Synthesis and characterization of titanium tetraisocyanide complexes, [CpTi(CNXyl)4E], E = I, SnPh3, and SnMe3

Article abstract of DOI:10.1016/j.jorganchem.2007.11.047

Oxidation of [CpTi(CO)₄]^- by I₂, Ph₃SnCl, and Me₃SnCl in the presence of four equivalents of CNXyl, Xyl = 2,6-dimethylphenyl, affords unprecedented titanium tetraisocyanide complexes, [CpTi(CNXyl)₄E], E = I, SnPh₃, SnMe₃. These have been isolated and characterized by spectroscopic methods as well as single-crystal X-ray crystallography. A by-product of the iodine reaction was the Ti(III) complex, [CpTi(CNXyl)₂I₂], which was also characterized by X-ray crystallography.

Full text of DOI:10.1016/j.jorganchem.2007.11.047

Available online at www.sciencedirect.com
Journal of Organometallic Chemistry 693 (2008) 1536–1542
www.elsevier.com/locate/jorganchem
Synthesis and characterization of titanium tetraisocyanide
q
complexes, [CpTi(CNXyl)4E], E = I, SnPh3, and SnMe3
Jessica M. Allen, John E. Ellis *
Department of Chemistry, University of Minnesota, Minneapolis, MN 55455, USA
Received 28 September 2007; accepted 5 November 2007
Available online 4 December 2007
Dedicated to the memory of Professor F. Albert Cotton, for his wonderful support of and contributions to fundamental research
in inorganic and organometallic chemistry.
Abstract
Oxidation of [CpTi(CO)4]ꢀ by I2, Ph3SnCl, and Me3SnCl in the presence of four equivalents of CNXyl, Xyl = 2,6-dimethylphenyl,
affords unprecedented titanium tetraisocyanide complexes, [CpTi(CNXyl)4E], E = I, SnPh3, SnMe3. These have been isolated and char-
acterized by spectroscopic methods as well as single-crystal X-ray crystallography. A by-product of the iodine reaction was the Ti(III)
complex, [CpTi(CNXyl)2I2], which was also characterized by X-ray crystallography.
Ó 2007 Elsevier B.V. All rights reserved.
Keywords: Titanium; Isocyanides; Tin; X-ray structure
1. Introduction
complexes of these metals have been reported previously,
including CpTi(CNR)2Cl2, R = C6H11 [7], Xyl [8]; Cp2Ti-
(CNXyl)Cl [9], Cp2M(CO)(CNR), M = Ti, Zr, Hf, R =
t-Bu; M = Ti, Zr, R = Xyl [10a], Cp2M(CNXyl)2, M =
Ti, Zr [10b], cis-TiCl4(CN-t-Bu)2 [11], [TiCl4(CNXyl)]2
[12], and trans-[Ti(TPP)(CN-t-Bu)2], TPP = meso-tetra-p-
tolylporphyrin [13]. In this article, we describe the first tita-
nium tetraisocyanide complexes, [CpTi(CNXyl)4E], E = I,
SnPh3, SnMe3.
Isocyanide complexes of metals date back to 1856, when
Meyer, and later Gautier, obtained silver monoisocyanide
complexes, of the composition Ag(CN)(CNR), by the
treatment of silver cyanide with alkyl iodides [2]. Subse-
quently, numerous di- and poly-isocyanide complexes have
been reported for most transition metals [3,4], but rela-
tively little progress in this area has been achieved for the
group 4 elements. For example, no poly-isocyanide
complexes of Ti, Zr, or Hf are known, nor have any isocy-
anide complexes containing these elements in zero- or
lower-valent states been described in the scientific litera-
ture. In view of recent progress in the synthesis of novel
group 5 isocyanide complexes, including [V(CNXyl)6]0
[5], [Ta(CNXyl)7]+ and [Nb(CNXyl)6]ꢀ [6], Xyl = 2,6-
dimethylphenyl, extension of this research to the group 4
elements was of interest. Mono- and di-isocyanide
2. Experimental
2.1. General information
All reactions were carried out under an argon atmo-
sphere with a double manifold vacuum line that has been
previously described [14]. See a prior paper for a discussion
of general procedures and the purification of solvents [15].
[Et4N][CpTi(CO)4] was prepared according to the literature
procedure [16]. All other reagents were obtained from com-
mercial sources and freed of dioxygen and moisture by
standard methods [17] before use.
q
Highly Reduced Organometallics, Part 64. For Part 63, see Ref. [1].
Corresponding author. Tel.: +1 612 6256391; fax: +1 612 6267541.
*
E-mail address: ellis@chem.umn.edu (J.E. Ellis).
0022-328X/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.11.047

J.M. Allen, J.E. Ellis / Journal of Organometallic Chemistry 693 (2008) 1536–1542
1537
2.2. Synthesis of [CpTi(CNXyl)4I] (1)
sequentially with cold (ꢀ70 °C) colorless solutions of
Ph3SnCl (0.571 g, 1.4 mmol) and CNXyl (0.744 g,
5.7 mmol) in toluene (30 mL and 20 mL, respectively) with
vigorous stirring. On warming the reaction mixture slowly
to room temperature over a period of about 10 h, it grad-
ually changed from bright red to bright purple. After stir-
ring for five days under ambient conditions, the mixture
was filtered and solvent was removed under vacuum from
the filtrate. Free CNXyl was separated by sublimation.
The product was then recrystallized from toluene to
provide satisfactorily pure deep purple 3 (0.606 g 44%),
m.p. 147–149 °C (dec). Elemental Anal. Calc. for
C59H56N4SnTi: C, 71.75; H, 5.71. Found: C, 71.13; H,
5.46%. IR (THF), m(CN): 2005 sh, 1977 vs cmꢀ1; IR (tolu-
Cold toluene (100 mL, ꢀ78 °C) was added to a solid mix-
ture of [Et4N][CpTi(CO)4] (1.00 g, 2.82 mmol) and CNXyl
(1.50 g, 11.4 mmol). A solution of I2 (0.716 g, 2.82 mmol) in
toluene (60 mL, 20 °C) was then added via cannula with stir-
ring. The mixture immediately turned brown and began
evolving carbon monoxide (Caution: carbon monoxide is a
very toxic gas, so this procedure and similar ones described
below, must be carried out in a well-ventilated hood). It
was then stirred overnight while slowly warming to room
temperature and after 48 h, the reaction mixture was royal
purple. Following filtration, the solvent was removed under
vacuum to afford an air-sensitive dark purple solid. Excess
CNXyl was removed by sublimation (20 °C, 0.1 torr). The
compound was recrystallized from toluene to afford satisfac-
torilypure dark purple microcrystals of1(2.013 g, 94%), m.p.
120–122 °C (dec). Elemental Anal. Calc. for C41H41IN4Ti: C,
64.16; H, 5.38. Found: C, 64.60; H, 5.23%. IR (toluene),
1
ene), m(CN): 2005 sh, 1976 vs cmꢀ1. H NMR (500 MHz,
C6D6, 25 °C): d 2.15 (s, 24 H, o-CH3), 5.93 (s, with satel-
lites, JSn–H = 11.5 Hz, 5H, C5H5), 6.7–6.9 (m, m- and p-H
3
in SnPh3, CNXyl), 7.85 (dd with satellites, JH–H = 8 Hz,
4JH–H = 1.5 Hz, JSn–H = 31.8 Hz, o-H in SnPh3). 13C{1H}
NMR (75 MHz, C6D6, 25 °C): d 19.6 (s, o-CH3), 95.5 (s,
C5H5), 126.5, 127.0, 128.3, 133.3 (s, phenyl in CNXyl),
127.6 (s with satellites, JSn–C = 22 Hz, o- or m-C in SnPh3),
130.0 (s with satellites, JSn–C = 14 Hz, p-C in SnPh3), 139.1
(s with satellites, JSn–C = 30 Hz, m or o-C in SnPh3), 153.5
(s with satellites, JSn–C = 48 Hz, i-C in SnPh3), 214.1 (s,
CNXyl) ppm. 119Sn NMR (112 MHz, CDCl3, 25 °C,
SnMe4 ref): d ꢀ69.0 s ppm. X-ray quality single-crystals
of 3 were grown as purple blocks from a pentane-layered
concentrated toluene solution at 20 °C over a four month
period.
1
m(CN): 2035 vs, 1996 w cmꢀ1. H NMR (500 MHz, C6D6,
25 °C): d 2.42 (s, 24 H, o-CH3), 5.50 (s, 5H, C5H5), 6.68
(m, 12 H, m and p-H). 13C{1H} NMR (125 MHz, C6D6,
24 °C): d 19.9 (s, o-CH3), 95.8 (s, C5H5), 127.1, 128.9, 129.7,
133.8 (s, phenyl carbons), 197.9 (s, CNXyl).
X-ray quality single-crystals of 1 ꢁ 1/2 (pentane) were
grown as dichroic plates, i.e., purple by reflected light
and green by transmitted light, from a pentane-layered
THF solution at 20 °C after several days under an argon
atmosphere. The formula unit contains 0.5 pentane sol-
vent molecules, disordered over an inversion center.
2.3. Synthesis of [CpTi(CNXyl)2I2] (2)
2.5. Synthesis of [CpTi(CNXyl)4(SnMe3)] (4)
Treatment of 1 (0.167 g, 0.218 mmol) with I2 (0.028 g,
0.11 mmol) in cold toluene (40 mL, ꢀ65 °C), followed by
warming to room temperature, with constant stirring, over
a period of about 5 h, resulted in an apple-green solution.
The solution was filtered and the solvent removed under
vacuum, giving a bright green solid. However, attempts
to separate the product from free CNXyl invariably
resulted in partial to nearly complete decomposition to give
a brownish solution, from which only small amounts of
brownish-green impure solid 2 was obtained. An IR spec-
trum of impure 2 in THF or toluene showed a weak broad
absorption centered at about 2156 cmꢀ1, due to product,
along with an intense band due to free CNXyl, m(CN):
2116 cmꢀ1 in THF. However, X-ray quality single-crystals
of 2 were grown as green plates from a pentane layered-
THF solution at room temperature and provided unambig-
uous evidence for the presence of this substance in the reac-
tion mixture. The IR spectrum of single-crystals of 2 in
solution were identical to those shown above.
By the same procedure employed to prepare 3, toluene
suspensions or solutions of [Et4N][CpTi(CO)4] (0.500 g,
1.4 mmol), Me3SnCl (0.279 g, 1.4 mmol), and CNXyl
(0.744 g, 5.7 mmol) were combined at ꢀ70 °C. After
warming to room temperature and stirring for 5 days,
the solution was deep purple. Following filtration and
purification as described for 3, satisfactorily pure black
(or intensely deep purple) microcrystals of 4 were
obtained (0.339 g, 31%), m.p. 133–135 °C (dec). Elemental
Anal. Calc. for C44H50N4SnTi: C, 65.94; H, 6.29. Found:
C, 66.62; H, 6.84%. IR (THF), m(CN): 2005 sh, 1969 vs
cmꢀ1; IR (toluene), m(CN): 2005 sh, 1969 vs cmꢀ1 1H
.
NMR (300 MHz, C6D6, 25 °C): d 0.42 (s with satellites,
JSn–H = 29 Hz, 9H, SnMe3) 2.37 (s, 24H, o-CH3), 5.37
(s with satellites, JSn–H = 11.4 Hz, 5H, C5H5), 6.79 (m,
12 H, m- and p-H) ppm. 13C{1H} NMR (75 MHz,
C6D6, 25 °C): d ꢀ1.8 (s with satellites, JSn–C = 45 Hz,
SnMe3), 19.7 (s, o-CH3), 94.4 (s, C5H5), 126.8, 128.5,
130.3, 132.5 (s, phenyl CNXyl), 222.2 (s, CNXyl) ppm.
119Sn{1H} NMR (186 MHz, C6D6, 25 °C, SnMe4 ref): d
ꢀ17.4 (s) ppm. X-ray quality single-crystals of 4 were
grown as black blocks from a pentane-layered concen-
trated diethyl ether solution at ꢀ20 °C over a period of
several weeks.
2.4. Synthesis of [CpTi(CNXyl)4(SnPh3)] (3)
A slurry of bright red [Et4N][CpTi(CO)4] (0.500 g,
1.4 mmol) in cold toluene (20 mL, ꢀ70 °C) was treated

1538
J.M. Allen, J.E. Ellis / Journal of Organometallic Chemistry 693 (2008) 1536–1542
2.6. X-ray crystallographic analyses of (1)–(4)
followed by the addition of a slight excess, seven equiva-
lents, of 2,6-dimethylphenyl isocyanide, CNXyl, to afford
70–80% isolated yields of the monovalent metal complexes,
[M(CNXyl)6I], M = Nb, Ta [6], which represented the first
fully substituted isocyanide derivatives of the unknown
[M(CO)6I] for these elements:
Crystallographic data for compounds 1–4 are summa-
rized in Table 1. Procedures for the growth of X-ray qual-
ity crystals are described under their syntheses. A selected
crystal was coated with viscous oil and attached to a glass
fiber under an inert atmosphere and mounted on a Siemens
SMART Platform CCD diffractometer for data collection
(Mo Ka radiation, graphite monochrometer). The intensity
data were integrated using SAINT [18] and corrected for
absorption and decay using SADABS [19]. Structures were
solved by direct methods using SHELXL-97 software [20].
All non-hydrogen atoms were refined with anisotropic dis-
placement parameters, and all hydrogen atoms were placed
in ideal positions and refined as riding atoms with relative
isotropic displacement parameters. Molecular structures
and selected interatomic data for 1–4 will be shown and
discussed in the following section.
½MðCOÞ ꢂꢀ þ I2 þ 6CNXyl
6
THF
½MðCNXylÞ Iꢂ þ Iꢀ þ 6CO
ð1Þ
ꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀ!
6
ꢀ78 to þ 20 ꢃC
60 h
We were surprised that the CNXyl ligand was able to
replace all carbonyls in view of prior reports that similar
reactions only gave mixed isocyanide carbonyl complexes,
such as [Nb(CO)2(CNXyl)4I] and [Ta(CO)3[(CN-t-Bu)3I]
[21].
Our attention then turned to the possibility of extending
this research to the group 4 elements, because, as men-
tioned in the introduction, many opportunities for exciting
new discoveries exist in this poorly explored area of tita-
nium, zirconium and hafnium chemistry, particularly in
low-valent systems. Herein we describe some new results
involving the first carbonyltitanate complex, [CpTi(CO)4]ꢀ,
originally reported in 1986 [22].
3. Results and discussion
3.1. Syntheses of 1–4
Recently we reported on the oxidation of the hexacar-
bonylmetalates (1-) of niobium and tantalum by I2,
Table 1
Crystal data, data collection, and solution refinement for 1–4
Compound
1
2
3
4
Empirical formula
Formula weight
Crystal color, habit
Crystal size (mm)
Crystal system
C
800.65
43.5H47IN4Ti
C23H23I2N2Ti
629.13
Green, plate
0.47 ꢄ 0.38 ꢄ 0.14
Triclinic
C59H59N4SnTi
987.67
Purple, block
0.40 ꢄ0.40 ꢄ 0.30
Triclinic
C44H50N4SnTi
801.47
Black, block
0.45 ꢄ 0.45 ꢄ 0.35
Triclinic
Purple/green plate
0.41 ꢄ 0.31 ꢄ 0.18
Monoclinic
P21/c
11.287(1)
16.460(2)
21.816(2)
90
93.843(2)
90
4044.0(7)
4
1.315
1.009
173(2)
1.55–27.53
ꢀ14 6 h 6 14,
ꢀ21 6 k 6 21,
ꢀ28 6 l 6 28
46276
ꢀ
ꢀ
ꢀ
Space group
P1
P1
P1
˚
a (A)
8.1280(9)
8.4359(9)
17.586(2)
99.646(2)
93.812(2)
96.739(2)
1176.0(2)
2
1.777
3.000
173(2)
2.47–27.52
ꢀ10 6 h 6 10,
ꢀ10 6 k 6 10,
ꢀ22 6 l 6 22
13546
15.703(3)
17.301(3)
18.432(3)
86.133(7)
89.420(6)
87.057(7)
4990(1)
14.643(3)
16.108(3)
20.490(6)
106.251(4)
103.020(4)
108.970(3)
4113(2)
˚
b (A)
˚
c (A)
a (°)
b (°)
c (°)
3
˚
V (A)
Z
4
4
qcalc (Mg mꢀ3
)
1.315
0.702
173(2)
1.18–27.51
ꢀ20 6 h 6 20,
ꢀ22 6 k 6 22,
ꢀ23 6 l 6 23
59051
1.294
0.835
173(2)
1.10–25.09
ꢀ17 6 h 6 17,
ꢀ19 6 h 6 19,
ꢀ24 6 l 6 24
39090
l(Mo Ka) (nmꢀ1
Temperature (K)
h Range (°)
)
Index ranges
Reflections collected
Unique reflections
Raint
9287
0.0377
5313
0.0340
22661
0.0220
14498
0.0445
Data/restraints/parameters
Reflections with I > 2r(I)
Rb1; wRc2 (I > 2r(I))
R1; wR2 (all data)
Goodness-of-fit on F2
9287/31/469
7714
0.0300; 0.0770
0.0406; 0.0832
1.028
5313/0/254
4789
0.0287; 0.0742
0.0328; 0.0766
1.019
22661/0/1171
19959
0.0344; 0.0762
0.0425; 0.0786
1.133
14498/21/934
13056
0.0598; 0.1507
0.0672; 0.1562
1.085
ꢀ3
˚
Largest difference peak/hole (e A
)
0.995/ꢀ0.331
1.187/ꢀ0.917
0.858/ꢀ0.446
3.057/ꢀ1.733
P
P
a
b
c
Rint
¼
jF 2o ꢀ hF 2cij= jF o2j.
P
P
R1 ¼ jF oj ꢀ jF cj= jF oj:
h
i
1=2
P
P
2
2
wR2 ¼
wðF 2o ꢀ F c2Þ = ðwðF 2oÞ Þ
:

J.M. Allen, J.E. Ellis / Journal of Organometallic Chemistry 693 (2008) 1536–1542
1539
Prior work showed that treatment of [CpTi(CO)4]ꢀ with
I2 at ꢀ70 °C in THF or CH2Cl2 produced solutions of an
exceedingly unstable deep red substance. This species
was tentatively identified as ‘‘CpTi(CO)4I” [16], on the
basis of its reaction with Na+C5H5ꢀ, which gave small
amounts of the prototypical titanium carbonyl complex
[Cp2Ti(CO)2] [23]:
by exchange of one or more CO groups by better r-donor
ligands, such as isocyanides. Thus, when a slight excess of
CNXyl was added to an initial mixture of [Et4N]-
[CpTi(CO)4] and R3SnCl, R = Ph, Me, in toluene, under
the same conditions employed in the synthesis of 1, see
above, the corresponding mixed tin–titanium complexes,
[CpTi(CNXyl)4SnR3], R = Ph, 3, and Me, 4, were obtained
as thermally stable dark purple and nearly black microcrys-
tals. Isolated yields of 3 and 4 were appreciably lower, ca.
30–40%, than those for 1 mainly due to the much greater
solubility of these species in hydrocarbon solvents and sub-
sequent difficulty in their separation from uncoordinated
CNXyl:
THF
½CpTiðCOÞ ꢂꢀ þ I2
“CpTiðCOÞ I”
ꢀꢀꢀꢀꢀ!
4
4
ꢀ70 ꢃC
ꢀIꢀ
THF;Cpꢀ
½Cp TiðCOÞ ꢂ þ 2CO
ꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀ!
2
2
ꢀ70 to þ 20 ꢃC
ꢀIꢀ
½CpTiðCOÞ ꢂꢀ þ R3SnCl þ 4CNXyl
4
toluene
ð2Þ
½CpTiðCNXylÞ SnR ꢂ þ 4CO
ð4Þ
ꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀ!
3
4
ꢀ70 to þ 20 ꢃC
To obtain more evidence for the existence of the presumed
precursor, [CpTi(CO)4I], an analogous iodination of
[CpTi(CO)4]ꢀ was carried out in the presence of slightly
more than four equivalents of CNXyl. Also, the reaction
was conducted in toluene, rather than THF or CH2Cl2,
previously employed in related reactions [16]. The latter
two solvents gave very poor conversion to product, likely
due to the instability of intermediates in these solutions.
Because [Et4N][CpTi(CO)4] is essentially insoluble in tolu-
ene, the reaction proceeded slowly and required about 2
days to achieve optimum results and provided about 90%
isolated yields of dark purple [CpTi(CNXyl)4I], 1:
ca: 5 days
ꢀIꢀ
Spectral and X-ray data for 3 and 4 will be discussed
below.
3.2. Characterizations of 1–4
Selected spectral data for the coordinated isocyanide
groups in 1–4 are shown in Table 2. IR spectra for
[CpTi(CNXyl)4E], E = I, 1; SnPh3, 3; SnMe4, 4, in the iso-
cyanide m(CN) region show one intense (and broad peak),
suggestive of pseudo-octahedral structures of the general
formula trans-LL0M(CNR)4, where the g5-C5H5 groups
are approximately opposite to the iodo or triorganotin
units, in accord with the molecular structures for these spe-
cies observed in the solid state, see below. Qualitatively
similar IR spectra in the m(CO) region were previously
reported for the related [Cp*Ti(CO)4E] complexes, E = I,
SnPh3, and AuPPh3, and analogous structures were
proposed for these carbonyls on this basis [16]. As
expected, the m(CN) positions of the most intense peak
for [CpTi(CNXyl)4E] decrease in the order E = I >
SnPh3 > SnMe3, which is just the opposite of the relative
donor abilities of these substituents. It is noteworthy that
the m(CN) values are well below that for free CNXyl,
2116 cmꢀ1, which indicates that the isocyanide groups in
1, 3, and 4 are functioning as net acceptor ligands in these
formally Ti(II) electron rich complexes. In contrast, the
infrared spectrum of [CpTi(CNXyl)2I2], 2, a formal Ti(III)
complex, shows a single broad m(CN) absorption at
½CpTiðCOÞ ꢂꢀ þ I2 þ 4CNXyl
4
toluene
½CpTiðCNXylÞ Iꢂ þ 4CO
ð3Þ
ꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀ!
4
ꢀ70 to þ 20 ꢃC
48 h
ꢀIꢀ
The product is of interest as a completely substituted isocy-
anide complex of the unknown [CpTi(CO)4I], and its for-
mation lends credence to the premise that the carbonyl
iodide complex is a thermally unstable precursor in this
reaction. A green by-product, obtained in small amounts
from this reaction, was the Ti(III) complex CpTi(CN-
Xyl)2I2, 2, an iodo-analog of the previously known and
also green CpTi(CNXyl)2Cl2 [8]. Compound 2 was also
obtained by the reaction of 1 with 0.5 equivalent of I2.
However, attempts to obtain acceptably pure bulk samples
of 2 from this reaction were unsuccessful due to its appar-
ent instability in THF or toluene. Spectral and single-crys-
tal X-ray crystallographic structural data for 1 and 2 will
be presented and discussed below.
Reactions of [CpTi(CO)4]ꢀ with the organotin electro-
philes, R3SnCl, R = Ph, Me, also provided thermally
unstable products proposed to have the corresponding for-
mulas [CpTi(CO)4SnR3]. IR and NMR spectral data for
the triphenylstannyl adduct have been previously reported
[16]. Both of these formally Ti(II) complexes are believed to
be unstable due to the lability of their CO groups and on
this basis it was anticipated that they would be stabilized
Table 2
Selected IR and NMR data for 1–4 and CNXyl in toluene
Compound
m(CN) (cmꢀ1
)
13C d(CN) (ppm)
1
2035 br vs, 1996 w
2156 br vs
2005 sh, 1976 br vs
2005 sh, 1969 br vs
2116 vs
197.9
–
214.1
222.2
165.4
2
3
4
CNXyl

1540
J.M. Allen, J.E. Ellis / Journal of Organometallic Chemistry 693 (2008) 1536–1542
2156 cmꢀ1, which is 40 cmꢀ1 above that of the free ligand.
Notably, the previously reported chloro-analog of 2,
[CpTi(CNXyl)2Cl2], which was not structurally character-
ized, is also a green substance and shows a single intense
m(CN) band at 2160 cmꢀ1 [18]. IR data for 2 and the
chloro-analog indicate that the isocyanide ligands in these
much less electron rich Ti(III) complexes function primar-
ily as donor ligands. These IR spectral changes for 1–4 are
entirely consistent qualitatively with expectations based on
the classic Chatt–Dewar–Duncanson model for back-
bonding [24]. However, the IR m(CN) peaks for 1, 3, and
4 are much broader than corresponding peaks for the
[Cp*Ti(CO)4E] complexes, due to the non-linearity of the
M-CNR units, which effectively decreases the symmetries
of isocyanide complexes compared to carbonyl analogs.
Cotton and Zingales first reported on this phenomenon
nearly 50 years ago [25]. A recent example involved com-
parison of IR spectra of [Ta(CNXyl)6]ꢀ and [Ta(CO)6]ꢀ
[6]. Prominent shoulders observed in the IR m(CN) spectral
region for 3 and 4, but not 1, suggests that the bulky char-
acter of the R3Sn groups in the former two complexes, also
contributes to further reduction in the local symmetry of
the titanium. A similar bifurcation of the IR m(CN) band
envelope in THF solution was recently reported for trans-
[Fe(CNXyl)4(SnPh3)2], which contains a Fe(CN)4Sn2 core
structure of nearly D4h symmetry in the solid state [26].
1H and 13C NMR spectra of diamagnetic 1, 3, and 4 are
summarized in Section 2 under the respective compounds
and show normal resonances and positions for the Cp
and aliphatic/aromatic portions of the coordinated CNXyl
ligands. In addition, unexceptional 1H, 13C and 119Sn
NMR spectra for the Ph3Sn and Me3Sn ligands in 3 and
4, respectively, were observed. In Table 2, the ligated isocy-
anide carbon 13C NMR resonances are collected for the
new complexes, along with that of free CNXyl in toluene.
They show the expected trend, i.e., the isocyanide dC values
become more positive as the transition metal center
becomes more electron rich [27]. Thus, these results are
consistent with the IR m(CN) data for [CpTi(CNXyl)4E],
where the relative donor ability of the substituents increase
in the order E = I < SnPh3 < SnMe3.
Fig. 1. Molecular structure of 1. Thermal ellipsoids are set at the 50%
probability level, with hydrogens omitted for clarity. Selected distances
˚
(A) and angles (°): Ti–C(1) 2.137(3), Ti–C(2) 2.122(3), Ti–C(3) 2.144(3),
Ti–C(4) 2.149(3), Ti–C(5) 2.427(3), Ti–C(6) 2.384(3), Ti–C(7) 2.326(3), Ti–
C(8) 2.333(3), Ti–C(9) 2.391(3), Ti–I 3.0541(6), C(1)–N(1) 1.162(4), C(2)–
N(2) 1.171(4), C(3)–N(3) 1.167(4), C(4)–N(4) 1.165(4), Cp(centroid)–Ti–I
179.6.
Fig. 2. Molecular structure of 2. Thermal ellipsoids are set at the 50%
probability level, with hydrogens omitted for clarity. Selected distances
˚
(A) and angles (°): Ti–C(1) 2.180(3), Ti–C(2) 2.185(3), Ti–C(3) 2.360(3),
Ti–C(4) 2.351(3), Ti–C(5) 2.334(3), Ti–C(6) 2.324, Ti–C(7) 2.344(3), Ti–
I(1) 2.7454(5), Ti–I(2) 2.7699(6), C(1)–N(1) 1.155(3), C(2)–N(2) 1.153(4),
I(1)–Ti–I(2) 115.19(2), av Cp(centroid)–Ti–C 106.4, av Cp(centroid)–Ti–I
122.7.
The latter has a K[NiCl3] lattice-type [31] containing poly-
nꢀ
meric [Ti(l-I)3]n units with bridging Ti–I bonds, which
Single-crystal X-ray structural characterizations were
carried out to confirm the formulations of 1–4 and to pro-
vide interatomic data for these unusual compounds.
Molecular structures of the iodo complexes, 1 and 2 are
shown in Figs. 1 and 2, respectively. Structural features
and interatomic data, which are common to 1–4 will be
summarized and discussed below and compared to previ-
ously reported structures of related organotitanium com-
are longer than terminal Ti–I bonds in similar environ-
ments. For example, in [{TiI2(PMe3)2}2(l-I)2], which con-
tains six-coordinate Ti(III), the average Ti–I (terminal)
˚
and Ti–I (bridging) distances are 2.69(1) and 2.86(2) A,
respectively [32]. It is also noteworthy that the Ti–I dis-
˚
tance in 1 is about 0.4 A longer than the sum of standard
˚
covalent radii of titanium and iodine, 2.65 A [33].
Undoubtedly the anomalously long Ti–I distance in 1 is
due to the presence of the four bulky xylylisocyanide
groups, which, along with the g5-Cp group, provide a very
crowded environment about the eight-coordinate titanium
center. Although 1 behaves as a non-electrolyte, for exam-
ple, it is soluble in saturated and unsaturated hydrocarbons
and its IR spectra are nearly identical in THF and toluene,
available data do not rule out the possibility that the Ti–I
interaction in 1 is predominantly electrostatic, in which
˚
plexes. However, the Ti–I distances in 1, 3.0541(6) A, and
˚
2, (av) 2.76(2) A, are quite different and warrant separate
consideration. The latter value is normal and statistically
identical to the corresponding distances reported for
à
[Cp2TiI2], 2.7690(7) [28] and [Cp2 TiI], 2.759(2), where
Cpà = C5Me4H [29]. In contrast the Ti–I distance in 1 is
˚
about 0.15 A longer than the previous ‘‘record” value
˚
observed in the Ti(II) complex, Cs[TiI3], 2.8992(7) A [30].

J.M. Allen, J.E. Ellis / Journal of Organometallic Chemistry 693 (2008) 1536–1542
1541
Table 3
case it would contain a 16 electron cation, [CpTi(CN-
Xyl)4]+, tightly ion-paired to Iꢀ. In this regard, attempts
to obtain bona fide salts containing this cation, such as
[CpTi(CNXyl)4][B(C6F5)4] will be of interest in shedding
more light on the constitution of 1.
˚
Selected average interatomic distances (A) and angles (°) for 1–4
1
2
3
4c
Ti–C(CNR)
Ti–C(Cp)
Ti–Cp(centroid)
C–N (CNR)a
Ti–C–N
2.14(1)
2.37(4)
2.047(1)
1.166(4)
176(2)
172(4)
86(3)
150.8(3)
75.5(3)
2.183(3)
2.34(2)
2.021(1)
1.154(4)
174(3)
174(2)
–
2.115(4)
2.37(4)
2.048(1)
1.170(3)
173(1)
173(2)
75(2)
2.10(1)
2.36(2)
2.035(3)
1.18(1)
176(1)
167(5)
72(4)
The molecular structures of 3 and 4 are extremely similar
and the one for 3 is shown in Fig. 3. Both structures are sim-
ilar to that of 1 and are fully compatible with the solution
IR and NMR data previously discussed for these species.
The Ti–Sn distances in [CpTi(CNXyl)4E], E = SnPh3,
C–N–C
cis-C–Ti–Ca
trans-C–Ti–Ca
C–Ti–Xb
147.0(1)
81(2)
149.8(6)
75(2)
142.4(8)
72(4)
˚
2.9839(6), and SnMe3, 2.949(1) A, are longer than values
a
For ligated C of isocyanides.
for prior compounds containing Ti–Sn bonds, but not
anomalously long, as in the case of the Ti–I bond in 1.
For example, Ti–Sn distances in previously known triorga-
b
c
For ligated C of isocyanides, X = I (1, 2) or Sn (3, 4).
Cp(centroid)–Ti–Sn 174.4°.
˚
notin–titanium complexes range from 2.83(2) A in
ꢀ
[Ti(CO)5(SnPh3)2]2ꢀ [34] to 2.921(1) A in [Ti(CO)6(SnR3)] ,
[CpTiCl3] [38] to 2.049 A in the Ti(0) complex
[CpTi(CO)4]ꢀ [22]. Corresponding Ti–Cp(centroid) dis-
tances for 1, 3 and 4 are somewhat longer than that of 2,
likely owing to both the higher coordination number (eight
versus seven) and lower oxidation state (II versus III) of the
titanium atoms in the former complexes. Interatomic data
for the isocyanide groups in 1–4 are qualitatively in accord
with expectations based on IR and NMR (for 1, 3, and 4)
spectra discussed previously. Thus, the longest average Ti–
C(isocyanide) and shortest ligated carbon C–N distances of
˚
˚
˚
R = cyclohexyl, [35], where the latter is only 0.03 A shorter
than that observed in 4. Also, in contrast to 1, for which
structural and spectroscopic data were inconclusive in
determining the nature of the Ti–I bond, NMR data indi-
cate that the corresponding Ti–Sn bonds in 3 and 4 are pre-
dominantly covalent in character. Thus, the 119Sn NMR
spectra for 3 and 4 show sharp resonances at ꢀ69.0 and
ꢀ17.4 ppm, respectively, which are significantly downfield
of the corresponding values for salts of the stannyl anions,
e.g., [Ph3SnLi], ꢀ110 ppm [36], and [Me3SnLi], ꢀ183 ppm
[37]. On this basis, there is no doubt that 3 and 4 contain
long, but otherwise unexceptional, titanium–tin bonds.
Table 3 shows interatomic data for structural units that
are common to 1–4. The only Ti(III) complex in this study,
2, has an overall four-legged ‘‘piano stool” geometry
(Fig. 2) that is related to ones previously observed for
[CpTi(CO)4]ꢀ [22] and [CpTi(PMe3)2Cl2] [8]. The
˚
2.183(3) and 1.154(4) A, respectively, were observed for 2
which had the highest IR m(CN) value among these com-
plexes, in accord with the Chatt–Dewar–Dunson model
for back-bonding [24]. Also of interest is comparison of 2
with the d0 Ti(IV) complex, [Ti(CNXyl)Cl4]2, which has
corresponding Ti–C and C–N distances of 2.235(2) and
˚
1.147(8) A, respectively, and an IR m(CN) absorption at
2200 cmꢀ1 [11], a value well above that of 2 or free CNXyl
(Table 2). If the isocyanide is functioning essentially as a
pure donor ligand in the Ti(IV) complex, it is evident that
some degree of Ti to p*(CNXyl) back-bonding is present in
2, in accord with both spectral and structural data. Other
data shown in Table 3 provide more details on the molec-
ular structures of 1–4, but these are largely unexceptional
and will not be discussed herein.
˚
Ti–Cp(centroid) value of 2.021(1) A for 2 lies within the
range of values reported for mono(cyclopentadienyl)tita-
˚
nium compounds, i.e., 2.01 A in the Ti(IV) complex,
4. Conclusion
Oxidation of [CpTi(CO)4]ꢀ by I2, Ph3SnCl and
Me3SnCl in the presence of CNXyl afford new eight-coor-
dinate Ti(II) complexes of the general formula [CpTi(CN-
Xyl)4E], E = I, SnPh3, SnMe3. Spectroscopic and X-ray
structural studies indicate that discrete coordinated isocy-
anides are present in these unprecedented examples of tita-
nium tetraisocyanide complexes. Extensions of this study
to alkyl isocyanides, the heavier group 4 elements, and
analogous oxidations of [Cp*Ti(CO)4]ꢀ [16] in the pres-
ence of CNXyl will be of significance to determine
whether related group 4 isocyanide complexes are possi-
ble. Use of Cp* instead of Cp in the titanium chemistry
is of particular interest because of a prior study by Teuben
and co-workers which showed that the reaction of the
Fig. 3. Molecular structure of 3. Thermal ellipsoids are set at the 50%
probability level, with hydrogens omitted for clarity. Selected distances
˚
(A) and angles (°): Ti–C(1) 2.121(2), Ti–C(2) 2.113(2), Ti–C(3) 2.113(2),
Ti–C(4) 2.112(2), Ti–C(5) 2.426(2), Ti–C(6) 2.378(2), Ti–C(7) 2.327(2), Ti–
C(8) 2.336(2), Ti–C(9) 2.393(2), Ti–Sn 2.9839(6), C(1)–N(1) 1.170(3),
C(2)–N(2), 1.169(3), C(3)–N(3) 1.168(3), C(1)–N(4) 1.172(3), Cp(cen-
troid)–Ti–Sn 176.8.

1542
J.M. Allen, J.E. Ellis / Journal of Organometallic Chemistry 693 (2008) 1536–1542
quite labile diene complex ([Cp*Ti(diene)Cl], diene = 2,
3-dimethyl-1,3-butadiene), with CNXyl afforded a reduc-
tively coupled isocyanide complex, [(Cp*TiCl)2(l-{N(Xyl)-
C}2)] [39], instead of one containing discrete isocyanide
ligands, as in the case of 1. In particular, did the reductive
coupling of CNXyl in Teuben’s system arise due to the
presence of the diene, the Cp* group, both ligands simul-
taneously, or other factors?
[10] (a) L.B. Kool, M.D. Rausch, M. Heberhold, H.G. Alt, U. Thewalt,
B. Honold, Organometallics 5 (1986) 2465;
(b) L.B. Kool, M.D. Rausch, H.G. Alt, H.E. Engelhardt, M.
Heberhold, J. Organometal. Chem. 317 (1986) C38.
[11] T. Carofiglio, C. Floriani, A. Chiesi-Villa, C. Guastini, Inorg. Chem.
28 (1989) 4417.
[12] T. Carofiglio, P.G. Cozzi, C. Floriani, A. Chiesi-Villa, C. Rizzoli,
Organometallics 12 (1993) 2726.
[13] X. Wang, S.D. Gray, J. Chen, L.K. Woo, Inorg. Chem. 37 (1998) 5.
[14] J.E. Ellis, in: A.L. Wayda, M.Y. Darensbourg (Eds.), Experimental
Organometallic Chemistry: A Practicum in Synthesis and Character-
ization, American Chemical Society, Washington, 1987, pp. 34–65.
[15] G. Tripepi, V.G. Young Jr., J.E. Ellis, J. Organometal. Chem. 593–
594 (2000) 354.
[16] J.E. Ellis, S.R. Frerichs, B.K. Stein, Organometallics 12 (1993) 1048.
[17] D.F. Shriver, M. Drezdzon, The Manipulation of Air-Sensitive
Compounds, second ed., Wiley-Interscience, New York, 1986.
[18] Saint V6.45A, Bruker Analytical X-ray Systems, Madison, WI, 2001.
[19] R. Blessing, Acta Crystallogr., Sect. A 51 (1995) 33.
[20] SHELXTL V6.12, Bruker Analytical X-ray Systems, Madison, WI, 2000.
Acknowledgements
Financial support from the US National Science Foun-
dation and the Petroleum Research Fund, administered by
the American Chemical Society, is gratefully acknowl-
edged. Also, we greatly thank Benjamin E. Kucera for help
with the X-ray crystallography studies and Ms. Christine
Lundby for her expert assistance in the preparation of
the manuscript.
[21] D. Rehder, C. Buttcher, C. Collazo, R. Hedelt, H. Schmidt, J.
¨
Organometal. Chem. 585 (1999) 294, and references cited therein.
[22] B.A. Kelsey, J.E. Ellis, J. Am. Chem. Soc. 108 (1986) 5016.
[23] J.G. Murray, J. Am. Chem. Soc. 83 (1961) 1287.
[24] C. Elschenbroich, Organometallics, third ed., Wiley–VCH, Wein-
heim, 2006, p. 399.
[25] F.A. Cotton, F. Zingales, J. Am. Chem. Soc. 83 (1961) 351.
[26] W.W. Brennessel, J.E. Ellis, Angew. Chem., Int. Ed. 46 (2007) 598.
[27] D.L. Cronin, J.R. Wilkinson, L.J. Todd, J. Magn. Reson. 17 (1975)
353.
[28] L.I. Strumkina, M.K. Minacheva, K.A. Lyssenko, V.V. Burlakov,
W. Baumann, P. Arndt, B.N. Strunin, V.B. Shur, J. Organometal.
Chem. 691 (2006) 557.
[29] S.I. Troyanov, V.B. Rybakov, U. Thewalt, V. Varga, K. Mach, J.
Organometal. Chem. 447 (1993) 221.
Appendix A. Supplementary material
CCDC 661872, 661873, 661874 and 661875 contain the
supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif. Supplementary data associated with this
article can be found, in the online version, at
doi:10.1016/j.jorganchem.2007.11.047.
References
[30] L. Jongen, T. Gloger, J. Beckhuizen, G. Meyer, Z. Anorg. Allg.
Chem. 631 (2005) 582.
[31] A.F. Wells, Structural Inorganic Chemistry, fifth ed., Oxford
University Press, Oxford, 1984, pp. 450–451.
[32] S.I. Troyanov, A. Meetsma, J.H. Teuben, Inorg. Chim. Acta 271
(1998) 180.
[33] J. Emsley, The Elements, third ed., Oxford University Press, Oxford,
1998.
[1] V.J. Sussman, J.E. Ellis, Angew. Chem., Int. Ed. 47 (2008) 484.
[2] (a) E. Meyer, J. Prakt. Chem. 68 (1856) 279;
(b) M. Gautier, Ann. Chim. Phys. 17 (1869) 203.
[3] L. Malatesta, Prog. Inorg. Chem. 1 (1959) 283.
[4] (a) E. Singleton, H.E. Oosthuizen, Adv. Organometal. Chem. 22
(1983) 209;
[34] K.-M. Chi, S.R. Frerichs, J.E. Ellis, J. Chem. Soc., Chem. Commun.
(1988) 1013.
[35] J.E. Ellis, P. Yuen, Inorg. Chem. 32 (1993) 4998.
[36] C. Elschenbroich, Organometallics, third ed., Wiley–VCH, Wein-
heim, 2006, p. 181.
[37] K. Kobayashi, M. Kawanisi, S. Kozima, T. Hitomi, H. Iwamura, T.
Sugawara, J. Organometal. Chem. 217 (1981) 315.
[38] L.M. Engelhardt, R.I. Papasergio, C.L. Raston, A.H. White, Orga-
nometallics 3 (1984) 18.
(b) L. Weber, Angew. Chem., Int. Ed. 37 (1998) 1515.
[5] M.V. Barybin, V.G. Young Jr., J.E. Ellis, J. Am. Chem. Soc. 122
(2000) 4678.
[6] M.V. Barybin, W.W. Brennessel, B.E. Kucera, M.E. Minyaev, V.J.
Sussman, V.G. Young Jr., J.E. Ellis, J. Am. Chem. Soc. 129 (2007)
1141.
[7] C. Floriani, G. Fachinetti, J. Chem. Soc., Dalton Trans. (1973) 1954.
[8] R. Razavi, D.T. Mallin, R.O. Day, M.D. Rausch, H.G. Alt, J.
Organometal. Chem. 333 (1987) C48.
[39] B. Hessen, J. Blenkers, J.H. Teuben, G. Helgesson, S. Jagner,
Organometallics 8 (1989) 830.
[9] E.J.M. DeBoer, J.H. Teuben, J. Organometal. Chem. 166 (1979)
193.