Doubly bonded systems between heavier Group 15 elements

Article abstract of DOI:10.1039/b715033d

There has been much interest in the synthesis and properties of doubly bonded systems between heavier Group 15 elements, i. e. heavier analogues of azo-compounds (dipnictenes), from the viewpoints of fundamental and material chemistry. Although such double-bond compounds between heavier main group elements are known to be highly reactive, too much so to be isolated as stable compounds, a number of reports on the synthesis of kinetically stabilized diphosphenes (RP=PR), diarsenes (RA=AsR), and phosphaarsenes (RP=AsR) bearing bulky substituent have been published since 1980. We have also succeeded in the synthesis of the first stable distibene (RSb=SbR) and dibismuthene (RBi=BiR) by taking advantage of efficient steric protection groups, 2,4,6- tris[bis(trimethylsilyl)methyl]phenyl (Tbt) and 2,6-bis[bis(trimethylsilyl) methyl]-4-[tris(trimethylsilyl)methyl]phenyl (Bbt), and revealed their structures and properties systematically. Thus, the doubly bonded compounds between heavier Group 15 elements are no longer imaginary species but are those with real existence which are stable, even in the case of the heaviest non-radioactive element bismuth, when they are appropriately protected by bulky substituents. This Perspective describes our research on the chemistry of kinetically stabilized double-bond compounds between heavier Group 15 elements.

Full text of DOI:10.1039/b715033d

View Article Online / Journal Homepage / Table of Contents for this issue
PERSPECTIVE
www.rsc.org/dalton | Dalton Transactions
Doubly bonded systems between heavier Group 15 elements†
Takahiro Sasamori and Norihiro Tokitoh*
Received 1st October 2007, Accepted 31st October 2007
First published as an Advance Article on the web 14th November 2007
DOI: 10.1039/b715033d
There has been much interest in the synthesis and properties of doubly bonded systems between heavier
Group 15 elements, i. e. heavier analogues of azo-compounds (dipnictenes), from the viewpoints of
fundamental and material chemistry. Although such double-bond compounds between heavier main
group elements are known to be highly reactive, too much so to be isolated as stable compounds, a
=
number of reports on the synthesis of kinetically stabilized diphosphenes (RP PR), diarsenes
=
=
(RAs AsR), and phosphaarsenes (RP AsR) bearing bulky substituent have been published since
=
1980. We have also succeeded in the synthesis of the first stable distibene (RSb SbR) and dibismuthene
=
(RBi BiR) by taking advantage of efficient steric protection groups, 2,4,6-tris[bis(trimethylsilyl)-
methyl]phenyl (Tbt) and 2,6-bis[bis(trimethylsilyl)methyl]-4-[tris(trimethylsilyl)methyl]phenyl (Bbt),
and revealed their structures and properties systematically. Thus, the doubly bonded compounds
between heavier Group 15 elements are no longer imaginary species but are those with real existence
which are stable, even in the case of the heaviest non-radioactive element bismuth, when they are
appropriately protected by bulky substituents. This Perspective describes our research on the chemistry
of kinetically stabilized double-bond compounds between heavier Group 15 elements.
7
=
(Mes = mesityl) and diphosphene Mes*P PMes* (Mes* = 2,4,6-
tri-tert-butylphenyl)8 in 1981. These compounds are significant
landmarks in the chemistry of doubly bonded systems between the
heavier main group elements. Nowadays, as for the double-bond
compounds between heavier Group 15 elements (dipnictenes), nu-
1. Introduction
The syntheses of stable compounds containing multiple bonds
between heavier main group elements have been among the fron-
tiers of great interest in chemical research.1 However, according
to the so-called “classical double bond rule”,2 stable molecules
featuring multiple bonding have been thought possible for a long
time only with elements of the second-row in the periodic table.
The statement that such multiple bonding in heavier elements
would be impossible was rationalized by the long bond distances
between the elements, which would not allow sufficient overlap
of p-orbitals. This statement was supported by a number of
unsuccessful attempts to synthesize compounds having a double
bond between two heavier main group elements. The history
of attempted syntheses of heavier congeners of azo-compounds
dates back to 1877, when Ko¨ler and Michaelis reported that
the condensation reaction of PhPCl2 and PhPH2 resulted in the
=
merous examples of kinetically stabilized diphosphenes (RP PR),
=
=
diarsenes (RAs AsR), and phosphaarsenes (RP AsR) have been
isolated and fully characterized.1a,c,9
Although theoretical calculations predicted that distibene
=
=
(RSb SbR) and dibismuthene (RBi BiR) should be accessible
by utilizing kinetic stabilization,10 no stable doubly bonded
compound between two antimony atoms or two bismuth atoms
had been reported except for a few transition-metal complexes,11
in which the multiple bond was supported by metal bridging,
until our successful syntheses and isolation of the first stable disti-
bene12,13 and dibismuthene13,14 utilizing very effective and bulky
substituents, 2,4,6-tris[bis(trimethylsilyl)methyl]phenyl (denoted
as Tbt) and 2,6-bis[bis(trimethylsilyl)methyl]-4-[tris(trimethyl-
silyl)methyl]phenyl (denoted as Bbt). In this Perspective, we
present our systematic studies on the syntheses and properties
of diphosphenes 1, distibenes 2, dibismuthenes 3, and related
compounds (Scheme 1), which are kinetically stabilized by Tbt
or Bbt groups, together with the recent progress of our research
on the chemistry of heavier dipnictenes.
=
formation of a compound formulated as “C6H5P PC6H5”, which
they dubbed “phosphobenzene”.3 However, subsequent X-ray
diffraction studies have revealed that this product has pentameric
or hexameric structures in the solid state.4 A similar mistake arose
in the case of arsenic. Although the structure of the chemother-
apeutic drug “Salvarsan” was first formulated as a monomeric
diarsene structure by Ehrlich,5 the X-ray crystallographic work
on the compound of empirical composition C6H5As revealed its
oligomeric character.6 Based on these failures, it was realized that
the introduction of sufficiently large substituents is necessary to
avoid the oligomerization of these reactive double-bond species.
The idea of kinetic stabilization of reactive multiple bonds by
=
bulky ligands led to the first syntheses of disilene Mes2Si SiMes2
Institute for Chemical Research, Kyoto University, Gokasho, Uji, Kyoto,
611-0011, Japan. E-mail: tokitoh@boc.kuicr.kyoto-u.ac.jp; Fax: +81-774-
38-3209; Tel: +81-774-38-3200
† Dedicated to Prof. Renji Okazaki on the occasion of his 70th birthday.
Scheme 1
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1395–1408 | 1395
©

View Article Online
Takahiro Sasamori, born in 1975, received his Ph.D. from Kyushu University in 2002. He joined Prof. Tokitoh’s group at Kyushu University
and the Institute for Chemical Research of Kyoto University in 1999 as Research Fellow of Japan Society for the Promotion of Science
(JSPS) and then got a position as Assistant Professor in 2003. His main research fields are organometallic and organoheteroatom chemistry.
He received the Bulletin of Chemical Society of Japan Awards in 2002, 2005, and 2007, Inoue Research Award for Young Scientists in 2005,
Daiichi Pharmaceutical Co., Ltd. Award in Synthetic Organic Chemistry of Japan in 2007.
Norihiro Tokitoh, born in 1957, received his Ph.D. from The University
of Tokyo in 1985. He became Research Associate (1986) and Assistant
Professor (1987) at Tsukuba University, and then Assistant Professor
(1989) and Associate Professor (1994) at The University of Tokyo.
He was promoted to Professor of Chemistry at the Institute for
Fundamental Research of Organic Chemistry, Kyushu University in
1998, and moved to Kyoto University as Professor of the Organoelement
Chemistry Laboratory at the Institute for Chemical Research (ICR)
in 2000. He was Visiting Professor of the Coordination Chemistry
Laboratories at the Institute for Molecular Sciences during the period
of 2001–2003, and is now serving as Vice-director of ICR (since 2005)
and Director of the Pioneering Research Unit for Next Generation,
Kyoto University (since 2006). Since 2004, he has served as Visiting
Professor at Technische Universita¨t Braunschweig, Germany. His main
research fields are organoheteroatom and organometallic chemistry, and
his current research interests are focused on the kinetic stabilization
of highly reactive, low-coordinate compounds of heavier main group
elements and the design and application of novel heteroatom ligands for
coordination chemistry. He received the Progress Award in Synthetic
Organic Chemistry of Japan in 1992, the Progress Award in Silicon
Chemistry of Japan in 1996, the Japan IBM Science Award in 1998,
the Bulletin of Chemical Society of Japan Awards in 2002, 2005, and
2007, the Chemical Society of Japan Award for Creative Work in 2003,
and the Alexander von Humboldt Research Award in 2003.
Takahiro Sasamori
Norihiro Tokitoh
Finally, we found an effective synthetic method for distibene
2a and dibismuthene 3a, that is, the deselenation reaction of
the corresponding 1,3,5,2,4,6-triselenatripnictanes, (TbtESe)3 (6a;
E = Sb, 7a; E = Bi), which were prepared by the reaction of
4a and 5a with Li2Se in THF at room temperature, respectively
(Scheme 2).12,13 When 6a and 7a were heated in toluene together
2. Synthesis of distibenes and dibismuthenes
The natural and simple approach to the synthesis of a distibene
may be that based on the methods used in the syntheses of
diphosphenes, that is, the coupling reactions of dihalostibines.
However, unsuccessful attempts to synthesize distibenes by taking
advantage of some traditional bulky substituents used in the
synthesis of stable diphosphenes and diarsenes15 prompted us to
choose much bulkier substituents as steric protection groups. We
chose a Tbt group as a steric protection group, since we have
already succeeded in the synthesis of a variety of compounds
having a double bond between the heavier main group elements
by taking advantage of kinetic stabilization using a Tbt group.
Dichlorostibine and dichlorobismuthine bearing a Tbt group
TbtECl2 (4a; E = Sb, 5a; E = Bi) were readily prepared by
the reaction of TbtLi with SbCl3 and BiCl3, respectively. The
attempted coupling reactions of 4a using various reducing reagents
such as magnesium, lithium, lithium naphthalenide, and so on
resulted in the recovery of a small amount of starting materials
together with the formation of a mixture of insoluble inorganic
salts and unidentified green powder. Although the inseparable
green powder was unidentified at this stage, it was found to be
distibene 2a after the successful synthesis of 2a by another method
(vide infra).
◦
with an excess amount of (Me2N)3P at 110 C in a sealed tube,
insoluble single crystals of distibene 2a and dibismuthene 3a were
precipitated from the mixture on cooling. Compounds 2a and 3a
were isolated by filtration in a glovebox filled with argon as deep
green and purple single crystals in 94 and 68% yields, respectively.
Fortunately, the extremely low solubility of 2a and 3a in toluene
made their isolation quite simple.
Scheme 2
Thus, the first stable distibene and dibismuthene, i.e., the long-
sought after doubly bonded compounds consisting of antimony
1396 | Dalton Trans., 2008, 1395–1408
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
and bismuth, were synthesized and isolated as stable crystalline
compounds, showing that a stable double-bond can be formed
between the heavier Group 15 elements even in the case of the
heaviest element, bismuth.
other hand, the deselenation reaction of 6b with (Me2N)3P in
the presence of an excess amount of 2,3-dimethyl-1,3-butadiene
afforded stibolene 10b in 70% yield together with 11 (30%)
(Scheme 4). As in the case of the generation of 2a by the
thermolysis of stibolene 9a, thermolysis of stibolene 10b gave
distibene 2b having a considerably high solubility in common
organic solvents, via a retro [1 + 4] cycloaddition reaction. It
was confirmed that the treatment of 2b, which was isolated by the
thermolysis of 10b, with (Me2N)3PSe afforded selenadistibirane 11
almost quantitatively. Anyway, Bbt-substituted distibene 2b was
found to be soluble in organic solvents in contrast to the case of
Tbt-substituted distibene 2a.13
In the deselenation reaction of 6a leading to the formation of
distibene 2a, the corresponding monovalent antimony species,
stibinidene 8a, can be thought as an intermediate.13,16 Heating
a toluene solution of 6a and (Me2N)3P in the presence of an
excess amount of isoprene at 110 ◦C in a sealed tube for 22 h
afforded stibolene 9a as a main product (62%) together with a
small amount of green crystals of 2a (26%) (Scheme 3). Stibolene
10a was also obtained (65%) in a similar deselenation reaction
of 6a using 2,3-dimethyl-1,3-butadiene instead of isoprene. Thus,
it could be suggested that stibinidene 8a is generated by the
thermal reaction of 6a with (Me2N)3P as an initial intermediate
along with (Me2N)3PSe, and then 8a, which can be trapped
by isoprene or 2,3-dimethyl-1,3-butadiene to afford 9a or 10a,
respectively, readily undergoes dimerization to give distibene 2a.
In addition, thermolysis of 9a in toluene-d8 at 120 ◦C for 19 h
afforded distibene 2a in 55% yield, suggesting the dimerization
of the stibinidene intermediate 8a is generated via the thermal
retro [1 + 4] cycloaddition reaction of 9a. It was evidenced that
thermolysis of a stibolene can be used as another synthetic method
for a distibene.
Scheme 4
Thus, it was found that Bbt-substituted distibene 2b cannot be
isolated by the deselenation reaction of 6b but can be prepared
and isolated as a stable compound by a stepwise transformation
via stibolene derivative 10b. However, it is necessary to investigate
other synthetic methods for the soluble dibismuthene 3b, since
1,3,5,2,4,6-triselenatribismutane (BbtBiSe)3 is not stable enough
to be isolated.
Since Bbt-substituted distibene 2b was found to be soluble in
hexane, the direct reductive coupling reaction was examined for
the synthesis of 2b. Bbt-substituted dibromostibine 4b was treated
with Mg metal in THF at room temperature for 30 min and
then the reaction mixture was extracted with hexane. Removal of
inorganic salts and subsequent evaporation of hexane gave the ex-
pected distibene 2b almost quantitatively (Scheme 5). This efficient
synthetic method was found to be applicable to Bbt-substituted
dibismuthene 3b. The treatment of dibromobismuthine 5b with
Mg in THF at room temperature resulted in the almost quantita-
tive formation of a purple powder (3b), which has a high solubility
in common organic solvents as expected. Thus, Bbt group makes
it easy to synthesize and isolate distibene and dibismuthene
derivatives. In addition, Tbt- and Bbt-substituted diphosphenes
Scheme 3
Although the first stable distibene 2a and dibismuthene 3a
have been isolated as crystalline compounds, the extremely low
solubility of 2a and 3a in common organic solvents (e.g., hexane,
benzene, toluene, CHCl3, CH2Cl2, THF, Et2O, etc.) prevented
further studies on their chemical properties in solution. On
the other hand, we have developed another bulky aromatic
substituent, Bbt, which is expected to be a potentially useful steric
protection group together with Tbt.13,17 Taking account of the
empirical fact that TbtH has a very low solubility while BbtH is
comparatively more soluble than TbtH, the synthesis of the Bbt-
substituted distibene 2b was examined in the hope of obtaining a
new distibene having higher solubility than 2a.
When 1,3,5,2,4,6-triselenatristibane (BbtSbSe)3 (6b), which was
prepared by the reaction of BbtSbBr2 4b with Li2Se in THF, was
treated with (Me2N)3P in toluene at 130 ◦C in a sealed tube in
a manner similar to that used for the transformation of the Tbt
analogue 6a into 2a, not distibene 2b but the selenadistibirane
derivative 11 was obtained as air-stable orange crystals in 85%
yield.13,18 Although the result seems to be strange, it was found
based on further studies (vide infra) that 11 was formed by
the reaction of the resulting distibene 2b, which would have
considerable solubility, with the resulting (Me2N)3PSe. On the
Scheme 5 Synthesis of diphosphenes, distibenes, and dibismuthenes.
Dalton Trans., 2008, 1395–1408 | 1397
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
a
=
Table 2 Theoretically optimized structural parameters for RE ER
1a,b can also be synthesized by the reductive coupling reaction
of the corresponding dichlorophosphines 13a,b, respectively.19
With 1a,b in hand, we can make a systematic comparison of the
properties between series of the heavier dipnictenes bearing the
same steric protection groups, Tbt and Bbt.
systems calculated at B3LYP/6-31G(d) (TZ(2d)+diffuse for E)b
=
RE ER
E = P
E = As
E = Sb
E = Bi
R = H
E–E–C/◦
˚
E–E/A
2.046
94.4
2.268
92.6
2.653
91.2
2.765
90.3
R =Me
E–E–C/◦
3. Structures of diphosphenes, distibenes, and
dibismuthenes bearing Tbt and Bbt groups
˚
E–E/A
2.043
101.4
2.265
99.3
2.654
97.4
2.767
96.6
R = Ph
Systematic comparison of molecular structures among a series
of heavier dipnictenes 1–3 is of great intrest and importance.
Particularly, it is very important to know the structures and
˚
E–E/A
2.059
101.9
2.280
99.5
2.664
97.1
2.775
95.6
E–E–C/◦
R = Mes
˚
E–E/A
2.050
101.0
2.272
99.0
2.659
97.0
2.771
95.6
=
definitive structural parameters for the Bi Bi double bond, since
E–E–C/◦
=
dibismuthenes 3a,b are the first compounds containg a Bi Bi
a E = P, As, Sb, Bi. R = H, Me, Ph, Mes. b Ref. 25.
double bond, which is the ’heaviest” double bond among those
consisting of non-radioactive elements in the periodic table.
The structural parameters of 1–3 obtained by their X-ray
crystallographic analyses (Fig. 1) are summarized in Table 1.13,19 In
all cases of distibenes 2a,b and dibismuthenes 3a,b, their molecular
=
=
the shorter C–P and P P bonds than C–E and E E (E = Sb and
Bi) bonds.19
=
All of the lengths of the E E bonds in 1–3 were found to
=
structures have a center of symmetry in the middle of the E E
be shorter by ca. 7% than those of the corresponding E–E single
bonds in Ph2E–EPh2 (E = P, Sb, Bi),20–22 indicating the considerable
double-bond character of 1–3. The observed E–E–C bond angles
in 1–3 deviate greatly from the ideal sp2 hybridized bond angle
(120◦). Although we have not extended our chemistry to that of
double bond (E = Sb, Bi) and have trans configurations with the
plane of the aryl rings of the Tbt or Bbt groups perpendicular to
their C–E–E–C (E = Sb, Bi) planes. Although diphosphenes 1a
and 1b also have trans configurations as well as the case of 2a,b
and 3a,b, diphosphenes 1a,b have no intramolecular center of
symmetry, and the two Tbt and Bbt groups in 1a,b are intertwined
with each other in sharp contrast to 2a,b and 3a,b probably due to
the severe steric repulsion between Tbt or Bbt groups caused by
=
arsenic analogues, such structural features of RE ER systems
were again evidenced by Power and co-workes, who successfully
applied their original m-terphenyl ligands (2,6-Ar2C6H3, Ar = Mes
and Tip) to the synthesis of all the series of heavier dipnictenes
RE ER (E = P, As, Sb, Bi).23 In addition, theoretical calculations
=
=
(Table 2) for the RE ER systems (R = H, Me, Ph, Mes; E = P,
As, Sb, Bi) also show that the E–E–H angles get closer to 90◦ as
the central element “E” goes from P to Bi, because the use of these
three orthogonal 6p orbitals without significant hybridization
leads to a bond angle of approximately 90◦ at bismuth. That is,
the experimentally observed structural features are most likely
interpreted in terms of the lower tendency of heavy atoms to
form a hybridized orbital and prefer to maintain the (ns)2(np)3
valence electron configuration.24 Although the experimentally
observed Bi–Bi–C bond angles in 3a,b were widened upto ca.
100◦ probably due to steric reasons and were found to be larger
than the optimized structural parameters of the less hindered
=
models, RBi BiR (Table 2), the narrow bond angles of 3a,b should
be important as experimental evidence for the non-hybridization
effect of bismuth.
Fig. 1 ORTEP drawing (30% probability) of (a) dibismuthene 3a and (b)
diphosphene 1a.
4. Physical properties of the dipnictenes
4-1. Electronic spectra
Table 1 Structural parameters of dipnictenes 1–3
E–E–C/◦
E–E (in Ph4E2)/A
˚
˚
=
ArE EAr
E–E/A
The heavier congeners of azo-compounds are intensely colored.
a
a
=
TbtP PTbt (1a)
2.051(2)
2.043(1)
106.4(2)
104.5(2)
97.8(1)
2.217c
2.217c
=
Their colors arise from the E E double-bond chromophores,
=
BbtP PBbt (1b)
which appear as two absorption maxima in each case assignable as
the symmetry allowed p–p* (intense) and symmetry forbidden n–
p* (weak) electron transitions at wavelengths longer than 300 nm.
The absorption maxima for 1–3 are summerized in Table 3.13,19
In both series of Tbt- (1a–3a) and Bbt- (1b–3b) derivatives, it
can be seen that their p–p* and n–p* transitions shift to longer
wavelength regions as the element row is descended. These trends
114.9(1)
b
a
2.837d
2.837d
2.990e
2.990e
=
TbtSb SbTbt (2a)
2.6422(7) 101.4(1)
2.7037(6) 105.4(1)
2.8206(8) 100.5(2)
2.8699(6) 104.2(1)
=
BbtSb SbBbt (2b)
b
=
TbtBi BiTbt (3a)
a
=
BbtBi BiBbt (1a)
a Measured at −170 ◦C. b Measured at 25 ◦C. c Ref. 20. d Ref. 21. e Ref. 22.
1398 | Dalton Trans., 2008, 1395–1408
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 3 Observed absorption maxima (nm) for 1–3 and calculated
electron transitioins (nm) for RE ER (R = H, Mes; E = P, As, Sb, Bi)
Table 5 Observed redox potentials for 1b, 2b, and 3b along with
=
=
calculated electron affinities and KS-LUMO levels for RE ER (R = H,
Me, Ph, Mes; E = P, As, Sb, Bi)
=
RE ER
E = P
E = As
E = Sb
E = Bi
=
RE ER
E = P
E = As
E = Sb
−1.65
E = Bi
−1.79
R = Tbta
p–p* (e)
n–p*(e)
R = Bbta
p–p*(e)
n–p*(e)
405 (13000)
530 (2000)
—
—
466 (5200) 525 (4000)
599 (170) 660 (100)
R = Bbt
E1/2a/V
−1.84
—
R = H
418 (12000)
532 (1000)
—
—
490 (6000) 537 (6000)
KS-LUMOb
EAb,c/kcal mol−1
R = Me
−3.12
−3.23
−3.30
−3.14
594 (200)
670 (20)
25.73
30.17
35.46
32.89
b
=
R = Mes
p–p*
n–p*
393
507
3.37
411
546
3.15
434
601
2.88
470
596
2.68
KS-LUMOb
EAb,c/kcal mol−1
R = Ph
−2.37
−2.59
−2.80
−2.69
14.29
19.87
26.94
25.27
c
DHOMO−LUMO
R = Hb
p–p*
n–p*
DHOMO−LUMO
KS-LUMOb
EAb,c/kcal mol−1
R = Mes
−2.61
−2.76
−2.91
−2.83
277
389
4.18
322
414
3.78
409
485
2.98
443
488
2.77
27.56
30.41
34.55
32.92
c
KS-LUMOb
EAb,c/kcal mol−1
−2.41
−2.60
−2.74
−2.66
23.34
27.61
32.00
30.51
a Observed in hexane solution. b Calculated at TD-B3LYP/6-31G(d)
(TZ(2d)+diffuse for E) level. c Energy gaps between KS-HOMO and KS-
LUMO in eV.
a Observed in THF solution with Bu4N+PF6 (0.1 M) at r.t. (vs. Ag/Ag+).
b Calculated at (U)B3LYP/6-31G(d) (TZ(2d)+diffuse for E) level. c EA is
adiabatic electron affinity, which represents the difference between the total
energies of the neutral and anionic species [EA = E(neutral) − E(anion
radical)].
−
for the electron transitions indicate the weakening of the p-bond
as the principal quantum number increases, and it is consistent
with the results of TDDFT calculations and the frontier orbital
are unstable under these conditions. On the other hand, 1b, 2b, and
3b showed reversible one-electron redox couples in the reduction
region, respectively (Table 5). It should be noted that distibene
2b showed the lowest E1/2 value as reduction potential among the
three dipnictenes 1b–3b. The experimental results observed here
indicate that a distibene system may undergo facile one-electron
reduction as compared with the other heavier dipnictenes. The
experimental results were supported by theoretical calculations
on the electron affinities and KS-LUMO levels (p*-orbital) for
25,26
=
energy gaps for RE ER (E = P, As, Sb, and Bi).
In addition,
similar trends are also observed for the absorption maxima of the
2,6-Mes2-C6H3 and 2,6-Tip2-C6H3 derivatives reported by Power
et al.23
4-2. Raman spectra
In order to identify the double-bond character of the central
=
E E bonds in heavier dipnictenes 1–3, the frequencies of the
=
model compounds (RE ER, R = H, Me, Ph, Mes; E = P, As, Sb,
=
E E stretching vibrations were measured by resonance Raman
spectroscopy. The results are summarized in Table 4,13,19 together
=
Bi), where the reduction of distibene systems, RSb SbR, should
be the most exothermic among those of the dipnictenes in all
cases R = H, Me, Ph, and Mes (Table 5). The unique tendency of
p-electron systems of heavier Group 15 elements observed here,
which may be contradictory to the intuitive expectation that the
=
with the calculated stretching vibrations for HE EH and H2E–
EH2. The observed strong Raman shifts for 1–3 (except for 3b)
=
assignable to the E E stretching vibrations (E = P, Sb, Bi) are
higher than those for the corresponding dipnictanes, Ph2E–EPh2,
=
p*-orbital level of an E E double-bond will be lowered as the
=
suggesting that dipnictenes 1–3 feature stronger E E bonds as
element row descends, can be most likely interpreted in terms of
the “relativistic effect of the 6th row elements”.31
compared with the corresponding E–E single bonds (E = P, Sb,
Bi).
4-3. Electrochemical properties
5. Reactivities of the dipnictenes
The redox behavior of dipnictenes 1b, 2b, and 3b, which possess
sufficient solubility in organic solvents, has been revealed by cyclic
voltammetry.25,30 In the oxidation region, irreversible oxidation
waves were observed for 1b, 2b, and 3b around Epa = +1.01,
+1.02, and +0.74 V vs. Ag/Ag+, respectively, suggesting that the
corresponding radical cations and/or dications of 1b, 2b, and 3b
5-1. Reduction reactions
The systematic studies on the redox behavior described in section
4-3 naturally prompted us to obtain the anion radical species
of the dipnictenes. Although there are a number of reports on
the ESR observation of anion radical species of kinetically stabi-
Table 4 Observed and calculated vibrational frequencies (cm−1) of E–E and E E stretchings (E = P, Sb, Bi)
=
a
a
b
b
=
=
=
E
TbtE ETbt
BbtE EBbt
Ph2E–EPh2
HE EH
H2E–EH2
P
Sb
Bi
609
207
135
603
196
—
530c
141d
103f
613
225
160
426
165
120
e
a Observed Raman shifts (solid; excitation, He-Ne laser 632.8 nm). b Calculated values without correction at B3LYP/6-31G(d) (TZ(2d)+diffuse for E)
level. c Ref. 27. d Ref. 28. e Decomposed under the measurement conditions. f Ref. 29.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1395–1408 | 1399
©

View Article Online
lized dipnictenes (diphosphene, diarsenes, and phosphaarsenes)
generated in situ,32 such radical species cannot be handled as
crystalline compounds most likely due to their extremely high
reactivity and/or instability. Furthermore, there is no report on
the attempted generation of the anion radical species of a distibene
or a dibismuthene until our research.
and Ph2Sb–SbPh2,28 and lower than those of the corresponding
neutral double-bonds, 1b and 2b (see, Table 4), respectively.
Those of [Mes2P2]•− and [Mes2Sb2]•− were estimated by theoretical
calculations to be 571 and 189 cm−1, respectively. These features of
the anion radical species 1b•− and 2b•− were reasonably explained
in terms of the introduction of one electron onto the anti-
bonding p*-orbital of the neutral double-bond compounds, 1b
and 2b, leading to the elongation of the E–E (E = P, Sb) bond
lengths. Finally, the structural parameters of [Li+(dme)3][2b−]
were determined by X-ray crystallographic analysis (Fig. 2(a)).
There should be no interaction between the [2b•−] and [Li+(dme)3]
moieties in the crystalline state as judged by the long Sb–Li
After several examinations, the reduction of diphosphene 1b
and distibene 2b with lithium metal in DME was found to afford
the corresponding anion radical species as stable compounds,
which were isolated as stable powders of [Li+(dme)3][1b•−] and
[Li+(dme)3][2b•−], respectively.25 The anion-radical species of 1b
was characterized by the ESR spectrum, which displayed a
hyperfine structure due to two equivalent phosphorus atoms [g =
2.009, A(31P) = 135 MHz] as well as the previously reported
anion-radical species of kinetically stabilized diphosphenes [g =
2.007–2.018, A(31P) = 120–150 MHz].32 On the other hand, the
THF solution of [Li+(dme)3][2b•−] showed a highly broadened
signal with g = 2.097 (Fig. 2(b)), while a complicated spectrum
was obtained from the powder-pattern ESR spectra. It should be
noted that the complicated spectrum can be reasonably explained
by coupling with two antimony atoms, which have two isotopes
of 121Sb (57.25%, I = 5/2) and 123Sb (42.75%, I = 7/2) with
parameters of gxx = 1.961, gyy = 2.030, gzz = 2.368, Axx = 644 MHz,
Ayy = 350 MHz, and Azz = 126 MHz (giso = 2.1197), indicating
that the bulkiness of the Bbt groups of 2b•− might prevent the
intermolecular interaction of the radical species.
•
−
˚
distance (ca. 6.7 A). The geometry of [2b ] is similar to that of 2b,
having a center of symmetry in the middle of its Sb–Sb bond with
atrans-configuration. The small Sb–Sb–C(Bbt) angle of 102.29(8)◦
can be explained in terms of the tendency to maintain the (5s)2(5p)3
configuration of the Sb atoms in the case of 2b. The Sb–Sb bond
•
−
˚
length of [2b ] is 2.7511(4) A, which is the medium value between
the Sb–Sb single and double bonds (see, Table 1), indicating the
introduction of one electron into the antibonding p*-orbital of 2b.
5-2. Oxidation and chalcogenation reactions
In contrast to the systematic studies on the oxidation and
chalcogenation reactions of dimetallenes (or metallylenes) of
heavier Group 14 elements,1,33 the reactivities of dipnictenes have
been concealed except for the case of the kinetically stabilized
diphosphenes and diarsenes due to the lack of stable examples of
a distibene and dibismuthene until our research. We have revealed
the unique reactivity of a series of stable dipnictenes 1–3 toward
oxidation and chalcogenation reactions with aerobic oxygen and
elemental chalcogens (Table 6).34
Although dipnictenes 1–3 are quite stable compounds under
ambient conditions in inert atmosphere, Tbt-substituted distibene
2a and dibismuthene 3a, which are almost insoluble in organic sol-
vents, were found to undergo gradual oxidation reactions with aer-
obic oxygen leading to the formation of 1,3,2,4-dioxadipnictanes
14a and 15a in the crystalline state, respectively. Interestingly,
these oxidation reactions are “single crystals to single crystals”
reactions, that is, they proceed maintianing crystallinity through
the oxidation process. Although similar oxidation reactions of
crystals of 2b also gave 14b under the same conditions, those
of a C6D6 solution of 2b afforded only BbtH. Whereas Mes*-
Fig. 2 (a) ORTEP drawing of [Li+(dme)3][2b•−] (50% probability). (b)
ESR spectrum of [Li+(dme)3][2b•−] in THF solution. (c) Observed and
simulated spectra of [Li+(dme)3][2b•−] in the solid state.
=
substituted diphosphene (Mes*P PMes*, 16) is quite stable in
air, 1b underwent a unique oxidation reaction with aerobic
oxygen in the crystalline state leading to the formation of novel
oxadiphosphirane 17, which was identified based on its NMR (1H
and 31P) and FAB-mass spectra and elemental analysis together
with preliminary X-ray crystallographic analysis results.35 Since
The UV-vis spectra of [Li+(dme)3][1b•−] and [Li+(dme)3][2b•−] in
DME showed characteristic absorption maxima in the region of
longer wavelengths at 539 and 812 nm, respectively, which can be
assigned to the corresponding p–p* transitions based on TDDFT
calculations on the model compounds, [Mes2P2]•− and [Mes2Sb2]•−.
Thus, their p–p* transitions are red-shifted as compared with those
of the corresponding neutral species (1b: 418 nm, 2b: 490 nm),
indicating their weakened p-bonds. Furthermore, Raman spectra
(in the solid state) of [Li+(dme)3][1b•−] and [Li+(dme)3][2b•−]
provided important information on the bonding character of the
anion radical species, showing strong Raman lines attributable
to the corresponding E–E (E = P and Sb) stretching vibrations
at 537 and 188 cm−1, respectively. These observed values are
=
Bbt and Tbt effectively prevent the E E moiety from dimerization
but do not obstruct the reaction with small molecules such as
oxygen, unique reactivity of a diphosphene with aerobic oxygen
was revealed in the case of 1b.
The sulfurization reaction of 1b proceeded under relatively
severe conditions, that is, heating of 1b with elemental sulfur
(S8, 10 eq. as S) in the presence of Et3N (10 eq.) at 120 ◦C in
C6D6 for 40 h afforded thiadiphosphirane 18 (88%)19 as in the
case of sulfurization of Mes*P PMes* with elemental sulfur.36 In
=
27
higher than those of the corresponding single bonds, Ph2P–PPh2
contrast to the case of 1b, the sulfurization reaction of 2b and
1400 | Dalton Trans., 2008, 1395–1408
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 6 Oxidation and chalcogenation reactions of the dipnictenes 1–3
=
=
=
BbtP PBbt
ArSb SbBbt
ArBi BiAr
Oxidation (aerobic oxygen)
Sulfurization (S8)
Selenization (Se)
Tellurization (Bu3PTe)
No reaction
3b with an excess amount of elemental sulfur (S8) in benzene
readily proceeds at room temperature to give the corresponding
heterocycles containing sulfur and pnictogen (Sb, Bi) atoms,
respectively.34,37 Although the sulfurization of 3b afforded exclu-
sively 1,2,4,3,5-trithiadibismolane derivative 19, that of 2b under
the same conditions gave three types of antimony-containing cyclic
polysulfides, i.e., 1,2,4,3,5-trithiadistibolane 20 (69%), 1,2,3,5,4,6-
tetrathiadistibinane 21 (17%), and 1,3,2,4-dithiadistibetane 22
(14%), without any other identifiable compound.
established by spectroscopic and X-ray crystallographic analyses
(Fig. 3).
On the other hand, the selenization reactions of 2b and 3b using
elemental selenium in the presence of Et3N in C6D6 at high temper-
ature (60 ◦C and 80 ◦C, respectively) afforded the corresponding
three-membered ring compounds, selenadipnictiranes 23 and 24 in
high yield,19 while that of 1b gave selenadiphosphirane 25 together
with diselenoxophosphorane BbtPSe2 (26), which could not be
isolated due to its high sensitivity toward light, as in the case of
Fig. 3 ORTEP drawings (50% probability) of (a) 1,2,4,3,5-trithiadibis-
molane 19 and (b) telluradibismirane 28.
Mes*P PMes*.38 In contrast to the sulfurization and selenization
=
reactions of 1b, the tellurization reaction of 1b using elemental
=
tellurium or (n-Bu)3P Te, which has been known as a good
tellurization reagent, ended up with no reaction even on heating at
120–150 ◦C in C6D6 in a sealed tube. On the other hand, heating
of 2b with elemental tellurium (10 eq.) in C6D6 in a sealed tube at
120 ◦C for 2 d afforded the corresponding telluradistibirane 27 as a
stable compound in 15% yield. At last, we found that a phosphine
telluride could work as a good tellurization reagent toward 2b and
3b under mild conditions. As a result, novel telluradipnictiranes
27 (50%) and 28 (43%) were successfully isolated by the treatment
5-3. Cycloaddition reactions
As descrived in section 5-2, it was demonstrated that the doubly
bonded systems between heavier Group 15 elements are good
precursors for unique and novel heterocycles containing pnic-
togen atoms. In addition, many examples have been reported
for the [2 + 4]-cycloaddition reactions of kinetically stabilized
diphosphenes with dienes giving the corresponding six-membered
ring compounds containing a P–P bond. Whereas 1–3 did not react
with 2,3-dimethyl-1,3-butadiene even on heating, it was found that
distibene 2b reacted with MesCNO as a 1,3-dipole to afford the
corresponding cycloadduct, 1,5,2,3-oxazadistibacyclopent-4-ene
29 (Scheme 6).40 It is unclear at present why only 2b can readily
undergo the [2 + 3]-cycloaddition with MesCNO in contrast to the
=
of 2b and 3b with (n-Bu)3P Te in C6H6 at room temperature
for 2 h.34,39 Telluradibismirane 28 is the “heaviest” example
among stable three-membered heterocyclic compounds isolated
so far. Thus, it was found that the doubly bonded systems
between heavier Group 15 elements are good precursors for
unique and novel heterocycles containing pnictogen atoms. The
molecular structures of the obtained heterocyclic compounds were
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1395–1408 | 1401
©

View Article Online
Scheme 6 Reaction of 2b with MesCNO.
case of diphosphene 1b and dibismuthene 3b, which were found
to be completely inert with MesCNO even on heating.
6. Synthesis of heteronuclear double-bond
compounds containing the bismuth atom
Since it was demonstrated that homonuclear doubly bonded
systems between heavier Group 15 elements are no longer
imaginary species even in the heaviest case of bismuth, the next
target molecules should be the heteronuclear doubly bonded
systems, which may be key compounds of great importance in the
systematic elucidation of the intrinsic nature of low-coordinated
compounds of heavier main group elements. Although some ex-
amples of phosphaarsenes and phosphastibenes have already been
reported,1,9,41 no heteronuclear double bonds between bismuth
and another Group 15 element have been discovered until our
research, probably due to the weakness of the p-bond between
Bi and another atom caused by the difference in the size of the
p-orbitals.
Fig. 4 ORTEP drawings (50% probability) of (a) stibabismuthene 31 (Sb
and Bi atoms are highly disordered) and (b) phosphabismuthene 33.
The UV/vis spectrum of 31 in hexane shows two absorption
maxima at 709 nm (e 200 dm3 mol−1 cm−1) and 516 nm (e
7500 dm3 mol−1 cm−1), which are most likely assignable to the
=
forbidden n–p* and the allowed p–p* transitions of the Sb Bi
chromophore, respectively. The kmax value for the p–p* transition
of 31 is between those of 2b and 3b.
Similarly, the condensation reaction of BbtBiBr2 (5b) with
Mes*PH2 (32) in the presence of DBU in hexane at room tem-
perature afforded a deep orange powder, the 31P NMR spectrum
of which showed a characteristic broad singlet at 612 ppm. The
broadening of the signal is most likely due to the 9/2 nuclear
spin of the adjacent Bi atom. Since the low-field chemical shift of
612 ppm is characteristic of a low-coordinated phosphorus atom,
this result strongly suggests the formation of phosphabismuthene
33 in this reaction.44 It is noteworthy that 33 is not only a
new member of a novel class of heteronuclear doubly bonded
systems between heavier Group 15 elements but also the first
example of a stable species with a double bond between the third
and sixth row main group elements. The molecular structure of
33 was determined by X-ray crystallographic analysis (Fig. 4),
showing its trans-configuration with a C–Bi–P–C torsion angle
of 179.79(9)◦. The length of the P–Bi bond of 33 is 2.4541(6)
The condensation reaction of dihydrostibine RSbH2 with
dibromobismuthine 5b was thought to be a good synthetic
route in view of the previous reports on the synthesis of un-
42
=
symmetrically substituted diphosphenes such as Mes*P PMes.
Although a dihydrostibine generally shows thermal instability and
decomposes in minutes or hours at ambient temperature, Bbt-
substituted dihydrostibine 30 was successfully synthesized as a
stable crystalline compound by the reaction of dibromostibine
4b with LiAlH4. Preliminarily, it was demonstarated as a novel
synthetic method for 2b that the condensation reaction of 30 with
4b in the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as
a base in THF afforded distibene 2b as expected.35 Furthermore,
condensation reaction of 30 with dibromobismuthine 5b in THF
at −95 ◦C afforded stibabismuthene 31 as red-purple crystals
quantitatively (Scheme 7).43 Unfortunately, the definite structural
parameters for 31 have not been obtained yet due to the inevitable
disorder of the antimony and bismuth atoms, which cannot
be solved by data collection with a number of different single
crystals of 31 even at −180 ◦C (Fig. 4). However, the structure
of 31 was distinctly characterized by spectroscopic analyses. In
the Raman spectrum (in the solid state) of 31, a strong line
attributable to Sb–Bi stretching was observed at 169 cm−1. This
˚
=
A, which is just intermediate between the E E bond lengths of
8
˚
=
the corresponding diphosphene 16 [Mes*P PMes*, 2.034(2) A]
and dibismuthene 3b, and is ca. 8% shorter than the sum of the
covalent radii of the phosphorus and bismuth atoms (2.66 A).
˚
Unfortunately, the Raman spectral data could not be obtained
due to sample decomposition under the measurement conditions.
The UV/vis spectrum of 33 in hexane shows characteristically red-
shifted absorption maxima at 455 (e 10000 dm3 mol−1 cm−1), 540
(sh, e 1000 dm3 mol−1 cm−1), and 670 (e 300 dm3 mol−1 cm−1)
nm, whereas the previously-known heavier congeners of azo-
compounds show only two absorption maxima corresponding to
the symmetry allowed (p–p*) and the symmetry forbidden (n–
p*) electron transitions. The kmax value of 455 nm attributable to
=
vibrational frequency lies between the values of the Sb Sb
=
stretching vibration in distibenes 2 and that of the Bi Bi stretching
vibration in dibismuthene 3a (see, Table 4), and is higher than
the Sb–Sb and Bi–Bi stretching frequencies for Ph2E–EPh2.28,29
=
the p–p* transition of the P Bi chromophore is between those
8
3
−1
of Mes*P PMes* (16) [340 nm (e 7690 dm mol cm−1)] and
=
=
BbtBi BiBbt (3b). Although the latter two kmax values observed
for 33 (540 and 670 nm) are most likely assignable to the n–p*
transitions and one of them at 540 nm is reasonably found between
the kmax value of 16 [460 nm (e 1360 dm3 mol−1 cm−1)] and that of
3b, another one at 670 nm lies at a strangely longer wavelength.
Although we have no clear explanation for the additional n–p*
Scheme 7 Synthesis of stibabismuthene 31 and phosphabismuthene 33.
1402 | Dalton Trans., 2008, 1395–1408
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
transition for 33 observed in the longer wavelength region, it may
be consistent to assume that heteronuclear doubly bonded systems
between heavier Group 15 elements should have two weak n–p*
transitions due to the two chemically distinct lone pairs on the
phosphorus and bismuth atoms. The outcomes for 33 discussed
above are also of great importance in the understanding of the
spectroscopic properties of stibabismuthene 31, that is, it can be
thought that only one n–p* transition at a longer wavelength
(709 nm) is observed for 31 but the other one might be hidden
by the intense p–p* transitions at 516 nm.
Now, as for the heavier Group 15 elements, not only the
homonuclear doubly bonded compounds but also the heteronu-
clear systems containing a bismuth atom are in hand as stable
compounds, even in the case between extremely different row
elements.
from 16, which can be initially generated by the photolysis of 33
along with 3b as in the case of the photolysis of 31. Although it is
important and interesting to elucidate the reaction mechanism, it
is still unclear at present.
7. Recent topics: functionalized diphosphenes
Recently, much attention has been focused on the unique
electrochemical properties of phosphaalkenes and diphosphenes
from the viewpoints of material science, since they are known
to be more easily reduced than olefins or azo-compounds.32,46
To date, a number of p-conjugated systems containing low-
coordinated phosphorus atoms have been reported.47 Particularly,
Gates et al. and Protasiewicz et al. reported their initial work
on the photochemical properties of the “phospha-PPVs”, poly(p-
phenylenephosphaalkene)s,48 which exhibit unique fluorescence
Interestingly, the themolysis of 31 and 33 in C6D6 (sealed
tube) at 80 ◦C gave the corresponding homonuclear double-bond
compounds, respectively, that is, 2b and 3b were obtained from 31
(Scheme 8), and 16 and 3b were generated from 33 (Scheme 9).35,43
In addition, the photoirradiation of 31 also afforded 2b and
3b. Although the formation of 2b and 3b may be intuitively
=
=
due to the p-conjugated P C units. Since a P P double bond
should possess an apparently smaller HOMO − LUMO energy
=
gap compared with those of a P C double bond, fundamental
studies on the properties of a p-electron extended system of a
diphosphene should also be interesting and important for begin-
ning new chemistry of material science. Recently, we have reported
the first stable 9-anthryldiphosphene,49 ferrocenyldiphosphene,50
and 1,1ꢀ-bis(diphosphenyl)ferrocenes51,52 as new candidates for a
family of p-conjugated diphosphenes. Furthermore, a novel b-
ketophosphenato ligand,53 which should feature unique properties
reflected by the low-coordinated phosphorus atom, was designed,
synthesized, and applied to transition metal complexation.
=
interpreted in terms of the dissociation of the Sb Bi bond of 31
followed by the dimerization of the resulting stibinidene BbtSb:
(8b) and bismuthinidene BbtBi: (34), the attempted trapping
reaction of the stibinidene intermediate 8b was unsuccessful. Thus,
no stibolene 10b, the expected [1 + 4]-cycloaddition product of
8b (see, Scheme 3), was obtained by the photolysis of 31 in the
presence of 2,3-dimethyl-1,3-butadiene under the same conditions,
suggesting the formation of 2b and 3b is most likely interpreted in
terms of the photochemical head-to-head [2 + 2]-dimerization of
31 followed by the retro [2 + 2]-reaction. On the other hand,
the photoirradiation of 33 gave 3b and phosphaindane 35 in
a ratio of 1 : 2. Phosphaindane 35 is known to be formed via
the intramolecular cyclization of the phosphinidene (Mes*P:)
generated by the photchemical dissociation of 16.35,45 Although
the generation of 35 may be explained by the dissociative cleavage
7-1. 9-Anthryldiphosphenes
Condensation reactions of ArPH2 (36a: Ar = Tbt, 36b: Ar =
Bbt), which were prepared by the reaction of ArPCl2 (13a,b)
with LiAlH4, AnthPCl2 (Anth = 9-anthryl) in the presence of
DBU in toluene at room temperature afforded the first stable 9-
anthryldiphosphenes 37a,b as stable red crystals (Scheme 10).49 It
should be noted that both 37a and 37b are quite stable under
an inert atmosphere even in solution (C6D6, 80 ◦C for 26 h)
reflecting the superior steric protecting ability of Tbt and Bbt
for preventing oligomerization. X-Ray crystallographic analysis of
37a (Fig. 5) revealed its trans-configured geometry along with the
=
of the P Bi bond of 33 under irradiation, 35 may also be generated
˚
intermolecular p–p interaction (ca. 3.55 A) between the anthryl
=
moieties of 37a in the packing structure. The P P bond length
and P–P–C angles were found to be 2.0352(16) A and ca. 101 ,
◦
˚
respectively, which are within the range of the previously reported
structural parameters of diaryldiphosphenes. The anthryl group
◦
=
is nearly perpendicular (81.7 ) to the P P axis probably due to
steric reasons, indicating the less effective p-electron conjugation
Scheme 8 Thermolysis and photolysis of 31.
=
between the P P and anthryl moieties in the solid state.
Scheme 9 Thermolysis and photolysis of 33.
This journal is The Royal Society of Chemistry 2008
Scheme 10 Synthesis of 9-anthryldiphosphenes 37.
Dalton Trans., 2008, 1395–1408 | 1403
©

View Article Online
Scheme 11 Synthesis of ferrocenyldiphosphenes 39a, 43a, and 43b.
suggesting a conjugative interaction between the p-electrons of the
=
=
P P unit and those of the Cp unit. However, the P P bond length
˚
of 39a [2.0285(15) A] seems to be little affected by the conjugation,
Fig. 5 (a) ORTEP drawing (50% probability) and (b) cyclic voltammo-
gram (0.1 M Bu4NBF4 in THF) of 37a.
since it is in the range of the previously reported non-conjugated
1
˚
diaryldiphosphenes (ca. 2.00–2.05 A). The UV/vis spectrum of
39a in hexane shows two characteristic absorption maxima at
371 (e 2500 dm3 mol−1 cm−1) and 542 (e 900 dm3 mol−1 cm−1)
nm, which are assignable to the p–p* electron transition of the
The electrochemical data for 37a were determined by cyclic
voltammetry, showing two reversible one-electron redox couples
at E1/2 = −1.73 and −2.32 V (vs. Ag/Ag+). It should be noted
that the first reduction potential in a less negative region than that
=
P P unit and the metal-to-ligand charge transfer (MLCT) band
of TbtP PTbt (1a, E1/2 = −1.93 V vs. Ag/Ag+) and anthracene
=
from the d-orbital of the Fe atom to the p*-orbital of the P P unit,
=
(E1/2 = −2.42 V vs. Ag/Ag+) evidenced the stability of the anion
radical species of 37a due to the delocalization of the unpaired
respectively. The red-shifted absorption of 39a as compared with
that of azoferrocenes (321–328 nm for p–p* transitions, 522–
539 nm for MLCT)55 can be explained by the lower level of the
=
electron on the P P and the anthryl moieties in solution. In
=
=
p*-orbital of the P P unit than that of the N N unit. In addition,
the cyclic voltammograms of 39a showed E1/2 values of + 0.36 (ox.)
and −1.93 (red.) V (vs. Ag/Ag+), which are in the region similar
to those of ferrocene and diphosphene 1a.
the UV/vis spectrum of 37a in hexane, a complicated pattern
=
due to the p–p* electron transitions of the anthryl and P P
moieties was observed in the longer wavelength region than that
of anthracene, indicating the extended p-electron conjugation in
solution. Interestingly, 9-anthryldiphosphene 37a shows weak but
apparent fluorescence emissions in the range of ca. 400–500 nm
(U = 1.2 × 10−3), though a p-conjugated polymer containing
We have also extend the chemistry of ferrocenyldiphosphenes
to that of 1,1ꢀ-bis(diphosphenyl)ferrocenes. The treatment of
ArPHLi (40a,b) with 1,1ꢀ-bis(dichlorophosphino)ferrocene af-
forded the first stable 1,1ꢀ-bis(diphosphenyl)ferrocenes 43a and
43b.51,52 At almost the same time, Pietschnig and co-workers
independently reported the synthesis of another type of stable 1,1ꢀ-
bis(dichlorophosphino)ferrocene bearing m-terphenyl ligands,
2,6-Mes2-C6H3.56 The structural and spectroscopic features of
=
P P units reported by Protasiewicz et al. and other kinetically
stabilized diphosphenes previously reported exhibit no appreciable
luminescence.48d The observed lifetime of the photoluminescence
of 37a is ca. 8–10 ns suggesting that the luminescence of 37a should
be not phosphorescence but fluorescence.
=
43a,b are similar to those of 39a. Interestingly, the P P vibrational
=
frequencies for the two P P units of 43a,b are both Raman and IR
7-2. Ferrocenyldiphosphenes
active due to the symmetric and asymmetric vibrations of the two
=
=
Ferrocenyldiphosphenes are expected to be a unique class of
intramolecular donor–acceptor system, where a diphosphene and
a ferrocene unit should work as an acceptor and a donor,
P P units, though the P P vibration of ferrocenyldiphosphene
39a is active for only Raman spectra. The redox behavior of 43a
and 43b has been furnished by cyclic voltammetry, showing two
reversible one-electron redox waves due to the intramolecular two
redox centers observed at −1.84 and −2.19 V (43a) and −1.78 and
−2.13 V (43b) vs. Ag/Ag+, respectively (see Table 7). In both cases,
the difference between the half-potentials (DE1/2 = 0.35 V) means
a comproportionation constant of Kc = 8.2 × 105 for the mixed-
valence state, indicating an effective electronic interaction through
the central ferrocene unit to some extent similar to the case of the
previously reported 1,4-bis(diphosphenyl)benzene [DE1/2 = 0.34 V
and Kc = 5.6 × 105].
=
respectively, since the P P double bond should be easily reduced
to give the corresponding anion radical when it is kinetically well
stabilized as shown in section 4-3. Pietschnig and Niecke began
the chemistry of ferrocenyldiphosphene and have reported their
initial study on the synthesis and spectroscopic characterization
54
=
of Mes*P PFc (38, Fc = ferrocenyl), though the structural
characterization and further investigation from the viewpoint of
its electrochemical properties have not been accomplished due to
its instability (stable only below −30 ◦C in the absence of daylight).
=
Tbt-substituted ferrocenyldiphosphene, TbtP PFc (39a), was
As the next stage, we applied 1,1ꢀ-bis(diphosphenyl)ferrocene
43b as a novel bidentate ligand bearing two P P units linked
by the ferrocenyl unit toward transition metal complexes in the
expectation of obtaining a unique d–p conjugated bimetallic
=
sucessfully synthesized and isolated as a stable crystalline com-
pound (mp 182–184 ◦C, stable up 100 ◦C in C6H6, inert to light) by
reaction of TbtPHLi (40a) with FcPCl2 (41) followed by treatment
with DBU in situ (Scheme 11).50 The remarkable stability of 39a
compared with 38 reflects the superior protecting ability of Tbt.
In the crystal structure, 39a shows a trans-configuration with a
C–P–P–C dihedral angle of ca. 180◦, and the C–P–P–C plane
is almost coplanar with the Cp ring of the ferrocenyl moiety,
=
systems containing P P units. Heating of 43b in the presence of an
excess amount of [M(CO)5(CH3CN)] (M = Cr, Mo, W) at 50 ◦C
in C6D6 for 2–3 h afforded novel dinuclear metal complexes 44–46
(Scheme 12),52 the structures of which were supported by their
spectroscopic analyses and the X-ray crystallographic analysis
1404 | Dalton Trans., 2008, 1395–1408
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 7 Redox potentials (E/V vs. Ag/Ag+) of ferrocenyldiphosphenes
moderate coordination ability towards a transition metal center
lying between b-diketiminato (stronger) and Schiff-base (weaker)
ligands to the same metal center and its unsymmetrical trans-
effect based on the difference of the coordinating atoms (N
and O).58 On the other hand, the coordination chemistry of
phosphaalkenes has also drawn a great deal of recent attention due
Compounds
E
pa/V
Epc/V
E
1/2/V
DE1/2/V
39a (ox.)a
39a (red.)b
43a (ox.) a
43a (red.) b
+0.45
−1.85
+0.17
−1.66
−2.02
+0.18
−1.67
−2.02
+0.28
−2.10
+0.10
−2.02
−2.35
+0.12
−1.89
−2.24
+0.36
−1.93
+0.14
−1.84
−2.19
+0.15
−1.78
−2.13
0.35
0.35
=
to the unique electronic properties of the C P double bond, whose
43b (ox.) a
43b (red.) b
characteristic low-lying p*-orbital make it possible to function as
a good p-accepter towards a transition metal center.59 During the
course of our investigations on the low-coordinated species of
the heavier Group 15 elements, we designed hitherto unknown
b-ketophosphenato ligands,60 which have both features of b-
ketoiminato and low-coordinated phosphorus ligands, and suc-
ceeded in the synthesis of stable rhodium–cod b-ketophosphenates
47 and 48 as touchstones for the elucidation of the properties of
b-ketophosphenato ligands.
a 0.1 M (n-Bu)4NBF4 in CH2Cl2, E1/2 (FcH/FcH+) = + 0.24 V vs. Ag/Ag+.
b 0.1 M (n-Bu)4NBF4 in THF.
=
for 46 (Fig. 6). In all cases of 44–46, the two P P units were
found to coordinate to the Group 6 metals in a cis-configuration
1
=
with g -fashion. Interestingly, all the P P units of 44–46 are
(Z,Z)-isomers, though the starting material, ligand 43b, is an
(E,E)-isomer. That is, the diphosphene units of 43b underwent
E-to-Z isomerization during complexation with Group 6 metals,
whereas kinetically stabilized diaryldiphosphenes are known to
undergo E-to-Z isomerization not thermally but photochemically.
Thus, it was demonstrated that 1,1ꢀ-bis(diphosphenyl)ferrocene
43b can work as a bidentate ligand toward Group 6 metals.
Further investigations on the physical properties, reactivities,
electrochemical properties, and catalytic ability of 44–46 are
currently in progress.
Scheme 13 Synthesis of rhodium b-ketophosphenates 47 and 48.
Deprotonation reactions of precursor 49 with LDA in THF
afforded lithium b-ketophosphenate [Li(OEt2)·50], the structure of
which was identified by spectroscopic and X-ray crystallographic
analyses. The reaction of [Li(OEt2)·50] with [RhCl(cod)]2 resulted
in the formation of rhodium b-ketophosphenate 47 in 98% yield
(Scheme 13).53 Similarly, rhodium b-ketophosphenate 48 was also
isolated as a stable compound using precursor 51, where the
intermediate lithium b-ketophosphenate could not be isolated.
Structures of the rhodium complex 48 and rhodium b-
ketoiminate 52, which was prepared as a comparative compound,
were revealed by X-ray crystallographic analysis (Fig. 7). It was
found that both 48 and 52 possessed an almost planar structure
for the central six-membered ring moiety, [–Rh–O–C–C–C–E–]
(E = N and P), where every bond length of the central hexagonal
ring was between those of the corresponding single and double
bonds, suggesting a delocalized cyclic p-electron system. It should
be noted that the two kinds of Rh–C(cod) distance of 48 were
apparently different from each other, though those of 52 were
Scheme 12 Synthesis of dinuclear metal complexes 44–46.
Fig. 6 Molecular structures of (a) 43b and (b) 46. Substituents on the Bbt
groups were omitted for clarity.
7-3. b-Ketophosphenato ligands
The coordination chemistry of b-diketiminato and b-ketoiminato
ligands has attracted much interest especially in the field of olefin
polymerization based on late transition metals.57 Particularly,
a b-ketoiminato ligand features unique properties due to its
Fig.
7
ORTEP drawings (50% probability) of (a) rhodium
b-ketophosphenate 48 and (b) rhodium b-ketoiminate 52.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1395–1408 | 1405
©

View Article Online
similar to each other, suggesting a stronger trans-influence of
the phosphorus moiety of the b-ketophosphenato ligand of 48
as compared with its oxygen moiety in contrast to the similar
degree of trans-influence of N and O atoms in the b-ketoiminto
ligand. Such a unique unsymmetrical electronic feature of the
b-ketophosphenato ligand is most likely interpreted in terms of a
high r-donating ability due to the electropositive phosphorus atom
and the effective p-electron accepting ability due to the low-lying
p*-orbital of the sp2-hybridized phosphorus atom. Furthermore,
heating of rhodium b-ketophosphenate 48 in the presence of
norbornadiene (10 eq.) in C6D6 at 40 ◦C for 5 h afforded the
corresponding diene-exchange product 53 almost quantitatively
(Scheme 14), though no diene-exchange reaction was observed in
the case of rhodium b-ketoiminate 52 under the same conditions.
The lower barrier of the diene-exchange reaction of 48 than that
of 51 might be due to the strong trans-effect of the P atoms of the
b-ketophosphenato ligand. Thus, it was demonstrated for the first
time that the b-ketophosphenato ligands should be unique and
useful ligands.
that element chemistry can play an important role in several fields
of material science in the near future.
Acknowledgements
The authors are grateful to the collaborators, especially Prof.
Dr Renji Okazaki (The University of Tokyo), Mr Yoshimitsu
Arai (The University of Tokyo), Dr Nobuhiro Takeda (Kyoto
University, Gunma University at present), Dr Noriyoshi Naga-
hora (Kyoto University), Ms Eiko Mieda (Kyoto University),
Mr Takayuki Yamasaki (Kyoto University), Mr Akihiro Tsu-
rusaki (Kyoto University), and Mr Teruyuki Matsumoto (Kyoto
University), for their invaluable contributions. We also wish to
express our gratitude to Prof. Dr Takeji Takui, Prof. Dr Kazunobu
Satoh, and Prof. Dr Daisuke Shiomi (Osaka City University)
for the measurement of EPR spectra, Prof. Dr Shigeru Nagase
and Dr Nozomi Takagi (Institute for Molecular Science) for
their kind advice on theoretical calculations, Prof. Dr Yukio
Furukawa (Waseda University) for the measurement of Raman
spectra, and Prof. Dr Yoshihiko Kanemitsu and Dr Kazunari
Matsuda (Kyoto University) for the measurement of luminescence
spectra. This work was partially supported by Grants-in-Aid for
Creative Scientific Research (No. 17GS0207), Science Research
on Priority Areas (No. 19027024, “Synergy of Elements”), Young
Scientist (B) (Nos. 18750030), the 21st Century COE Program on
Kyoto University Alliance for Chemistry (B14), and the Global
COE Program (B09, “Integrated Material Science”) from the
Ministry of Education, Culture, Sports, Science and Technology,
Japan.
Scheme 14 Diene exchange reaction of 48.
8. Conclusion and outlook
The isolation and characterization of kinetically stabilized heavier
dipnictenes have been almost accomplished, though some un-
precedented double bonds still remain as interesting synthetic
Notes and references
1 (a) P. P. Power, Chem. Rev., 1999, 99, 3463; (b) G. Raabe and J. Michl,
Chem. Rev., 1985, 85, 419; (c) N. Tokitoh, J. Organomet. Chem., 2000,
611, 217; (d) N. Tokitoh and R. Okazaki, Coord. Chem. Rev., 2000, 210,
251; (e) N. Tokitoh and R. Okazaki, Adv. Organomet. Chem., 2001, 47,
121; (f) N. Tokitoh, Acc. Chem. Res., 2004, 37, 86; (g) P. P. Power,
J. Organomet. Chem., 2004, 689, 3904.
=
=
targets, e.g., Sb As and Bi As. Now, it has been evidenced
that the doubly bonded compounds between heavier Group 15
elements can be isolated in hand as “bottlable” compounds even
in the heaviest non-radioactive element bismuth, when they are
appropriately protected by bulky substituents such as Tbt and
Bbt. While the concept of kinetic stabilization should certainly be
of great use for the construction of these unprecedented chemical
bonds, in the case of not only heavier Group 15 elements but also
other main group elements, a number of homo- and heteronuclear
double bonds between heavier main group elements have been
constructed and their unique properties have been revealed. Thus,
this is a turning point in the chemistry of main group elements,
where the synthetic work for the doubly bonded systems and/or
low-coordinated species of heavier main group elements have
been almost established. Now, there will be much interest in
the expansion of the chemistry of low-coordinated heavier main
group elements to functional chemistry and material science from
the standpoints of element chemistry. As we demonstrated the
application of the low-coordinated phosphorus species toward the
synthesis of p-conjugated systems, such as anthryl- and ferrocenyl-
substituted diphosphenes, and also toward the development of
novel heteroatom ligands for transition metals, the chemistry is just
starting as a basis at the frontier of element chemistry. We believe
2 K. Pitzer, J. Am. Chem. Soc., 1948, 70, 2140.
3 H. Ko¨ler and A. Michaelis, Ber. Dtsch. Chem. Ges., 1877, 10, 807.
4 (a) W. Kuchen and H. Buchwald, Chem. Ber., 1958, 91, 2296; (b) J. J.
Daly and L. Maier, Nature, 1964, 203, 1167; (c) J. J. Daly and L. Maier,
Nature, 1965, 208, 383; (d) J. J. Daly, J. Chem. Soc., 1964, 6147; (e) J. J.
Daly, J. Chem. Soc., 1965, 4789; (f) J. J. Daly, J. Chem. Soc. A, 1966,
428.
5 (a) P. Ehrlich, Lancet, 1907, 173, 351; (b) P. Ehrlich and A. Bertheim,
Ber. Dtsch. Chem. Ges., 1911, 44, 1265; (c) P. Ehrlich and A. Bertheim,
Ber. Dtsch. Chem. Ges., 1912, 45, 761.
6 (a) K. Hedberg, E. W. Hughes and J. Waser, Acta Crystallogr., 1961,
14, 369; (b) A. C. Rheingold and P. J. Sullivan, Organometallics, 1983,
2, 327; (c) N. C. Lloyd, H. W. Morgan, B. K. Nicholson and R. S.
Ronimus, Angew. Chem., Int. Ed., 2005, 44, 941.
7 R. West and M. J. Fink, Science, 1981, 214, 1343.
8 M. Yoshifuji, I. Shima, N. Inamoto, K. Hirotsu and T. Higuchi, J. Am.
Chem. Soc., 1981, 103, 4587.
9 For the first stable diarsenes and phosphaarsenes, see: (a) A. H. Cowley,
J. G. Lasch, N. C. Norman and M. Pakulski, J. Am. Chem. Soc., 1983,
105, 5506; A. H. Cowley, J. G. Lasch, N. C. Norman, M. Pakulski and
B. R. Whittlesey, J. Chem. Soc., Chem. Commun., 1983, 881; J. Escudie´,
C. Couret, H. Ranaivonjatovo and J. G. Wolf, Tetrahedron Lett., 1983,
24, 3625; C. Couret, J. Escudie´, Y. Madaule, H. Ranaivonjatovo and
J. G. Wolf, Tetrahedron Lett., 1983, 24, 2769; (b) L. Weber, Chem. Rev.,
1992, 92, 1839 and references cited therein.
1406 | Dalton Trans., 2008, 1395–1408
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
10 S. Nagase, S. Suzuki and T. Kurakake, J. Chem. Soc., Chem. Commun.,
1990, 1724.
21, 563; (d) M. Driess, S. Rell and H. Pritzkow, J. Chem. Soc., Chem.
Commun., 1995, 253; (e) K. L. Mckillop, G. R. Gillette, D. R. Powell
and R. West, J. Am. Chem. Soc., 1992, 114, 5203; (f) M. J. Fink, K. J.
Haller, R. West and J. Michl, J. Am. Chem. Soc., 1984, 106, 822; (g) A. J.
Millevolte, D. R. Powell, S. G. Johnson and R. West, Organometallics,
1992, 11, 1091; (h) A. Scha¨fer, M. Weidenbruch, W. Saak, S. Pohl
and H. Marsmann, Angew. Chem., Int. Ed. Engl., 1991, 30, 962; (i) A.
Scha¨fer, M. Weidenbruch, W. Saak, S. Pohl and H. Marsmann, Angew.
Chem., Int. Ed. Engl., 1991, 30, 834; (j) H. Suzuki, N. Tokitoh and
R. Okazaki, Bull. Chem. Soc. Jpn., 1995, 68, 2471; (k) S. Masamune,
S. A. Batcheller, J. Park, W. M. Davis, O. Yamashita, Y. Ohta and Y.
Kabe, J. Am. Chem. Soc., 1989, 111, 1888; (l) H. B. Yokelson, A. J.
Millevolte, G. R. Gillette and R. West, J. Am. Chem. Soc., 1987, 109,
6865; (m) R. West, D. J. Deyoung and K. J. Haller, J. Am. Chem. Soc.,
1985, 107, 4942; (n) R. P.-K. Tan, G. R. Gillette, D. R. Powell and R.
West, Organometallics, 1991, 10, 546.
11 (a) C. Jones, Coord. Chem. Rev., 2001, 215, 151; (b) S. J. Black, D. E.
Hibbs, M. B. Hursthouse, C. Jones and J. W. Steed, Chem. Commun.,
1998, 2199; (c) K. Plossl, G. Huttner and L. Zsolnai, Angew. Chem.,
Int. Ed. Engl., 1989, 28, 446; (d) U. Weber, G. Huttner, O. Scheidsteger
and L. Zsolnai, J. Organomet. Chem., 1985, 289, 357; (e) A. H. Cowley,
N. C. Norman and M. Pakulski, J. Am. Chem. Soc., 1984, 106, 6844;
(f) A. H. Cowley, N. C. Norman, M. Pakulski, D. L. Bricker and D. H.
Russell, J. Am. Chem. Soc., 1985, 107, 8211; (g) G. Huttner, U. Weber,
B. Sigwarth and O. Scheidsteger, Angew. Chem., Int. Ed. Engl., 1982,
21, 215.
12 N. Tokitoh, Y. Arai, T. Sasamori, R. Okazaki, S. Nagase, H. Uekusa
and Y. Ohashi, J. Am. Chem. Soc., 1998, 120, 433.
13 T. Sasamori, Y. Arai, N. Takeda, R. Okazaki, Y. Furukawa, M.
Kimura, S. Nagase and N. Tokitoh, Bull. Chem. Soc. Jpn., 2002, 75,
661.
34 T. Sasamori, E. Mieda, N. Tokitoh, Bull. Chem. Soc. Jpn., in
14 N. Tokitoh, Y. Arai, R. Okazaki and S. Nagase, Science, 1997, 277,
press.
78.
35 T. Sasamori, Doctor Thesis, Kyushu University, Japan,
2002.
15 (a) M. Ates, H. J. Breunig, S. Gulec, W. Offermann, K. Ha¨berle and
M Dra¨ger, Chem. Ber., 1989, 122, 473; (b) M. Ates, H. J. Breunig, K.
Ebert, S. Gu¨lec, R. Kaller and M. Dra¨ger, Organometallics, 1992, 11,
145.
=
36 In the sulfurization reaction of Mes*P PMes*, the corresponding
=
diphosphene-1-sulfide, Mes*P(S) PMes*, was obtained as an isolable
intermediate. (a) M. Yoshifuji, K. Shibayama, I. Shima and N. Inamoto,
Phosphorus Sulfur Relat. Elem., 1983, 18, 11; (b) M. Yoshifuji, K.
Shibayama, N. Inamoto, K. Hirotsu and T. Higuchi, J. Chem. Soc.,
Chem. Commun., 1983, 862.
16 T. Sasamori, Y. Arai, N. Takeda, R. Okazaki and N. Tokitoh, Chem.
Lett., 2001, 30, 42.
17 N. Kano, N. Tokitoh and R. Okazaki, Organometallics, 1998, 17,
1241.
37 T. Sasamori, E. Mieda, N. Takeda and N. Tokitoh, Chem. Lett., 2004,
33, 104.
18 N. Tokitoh, T. Sasamori and R. Okazaki, Chem. Lett., 1998, 27,
725.
38 M. Yoshifuji, K. Shibayama and N. Inamoto, Chem. Lett., 1984, 1984,
19 T. Sasamori, N. Takeda and N. Tokitoh, J. Phys. Org. Chem., 2003, 16,
450.
603.
39 T. Sasamori, E. Mieda, N. Takeda and N. Tokitoh, Angew. Chem., Int.
Ed., 2005, 44, 3717.
40 T. Sasamori, E. Mieda, N. Nagahora and N. Tokitoh, Chem. Heterocycl.
Compd. (N. Y.), 2006, 42, 1603.
41 (a) B. Twamley and P. P. Power, Chem. Commun., 1998, 1979; (b) A. H.
Cowley, J. G. Lasch, N. C. Norman and M. Pakulski, J. Am. Chem.
Soc., 1983, 105, 5506; (c) A. H. Cowley, J. G. Lasch, N. C. Norman, M.
Pakulski and B. R. Whittlesey, J. Chem. Soc., Chem. Commun., 1983,
881.
42 M. Yoshifuji, K. Shibayama, N. Inamoto, T. Matsushita and K.
Nishimoto, J. Am. Chem. Soc., 1983, 105, 2495.
20 A. Dashti-Mommertz and B. Neumu¨ller, Z. Anorg. Allg. Chem., 1999,
625, 954.
21 (a) K. von Deuten and D. Rehder, Cryst. Struct. Commun., 1980, 9,
167; (b) G. Becker, H. Freudenblum and C. Witthauer, Z. Anorg. Allg.
Chem., 1982, 492, 37.
22 F. Calderazzo, R. Poli and G. Pelizzi, J. Chem. Soc., Dalton Trans.,
1984, 11, 2365.
23 B. Twamley, C. D. Sofield, M. M. Olmstead and P. P. Power, J. Am.
Chem. Soc., 1999, 121, 3357.
24 S. Nagase, in The Chemistry of Organic Arsenic, Antimony and Bismuth
compounds, ed. S. Patai, Wiley, New York, 1994, ch. 1, p. 1.
25 T. Sasamori, E. Mieda, N. Nagahora, K. Sato, D. Shiomi, T. Takui,
Y. Hosoi, Y. Furukawa, N. Takagi, S. Nagase and N. Tokitoh, J. Am.
Chem. Soc., 2006, 128, 12582.
43 T. Sasamori, N. Takeda and N. Tokitoh, Chem. Commun., 2000,
1353.
44 T. Sasamori, N. Takeda, M. Fujio, M. Kimura, S. Nagase and N.
Tokitoh, Angew. Chem., Int. Ed., 2002, 41, 139.
26 T. Sasamori and N. Tokitoh, unpublished results.
27 K. Hassler and F. Hofler, Z. Anorg. Allg. Chem., 1978, 443, 125.
28 A. R. Eujen, G. Becker, O. Mundt, M. Westerhausen and C. Witthauer,
J. Mol. Struct., 1983, 98, 265.
29 A. J. Ashe, E. G. Ludwig and J. Oleksyszyn, Organometallics, 1983, 2,
1859.
30 T. Sasamori, E. Mieda, N. Nagahora, N. Takeda, N. Takagi, S. Nagase
and N. Tokitoh, Chem. Lett., 2005, 34, 166.
31 P. Pyykko¨, Chem. Rev., 1988, 88, 563.
32 (a) B. C¸ etinkaya, A. Hudson, M. F. Lappert and H. Goldwhite,
J. Chem. Soc., Chem. Commun., 1982, 609; (b) B. C¸ etinkaya, P. B.
Hitchcock, M. F. Lappert, A. J. Thorne and H. Goldwhite, J. Chem.
Soc., Chem. Commun., 1982, 691; (c) M. Culcasi, G. Gronchi, J. Escudie,
C. Couret, L. Pujol and P. Tordo, J. Am. Chem. Soc., 1986, 108,
3130; (d) S. Shah, S. C. Burdette, S. Swavey, F. L. Urbach and J. D.
Protasiewicz, Organometallics, 1997, 16, 3395; (e) T. L. Allen, A. C.
Scheiner and H. F. Schaefer, J. Phys. Chem., 1990, 94, 7780; (f) H.
Binder, B. Riegel, G. Heckmann, M. Moscherosch, W. Kaim, H. G.
vonSchnering, W. Honle, H. J. Flad and A. Savin, Inorg. Chem., 1996,
35, 2119; (g) J. Geier, J. Harmer and H. Gru¨tzmacher, Angew. Chem.,
Int. Ed., 2004, 43, 4093; (h) A. J. Bard, A. H. Cowley, J. E. Kilduff,
J. K. Leland, N. C. Norman, M. Pakulski and G. A. Heath, J. Chem.
Soc., Dalton Trans., 1987, 249; (i) M. Geoffroy, A. Jouaiti, G. Terron,
M. Cattanilorente and Y. Ellinger, J. Phys. Chem., 1992, 96, 8241;
(j) K. Tsuji, Y. Fujii, S. Sasaki and M. Yoshifuji, Chem. Lett., 1997, 26,
855.
45 M. Yoshifuji, T. Sato and N. Inamoto, Chem. Lett., 1988, 1735.
46 K. Ito and S. Nagase, Chem. Phys. Lett., 1986, 126, 531.
47 For a review, see: T. Baumgartner and R. Reau, Chem. Rev., 2006, 106,
4681.
48 (a) V. A. Wright and D. P. Gates, Angew. Chem., Int. Ed., 2002, 41,
2389; (b) V. A. Wright, B. O. Patrick, C. Schneider and D. P. Gates,
J. Am. Chem. Soc., 2006, 128, 8836; (c) R. C. Smith, X. F. Chen and
J. D. Protasiewicz, Inorg. Chem., 2003, 42, 5468; (d) R. C. Smith and
J. D. Protasiewicz, J. Am. Chem. Soc., 2004, 126, 2268; (e) R. C. Smith
and J. D. Protasiewicz, Eur. J. Inorg. Chem., 2004, 998; (f) S. Shah, T.
Concolino, A. L. Rheingold and J. D. Protasiewicz, Inorg. Chem., 2000,
39, 3860; (g) C. Dutan, S. Shah, R. C. Smith, S. Choua, T. Berclaz, M.
Geoffroy and J. D. Protasiewicz, Inorg. Chem., 2003, 42, 6241; (h) V. A.
Wright, B. O. Patrick, C. Schneider and D. P. Gates, J. Am. Chem. Soc.,
2006, 128, 8836.
49 T. Sasamori, A. Tsurusaki, N. Nagahora, K. Matsuda, Y. Kanemitsu,
Y. Watanabe, Y. Furukawa and N. Tokitoh, Chem. Lett., 2006, 35,
1382.
50 (a) N. Nagahora, T. Sasamori, N. Takeda and N. Tokitoh, Chem.–
Eur. J., 2004, 10, 6146; (b) N. Nagahora, T. Sasamori, N. Takeda and
N. Tokitoh, Organometallics, 2005, 24, 3074.
51 N. Nagahora, T. Sasamori and N. Tokitoh, Chem. Lett., 2006, 35,
220.
52 N. Nagahora, T. Sasamori, Y. Watanabe, Y. Furukawa and N. Tokitoh,
Bull. Chem. Soc. Jpn., 2007, 80, 1884.
53 T. Sasamori, T. Matsumoto, N. Takeda and N. Tokitoh,
Organometallics, 2007, 26, 3621.
54 R. Pietschnig and E. Niecke, Organometallics, 1996, 15, 891.
55 (a) T. Tsutsumi, M. Okubo, N. Yasuoka and Y. Katsube, Bull.
Chem. Soc. Jpn., 1988, 61, 237; (b) M. Kurihara, T. Matsuda, A.
33 (a) T. Tsumuraya, Y. Kabe and W. Ando, J. Chem. Soc., Chem.
Commun., 1990, 1159; (b) T. Tsumuraya, S. Sato and W. Ando,
Organometallics, 1988, 7, 2015; (c) N. Tokitoh, K. Kishikawa, R.
Okazaki, T. Sasamori, N. Nakata and N. Takeda, Polyhedron, 2002,
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 1395–1408 | 1407
©

View Article Online
Hirooka, T. Yutaka and H. Nishihara, J. Am. Chem. Soc., 2000, 122,
12373.
56 C.Moser, M.Niegerand R.Pietschnig, Organometallics, 2006, 25, 2667.
57 L. Bourget-Merle, M. F. Lappert and J. R. Severn, Chem. Rev., 2002,
102, 3031.
58 (a) B. Y. Liu, C. Y. Tian, L. Zhang, W. D. Yan and W. J. Zhang, J. Polym.
Sci., Part A: Polym. Chem., 2006, 44, 6243; (b) D. Zhang, G. X. Jin,
L. H. Weng and F. S. Wang, Organometallics, 2004, 23, 3270; (c) X. F.
Li, K. Dai, W. P. Ye, L. Pan and Y. S. Li, Organometallics, 2004, 23,
1223.
59 (a) P. L. Floch, Cood. Chem. Rev., 2006, 250, 627; (b) F. Ozawa, S.
Kawagishi, T. Ishiyama and M. Yoshifuji, Organometallics, 2004, 23,
1325.
60 Although there have been a few reports on the theoretical calculations
for a phosphorus analogue of a Shiff-base ligand from the viewpoint
of catalytic activities, no example of such a unique ligand has been
synthesized so far. (a) M. S. W. Chan, L. Q. Deng and T. Ziegler,
Organometallics, 2000, 19, 2741; (b) T. Z. Zhang, D. W. Guo, S. Y. Jie,
W. H. Sun, T. Li and X. Z. Yang, J. Polym. Sci., Part A: Polym. Chem.,
2004, 42, 4765.
1408 | Dalton Trans., 2008, 1395–1408
This journal is
The Royal Society of Chemistry 2008
©