
used.
When PPh3 was chosen as the nucleophilic agent to react
with 1, the corresponding 2,4,6-triphenyl pyrylium salt 7 was
obtained as the sole product. A comparison of the resonance
structures of 2,4,5- and 2,4,6-triphenylpyrylium cations
shows that the presence of a phenyl group at the 6-position
makes cation 7 more stable than cation 1. Accordingly, it
is thought that PPh3 herein might act as the catalyst to
rearrange the 2,4,5-triphenylpyrylium salt 1 to the more stable
2,4,6-triphenylpyrylium salt 7. The details of this conversion
need further investigation.
Significantly, when 1 was reacted with PMe3, PEt3,
PMe2Ph, and PMePh2, which are relatively less sterically
demanding compared to PPh3, the ring contraction prod-
ucts, the triphenylcyclopentadienyl phosphonium salts 4,
were obtained in good yields. The structures of 4 were fully
characterized by 1H NMR, 13C NMR, 31P NMR, MS, and
elemental analyses. Luckily, a single crystal of 4a suitable
for X-ray diffraction determination could be grown by
slow diffusion of n-hexane into a CH2Cl2 solution of 4a
(Figure 1).
a Isolated yields.
Treatment of compounds 4 with sodium hydroxide gives
the target PCPY ligands 5 in excellent yields.11 The synthetic
results are summarized in Table 1.
It is clear that the ring contraction of pyrylium salt 1 reported
herein provides a novel synthetic route to PCPY ligands,
especially PCPY ligands with different substituents on phos-
phorus, which are useful in exploring the effects of ligand
substitution on metal complex structures and reactivities.
In order to know more details about the reaction process,
(10) (a) Brownie, J. H.; Baird, M. C. J. Organomet. Chem. 2008, 693,
2812. (b) Brownie, J. H.; Baird, M. C.; Laws, D. R.; Geiger, W. E.
Organometallics 2007, 26, 5890. (c) Brownie, J. H.; Baird, M. C.; Schmider,
H. Organometallics 2007, 26, 1433. (d) Brownie, J. H.; Baird, M. C.;
Zakharov, L. N.; Rheingold, A. L. Organometallics 2003, 22, 33.
(11) Lloyd, D.; Singer, M. I. C. J. Chem. Soc. C 1971, 2941.

Article
Organometallics, Vol. 29, No. 24, 2010 6747
Figure 2. Reaction of 1 with equimolar PR3 (left) and the crystal structure of compound 3a (right). The anion has been omitted for clarity.
Scheme 2. Proposed Mechanism for Ring Transformation from Pyrylium Salt 1 to Cyclopentadienyl Phosphonium Salt 4
further investigation was done accordingly. It was found that
not only the structural properties but also the amount of
phosphines used could affect the formation of final products.
For example, use of over 2 equiv of η3-phosphines could
completely convert 1 to the ring-contraction product 4. In
contrast, the reaction of an equimolar amount of η3-phos-
phines with 1 yielded only the ring-opening product 3. An
X-ray-quality single crystal of 3a was also obtained by
diffusion of n-hexane into a CH2Cl2 solution of 3a (Figure 2).
Further work showed significantly that the reaction of
compound 3 with additional η3-phosphines gave solely the
cyclopentadienylphosphonium salts 4. This result clearly
indicated that compound 3 was the key intermediate of the
transformation from pyrylium salt 1 to cyclopentadienyl-
phosphonium salts 4. In this case, a plausible mechanism for
this transformation is proposed (Scheme 2).
At the first step, due to the more concentrated positive
charge at the R-position, attack by the first equivalent of PR3
should occur at this position to give the 2H-pyran 2, which
readily undergoes ring opening to its more stable valence
tautomer 3. This process is accordance with the well-known
ring-opening reaction of a general pyrylium ring induced by
nucleophilic agents.12 The subsequent conjugated addition
to 3 by another equivalent of phosphine gives the acyclic
phosphorus ylide 6, which undergoes an intramolecular
Wittig reaction to a cyclopentadiene ring, producing the
final product, the cyclopentadienyl phosphonium salt 4.
Our attempts to isolate compound 6 were unsuccessful, due
to its rapid conversion to compound 4.
Scheme 3. Reaction of 2,4,6-Triphenylpyrylium
Salt 7 with 2 Equiv of PMe3
To confirm the effect of R-reactivity of pyrylium salt 1 on
this novel ring transformation, we investigated the reaction
of the 2,4,6-triphenylpyrylium salt 7 with PMe3. As expected,
no ring contraction to a cyclopentadienyl phosphonium salt
was observed. Instead, the acyclic compound 8 was obtained
as the sole product (Scheme 3). Accordingly, it is apparent
that the steric hindrance of the phenyl group at the 6-position
inhibits the intramolecular Wittig reaction. This result indicates
the important role of R-reactivity of the pyrylium cation
and provides further evidence for the proposed mechanism
shown in Scheme 2.
In comparison with the well-documented C5H4PPh3 ligand,
the coordination properties of PCPY ligands with other
substituents on phosphorus and the C5H4 ring have been
less investigated. In this regard, ring transformation of the
pyrylium salt 1 to new PCPY ligands provides the chance
to do further research in this field. Luckily, we obtained
crystals of 5c, and on the basis of this crystal structure, its
electronic structure has been primarily investigated using
DFT methodologies (Figure 3). The calculated results demon-
strate that the almost degenerate HOMO orbitals have
symmetries rather similar to those of the corresponding
(12) Markl, G.; Lieb, F.; Merz, A. Angew. Chem., Int. Ed. Engl. 1967,
6, 944.

6748 Organometallics, Vol. 29, No. 24, 2010
Xu et al.
Figure 3. Contour plot of the HOMO of 5c (left; contour values are (0.05) and the X-ray crystal structure of 5c (right).
HOMO (doubly degenerate; E1 symmetry) of the cyclo-
pentadienyl anion. This is consistent with the electronic
structure implied by resonance for C5H4PPh3 and C5H4PMePh2
which has been reported.1,10 The coordination properties
of these new aryl- and alkylphosphonium cyclopentadi-
enylides 5 toward various transition metals are still being
conducted in our laboratory.
Those findings are valuable in terms of the design and syn-
thesis of new PCPY ligands with different substituents and
the exploitation of their coordination chemistry with various
metal ions.
Acknowledgment. This research has been supported by
the National Natural Science Foundation of China (Nos.
20772014 and 20923006) and the Fundamental Research
Funds for the Central Universities (No. 1000-893116).
We are grateful to Prof. Suning Wang (Queen’s University,
Canada) for helpful discussions and proofreading of
the manuscript.
Summary
In summary, we have reported an unprecedented ring
contraction of the R-reactive 2,4,5-triphenylpyrylium salt 1
with η3-phosphines and its utility in the efficient synthesis of
new PCPY ligands 5. The preliminary results supported the
plausible mechanism of a tandem conjugate addition and
intramolecular Wittig reaction, during which the R-reactivity
of pyrylium salt 1 has a profound effect on this ring contraction.
Supporting Information Available: Figures, tables, and CIF files
giving all spectra for synthesized compounds and single-
crystal data for 3a, 4a, and 5c. This material is available free
of charge via the Internet at http://pubs.acs.org.