Organometallic modification approach to the control of push-pull architectures: Synthesis, properties and radical polymerization of chromium N-vinylcarbazole derivatives

Article abstract of DOI:10.1016/j.jorganchem.2005.10.052

In an effort to combine the donor character of the carbazole unit with the electron-withdrawing nature of the chromiumtricarbonyl moiety, which can be further modified by exchange of the CO ligand, a new series of vinyl-type monomers with organometallic push-pull chromophores, (N-vinylcarbazole)Cr(CO) ₂L (L = CO (2), PPh₃ (4), (C3H5)+BF4- (5)) and (N-vinylcarbazole)bisCr(CO)₃ (3), have been synthesized and isolated, wherein 2 could be converted into 3 under appropriate conditions. All compounds were characterized spectroscopically, and X-ray crystal structure analyses were performed for 2-4. The coordination-induced geometrical changes occur predominantly on the N-center and its attached double bond. PPh₃ replaces a CO group leading to the N-vinyl double bond length shortening from 1.274 (12) to 1.245 (9) ?. The distance of Cr from the carbazole plane, which allows estimates of the strength of the metal-ligand π-bonding, is shorter in 3 than in 2 as a result of an electronic communication between the two Cr(CO)₃ groups on each aryl ring of carbazole molecule. Electronic absorption and fluorescence spectral features of these complexes have been studied in terms of the electronic nature of the ligands. The change from 2 to 3 causes a subtle red-shift of the absorption bands due to the electronic transitions within dinuclear-carbazole ring system. The polymerization studies of these complexes under free-radical conditions lead to a better understanding of how the organometallic moiety affects the vinyl polymerization. The intriguing effect of the organometallic moiety on the intrachain excimer formation in the resultant polymers have been shown to be likely.

Full text of DOI:10.1016/j.jorganchem.2005.10.052

Journal of Organometallic Chemistry 691 (2006) 994–1006
www.elsevier.com/locate/jorganchem
Organometallic modification approach to the control
of push–pull architectures: Synthesis, properties
and radical polymerization of chromium N-vinylcarbazole derivatives
*
Xiaohui Tian , Wei Shi, Kanyi Shen, Chenyan Li, Jiaping Lin, Yanchao Che, Ping Zhang
School of Materials Science and Engineering, East China University of Science and Technology, 130 Meilong Road, Shanghai 200237, China
Received 30 September 2005; received in revised form 29 October 2005; accepted 29 October 2005
Available online 5 December 2005
Abstract
In an effort to combine the donor character of the carbazole unit with the electron-withdrawing nature of the chromiumtricarbonyl
moiety, which can be further modified by exchange of the CO ligand, a new sþeries of vinyl-type monomers with organometallic push–pull
chromophores, (N-vinylcarbazole)Cr(CO)2L (L = CO (2), PPh3 (4), ðC3H5Þ BFꢀ4 (5)) and (N-vinylcarbazole)bisCr(CO)3 (3), have been
synthesized and isolated, wherein 2 could be converted into 3 under appropriate conditions. All compounds were characterized spectro-
scopically, and X-ray crystal structure analyses were performed for 2–4. The coordination-induced geometrical changes occur predom-
inantly on the N-center and its attached double bond. PPh3 replaces a CO group leading to the N-vinyl double bond length shortening
˚
from 1.274 (12) to 1.245 (9) A. The distance of Cr from the carbazole plane, which allows estimates of the strength of the metal–ligand p-
bonding, is shorter in 3 than in 2 as a result of an electronic communication between the two Cr(CO)3 groups on each aryl ring of car-
bazole molecule. Electronic absorption and fluorescence spectral features of these complexes have been studied in terms of the electronic
nature of the ligands. The change from 2 to 3 causes a subtle red-shift of the absorption bands due to the electronic transitions within
dinuclear-carbazole ring system. The polymerization studies of these complexes under free-radical conditions lead to a better understand-
ing of how the organometallic moiety affects the vinyl polymerization. The intriguing effect of the organometallic moiety on the intra-
chain excimer formation in the resultant polymers have been shown to be likely.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Chromium; N-vinylcarbazole; Push–pull architecture; Ligand exchange; Polymerization
1. Introduction
tional organic systems [3]. Far fewer systems are known
that contain organometallic subunits to control polymer
properties. Incorporation of metallic center into a car-
bazole system through a variety of novel attachment
profiles seems to be another route in designing and syn-
the- sizing new types of carbazole-based polymers with
improved materials properties unobtainable in simple
p-organic structure. The two coplanar aryl rings of car-
bazole unit should be able to provide a platform for
transition metal p-coordination since they bond tenaci-
ously to a wide variety of transition metal complexes
[4]. In regard to the electron-withdrawing nature of the
Cr(CO)3 moiety which is comparable to that of a nitro
substituent, coordination of Cr(CO)3 to carbazole ligand
The driving force to pursue research on carbazole-con-
taining polymers comes from both fundamental interests
and practical expectations [1]. It is known that these
polymers, especially those with carbazole as pendant unit
or copolymers, are sensitive only in the UV region of
spectra [2]. The applications of such polymers in electro-
photography and related process need sensitization to the
visible wavelengths. Most of these efforts have been
focused on charge-transfer (CT) formation in conven-
*
Corresponding author. Tel./fax: +86 21 5428 8175.
E-mail address: tianxh@263.net (X. Tian).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.10.052

X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
995
could yield an organometallic push–pull structure [5] and
thus, the electron cloud movement or the charge-transfer
electron density from the carbazole moiety to the car-
bonyl groups through the metal center will make the
molecule exhibit high polarizabilities, facile excitation
upon irradiation with visible light and stable species of
higher oxidation states. As shown by the recent reports
of Merlic and other researchers [6], chromium complexa-
tion leads to strong stabilization of both benzylic anions
and cations, and a high quantum yield of charge separa-
tion is achieved through field and inductive effects allow-
ing the transition metal to participate, yet maintain an
18-electron configuration. Kobayashi et al. [7] reported
on the CT complexes of areneCr(CO)3 with various p-
acceptors (e.g., 1,3,5-trinitro-benzene and tetracyanoeth-
ylene), in which the areneCr(CO)3 behaved as electron
donor with the ionization potential being ꢁ1.5 eV lower
than that of the free ligand mainly due to metal d-lone
pair. Considering the great structural and electronic
diversity available in the metal–organic environments, a
profound understanding of the unique mode of (carba-
zole)Cr(CO)3 and other synthetically relevant systems
for the development of advanced electroactive, optical,
magnetic, or catalytic materials is desired. On the other
hand, the chromium atom might be tuned to act as an
intermediary between donor and acceptor moieties by
replacing a CO grþoup with other ligands, for example,
PPh3 and ðC3H5Þ BFꢀ4 ligands are efficient at either
donating or accepting electrons relative to CO [8]. In
the present paper, we give full details of the syntheses
of the (N-vinylcarbazole)bisCr(CO)3 (3) and (N-vinylcar-
bazole)- Cr(CO)2L, where L = CO (2), PPh3 (4) and
ðC3H5ÞþBFꢀ4 (5) (Scheme 1). The structural features of
such complexes with various ligands are discussed in
relation to their behavior in photophysics and radical
polymerization. In view of the fact that some of these
complexes are still active for vinyl-type polymerization
[9], an extension of this research to the macromolecular
level was also pursued.
2. Experimental
2.1. General procedures
Unless otherwise noted, all manipulations with air-sen-
sitive materials were performed under a nitrogen atmo-
sphere using standard Schlenk techniques and
a
Vacuum Atmospheres drybox. Solvents were freshly dis-
tilled over appropriate drying reagents, and all Schlenk
flasks were flame-dried under vacuum prior to use.
Cr(CO)6 (99%, Strem Chemicals) was sublimed prior to
use. Commercial azobisisobutyronitrile (AIBN) was
recrystallized three times from methanol before use
(m.p.: 102–103 ꢁC (dec)). N-Vinylcarbazole (1) (99%,
Aldrich) was used as received. Commercially available
poly(N-vinylcarbazole) (PVK) was dissolved in benzene
and precipitated out with methanol numerous times to
free it from residual monomer and anthracene. Column
chromatography was performed on Matrix silica 60
(30–70 lm). Melting points were measured using a Reic-
hert Thermopan melting point microscope and are
uncorrected. All the crystalline chromium complexes
can be handled in air. NMR spectra were recorded on
a Bruker AVANCE 500 spectrometer. Chemical shifts
are referenced in ppm to internal Me4Si for the 1H
and 13C NMR spectra and to external 85% H3PO4 and
80% CF3COOH for the 31P and 19F NMR spectra,
respectively. IR spectra were recorded on a Nicolet
OC
CO
OC
HC
N
HC
CH2
N
CH2
HC
N
CH2
Cr
(NH3)3Cr(CO)3
OC Cr
+
OC Cr
CO
CO
CO
CO
3
1
2
hv
HC
N
CH2
inversion
PPh3
-1
C3H5OH
HBF4
hv
OC Cr
CO
toluene
AIBN
PPh3
4
toluene AIBN
+
CH CH2
CH CH2
HC
CH2
N
N
N
OC Cr
CO
OC Cr
CO
OC Cr
CO
PPh3
-
CO
BF4
n
n
6
5
7
Scheme 1.

996
X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
Magna-IR 550 instrument in KBr pellets. Elemental
analyses were carried out on an Elementar Vario EL
III analyzer (C, H and N) and TJARIS 1000 (Cr and
P), respectively. Electron-impact mass spectra were
recorded on a Micromass GCT (EI, 70 eV) mass spec-
trometer. Molecular weight measurements were carried
out in a PE Series 200 gel permeation chromatograph
(GPC) with polystyrene as the standard, using THF as
the eluent (30 ꢁC) at a flow rate of 1 mL/min. Thermo-
gravimetric analyses (TGA) were carried out using a
TGA SDTA 851e instruments with a heating rate of
10 ꢁC/min in a nitrogen flow (20 mL/min). The DSC-
200 PC of TA instruments were used to measure the
glass transition temperature (Tg) of the samples under
nitrogen at a heating rate of 10 ꢁC/min from ꢀ5 to
300 ꢁC. UV–Vis spectra were obtained on a Varian Cary
500 spectrophotometer rebuilt by OLIS to incorporate
computer control. Fluorescence spectra were recorded
on a FLS920 spectrophotometer (EDINBURGH Instru-
ments Ltd.) under the control of a Pentium-based PC
running the SpectraCalc software package. Excitation
was provided by a Xe 900 lamp operating at a current
of 5 A. Fluorescence quantum yields were obtained by
comparison of the integrated area of the corrected emis-
sion spectrum of the samples with that of a solutions of
quinine sulfate in 1 N H2SO4, which has a quantum effi-
ciency of 0.55. The concentration of the reference was
adjusted to match the absorbance of the test sample.
Emissions for 1–4 and 6 were integrated from 340 to
450 nm, and emissions for 5, 7 and PVK were integrated
from 340 to 550 nm. Measurements were made on argon
saturated samples at room temperature.
2.2. Crystallography
Single crystals of 1–4 suitable for X–ray diffraction
were obtained by recrystallization from CH2Cl2/n-hexane
at ꢀ5 ꢁC under an inert atmosphere and were sealed in
thin-walled glass capillaries. Data were collected on a
Bruker Smart Apex CCD diffractometer with graphite-
˚
monochromated Mo Ka radiation, k = 0.71073 A, using
the x–2h technique at 293 K. All structures were solved
by direct methods using SHELXS97 and refined by the
full-matrix least-squares method on all F2 data using
the program SHELXL97 [10]. There are two independent
molecules in the asymmetric unit of 1. All non-hydrogen
atoms were refined anisotropically. Hydrogen atoms were
inserted in idealized positions and refined using a riding
model. Relevant crystallographic information is summa-
rized in Table 1.
Table 1
Details of data collection and structure refinement for 2–4
2
3
4
Empirical formula
Formula weight
Crystal system
Space group
C
329.27
Monoclinic
C2
17H11NO3Cr
C20H11NO6Cr2
465.30
Triclinic
C34H26NO2PCr
563.53
Triclinic
ꢀ
ꢀ
P1
P1
Color, form
Orange plates
13.1714(17)
13.8190(17)
9.4899(12)
90
121.947(3)
90
1465.7(3)
4
Orange needles
7.5498(15)
9.826(2)
13.705(3)
69.422(2)
74.322(4)
86.268(4)
915.9(3)
2
Violet blocks
7.9481(8)
9.2605(10)
19.818(2)
83.487(2)
80.757(2)
72.148(2)
1367.2(2)
2
˚
a (A)
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
3
˚
V (A )
Z
d (calcd) (g Æ cmꢀ3
Absorption coefficient (mmꢀ1
F(000)
)
1.492
0.791
672
1.687
1.225
468
1.369
0.509
584
)
Crystal size (mm)
h Range (ꢁ)
0.356 · 0.321 · 0.231
2.34–28.85
4515
0.516 · 0.167 · 0.060
1.65–27.00
5449
0.495 · 0.338 · 0.207
1.04–27.00
8024
No. of reflections collected
No. of unique data
3130
3898
5775
No. of data/restraints/parameters
Rint
GOF/F2
3130/8/199
0.0345
0.928
3098/0/262
0.0884
1.170
5775/0/351
0.0642
0.965
R indices [I > 2r(I)]
R indices (all data)a,b
R1 = 0.0593
xR2 = 0.1287
R1 = 0.0636
xR2 = 0.1646
R1 = 0.0538
xR2 = 0.1372
R1 = 0.0868
R1 = 0.0731
R1 = 0.0730
xR2 = 0.1422
xR2 = 0.1746
xR2 = 0.1516
ꢀ3
˚
Largest diff. peak and hole (e A
)
0.548/ꢀ0.235
1.310/ꢀ0.584
0.910/ꢀ0.397
P
P
a
R1 ¼ jjF oj ꢀ jF cjj= jF oj.
n
o
1=2
P
P
2
2
b
xR2 ¼
½xðF 2o ꢀ F c2Þ ꢂ= ½xðF 2oÞ ꢂ
.

X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
997
2.3. Synthesis of mono- (2) and bis-Cr(CO)3 (3) complexes
of N-vinylcarbazole
2.6. Conversion of 2 into 3 in dioxane
2 (10.0 mmol, 3.3 g) was dissolved in 800 mL of
degassed dioxane, and the solution was then introduced
in equal portions into eight glass tubes, which were sealed
and left to stand for different time intervals at high temper-
ature (80 ꢁC) with avoiding sunlight. After respective time
intervals, volatiles were removed in vacuo and the crude
yellow solid was separated by column chromatography.
Three fractions (1–3) were obtained, quantified and identi-
Cr(CO)6 was reacted to give (NH3)3Cr(CO)3 according
to a literature method [5a]. A mixture of 10.3 g (55.0 mmol)
of (NH3)3Cr(CO)3 and 3.0 g (15.5 mmol) of 1 was heated
under reflux in 200 mL of dioxane for 6 h. After being
cooled, the solution was filtered through a celite filter and
the residue was extracted into 5 mL of CH2Cl2 and passed
through a 5 cm · 35 cm column of silica gel, eluting with a
CH2Cl2–petroleum ether mixture (1:3). Two bands were
observed: the first yellow band (mononuclear complex 2;
MS (m/z): 329 [M+]) was collected with ꢁ500 mL of the
eluent, and the second orange band (dinuclear complex 3;
MS (m/z): 465 [M+]) with ꢁ300 mL of CH2Cl2. The prod-
ucts were recrystallized five times from CH2Cl2/n-hexane
(3:1), respectively, to obtain two kinds of light orange
crystals.
1
fied by H NMR, respectively.
2.7. Attempted conversion of 3 into 2 in the solution
Treatment of 3 was performed analogously to that of 2.
Even after heating 3 in dioxane at 80 ꢁC for more than
120 h, spectral evidence for the process of 3 ! 2 was
absent.
2.8. Synthesis of N-vinylcarbazole Cr(CO)2PPh3 complex
(4)
2.4. Complex 2
Yield: 2.19 g (43.1%); m.p.: 125 ꢁC. Anal. Calc. for
2 (0.33 g, 1 mmol) and triphenylphosphine (0.79 g,
3.0 mmol, recrystallized from ethanol) were irradiated in
45 mL of dry benzene in a Schlenk tube under a Hanovia
ultraviolet lamp for 8 h. The heat from the lamp was
allowed to reflux the reaction mixture, and good stirring
was maintained under a nitrogen atmosphere. The dark
violet solution was filtered hot through a glass frit and
rinsed with n-hexane to obtain 0.17 g (31%) of a violet solid
which is stable to air. Recrystallization in dry, degassed
CH2Cl2/n-hexane (3:1) yielded large, dark violet crystals;
m.p.: 168 ꢁC (dec); Anal. Calc. for C34H26NO2PCr: C,
72.47; H, 4.62; N, 2.49; Cr, 9.24. Found: C, 72.42; H
4.45; N 2.28; Cr 9.03. 31P NMR (C6D6, 202 MHz):
C17H11NO3Cr: C, 62.01; H, 3.37; N, 4.25; Cr, 15.79.
1
Found: C, 62.21; H, 3.33; N, 4.11; Cr, 16.02. H NMR
(CDCl3, 500 MHz): d = 5.39 (d, J = 8.5 Hz, 1 H), 5.45 (t,
J = 6.0 Hz, 1 H), 5.72 (d, J = 16.0 Hz, 1 H), 5.92 (t.
J = 6.1 Hz, 1 H), 6.56 (d, J = 6.5 Hz, 1 H), 7.06 (d,
J = 6.1 Hz, 1 H), 7.31 (t, J = 7.7 Hz, 1 H), 7.37 (t, J =
6.5 Hz, 1 H), 7.57 (t, J = 7.5 Hz, 1 H), 7.74 (d, J =
8.0 Hz. 1 H), 13C NMR (CDCl3, 125 MHz): d = 77.65
(CH2), 87.47, 89.77, 93.79, 94.50 (Ar-C), 106.12, 111.67
(Ar-Cq), 120.69 (CH), 121.21, 122.55, 123.98, 128.50 (Ar-
C), 129.00, 141.49 (Ar-Cq), 234.53 (CO). EI MS (70 eV.
m/z (%)): 329 (M+, 43), 301 (M+ ꢀ CO, 4), 273
(M+ ꢀ 2CO, 52), 245 (M+ ꢀ 3CO, 100), 193 (M+ ꢀ
Cr(CO)3, 35), 52 (Cr+, 79). IR (KBr): m = 1966, 1872,
1
d = 90.82 (s, br). H NMR (C6D6, 500 MHz): d = 4.25 (s,
1 H), 4.66 (s, 1 H), 4.72 (d, J = 8.7 Hz. 1 H), 4.97 (d,
J = 5.1 Hz, 1 H), 5.15 (s, 1 H), 5.19 (d, J = 13.0 Hz, 1
H), 6.49 (d, J = 9.0 Hz, 1 H), 6.95 (s, 10 H), 7.15 (s, 2
H), 7.23 (d, J = 7.2 Hz, 1 H), 7.38 (s, 6 H). 13C NMR
(C6D6, 125 MHz): d = 71.90, 84.72, 86.72, 88.59 (Ar-C),
90.31 (CH2), 104.59, 111.09 (Ar-Cq), 118.72 (CH), 120.84,
121.87, 125.85 (Ar-C), 128.50 (Ar-Cq), 129.32 (Ar-C),
133.83 (Ar-Cq), 133.92, 140.40 (Ph-C). 241.70 (CO). EI
MS (70 eV, m/z (%)): 563 (M+, 5), 507 (M+ ꢀ 2CO, 49),
314 (CrPPhþ3 , 100), 262 (PPhþ3 , 20), 193 (M+ ꢀ
Cr(CO)2PPh3, 46), 52 (Cr+, 38), IR (KBr): m = 1865,
1639, 1612, 1544, 1466, 1442, 1010, 741 cmꢀ1
.
2.5. Complex 3
Yield: 0.28 g (4.0%); m.p.: 220 ꢁC (dec); Anal. Calc. for
C20H11NO6Cr2: C, 51.58; H, 2.36; N, 3.01; Cr, 22.35.
1
Found: C, 51.54; H, 2.31; N, 2.29; Cr, 22.17. H NMR
(CDCl3, 500 MHz): d = 5.10 (t, J = 6.3 Hz, 2 H), 5.57–
5.60 (m, 3 H), 5.67 (dd, J = 14.2 Hz, 1 H), 5.88 (d,
J = 6.8 Hz, 2 H), 6.38 (d, J = 6.3 Hz, 2 H), 6.75 (dd,
H). 13C NMR (CDCl3, 125 MHz):
1822, 1647, 1546, 1465, 1430, 1295, 1096, 747 cmꢀ1
.
J = 8.5 Hz,
1
d = 75.12, 85.68, 86.32 (Ar-C), 91.73 (CH2), 92.65 (Ar-
Cq), 112.53 (Ar-C), 122.13 (CH), 127.08 (Ar-Cq), 232.30
(CO). EI MS (70 eV. m/z (%)): 465 (M+, 9), 409
(M+ ꢀ 2CO, 6), 381 (M+ ꢀ 3CO, 7), 353 (M+ ꢀ 4CO, 7),
325 (M+ ꢀ 5CO, 29), 297 (M+ ꢀ 6CO. 23), 273
(M+ ꢀ 2CO-Cr(CO)3, 8), 245 (M+ ꢀ 3CO ꢀ Cr(CO)3,
100), 193 (M+ ꢀ 2Cr(CO)3, 16), 52.0 (Cr+, 96). IR (KBr):
m = 1937, 1899, 1881, 1851, 1639, 1564, 1473, 1428,
2.9. Synthesis of N-vinylcarbazole Cr(CO)2(C3H5)þBF ꢀ
(5)
4
2 (0.53 g, 1.6 mmol) and dry ethyl ether (80 mL) were
placed into a 250-mL round-bottom flask. A solution of
50% aqueous HBF4 (0.28 g, 1.6 mmol) and allyl alcohol
(0.12 g, 2.0 mmol) in 40 mL of ethyl ether was added. The
solution was irradiated with a Hanovia ultraviolet lamp
815 cmꢀ1
.

998
X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
under a nitrogen atmosphere. The flask was cooled with a
rapid stream of air to prevent any decomposition from
occurring. After 3–4 h of irradiation, the solution was fil-
tered under nitrogen leaving insoluble 5 as a brick red solid
(ca. 0.16 g, 23% yield), which was completely redissolved in
about 15 mL of dry acetone. Analytically pure samples were
obtained by numerous reprecipitations from the acetone/
ether mixture; m.p.: 155–157 ꢁC decompose without melt-
ing. Anal. Calc. for C19H16O2NBF4Cr: C, 53.15; H, 3.73;
N, 3.26; Cr, 12.12. Found: C, 53.10; H, 3.12; N, 3.13; Cr,
11.96. 19F NMR ((CD3)2CO, 471 MHz): d = 151.30 (s, br).
1H NMR ((CD3)2CO, 500 MHz): d = 1.10 (d, J = 6.7 Hz,
1 H), 1.22 (d, J = 10.4 Hz, 1 H), 3.52 (dd, J = 14.0 Hz, 2
H), 3.77 (s, 1 H), 5.80 (d, J = 8.2 Hz. 1 H), 6.04 (d,
J = 15.7 Hz, 1 H), 6.63 (s, 1 H), 6.94 (s, 1 H), 7.55 (dd,
J = 8.9 Hz, 1 H), 7.69 (s, 1 H), 7.76 (d, J = 6.7 Hz, 1 H),
7.96 (t, J = 6.7 Hz, 1 H), 8.08 (t, J = 10.1 Hz, 2 H), 8.52 (d,
J = 7.1 Hz, 1 H). 13C NMR ((CD3)2CO, 125 MHz):
d = 56.96, 57.25 (CH2), 88.72 (CH), 98.77 (CH2), 103.50,
111.60 (Ar-C), 112.66 (CH), 113.76, 121.07 (Ar-Cq),
121.64, 124.33, 124.81, 125.17, 127.30, 128.70 (Ar-C),
132.98, 140.18 (Ar-Cq), 224.69 (CO). ESI-MS: 342.1
½M ꢀ BFꢀ4 ꢂ, 771.2 ½2M ꢀ BFꢀ4 ꢂ. IR (KBr): m = 1997, 1928,
(m, 2 H), 7.45 (d, J = 7.7 Hz, 1 H). IR (KBr): m = 1952,
1860, 1545, 1460, 1425, 1310, 1202, 1165, 743 cmꢀ1
.
2.12. Poly(N-vinylcarbazole)Cr(CO)2PPh3 (7)
Anal. Calc. for C34H26NO2PCr: C, 72.52; H, 4.45; N,
2.28; P, 4.75; Cr, 9.43. Found: C, 72.47; H, 4.52; N, 2.09;
1
P, 4.70; Cr, 9.24. H NMR (C6D6, 500 MHz): d = 1.40 (s,
1 H), 2.11 (s, 1 H), 4.26 (t, J = 5.9 Hz, 1 H), 4.67–4.74
(m, 1 H), 5.16–5.23 (m. 1 H), 6.50 (dd, J = 8.9 Hz, 1 H),
6.8–7.4 (several broad peaks, 19 H), 7.82 (t, J = 7.9 Hz, 1
H). IR (KBr): m = 1951, 1860, 1815, 1550, 1480, 1445,
1315, 1160, 1095, 745 cmꢀ1
.
3. Results and discussion
3.1. Synthesis and Structures of mono- (2) and bis-Cr(CO)3
(3) complexes of N-vinylcarbazole
Treatment of N-vinylcarbazole (1) with 3 equiv of
(NH3)3Cr(CO)3 in refluxing dioxane under protection
against incandescent light for 6 h lead to the precipitation
of inorganic chromium and a clear deep orange solution.
Filtration and partial evaporation of the solvents from
the filtrate under vacuum gave a mixture of 2 and 3 with
92/8% occupancy, and they could be separated by column
chromatography. An attempt to improve the yield of 3
using an excessive Cr reagent, a high boiling point solvent
and longer reaction time failed. It is unclear in this reaction
what factors control the yield of 3 vs. 2, but we suspect that
some of reaction mixtures (or a trace of O2 due to non-
sealed system) decelerate the formation of 3. It is notewor-
thy that heating pure 2 in appropriate solvents under an
inert atmosphere for several hours gives rise to a conver-
sion to 3. Dioxane, THF and pyridine are found to be good
solvents for such conversion, probably due to their ability
to induce the decomplexation of 2, generating a labile inter-
mediate (solvent)3Cr(CO)3, which further affords ‘‘hot’’
Cr(CO)3 particles and randomly attacks 1 and the remain-
ing 2 [5a,11]. Under similar conditions, we are unaware of
the conversion from 3 to 2 and moreover, 3 is completely
stable even at higher temperature (100 ꢁC) for days. Kinetic
curves for the conversion of 2 into 3 in dioxane at 80 ꢁC are
shown in Fig. 1. An accumulation of 1 in the system is
almost comparable to equimolar amount of 3, indicating
that there is a simple equilibrium involved with little or
no loss of the Cr(CO)3 particles [12]. Accordingly, 2 repre-
sents the kinetically controlled product and 3 the thermo-
dynamically controlled one. This aspect allows us to hold
a new approach to synthesize the novel binuclear com-
plexes of the type 3.
1644, 1613, 1559, 1495, 1445, 1057, 751 cmꢀ1
.
2.10. Polymerization reactions
The procedures for both polymerizations of 2 and 4
were essentially the same. Polymerization was carried out
in freshly distilled toluene. AIBN was used as the free rad-
ical initiator. Monomers were stored at ꢀ20 ꢁC in the dark
and were freshly prepared before use.
Fischer–Porter aerosol compatible tubes with valves
were used for polymerization. Weighed amounts of mono-
mer, solvent, and initiator were charged to the tubes; the
tubes were sealed and degassed at 10ꢀ3 mm by three alter-
nate freeze–thaw cycles. Liquid nitrogen was used as the
freezing medium. After degassing, the tubes were placed
in a thermostat for a predetermined amount of time. After
removal from the bath, the polymer was precipitated in
rapidly swirling large excess of petroleum ether. When nec-
essary, the polymer solution was diluted with more solvent
before precipitation to obtain fine polymer particles rather
than large chunks. This process was repeated three times to
ensure that all monomers and other contaminants had been
washed from the polymer. After precipitation the polymers
were filtered and dried under vacuum. Yields were deter-
mined gravimetrically.
2.11. Poly(N-vinylcarbazole)Cr(CO)3 (6)
Anal. Calc. for C17H11NO3Cr: C, 62.21; H, 3.03; N,
4.20; Cr, 16.11. Found: C, 62.13; H, 3.07; N, 4.16; Cr,
15.99. 1H NMR (CDCl3, 500 MHz): d = 4.26 (s, 1 H),
4.30 (t, J = 6.4 Hz, 1 H), 4.72–4.78 (m, 2 H), 5.08 (dd,
J = 14.6 Hz, 1 H), 5.18 (d, J = 6.9 Hz, 1 H), 5.78 (d,
J = 6.4 Hz, 1 H), 6.33 (dd, J = 8.9 Hz, 1 H), 6.97–7.02
EI-mass spectra of 1 and 2 show a peak of the highest
mass corresponding to their monomeric formulas, respec-
tively. The infrared spectra of 2 and 3 in the carbonyl
stretching region are very different – two bands are
seen in 2 (mco = 1966, 1872 cmꢀ1), while 3 has three bands
(mco = 1937, 1899 and 1881 cmꢀ1), indicating some

X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
999
100
80
60
40
20
0
are provided in Figs. 2–4. Crystals of 1 belong to ortho-
rhombic system with space group Pna21, the lattice is com-
posed of two independent molecules with distinction of the
N-vinyl group, for example, the bond lengths of C13–C14
˚
and C27–C28 are 1.231 (6) and 1.270 (5) A, respectively;
2
˚
C14 lies 0.16 A above the carbazole plane which is planar
˚
˚
to within 0.03 A, while C28 lies 0.42 A above its carbazole
˚
3
plane (planar to within 0.02 A). There are no apparent
short-range intermolecular interactions between the carba-
1
zole rings. Complexes 2 and 3 crystallize in the monoclinic
ꢀ
and triclinic system with space group C2 and P1, respec-
tively. Owing to the repulsive interaction between the
Cr(CO)3 orbitals and the central heterocycle p orbitals
[4d,14], they adopt a similar exo (staggered) conformation
about the Cr–c(0) axis [c(0) = Cr-ring centroid] such that
the carbonyl ligands do not eclipse the junction C–C bond.
The Cr(CO)3 groups are shifted somewhat from an ideal-
ized g6 position away from the ring junctions with the tilt
angle (2.95ꢁ for 2; 2.09ꢁ and 1.19ꢁ for 3) between the nor-
mal to the ring plane and the line Cr–c(0). Two Cr(CO)3
groups in 3 are g6-coordinated to opposite faces of an
essentially planar carbazole ring, and the distances of
Cr(1) and Cr(2) from the mean plane are ꢀ1.731 and
0
18
36
54
72
Time (hr)
Fig. 1. Kinetic curves of conversion of 2 into 3 in dioxane at 80 ꢁC.
coupling of the CO vibration of the two Cr(CO)3 units in
the latter case. NMR experiments show the general upfield
shifts for all the aryl resonances in 3, whereas 2 displays the
highfield shifts only for one aryl ring. Moreover, the differ-
ence of the highfield shifts from metal coordination
between 2 and 3 was noticed, further illustrating the pres-
ence of ligand-mediated metal–metal interaction for the
dichromium species [13].
Complexes 2 and 3 were also characterized by X-ray
crystallography, together with their organic counterpart 1
for comparative purposes. Selected bond lengths and
angles are given in Table 2, and ORTEP diagrams of 1–3
˚
˚
1.734 A, which are relatively shorter than that of 1.741 A
in 2. This may be attributed to an indirect metal–metal
electronic interaction in 3 [13]. Probably the compromise
Table 2
˚
Selected bond lengths (A) and angles (ꢁ) for 2–4
2
N(1)–C(4)
N(1)–C(16)
C(4)–C(9)
C(16)–C(17)
1.436(12)
1.528(13)
1.408(6)
N(1)–C(15)
C(4)–C(5)
C(10)–C(15)
1.439(14)
1.437(12)
1.329(12)
1.274(12)
C(4)–N(1)–C(15)
C(4)–C(9)–C(8)
C(10)–C(15)–C(14)
109.0(8)
115.7(11)
124.5(13)
C(5)–C(4)–C(9)
C(11)–C(10)–C(15)
N(1)–C(16)–C(17)
70.7(5)
123.5(9)
122.7(12)
3
N(1)–C(7)
N(1)–C(19)
C(7)–C(12)
C(13)–C(18)
C(19)–C(20)
1.376(5)
1.425(5)
1.419(5)
1.419(5)
1.275(6)
N(1)–C(18)
C(7)–C(8)
C(12)–C(13)
C(17)–C(18)
1.392(18)
1.414(5)
1.450(5)
1.412(5)
N(1)–C(7)–C(12)
C(7)–C(12)–C(11)
N(1)–C(18)–C(13)
110.0(3)
120.8(3)
109.3(3)
C(8)–C(7)–C(12)
C(14)–C(13)–C(18)
C(13)–C(18)–C(17)
120.2(4)
120.3(3)
121.0(3)
4
Cr–P
2.2967(9)
1.376(5)
1.850(3)
1.245(9)
N(1)–C(3)
P–C(17)
P–C(29)
1.417(4)
1.840(3)
1.848(3)
N(1)–C(14)
P–C(23)
C(15)–C(16)
Cr–P–C(17)
Cr–P–C(29)
C(24)–C(23)–C(28)
116.83(10)
117.78(9)
117.4(3)
Cr–P–C(23)
C(18)–C(17)–C(22)
C(30)–C(29)–C(34)
116.68(10)
117.4(3)
118.1(3)
Fig. 2. ORTEP drawing of
probability level.
1 with thermal ellipsoids at the 30%

1000
X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
some extent of electron transfer to the Cr(CO)3 from the
double bond via zwitterionic resonance form [5a,15].
3.2. Synthesis and Structures of (N-vinylcarbazole)Cr-
(CO)2PPh3 (4)
Compound 2 reacted with triphenylphosphine in ben-
zene upon simultaneous UV irradiation for 6 h to afford
(N-vinylcarbazole)Cr(CO)2PPh3 (4). Analytically pure
crystals were obtained in 30% yield by fractional crystalli-
zation from a mixture of dichloromethane and hexane at
ꢀ30 ꢁC. The reaction of 3 with triphenylphosphine in a
similar manner has also been carried out resulting in a
complex mixture which is difficult to separate. The multiple
potential displacement sites are suspected to be the prob-
lems. Complex 4 is stable to air and moisture in the solid
state but rapidly decomposed in solution at room
temperature. The EI-mass spectrum of 4 shows a peak of
the highest mass at the molecular ion [M+] with a relative
intensity of 5%. The 31P NMR shows a broad singlet at
90.82 ppm. The IR spectrum gives two strong carbonyl
vibronic bands at 1865 and 1822 cmꢀ1 which are shifted
toward lower energies (vs 2) [8,16], suggesting that the
phosphine ligand exerts its influence on the Cr center. Sim-
ilar electronic effects are also noted in the 13C and 1H NMR
chemical shifts of the Cr-ring, i.e., shifted to high field,
while the carbonyl carbon resonances are slightly shifted
downfield.
Fig. 3. ORTEP drawing of
probability level.
2 with thermal ellipsoids at the 30%
The structure of 4 was also determined by X-ray analy-
ꢀ
sis. The crystals are triclinic, space group P1. Selected bond
lengths and angles are given in Table 2, and ORTEP dia-
grams of 4 are provided in Fig. 5. A distinguishing feature
of 4 is that the rotation about the Cr–c(0) axis forces the
Cr–P bond to the opposite position to the N-vinyl group
in response to the steric requirements of the PPh3 group,
Fig. 4. ORTEP drawing of
3 with thermal ellipsoids at the 30%
probability level. Hydrogen atom have been omitted for clarity.
between field and inductive effects by both metal centers is
better and serves to stabilize its dinuclear-structure [6a,6b].
The carbazole fragment is found keeping the ring planar to
˚
˚
within 0.05 A for 2 and 0.06 A for 3, and the N-vinyl group
and its attached ring in 2 are coplanar due to favorable
orbital interaction. The steric crowdedness around N center
resulting from the two Cr(CO)3 groups in 3 is evidenced by
the torsion angle of 146.0(5)ꢁ for C7–N–C19–C20. The N
bond lengths fall into the different range (1.436(12)–
˚
˚
1.528(13) A for 2; 1.376(5)–1.428(5) A for 3), but their
˚
attached double bond lengths (1.274 (12) A for 2; 1.275
(6) A for 3) are equal to each other and consistently longer
˚
Fig. 5. ORTEP drawing of
4 with thermal ellipsoids at the 30%
probability level. Hydrogen atom have been omitted for clarity.
than that observed in 1. This is plausibly rationalized by

X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
1001
one phenyl ring of which is less susceptible to the steric
shifted toward higher frequencies at 1997 and 1928 cmꢀ1
.
bulk and electronic influences of the Cr-ring than the other
two. The Cr atom is offset slightly toward the four unsub-
stituted Cr-ring carbon atoms with the tilt angle of 1.94ꢁ
between the normal to the ring plane and the line Cr–
Crystals suitable for X-ray diffraction were not obtained,
despite repeated attempts. We are therefore unable to pro-
vide a reliable estimate of the metal–organic architecture
for 5. However, by analogy with the structures of 2 and
4, it is plausible that insertion of the strongly p-accepting
ðC3H5ÞþBFꢀ4 ligand permit this species to function as
extended organometallic push–pull chromophores.
˚
c(0). The distance of 1.736 A from Cr to the least-squares
plane is slightly longer than that found in 3 but just shorter
than that found in 2. The Cr-ring carbon bond lengths (av.
˚
1.340 A) are within typical ranges, which are not signifi-
cantly different from that for 2. However, the double bond
3.4. Radical polymerizations
˚
length of 1.245 (9) A in 4 is exceptionally shorter (by ca.
˚
0.03 A) than that in 2 and 3. This suggests that ligand
Complexes 2 and 4 were successfully polymerized in
degassed toluene using recrystallized AIBN as the initiator
at 80 ꢁC to yield the corresponding well-defined poly(N-
vinylcarbazole)Cr(CO)3 (6) and poly(N-vinylcarbazole)Cr-
(CO)2PPh3 (7). No evidence for the above-mentioned
conversion of 2 into 3 has been found in toluene even after
prolonged periods at elevated temperature (100 ꢁC, 80 h).
Care must be exercised at this stage to conduct the poly-
merization in an evacuated and sealed system to avoid
decomplexation of the chromium from the carbazole ligand
by photolysis under O2. The resultant polymers 6 and 7 are
moderately soluble in benzene, toluene, and chlorine-con-
taining solvents and are film-forming materials. Addition-
ally, these polymers are very stable toward air and
moisture in the solid state and can be stored in solution
in contact with air for an extended period of time without
decomposition. The 1H NMR and IR spectra of these
polymers deserve comment consulting with their corre-
sponding parent complexes. Elemental analysis data, which
are included in Section 2, confirmed no loss of the metal
unit from the pendant carbazole groups for each polymer.
Notice in Fig. 6 that a linear Mn progression with the con-
version and low polydispersity indices (PDIs) lead us to
immediately suppose a stable free radical polymerization
(SFRP) mechanism. To further clarify this point, we have
simulated a stop-and-go experiment with these two com-
plexes, where the polymerization was stopped by liquid
exchange on the metal moiety imposes strong influence
on the double bond character, namely, the PPh3 ligand
enhances the electron density onto the N-vinyl group
through the Cr center and thus strengthens the double
bond. Further support for this viewpoint comes from an
1
upfield shift of the @CH2 group in the H and 13C NMR
spectra (vs 2). The out-of-plane deviation of the N-vinyl
group is registered for its terminal carbon atom bending
˚
towards the chromium by about 0.124 A, probably for
the same reason.
3.3. Synthesis of (N-vinylcarbazole)Cr(CO)2(C3H5)þBF ꢀ
(5)
4
The synthesis of ðN-vinylcarbazoleÞ-CrðCOÞ ðC3H5Þþ
2
BFꢀ4 (5) was accomplished by the photochemical reaction
of 2 with equimolar amounts of allyl alcohol and fluoboric
acid in anhydrous ether solution for 8 h. Choice of anhy-
drous ether as the reaction solvent was essential because
the ionic complex formed is practically insoluble in this sol-
vent and precipitates, thus avoiding decomposition under
the action of UV irradiation. After work-up, complex 5
as a brick red solid is insoluble in most of organic solvents
(e.g., benzene, CH2Cl2, and THF) but soluble in acetone,
DMSO and pyridine. The recrystallization and chromatog-
raphy did not lead to further purification due to its extreme
sensitivity toward air and moisture. However, its pure form
could be obtained by numerous reprecipitations from the
acetone/ether mixture. We tried to replace 2 by 3 in this
20000
15000
10000
5000
0
2.0
1.0
0.0
reaction, but no bisðN-vinylcarbazoleÞCrðCOÞ ðC3H5Þþ
2
BFꢀ4 complex was found as a product and the reactions
involved appear to be non-specific. Complex 5 resonates
at 151.3 ppm in the 19F NMR spectrum. Elemental analysis
is consistent with the given formulas and the 13C, 1H NMR
signals for each ligand appear in their characteristic
regions. Two doublets at ꢁ1.22 ppm (J = 6.7–10.4 Hz)
and ꢁ3.52 (J = 10.4 Hz) ppm were observed for the termi-
nal protons of the p-allyl ligand, which should be assigned
to the sin-protons and anti-protons, respectively. The cen-
tral proton shows itself as a broad singlet at 3.77 ppm.
Compared to 2 and 4, the resonances of the Cr-ring proton
and carbon in 5 are shifted downfield and the carbonyl sig-
nals are shifted upfield due to the strong electron-with-
drawing effect of the ðC3H5ÞþBF4ꢀ ligand. Correspond-
ingly, the mco bands in the IR spectrum are markedly
0
10
20
30
40
50
60
70
Conversion (%)
Fig. 6. Mn (left axis: open, circles for 2 and triangles for 4) and PDI (right
axis: solid, circles for 2 and triangles for 4) as a function of conversion for
The SFRP at 80 ꢁC.

1002
X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
nitrogen freezing for a day and then restarted after adding
a fresh aliquot of the complexes 2 and 4, respectively
(Fig. 7). The slope of the ln([M]0/[M]) vs. time plot is iden-
tical before and after the interruption within experimental
error, indicating that the total number of chains is con-
served. The other evidence for SFRP is the reduction of
the polymerization rate and complete disappearance of
the autoacceleration, which did not occur in the case of 1
(Fig. 8). This phenomenon is one of the general features
of controlled radical polymerization [17]. Indeed, the poly-
mer chain growth control is connected with the known sta-
bilization of a free radical in the coordination sphere of the
metal [18]. At the beginning of the reaction, a fast revers-
ible equilibrium is established between a reactive free radi-
cal and a stable free radical (or persistent radical), which
can be represented as a resonance structure containing an
unpaired electron at the chromium atom (see Scheme 2).
In situ the carbonyl groups and, even more strikingly, the
PPh3 group appear to form a very efficient ‘‘umbrella’’ over
the chromium atom, making its coordination sphere
practically inaccessible to attack by another reactant. For
the unpaired electron transfer (through-bond) in the orga-
nometallic radical, the exocyclic N–Ca bond is essentially
coplanar with carbazole ring to maximize orbital overlap
and electron delocalization between the two. It is also inter-
esting to point out that, with regard to the neighboring
group participation by chromium in the N-vinyl reaction
[8a,19], the N–Ca bond may be inclined from the carbazole
plane to afford closed-shell electron configurations
(through-space) at both chromium and the active carbons.
As a result of the competing direct chain propagation, the
Mn values of the polymers formed are consistently below
20,000. Attempts to polymerize the dinuclear complex 3
were unsuccessful even after prolonged periods at elevated
temperature (100 ꢁC, 56 h). Two Cr(CO)3 groups on one
carbazole ring increase steric compression and stabilization
of the resulting radical to prevent the chain propagation.
Complex 5 is not sufficiently stable to carry out the poly-
merization under those conditions. This reaction was not
studied in further detail.
1.2
1.0
0.8
Polymers 6 and 7 were subjected to thermal gravimetri-
cal analysis (TGA) under a nitrogen atmosphere. The
decomposition of 7 took place in three well-separated
stages with maximum rate at 240, 280 and 440 ꢁC, in con-
trast to 6, which decomposed in two stages with maximum
rate at 290 and 455 ꢁC. The IR spectrum of 7 after heating
to 250 ꢁC under nitrogen and then cooling shows no bands
assignable to the CO groups, indicating that its initial
weight loss was attributed to the liberation of carbon mon-
oxide from the chromium center. Correspondingly, the sec-
ond stage can be explained to result from the disruption of
Cr–C (ring) bonds concomitant with the cleavage of partial
P–Cr bonds [20]. Other features noted are that no clear
phase transition was observed in DSC scans up to
300 ꢁC. As with the polar, optically anisotropic chromoph-
ores as pendant groups in the side chains, these metallo-
polymers possess high glass transition temperature and
therefore the electric field induced dipole orientation can
conceivably be fixed after the film-forming, corresponding
to the lowest degree of orientational relaxation. Assembled
arrays of Cr-bound carbazole groups obtained in the more
stable poled film make this class of materials a very com-
petitive one for optoelectronics.
0.6
fresh 2 and 4 added
0.4
0.2
0.0
0
500
1000
1500
2000
2500
Time (min)
Fig. 7. Plots of ln([M0]/[M]) vs. time for the SFRP with 2 (open circles)
and 4 (solid symbols).
100
1
2
50
4
3.5. Electronic absorption and fluorescence spectroscopy
The electronic absorption spectra of the complexes have
been determined to elucidate the effects of donor–acceptor
properties of various ligands. The measurements were car-
ried out using a solution of their analytically pure samples
under a nitrogen atmosphere. The spectra of complexes 2
and 3 are characterized by two rather strong bands in the
region of 340–355 and 395–440 nm (Fig. 9). The lower
energy bands are assigned to the metal-to-ligand charge-
0
0
200
400
Time (min)
600
800
1000
Fig. 8. Conversion vs. time for the solution polymerization at 80 ꢁC with 1
(squares), 2 (open circles) and 4 (solid circles).

X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
1003
Mn
CH2
CH
CH2
CH
Mn
CH2
CH
N
N
N
Mn
+
OC
Cr
OC
Cr
OC
Cr
L
L
L
CO
CO
CO
L = CO (6); PPh3 (7)
Scheme 2.
energy bands immediately adjacent have higher extinction
coefficients and can be regarded as the intraligand p ꢀ p*
transition (IL). The change from 2 to 3 causes a subtle
red-shift both in the lower energy band by 35 nm and the
higher energy band by 8 nm. A plausible explanation is
the electronic transitions within the dinuclear-carbazole
ring system accompanied by the electronic communication
between the two metal centers. Additional electron density
from each chromium center may interact with the carba-
zole ligand in order to decrease the coordinative unsatura-
tion, thereby increasing delocalization of the p electrons
into the carbazole system. The differences in red-shift
extent of MLCT and IL bands can be viewed as the popu-
lation redistribution associated with the charge-transfer
transition. These results would assist us in evaluating the
question how well the electronic behavior of the coordi-
nated carbazole ring was ‘‘matched’’ to that of the Cr(CO)3
center. It seems likely that 3 would prove to be better
matched because of a red-shift of its absorption bands rel-
ative to 2, again consistent with the aforementioned con-
version of 2 into 3 in solution. Replacing a CO with a
phosphine ligand as in 4, causes the low-energy band to
have a large molar absorptivity and extend past 600 nm,
but no observable changes occur in the high-energy band.
This can be attributed to the addition effect of the extinc-
tion coefficients of the PPh3 unit and the mixing of the
intraligand (IL) p ꢀ p* (PPh3) transition into the MLCT
transition due to the lone-pair contribution from the phos-
phorus atom. Complex 5 exhibits a high-energy absorption
band at about 330 nm with a long tail extending far into
the visible. Perhaps the greater extent of the blue-shift for
the MLCT band than the IL band, which was causeþd by
the strong electron-withdrawing effect of ðC3H5Þ BF4ꢀ
ligand, makes these two bands be substantially overlap-
ping, and it is not accessible to accurately assign their posi-
tions. The electronic spectra of polymer 6 and 7 were also
examined under the same conditions and show no clear
difference in energy compared to their corresponding
monomers (Fig. 10), suggesting that Cr-bound carbazole
served as the pendant chromophore is practically un-
affected by the vinyl-type polymerization due to their
unconjugated polymer backbone.
2.0
1.5
1.0
0.5
0.0
4
3
2
5
1
250
300
350
400
450
500
550
600
Wavelength (nm)
Fig. 9. Electronic absorption spectra of 1–4 in dichloromethane and 5 in
acetone at 298 K.
transfer (MLCT) transitions (both M ! pꢃco and M ! pLꢃ )
in accord with their negative solvatochromium and appar-
ent sensitivity to the electronic nature of the ligands. As
reported on the related system [5a,21], the MLCT transi-
tion band is expected to be smaller in its intensity owing
to the non-coincidence of the ground-state dipole moment
and the charge-transfer directions, it is reasonable to sug-
gest that these transitions of the lower energy bands may
contain considerable ligand field character on the basis of
their fairly large molar absorptivities (Table 3). The higher
Table 3
Electronic absorption and emission data in dichloromethane at 298 K
unless otherwise stated
Compound
k
max, nm ( 10ꢀ3e, Mꢀ1 cmꢀ1
)
kemb, nm (102U)
1
2
3
4
328 (38), 341 (40)
337 (30), 395 (15)
345 (30), 430 (16)
329 (41), 342 (42),
365 (32), 430sh (11)
333 (33), 370sh (10)
334 (30), 402 (15)
332 (29), 343 (30)
330 (30), 344 (30),
363 (29), 398sh (22)
353, 366sh (42)
353sh, 369 (3.8)
349, 364sh (1.7)
353, 367sh (7.5)
5a
394 (ꢁ0.09)
6
351, 366 (3.1)
354, 420sh (2.9)
380, 420sh (56)
7
All complexes in this work are emissive in a degassed
solution at room temperature, and the emissive behaviors
are sensitive to their concentration in the range from
10ꢀ3 to 10ꢀ5 mol dmꢀ3, but independent of the excitation
PVK
a
Measured in acetone.
The excitation wavelength is 315 nm.
b

1004
X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
2.0
1.5
1.0
0.5
0.0
PVK
6
7
7
6
PVK
350
400
450
500
550
600
250
300
350
400
450
500
550
600
Wavelength (nm)
Wavelength (nm)
Fig. 12. Emission spectra of 6, 7 and PVK in dichloromethane at 298 K.
Fig. 10. Electronic absorption spectra of 6, 7 and PVK in dichlorometh-
ane at 298 K.
from 1 to 3 is consistent with the increased heavy-atom
effect. It is also clear that variation of the ligands has some-
what of an influence on the emission intensity, wherein an
order of 4 > 2 > 5 is expected on the basis of the different
accepting capacities of the ligands, i.e., PPh3 < CO <
ðC3H5ÞþBFꢀ4 . Extension of the observations on these
small-molecules to the macromolecular level is straightfor-
ward in Fig. 12. Polymer 6 totally manifests the emissive
behavior of its parent complex 2. No new emission bands
appeared at a relatively longer wavelength excludes the
possibility of an overlapping, eclipsed, sandwich-like con-
formation concerning excimer formation [24]. To our sur-
prise, polymer 7 exhibits, in addition to the 350 nm
wavelength used. The emissions of 2–4 appear in the same
region and have structural features very similar to that of 1
(Fig. 11), suggesting the same source from the carbazole
ligand-dominating emissive state. Thus their lowest emis-
sive states can tentatively be assigned as metal-perturbed
p ꢀ p* transitions. However, complex 5 shows broad emis-
sion with a maximum at 394 nm, whose behavior of the
change in spectral shape and position is considered to orig-
inate from the significant perturbation of the cationic p-
allyl ligand. In addition, we observed a notable but not
dramatic quenching effect for these complexes as far as
fluorescence quantum yields are concerned [22]. We can
possibly attribute the quenching in the emission intensity
to the so-called heavy-atom effect of the central chromium,
which triggers the non-radiative deactivation of excited
states of the fluorophore. Due to the lack of available
low-lying metal-localized excited states [23], it is unlikely
that energy or electron-transfer quenching mechanism
operates here. The trend of the emission maxima decrease
monomer fluorescence,
a
structureless emission at
k > 380 nm which compares well with that observed for
PVK. This is tempting to be excimer-like in its behavior
whose origin undoubtedly results from the formation of
an intrachain excimer state. Apparently, the very bulky
Cr(CO)2PPh3 fragment does not sterically avert overlap
of the pendant organometallic chromophores. These
inspections led us to speculate that, irrespective of the steric
compression, the electronic property of the ligand is also a
crucial factor, which determines the appropriate disposi-
tion of two Cr-bound carbazole groups and the formation
of excimers. Encouraged by these results, we are continuing
our studies on these metallopolymers, in particular, their
photophysical properties concerning themselves with the
configurational requirements.
1
4
5
3
2
4. Conclusion
We have shown the synthesis and isolation of mono-
(2) and bis-Cr(CO)3 (3) complexes of N-vinylcarbazole,
which are well promising for optoelectronic applications
because of their organometallic push–pull structures,
ability of vinyl-type polymerization and versatility in
structural modification following a rational exchange of
a CO group. To our knowledge, it is also the first
instance that conversion from 2 to 3 occurs upon heating
350
400
450
500
550
600
Wavelength (nm)
Fig. 11. Emission spectra of 1–4 in dichloromethane and 5 in acetone at
298 K.

X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
1005
in coordinating solvents because 3 is thermodynamically
more stable than 2. The N-vinylcarbazoleCr(CO)2L com-
plexes, where L = PPh3 (4), ðC3H5ÞþBF4ꢀ (5), are
acquired as comparable products, which possess varying
physico-chemical properties as these ligands exhibit dif-
ferent donor–acceptor properties (ðC3H5ÞþBF4ꢀ is a very
strong acceptor, CO is a strong p-acceptor and PPh3 is
a moderate donor and acceptor). Single-crystal X-ray
structures were obtained for 2–4. By structure–property
correlation, the followings can be summarized as own
conclusion. The N-vinyl double bond length seems rather
sensitive to the electronic nature of the ligand, shortening
for compound 3, No. 285131 for compound 4 and No.
285132 for compound 1. Copies of this information may
be obtained free of charge from The Director, CCDC, 12
Union Road, Cambridge, CB2 1EZ, UK (fax: +44 1223
336033; e-mail: deposit@ccdc.cam.ac.uk or http://www.
ccdc.cam.ac.uk). Supplementary data associated with this
article can be found, in the online version, at
doi:10.1016/j.jorganchem.2005.10.052.
References
[1] J. Mort, G. Pfister, in: J. Mort, G. Pfister (Eds.), Electronic Properties
of Polymers, Wiley Interscience, New York, 1982.
[2] V.S. Mylnikov, Adv. Polym. Sci. 115 (1994) 1.
[3] (a) G. Zotti, G. Schiavon, S. Zecchin, J.-F. Morin, M. Leclerc,
Macromolecules 35 (2002) 2122;
˚
by more than 0.029 A upon the replacement of one CO
group with PPh3 ligand. The distances of Cr from the
carbazole plane, which allows estimates of the strength
of the metal–ligand p-bonding, are equal to each other
for 2 and 4 but shorter for 3. Electronic absorptions of
these complexes extend over a wide range and can be
varied by modifying the ligands. Furthermore, a subtle
red-shift in absorption bands caused by change from 2
to 3 provides support for the electronic communication
between the two Cr(CO)3 groups on one carbazole unit.
Despite a notable but not dramatic quenching effect, the
emission properties of the complexes 2–4 hardly differ
from each other and are similar to that of their organic
counterpart 1, suggesting the carbazole ligand-dominat-
ing emissive state. However, the emission profile of com-
(b) S.-W. Hwang, Y. Chen, Macromolecules 34 (2001) 2981;
(c) B. Andre, R. Lever, J.Y. Moisan, J. Chem. Phys. 137 (1989) 281;
(d) M. Yokayama, S. Shimokihara, A. Matsubara, H. Mikawa, J.
Chem. Phys. 76 (1982) 724;
(e) C.I. Simionescu, V. Percec, G. Ioanid, Rev. Roum. Chim. 24
(1979) 1077.
[4] (a) J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Principles
and Applications of Organotransition Metal Chemistry, University
Science Books, Mill Valley, CA, 1987;
(b) W.E. Silxerthorn, in: F.G.A. Stone, R. West (Eds.), Advances in
Organometallic Chemistry, vol. 13, Academic, New York, 1975, p. 48;
(c) J.E. Sheats, C.E. Carraher, C.U. Pittman Jr., Metal-containing
Polymeric Systems, Plenum Press, New York, 1985;
(d) E.L. Muetterties, J.R. Bleeke, E.J. Wucherer, T.A. Albright,
Chem. Rev. 82 (1982) 499;
plex
5 changes dramatically due to a significant
(e) R.H. Crabtree, Chem. Rev. 85 (1985) 245.
perturbation of the cationic p-allyl ligand. The polymer-
ization behavior is proposed to be associated with the
influence of the metal atom, which makes the organome-
tallic radical or corre- sponding macroradical inert and
of low polymerizability. Under radical initiation condi-
tions, complexes 3 and 5 could not be polymerized, while
complexes 2 and 4 displayed a controlled polymerization
by the stable organometallic radical polymerization
mechanism and gave the corresponding well-defined
polymers 6 and 7. An emission spectral examination of
the resultant polymers provides notable insight into the
fact that electronic effects of the ligands dominate the
intrachain excimer formation, even those such as PPh3
which are largely bulky. Further photophysical studies
of these chromium–carbazole complexes and their poly-
mers will be reported in due course.
[5] (a) K. Shen, X. Tian, J. Zhong, J. Lin, Y. Shen, P. Wu, Organomet-
allics 24 (2005) 127;
(b) T.J.J. Mu¨ller, M. Ansorge, H.J. Lindner, Chem. Ber. 129 (1996)
1433.
[6] (a) C.A. Merlic, M.M. Miller, B.N. Hietbrink, K.N. Houk, J. Am.
Chem. Soc. 123 (2001) 4904;
(b) C.A. Merlic, B.N. Hietbrink, K.N. Houk, J. Org. Chem. 66
(2001) 6738;
(c) C.A. Merlic, J.C. Walsh, D.J. Tantillo, K.N. Houk, J. Am. Chem.
Soc. 121 (1999) 3596;
(d) S.G. Davies, T.J. Donohoe, Synlett (1993) 323;
(e) C. Michon, J.-P. Djukic, Z. Ratkovic, J.-P. Collin, M. Pfeffer, A.
De Cian, J. Fischer, D. Heiser, K.H. Do¨tz, M. Nieger, Organomet-
allics 21 (2002) 3519.
[7] (a) H. Kobayashi, M. Kobayashi, Y. Kaizu, Bull. Chem. Soc. Jpn. 46
(1973) 3109;
(b) H. Kobayashi, M. Kobayashi, Y. Kaizu, Bull. Chem. Soc. Jpn. 48
(1975) 1222;
(c) S.D. Worley, T.R. Webb, J. Organomet. Chem. 192 (1980) 139;
(d) P.G. Sennikov, V.A. Kuznetsov, A.N. Egorochkin, N.I. Sirotkin,
R.G. Nazarova, G.A. Razuvaev, J. Organomet. Chem. 190 (1980)
167;
Acknowledgments
(e) R. Lal De, J. von Seyerl, L. Zsolnai, G. Huttner, J. Organomet.
Chem. 175 (1979) 185.
[8] (a) M. Cais, M. Kaftory, D.H. Kohn, D. Tatarsky, J. Organomet.
Chem. 184 (1979) 103;
(b) R. Dabard, G. Jaouen, G. Simonneaux, M. Cais, D.H. Kohn, A.
Lapid, D. Tatarsky, J. Organomet. Chem. 184 (1980) 91;
(c) M.J. Wovkulich, J.D. Atwood, J. Organomet. Chem. 184 (1979) 77;
(d) G. Jaouen, R. Dabard, J. Organomet. Chem. 72 (1974) 377.
[9] (a) C.U. Pittman Jr., O.E. Ayers, S.P. McManus, Macromolecules 7
(1974) 737;
We are grateful to the National Natural Science Foun-
dation of China (Grant No. 20274011) for generous finan-
cial support. We thank Dr. Jie Sun for assistance in X-ray
data collection and refinement of compounds 1–4.
Appendix A. Supplementary data
Crystallographic data for the structural analysis has
been deposited with the Cambridge Crystallographic Data
Centre, CCDC No. 285129 for compound 2, No. 285130
(b) C.U. Pittman Jr., R.L. Voges, J. Elder, Macromolecules 4 (1974)
302;

1006
X. Tian et al. / Journal of Organometallic Chemistry 691 (2006) 994–1006
(c) C.U. Pittman Jr., G.V. Marlin, J. Polym. Sci. 11 (1973) 2753;
(d) C.U. Pittman Jr., P.L. Grube, O.E. Ayers, S.P. McManus, M.D.
Rausch, G.A. Moser, J. Polym. Sci. A 10 (1972) 379.
[17] (a) D.F. Grishin, L.L. Semyonycheva, A.N. Artemov, E.V. Telegina,
N.B. Valetova, I.S. Illichev, Appl. Organometal. Chem. 17 (2003) 717;
(b) D.F. Grishin, L.L. Semyonycheva, I.S. Illichev, A.N. Artemov,
Appl. Organometal. Chem. 15 (2001) 169.
[18] (a) A.D. Pomogailo, V.S. Savostyanov, Synthesis and Polymerization
of Metal-containing Monomers, CRC Press, Boca Raton, 1994;
(b) D.F. Grishin, Russ. Chem. Rev. 62 (1993) 951.
[10] (a) G.M. Sheldrick, SHELXS-97, Program for Crystal structure Solu-
tion, Gottingen University, Germany, 1997;
(b) G.M. Sheldrick, SHELXS-97, Program for Crystal structure
Refinement, Gottingen University, Germany, 1997.
[11] Y. Oprunenko, S. Malyugina, P. Nesterenko, D. Mityuk, O.
Malyshev, J. Organomet. Chem. 597 (2000) 42.
[12] (a) T.A. Albright, P. Hofmann, R. Hoffmann, C.P. Lillya, P.A.
Dobosh, J. Am. Chem. Soc. 105 (1983) 3396;
[19] (a) C.A. Merlic, M.M. Miller, Organometallics 20 (2001) 373;
(b) D.J. Tantillo, B.N. Hietbrink, C.A. Merlic, K.N. Houk, J. Am.
Chem. Soc. 122 (2000) 7136;
(c) T.G. Traylor, M.J. Goldberg, J. Am. Chem. Soc. 109 (1987) 3968.
[20] The EI-MS spectrum of the parent complex 4 shows the (CrPPh3+
)
(b) E.P. Kundig, V. Grivet, C. Desobry, B. Spichiger, S. Rudolph,
¨
Organometallics 6 (1987) 1173;
(c) K.H. Do¨tz, J. Stendel Jr., S. Muller, M. Nieger, S. Ketrat, M.
¨
Dolg, Organometallics 24 (2005) 3219.
ion as the thermodynamically more stable fragment with the
highest intensity, implying the strength of the phosphorus-binding
pattern.
[13] (a) S.C. Jones, T. Hascall, S. Barlow, D. OꢀHare, J. Am. Chem. Soc.
124 (1998) 11610;
[21] (a) D.R. Kanis, M.A. Ratner, T.J. Marks, J. Am. Chem. Soc. 114
(1992) 10338;
(b) S. Barlow, D. OꢀHare, Chem. Rev. 97 (1997) 637.
[14] R.D. Rogers, J.L. Atwood, T.A. Albright, W.A. Lee, M.D. Rausch,
Organometallics 3 (1984) 263.
(b) T.M. Gilbert, F.J. Hadley, C.B. Bauer, R.D. Rogers, Organo-
metallics 13 (1994) 2024;
(c) T.J.J. Muller, J. Blumel, J. Organomet. Chem. 683 (2003) 354.
¨
¨
[15] (a) A.D. Hunter, L. Shilliday, W.S. Furey, M.J. Zaworotko,
Organometallics 11 (1992) 1550;
[22] Y. Morisaki, H. Chen, Y. Chujo, J. Organomet. Chem. 689 (2004)
1271.
(b) A.D. Hunter, V. Mozol, S.D. Tsai, Organometallics 11 (1992) 2251;
(c) J. Li, A.D. Hunter, R. McDonald, B.D. Santarsiero, S.G. Bott, J.L.
Atwood, Organometallics 11 (1992) 3050.
[23] Excitation into the MLCT absorption (k > 450 nm) band does not
result in any emission.
[24] (a) A. Karali, P. Dais, E. Mikros, F. Heatley, Macromolecules 34
(2001) 5547;
[16] (a) M. Ansorge, T.J.J. Muller, J. Organomet. Chem. 585 (1999) 174;
¨
(b) Y. Huang, H.L. Uhm, D.F.R. Gilson, I.S. Butler, Inorg. Chem.
36 (1997) 435;
(b) H. Sahai, A. Itaya, H. Masuhara, K. Sasaki, S. Kawata, Polymer
37 (1996) 31;
(c) T.E. Bitterwolf, Polyhedron 7 (1988) 1377.
(c) G.E. Johnson, J. Chem. Phys. 62 (1982) 4697.