Characterization of the μ-(η1-C: Η2-S,S′) dithiocarboxylate complexes Cp(CO)2Fe-CS2-Zr(X)Cp2 (X=Cl, OCMe3); CS2 insertion into the FeZr bond of the heterobimetallic complex Cp(CO)2Fe-Zr(OCMe3)Cp2

Article abstract of DOI:10.1016/S0022-328X(98)00752-9

Treatment of the carbon disulfide adducts FpCS₂K and Fp′CS₂K [Fp′=(η⁵-C₅H₄CH ₃)Fe(CO)₂] with Cp₂ZrCl₂ affords the μ-(η¹-C: η²-S,S′) dithiocarboxylate complexes FpCS₂ZrClCp₂ (1) and Fp′CS₂ZrClCp₂ (2). Both stable products were fully characterized. Metathesis between FpCS₂K and Cp₂ZrCl(OCMe₃) provided FpCS₂Zr(OCMe₃)Cp₂ (3), which was not obtained analytically pure. This product was characterized by comparison of its IR and ¹H-, ¹³C{¹H}-NMR spectral data with that for 1 and 2. The iron-zirconium complex FpZr(OCMe₃)Cp₂ (4) was transformed by one equivalent of CS₂ to 3 (75% spectroscopic yield), a reaction that did not occur for FpZrClCp₂. An insertion pathway is discussed for incorporating the CS₂ into the Fe-Zr bond of FpZr(OCMe₃)Cp₂.

Full text of DOI:10.1016/S0022-328X(98)00752-9

Journal of Organometallic Chemistry 566 (1998) 1–7
Characterization of the v-(p1-C: p2-S,S%) dithiocarboxylate complexes
Cp(CO)2Fe–CS2–Zr(X)Cp2 (X=Cl, OCMe3); CS2 insertion into the
FeZr bond of the heterobimetallic complex
Cp(CO)2Fe–Zr(OCMe3)Cp2
John R. Pinkes, Stephen M. Tetrick, Bruce E. Landrum, Alan R. Cutler *
Department of Chemistry, Rensselaer Polytechnic Institute, Troy, NY 12180, USA
Received 4 June 1997; received in revised form 18 May 1998
Abstract
Treatment of the carbon disulfide adducts FpCS2K and Fp%CS2K [Fp%=(p5-C5H4CH3)Fe(CO)2] with Cp2ZrCl2 affords the
v-(p1-C: p2-S,S%) dithiocarboxylate complexes FpCS2ZrClCp2 (1) and Fp%CS2ZrClCp2 (2). Both stable products were fully
characterized. Metathesis between FpCS2K and Cp2ZrCl(OCMe3) provided FpCS2Zr(OCMe3)Cp2 (3), which was not obtained
1
analytically pure. This product was characterized by comparison of its IR and H-, 13C{1H}-NMR spectral data with that for 1
and 2. The iron–zirconium complex FpZr(OCMe3)Cp2 (4) was transformed by one equivalent of CS2 to 3 (75% spectroscopic
yield), a reaction that did not occur for FpZrClCp2. An insertion pathway is discussed for incorporating the CS2 into the Fe–Zr
bond of FpZr(OCMe3)Cp2. © 1998 Elsevier Science S.A. All rights reserved.
1. Introduction
CO2 as FpCO2− [3], which then trapped the zirconocene
electrophile and produced the observed
FpCO2ZrClCp2. This ionization pathway, however, was
inconsistent with the results of carbon disulfide trap-
ping experiments. If Fp–ZrClCp2 did ionize under the
conditions of the CO2 insertion experiments (THF,
0°C), then it should intercept CS2 as the relatively
stable dithiocarboxylate FpCS2− [4] and perhaps form
the (then unreported) FpCS2ZrClCp2 (1). We observed
however that Fp–ZrClCp2 was inert towards CS2 under
comparable reaction conditions.
Recently we reported that Fp–ZrClCp2 [Fp=
Fe(CO)2Cp; Cp=p5-C5H5] incorporates carbon diox-
ide and efficiently gives the known [1] FeZr v-(p1-C:
p2-O,O%) bimetallocarboxylate FpCO2ZrClCp2, Eq. 1
[2]. A CO2 insertion pathway requiring bifunctional
activation and insertion of the CO2 was favored. An
alternative pathway required prior ionization of the
iron–zirconium bond on Fp–ZrClCp2 to give Fp− and
Cp2ZrCl(THF)+; the resulting Fp− could intercept the
Other examples of CO2 incorporation into metal–
metal bonds of heterobimetallic complexes have been
reported [5]. Several examples of CS2 insertion into
* Corresponding author. Tel.: +1 518 2768447. Fax: +1514
2764045.
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00752-9

2
J.R. Pinkes et al. / Journal of Organometallic Chemistry 566 (1998) 1–7
metal–metal bonds also are known ([5]b,c), the most
recent example having been reported for the Fe–Zr
complex HC(SiMe2NC6H4F-2)3ZrFe(CO)2Cp ([5]c,
[6]). We now report the synthesis and characterization
of the v-(p1-C: p2-S,S%) dithiocarboxylate complexes
FpCS2ZrClCp2 (1) and Fp%CS2ZrClCp2 (2) [Fp%=
Fe(CO)2(p5-C5H4CH3)], the partial characterization of
FpCS2Zr(OCMe3)Cp2 (3), and details on the reactions
of CS2 with FpZr(X)Cp2 (X=Cl, OCMe3).
greenish brown solid having a yellow hue and left a
light orange supernatant solution, which was re-
moved. The remaining solid was extracted with 75 ml
of benzene, and the filtrate was evaporated to a
green-brown solid. This solid was suspended in ben-
zene (15 ml), treated with hexane (100 ml), and
filtered. The 380 mg of greenish brown powder (with
a yellowish hue) that remained after washing with
hexane and vacuum drying was identified as
FpCS2Zr(Cl)Cp2 (1), yield 70% and m.p. 187–191°C
(decompositon). IR (CH2Cl2) 2034, 1987 cm−1; IR
(KBr) 2037, 1968 cm−1 (CO), 934, 851 cm−1
(SCS); 1H-NMR (C6D6) l 5.92 (s, Cp2Zr), 4.13 (s,
CpFe); 13C{1H}-NMR (CDCl3) 323.62 (s, CS2),
212.05 (s, CO), 112.32 (Cp2Zr), 87.84 (CpFe). Anal.
Calc. for C18H15O2S2FeZrCl: %C, 42.40; %H, 2.96.
Found: %C, 42.43; %H, 2.86. A somewhat lower
yield (39%) was realized by working up the reaction
by benzene extraction (5×5 ml) and precipitation
from THF-pentane (5–25 ml, −20°C).
2. Experimental section
2.1. Materials
Synthetic manipulations were performed in a nitro-
gen atmosphere using a combination of standard
Schlenk line, glovebox, and vacuum line procedures
[7]. Infrared spectra were recorded on a Perkin-Elmer
Model 1600 spectrophotometer. 1H- and 13C{1H}-
NMR spectra were recorded in C6D6 (which had been
stored over 3A molecular sieves) using a Varian
Unity 500 spectrometer, and the data were reported
as l values relative to residual C6D5H (1H: 7.15 ppm)
and C6D6 (13C: 128.00 ppm). Tetrahydrofuran (THF),
diethyl ether, pentane, and hexane were distilled from
sodium/benzophenone ketyl, and methylene chloride
was distilled from P2O5. The organometallic com-
2.3. Synthesis of
(p5-C5H4CH3)(CO)2FeCS2Zr(Cl)(p5-C5H5)2 (2)
A THF solution of Fp%K (1.0 mmol, 10 ml) was
generated by sonication (1 h) of Fp% (192 mg, 0.50
2
mmol) with potassium metal (150 mg) ([8]c), (CO)
1866, 1792 cm−1. The dark red centrifugate was
cooled to −23°C and treated with CS2 (60 ml, 1.0
mmol); an IR spectrum of the resulting orange-red
solution after 20 min indicated quantitative formation
of Fp%CS2K ([4]a, [12]), (CO) 1966, 1942 cm−1. Ad-
dition of a THF solution (10 ml) of Cp2ZrCl2 (293
mg, 1.0 mmol) to this precooled solution (−23°C)
provided a dark red solution within 20 min, (CO)
2026, 1979 cm−1. The reaction mixture was evapo-
rated at room temperature and the green-brown
residue was slurried with 5 ml of benzene before 100
ml of hexane was added. A light greenish brown solid
and a light yellow-brown supernatant solution re-
sulted, which was filtered. The remaining solid was
extracted with 75 ml of benzene (light golden brown),
filtered, and the filtrate was evaporated to a golden
brown solid. This solid was suspended in 5 ml of
benzene, diluted with excess hexane, and filtered. The
remaining mustard-colored powder was washed with
hexane (3×5 ml) and dried in vacuo, yielding 310
mg (59% yield) of Fp%CS2Zr(Cl)Cp2, m.p. 170–172°C
(decomposition); IR (KBr) 2023, 1972 cm−1 (CO),
922, 848 cm−1 (SCS); 1H-NMR (C6D6) l 5.94 (s,
Cp2Zr), 4.13 (m), 4.00 (m) (Cp%Fe), 1.43 (s) (CH3–
Cp%); 13C{1H}-NMR (C6D6) l 325.59 (CS2), 213.64
(CO), 112.58 (Cp2Zr), 87.78, 86.72 (Cp%Fe), 12.38
(CH3). Anal. Calc. for C19H17O2S2FeZrCl: %C, 43.55;
%H, 3.27. Found: %C, 43.72;%H, 3.24.
pounds
Cp(CO)2FeK
(FpK)
[8],
[(p5-
C5H4CH3)Fe(CO)2]2 or (Fp%) [9], Cp2ZrCl(OCMe3)
2
[10], and FpZr(X)Cp2 (X=Cl, OCMe3) [11] were pre-
pared by literature procedures and judged pure by IR
and
1H-NMR
spectroscopy.
Samples
of
FpCS2Fe(CO)Cp and FpC(S)SFp were available from
a previous study [12]. All other reagents were used as
received. Elemental microanalyses were performed by
Quantitative Technologies, N.J.
2.2. Synthesis of
(p5-C5H5)(CO)2FeCS2Zr(Cl)(p5-C5H5)2 (1)
To a dark orange solution of FpK (218 mg, 1.0
mmol) in 10 ml of THF, precooled to −23°C, was
added 60 ml of carbon disulfide (1.0 mmol). An IR
spectrum of the orange-red solution after 20 min was
consistent with quantitative transformation of FpK to
FpCS2K ([4]a, [12]), (CO) 1999, 1945 cm−1. To this
solution was added Cp2ZrCl2 (293 mg, 1.0 mmol) in
10 ml of THF, and an IR spectrum of the crimson
solution after 20 min showed only absorptions that
were attributed to FpCS2Zr(Cl)Cp2 (1), (CO) 2029,
1982 cm−1, along with residual CS2, 1520 cm−1. The
THF was evaporated at room temperature and the
resulting brown solid was suspended in 15 ml of ben-
zene. Addition of 100 ml of hexane precipitated a

J.R. Pinkes et al. / Journal of Organometallic Chemistry 566 (1998) 1–7
3
2.4. Preparation of
ence of FpZr(OCMe3)Cp2 (4), (CO) 1942, 1887 cm−1
[11] along with at most traces of Fp2 (1992, 1954,
1782 cm−1) and FpH (2013, 1953 cm−1). The solu-
tion was evaporated and benzene extracts of the
residue (3×5 ml) were filtered through oven-dried
Celite. After evaporation of the combined extracts,
the residue was dissolved in 5 ml of toluene and the
solution was cooled to −78°C. Bright yellow crystals
deposited of 4 formed, which were separated from the
supernatant solution, washed with ether (3 ml), and
dried under vacuum. Yield 4 30 mg (19%); IR (KBr)
1937, 1870 cm−1 (CO), 1358, 1182, 1002, 829, 797,
(p5-C5H5)(CO)2FeCS2Zr[OCMe3](p5-C5H5)2 (3)
A precooled (−23°C) solution of FpK (0.243 g,
1.12 mmol) in THF (10 ml) was treated with CS2 (70
ml, 1.16 mmol) for 20 min before a 5 ml THF solu-
tion of Cp2ZrCl(OCMe3) (0.371 g, 1.12 mmol) was
added. An IR spectrum of the orange-red solution
after 20 min was consistent with quantitative forma-
tion of FpCS2Zr(OCMe3)Cp2 (3), (CO) 2024, 1976
cm−1. The solution was warmed to room temperature
and the solvent was evaporated. Benzene extraction
(4×5 ml) and filtration followed by concentration of
the solution to 5 ml left a red-brown suspension,
which was diluted with 100 ml of pentane. The result-
ing brown precipitate was filtered, washed with pen-
tane (3×10 ml), and dried in vacuo: yield 530 mg of
material that contains 82% FpCS2Zr(OCMe3)Cp2 (3)
(as ascertained by 1H-NMR spectral integration ver-
sus PhOMe internal standard) along with low concen-
trations of several unidentified Cp2Zr-containing and
other organic residues. IR (THF) 2024, 1976(vs),
738 cm−1 1H-NMR (C6D6) l 5.91 (Cp2Zr), 4.10
;
(CpFe), 1.14 (Me); 13C{1H}-NMR (C6D6) l 219.81
(CO), 110.58 (Cp2Zr), 82.43 (CpFe), 80.67 (OCMe3),
31.72 (Me).
2.6. Reaction of (p5-C5H5)(CO)2FeZr(OCMe3)Cp2 (4)
with CS2
A THF solution of FpZr(OCMe3)Cp2 (4) (1.0
mmol, 20 ml) was generated from FpK (218 mg) and
Cp2ZrCl(OCMe3) (330 mg); an IR spectrum of the
yellow solution after 10 min was consistent with the
presence of only 4, (CO) 1942, 1887 cm−1 [11],
along with at most trace amounts of Fp2 (1992, 1950,
1782 cm−1) and FpH (2012, 1990 cm−1). CS2 (60.0
ml, 1.0 mmol) was added by syringe, and the solution
was stirred for 1 h at room temperature. IR spectral
monitoring established that the (CO) bands at 1942,
1887 cm−1 had been replaced by two new bands at
2024, 1976 cm−1, ca. 80% of the original intensity
(absorbance), plus (CO) bands for Fp2, the only
other IR-detectable Fp compound (15% yield). The
resulting reaction mixture was worked up by benzene
extraction and precipitation from THF-pentane
(−20°C) to give 530 mg of a gummy red-brown
1536(w), 1190(m), 1015(s), 834(m), 798(s) cm−1
,6
(KBr) 2031, 1960 cm−1 (CO); 1H-NMR (C6D6) l
6.03 (Cp2Zr), 4.26 (CpFe), 1.18 (Me); 13C{1H}-NMR
(C6D6)
l 312.44 (s, CS2), 214.27 (CO), 111.65
(Cp2Zr), 87.84 (CpFe), 79.34 (OCMe3), 32.46 (Me).
Attempts to further purify 3 by precipitation from
THF-pentane or toluene-pentane (−78°C to room
temperature) inevitably reduced the purity of this ma-
terial such that multiple impurity resonances also
were detected in the Cp2Zr, CpFe, and OCMe3 re-
gions.
1H-NMR spectra of FpCS2Zr(OCMe3)Cp2 (3) ex-
hibited intense singlets (relative intensities, 10:5:9) in
the appropriate Cp2Zr, CpFe, and OCMe3 regions,
respectively. Although no other organometallic prod-
ucts were identified, the NMR spectra typically had
several weak intensity singlets flanking the Fp and
Cp2Zr Cp singlets. The purity of 82% 3 was quanti-
tated by integration of these spectra versus the an-
isole internal standard. All attempts to crystallize or
precipitate 3 inevitably degraded it as evidenced by
reduced yields and often an increased presence of im-
purities.
1
solid. It was assayed by H-NMR spectroscopy using
an anisole internal standard: 4 was the main product
(75% yield), with the balance being Fp2 (15%) plus
the usual degradation residues.
2.7. Reaction of (p5-C5H5)(CO)2FeZr(OCMe3)Cp2 (4)
with CS2 in C6D6
2.5. Preparation of (p5-C5H5)(CO)2Fe–Zr(OCMe3)Cp2
(4)
An NMR tube was loaded with a C6D6 solution
(0.75 ml) of FpZr(OCMe3)Cp2 (4) and anisole (3.0 ml,
1
27.6 mmol). H-NMR spectroscopic integration of this
A 50 ml Schlenk flask was charged with FpK (72
mg, 0.333 mmol) and Cp2ZrCl(OCMe3) (110 mg,
0.333 mmol) in the glove box. A magnetic stirbar was
added, and the flask was attached to a vacuum line;
THF (10 ml) was added to the flask and the yellow-
brown slurry was stirred for 30 min. IR spectra of
the resulting yellow-brown solution showed the pres-
sample was used to quantify the 4 (15.4 mmol), and
degassed CS2 (1.0 ml, 16.6 mmol) then was injected.
The reaction was monitored by NMR spectroscopy as
the initial yellow solution gradually turned orange
and finally dark red over 48 h. At this time, 15% of 4
remained, and a 61% yield of FpCS2Zr(OCMe3)Cp2
(3) was quantified.

4
J.R. Pinkes et al. / Journal of Organometallic Chemistry 566 (1998) 1–7
2.8. Attempted reaction of (p5-C5H5)(CO)2FeZr(Cl)Cp2
spectrum ((CO) 1869, 1792, 1773 to 1991, 1941 cm−1
)
with CS2
were consistent with quantitative conversion of FpK to
the well known [13] FpCH2CH3.
A THF solution (5 ml) of FpZrClCp2 was generated
from the reaction between FpK (54 mg, 0.25 mmol)
and Cp2ZrCl2 (73 mg, 0.25 mmol). An IR spectrum of
the resulting orange solution after 20 min registered
intense (CO) absorptions of FpZrClCp2 [11] at 1955,
1905 cm−1, along with absorptions indicative of the
presence of 4% Fp2. Carbon disulfide (38.0 ml, 0.50
mmol) was added by syringe, and the solution was
monitored by IR spectroscopy. The only reaction that
was observed was the gradual conversion of FpZrClCp2
into Fp2 over 3 h. When the same reaction was con-
ducted at 0°C, only a 10% conversion of FpZrClCp2 to
Fp2 was noted; FpCS2Zr(Cl)Cp2 (1) was not detected
during either reaction.
2.11. Thermal degradation of
(p5-C5H5)(CO)2FeCS2Zr(Cl)(p5-C5H5)2 (1)
A solution of FpCS2Zr(Cl)Cp2 (1) (50 mg, 0.098
mmol) in 2.5 ml of THF was maintained at room
temperature for 12 h as its color changed from green to
crimson. An infrared spectrum obtained after this time
showed bands which correspond to an approximate 3:1
ratio of FpCS2Fe(CO)Cp (5) [12] (2029, 2013, 1977
cm−1) to FpC(S)SFp (6) (2029, 1982, 1937 cm−1) [12],
based upon the (CO) absorption intensities. The THF
1
was evaporated and the H-NMR spectrum that was
recorded (C6D6) for the dark red residue had a 4:3:1
ratio of (Cp2ClZr)2O [l 6.02 (s, Cp2Zr)], 5 [l 4.27, 4.14
(s, CpFe)], and 6 [l 4.17, 4.03 (s, CpFe)]. Only trace
amounts of the (independently prepared) (Cp2ClZr)2S
[l 5.87 (s, Cp2Zr)] [14] were detected.
2.9. Attempted reactions of
(p5-C5H5)(CO)2FeZr(Cl)Cp2 with Cp2Zr(OCMe3)Cl
and of (p5-C5H5)(CO)2FeZr(OCMe3)Cp2 (4) with
Cp2ZrCl2
FpZrClCp2 was generated in a THF solution (20 ml)
from FpK (75 mg, 0.35 mmol) and Cp2ZrCl2 (101 mg,
0.35 mmol). This solution was transferred via cannula
to a second 100-ml Schlenk flask containing a magnetic
stir bar and Cp2Zr(OCMe3)Cl (121 mg, 0.35 mmol).
The resulting orange solution showed no formation of
4, as adduced via IR spectroscopy, over 1 h. (varying
mixtures of both, Fp–zirconocene complexes easily
could be distinguished, particularly by their lower fre-
quency (CO) absorptions, 1905 and 1887 cm−1, re-
spectively; 5–10% quantities of either bimetallic
compound was detected in the presence of the other.)
The only discernible change for this reaction was a
gradual darkening to orange-brown as increasing
amounts of Fp2 appeared. Essentially the same results
3. Results and discussion
The CS2 adduct FpCS2K is available in essentially
quantitative yield by adding one equivalent (or more)
of CS2 to Fp− in THF at ca. 0°C. Although this
metallodithiocarboxylate, originally prepared by Ellis
([4]a), is stable up to at least at 0°C, it has not been
isolated. Rather, it has been derivatized via its reactions
with numerous organic and inorganic electrophiles that
produce fully characterized dithiocarboxylate esters,
e.g. FpCS2Fp with FpI [4,15] and the v-(p1-C: p2-S,S%)
adduct FpCS2Fe(CO)Cp with CpFe(CO)(CH3CN)2+
BF4− [12].
were
obtained
after
treating
(p5-
C5H5)(CO)2FeZr(OCMe3)Cp2 (4) (0.52 mmol in 20 ml
of THF) with 1.0 equivalent of Cp2ZrCl2 under identi-
cal conditions.
2.10. Attempted reaction of
(p5-C5H5)(CO)2FeZr(OCMe3)Cp2 (4) with ethyl
bromide
Treatment of THF solutions of FpCS2K or its
methylcyclopentadienyl analog Fp%CS2K at −23°C
with one equivalent of zirconocene dichloride cleanly
produced the v-(p1-C: p2-S,S%) dithiocarboxylate com-
plexes FpCS2ZrClCp2 (1) and Fp%CS2ZrClCp2 (2), Eq.
2. Monitoring the IR spectra of these reactions was
especially useful: with the former reaction the intense
terminal carbonyl (CO) bands of FpCS2K, 1999, 1945
To a THF solution of 4 (0.054 mmol in 3.0 ml) was
added excess ethyl bromide (10 ml, 0.13 mmol). The
results of IR spectral examination over three quarters
of an hour were consistent with the absence of
FpCH2CH3, although a gradual increase of Fp2 (20%)
was noted. Control reaction: a THF solution of FpK
(108 mg, 0.50 mmol in 20 ml) was treated with
CH3CH2Br (40 ml, 0.52 mmol). Both the immediate
change in color (dark orange to yellow) and in the IR
cm−1, were replaced by those for 1, 2029, 1982 cm−1
.
Subsequent workup of the reaction by extraction and
precipitation from benzene/hexane left analytically pure
1 as a greenish brown powder in 70% yield. A

J.R. Pinkes et al. / Journal of Organometallic Chemistry 566 (1998) 1–7
5
environment
resembling
that
observed
for
somewhat lower isolated yield of 2 (59%) was realized
from an otherwise similar reaction.
FpCO2Zr(Cl)Cp2 (2032, 1977 cm−1) [2]. Instead of the
appearance of the carboxylate (OCO) absorptions in the
l 1375–1250 cm−1 region for FpCO2Zr(Cl)Cp2, 1 and
2 exhibit two medium intensity (SCS) absorptions ([6]c)
at 930, 850 cm−1 (KBr pellet); only the latter absorption
could be assigned in THF solution (947–849 cm−1) due
to solvent interference. These dithiocarboxylate stretch-
ing frequencies were assigned only after also examining
analogous spectral data for FpZr(Cl)Cp2 and for
FpCS2K. Related (SCS) data (KBr) is available for
FpCS2Fe(CO)Cp (914, 876 cm−1, medium) and for
FpC(S)SFp (1000 cm−1, strong), which is consistent with
the usual appearance of two (SCS) absorptions for
v-(p1-C: p2-S,S%) dithiocarboxylates and one for
nonchelating v-(p1-C: p1-S) dithiocarboxylates ([6]c,
[12]).
Although both 1 and 2 appeared to be thermally
stable as solids at room temperature, their THF
solutions slowly degraded. Over 12 h, solutions of
1 at room temperature converted to 3:1 mixtures of
FpCS2Fe(CO)Cp and FpC(S)SFp plus (Cp2ClZr)2O as
the major Cp2Zr-containing species. The presence of
both Fe2 v-CS2 complexes was confirmed by IR and by
1H- and 13C{1H}-NMR spectroscopy, but curiously we
detected only trace amounts of the (independently pre-
pared) v-sulfido zirconocene complex (Cp2ClZr)2S [14].
This obviously complex degradation of 1 was not further
investigated.
The IR and NMR spectroscopy of 1 and 2 is consistent
with the presence of isolated Fp (or Fp%) and Zr(Cl)Cp2
fragments that are bridged by v-(p1-C: p2-S,S%) CS2
ligands. Such structures resemble that proposed for the
carbon dioxide analog FpCO2Zr(Cl)Cp2 [1,2] and sub-
stantiated for the isolobal heterobimetallic v-(p1-C:
p2-O,O%) CO2 compounds Cp*(CO)2RuCO2ZrClCp2 and
Cp*(CO)(NO)ReCO2ZrClCp2 by X-ray crystallographic
structure determinations [16]. 1H-NMR spectra for 1 and
2 accordingly feature a 2:1 intensity relationship between
Cp2Zr resonance and the CpFe singlet (or Cp%Fe multi-
plets); for 1, their chemical shifts (l 5.92, 4.13, respec-
tively) resemble those for FpCO2Zr(Cl)Cp2 (l 6.09, 4.15)
in C6D6. 13C{1H}-NMR spectra of 1 and 2 feature a CS2
absorption at l 324–326, which is not far removed from
that previously recorded for FpCS2K (l 309) [12]. Other
The
t-butoxy-substituted
dithiocarboxylate
FpCS2Zr(OCMe3)Cp2 (3) also can be generated from
metathesis of FpCS2K and the appropriate zirconocene
chloride (Eq. 3), although isolating and working with 3
is limited by its relatively low thermal stability. IR
spectral monitoring of the reaction mixture appeared
similar to that observed for generating 1 and 2: 3 was the
only Fp-containing material present, other than for trace
amounts of Fp2. Workup of the reaction mixtures
entailed benzene extraction and precipitation with pen-
1
tane in order to produce a brown powder. H-NMR
spectra of this material were consistent with a single set
of dominant Cp2Zr, CpFe, and OCMe3 resonances,
although several very weak singlets appeared near each
resonance for 3. The 83% yield of 3 was quantitated by
integration versus an internal standard. The purest
samples resulted only after minimal handling and a brief
workup: further attempts at purifying this material or
indeed leaving it in THF or benzene solutions further
degraded it to multiple unidentified products.
related CS2 absorptions occur at
l
298 for
FpCS2Fe(CO)Cp and at l 306 FpC(S)SFp [12].
IR spectral data for 1 and 2 also can be compared with
that for FpCO2Zr(Cl)Cp2. Thus, the two intense (CO)
absorptions for 1 (2029, 1982 cm−1) and for 2 (2026,
1979 cm−1) attests to the presence of an electronic Fp

6
J.R. Pinkes et al. / Journal of Organometallic Chemistry 566 (1998) 1–7
Identification of FpCS2Zr(OCMe3)Cp2 (3) rests upon
spectroscopy. In contrast, FpK immediately and quan-
titatively underwent alkylation by one equivalent of
ethyl bromide under comparable reaction conditions to
yield FpCH2CH3. The second observation pertains to
the absence of any zirconocene exchange upon treating
FpZrClCp2 with one equivalent of Cp2ZrCl(OCMe3),
Eq. 4. Likewise, attempting the reaction in its reverse
direction, mixing 4 with one equivalent of Cp2ZrCl2,
likewise did not afford any detectable (5% limit by IR
spectroscopy) FpZrClCp2. If FpZrClCp2 had ionized in
THF solution in the presence of Cp2ZrCl(OCMe3) (for-
ward direction, Eq. 4), then some of the resulting Fp−
should have been intercepted and converted to 4, which
corresponds to its synthetic procedure. We conclude
that 4 does not ionize in THF solution to give accessi-
ble Fp−, consonant with our previous observations for
FpZrClCp2 [2].
the resemblance of its IR and 1H-, 13C{1H}-NMR
spectral data with that for 1 and 2. The NMR chemical
shifts for the Cp resonances of 3 are comparable to
those for 1, in addition to the 2:1 intensity relationship
1
between Cp2Zr and the CpFe singlets in the H-NMR
1
spectra. H- and 13C-NMR spectral absorptions for the
zirconocene Cp and OCMe3 groups on 3 (l 6.03, 1.18
and 112, 79, 32) nevertheless can be differentiated from
those observed for Cp2ZrCl(OCMe3) (l 5.99, 1.06 and
113, 80, 32). Although the presence of the bridging CS2
ligand was established by its 13C-NMR spectral absorp-
tion at l 312, the assignment of the corresponding
dithiocarboxylate IR (SCS) absorptions was precluded
by the presence of interfering, broad OCMe3 absorp-
tions.
The dithiocarboxylate 3 also can be generated from
the reaction between Casey’s FpZr(OCMe3)Cp2 (4) [11]
and CS2, Eq. 3. One equivalent of CS2 transformed 4 in
THF solution at room temperature to 3 (75%) plus Fp2
(15%). We used IR spectroscopy to monitor this clean
4. Conclusions
We have synthesized the v-(p1-C: p2-S,S%) dithiocar-
boxylate
complexes
FpCS2ZrClCp2
(1)
and
1
transformation; H-NMR spectra of the isolated mate-
Fp%CS2ZrClCp2 (2) from the reactions between FpCS2K
and zirconocene dichloride; both stable products were
fully characterized. Although the iron-zirconium com-
plex FpZrClCp2 inserts carbon dioxide to give
FpCO2ZrClCp2 [2], CS2 incorporation into the same
FeZr bimetallic complex was not observed. Metathesis
between FpCS2K and Cp2ZrCl(OCMe3) provided
FpCS2Zr(OCMe3)Cp2 (3), which was not obtained ana-
lytically pure. This product was characterized by com-
rial were consistent with a 75% yield of 3. The same
reaction in C6D6 solution took much longer, with a
maximum 61% conversion recorded after 48 h. Prelimi-
nary observations likewise are available for converting
4 and carbon dioxide to a similar CO2 adduct [17].
Surprisingly, FpZrClCp2 [11] did not incorporate CS2
under similar conditions. Reaction between FpZrClCp2
and CS2 at room temperature yielded Fp2 promptly and
quantitatively, whereas no detectable reaction occurred
at 0°C over at least 3 h. Under either set of reaction
conditions, FpCS2Zr(Cl)Cp2 (1) was not detected.
An unresolved issue concerns the facility with which
CO2 adds to both FpZr(X)Cp2 (X=Cl, OCMe3 (4)),
whereas CS2 incorporates into 4 but not FpZrClCp2.
One possible rational is that the CS2 reaction requires
ionization of the iron–zirconium bond and formation
of FpCS2−, Cp2Zr(OCMe3)(THF)+ (which converts to
the observed 3), whereas CO2 insertion entails a direct,
perhaps bimetallic, pathway [18]. Analogous autoion-
ization of FpZrClCp2 evidently would not be available
under these conditions. The ionization pathway for 4,
however, can be ruled out on the basis of two
observations.
parison of its IR and H-, 13C{1H}-NMR spectral data
1
with that for 1 and 2. The iron–zirconium complex
FpZr(OCMe3)Cp2 (4) adds carbon disulfide to give
FpCS2Zr(OCMe3)Cp2 (3) in at least 75% spectroscopic
yield.
We are unable to resolve the dilemma concerning the
ease with which CO2 adds to both FpZr(X)Cp2 (X=
Cl, OCMe3 (4)), forming the FeZr v-CO2 complexes,
whereas only 4 interposes CS2, giving 3. The unique
reactivity of 4 towards CS2, however, can not be at-
tributed to autoionization of
4
to Fp− and
Cp2Zr(OCMe3)(THF)+ in order for the Fp− to bind
the CS2 as FpCS2− prior to reacting with the zirconium
electrophile. All attempts to spectroscopically probe or
otherwise encourage an ionization equilibrium involv-
ing FpZr(X)Cp2 (via trapping of the Fp−) have failed.
Studies in progress address the reconfiguration of the
coordination environments of heterobimetallic com-
plexes in order to facilitate CO2 and CS2 ‘insertion’ into
the metal–metal bonds.
First, treatment of 4 with ethyl bromide did not
produce any FpCH2CH3 [13], which would have been
expected if 4 had ionized in THF solution to Fp−,
Cp2Zr(OCMe3)(THF)+. A 3–5% conversion of 4 to
FpCH2CH3 [13] easily would have been detected by IR

J.R. Pinkes et al. / Journal of Organometallic Chemistry 566 (1998) 1–7
7
Bianchini, C. Mealli, A. Meli, M. Sabat, in: I. Bernal (Ed.),
Stereochemistry of Organometallic and Inorganic Compounds,
vol. 1, Elsevier, Amsterdam, 1986, pp. 146.
Acknowledgements
Support from the National Science Foundation,
Grant CHE–9108591, is gratefully acknowledged.
[7] D.F. Shriver, M.A. Drezdon, The Manipulation of Air-Sensitive
Compounds, 2nd, Wiley, New York, 1986.
[8] (a) J.A. Gladysz, G.M. Williams, W. Tam, D.L. Johnson, D.W.
Parker, J.C. Selover, Inorg. Chem. 18 (1979) 553. (b) A similar
procedure was used to prepare Fp%K.
[9] L.T. Reynolds, G. Wilkinson, J. Inorg. Nucl. Chem. 9 (1959) 86.
[10] Cp2ZrCl(OCMe3) was prepared by treatment of Cp2ZrCl2 with
KOCMe3, A. Hey-Hawkins, Chem. Rev. 94 (1994) 1661.
[11] (a) C.P. Casey, R.F. Jordan, A.L. Rheingold, J. Am. Chem. Soc.
105 (1983) 665. (b) C.P. Casey, J. Organomet. Chem. 400 (1990)
205.
[12] M.E. Giuseppetti-Dery, B.E. Landrum, J.L. Shibley, A.R. Cut-
ler, J. Organomet. Chem. 378 (1989) 421.
[13] (a) T.S. Piper, G. Wilkinson, J. Inorg. Nucl. Chem. 3 (1956) 104.
(b) M.L.H. Green, P. Nagy, J. Organometal. Chem. 1 (1963) 58.
[14] W. Mei, L. Shiwei, B. Meizhi, G. Hefu, J. Organomet. Chem.
447 (1993) 227.
References
[1] (a) J.C. Vites, B.D. Steffey, M.E. Giuseppetti-Dery, A.R. Cutler,
Organometallics 10, (1991) 2827. (b) D.H. Gibson, Chem. Rev.
96 (1996) 2063.
[2] J.R. Pinkes, B.D. Steffey, J.C. Vites, A.R. Cutler, Organometal-
lics 13 (1994) 21.
[3] (a) G.O. Evans, W.F. Walter, D.R. Mills, C.A. Streit, J.
Organomet. Chem. 144 (1978) C34. (b) J.R. Pinkes, C.J. Masi,
R. Chiulli, B.D. Steffey, A.R. Cutler, Inorg. Chem. 36 (1997) 70.
[4] (a) J.E. Ellis, R.W. Fennell, E.A. Flom, Inorg. Chem. 15 (1976)
2031. (b) H. Stolzenberg, W.P. Fehlhammer, M. Monari, V.
Zanotti, L. Busetto, J. Organomet. Chem. 272 (1984) 73. (c) L.
Busetto, M. Monari, A. Palazzi, V. Albano, F. Demartin, J.
Chem. Soc. Dalton Trans (1983) 1849.
[5] (a) C.P. Kubiak, C. Woodcock, R. Eisenberg, Inorg. Chem. 21
(1982) 2119. (b) T.A. Hanna, A.M, Baranger, R.G. Bergman, J.
Am. Chem. Soc. 117 (1995) 11363. (c) H. Memmler, U. Kauper,
L.H. Gade, I.J. Scowen, M. McPartlin, J. Chem. Soc. Chem.
Commun. (1996) 1751.
[15] A.R. Cutler, P.K. Hanna, J.C. Vites, Chem. Rev. 88 (1988) 1363.
[16] D.H. Gibson, J.M. Mehta, B.A. Sleadd, M.S. Mashuta, J.F.
Richardson, Organometallics 14 (1995) 4886.
[17] Treatment of Cp(CO)2FeZr(OCMe3)Cp2 with one equivalent or
excess of CO2 in THF solution between −10°C and room
temperature rapidly affords a compound that we tentatively
formulated on the basis of IR and NMR spectral data as
Cp(CO)2FeCO2Zr(OCMe3)Cp2. Although this relatively unstable
material evidently formed in 60–85% yields, we were unable to
isolate it sufficiently pure for unambiguous characterization.
[18] We thank a referee for this suggestion.
[6] Reviews on CS2 complexes. (a) K.K. Pandey, Coord. Chem.
Rev. 40 (1995) 37. (b) R.P.A. Sneedon, in: G. Wilkinson, F.G.A.
Stone, F.W. Abel (Eds.), Comprehensive Organometallic Chem-
istry, vol. 8, Pergamon, New York, 1982, Chp. 50.4. (c) C.
.