Synthesis of Fe-C≡C-M complexes through condensation of Fe-C≡C-H with M-H or M-NMe2

Article abstract of DOI:10.1021/om9800187

The iron acetylide Cp*(dppe)FeC≡CH (Cp* = η⁵-C⁵Me⁵, dppe = Ph₂PCH ₂CH₂PPh₂) reacted with [Cp₂Zr(H)Cl] _n (Schwartz's reagent, Cp = η⁵-C⁵

Full text of DOI:10.1021/om9800187

5920
Organometallics 1998, 17, 5920-5923
Syn th esis of F e-CtC-M Com p lexes th r ou gh
Con d en sa tion of F e-CtC-H w ith M-H or M-NMe2
Xinhong (Simon) Gu and Michael B. Sponsler*
Department of Chemistry and W. M. Keck Center for Molecular Electronics,
Syracuse University, Syracuse, New York 13244
Received J anuary 15, 1998
Summary: The iron acetylide Cp*(dppe)FeCtCH (Cp*
) η5-C5Me5, dppe ) Ph2PCH2CH2PPh2) reacted with
[Cp2Zr(H)Cl]n (Schwartz’s reagent, Cp ) η5-C5H5) to
produce the CtC-bridged complex Cp*(dppe)Fe-
CtCZrClCp2 in nearly quantitative yield instead of the
expected CHdCH-bridged hydrozirconation product. The
bulky ligand set around iron is postulated to prevent the
addition of hydride at the carbon R to the metal center.
Reaction of Cp*(dppe)FeCtCH with (dimethylamino)-
trimethyltin gave the condensed product Cp*(dppe)Fe-
CtCSnMe3 in nearly quantitative yield.
at room temperature gave rise to a dark red solution in
about 10 min. The reaction progress was monitored by
1H NMR, and the starting material 1 was consumed in
6 h. NMR showed an almost quantitative conversion of
1 to a new compound showing a single peak (1.52 ppm)
in the Cp* region and no olefinic signals. A preparative-
scale reaction afforded the new compound in 98% yield.
The IR spectrum displayed a strong asymmetric carbon-
carbon triple-bond absorption at 1843 cm-1. When
mixed with wet benzene, the dark red benzene solu-
tion turned yellow-orange instantly, and 1H NMR
showed that iron acetylide 1 was formed. On the basis
of the spectroscopic data and the hydrolysis reaction,
we assigned the structure of the new compound as
Cp*(dppe)FeCtCZrClCp2 (4, eq 2), one of relatively few
complexes with a C2-bridge connecting an electron-
deficient early transition metal to an electron-rich late
transition metal. Similar condensed products were
apparently also produced upon mixing 1 with diisobu-
tylaluminum hydride (DIBAL-H)8 or catecholborane
(HBcat),9 although these reactions were not as clean.
Hydrolysis of both products also led smoothly to 1.
We have reported experimental and theoretical in-
vestigations of a series of half-sandwich butadienediyl-
bridged diiron complexes.1 In search of an efficient way
to prepare [Cp*(dppe)Fe]2(µ-CHdCHCHdCH)2 (Cp* )
η5-C5Me5, dppe ) Ph2PCH2CH2PPh2) and longer bridge
analogues, we attempted the hydrometalation of Cp*-
(dppe)FeCtCH (1)3 with the intention of coupling the
Cp*(dppe)FeCHdCHMLn products. Organic terminal
acetylenes readily undergo hydrozirconization,4 hy-
droalumination,5 and hydroboration6 reactions. In ad-
dition, Cp(PMe3)2RuCtCH (2, Cp ) η5-C5H5) was
reported by Bullock and co-workers to react with [Cp2-
Zr(H)Cl]n (3) to give Cp(PMe3)2RuCHdCHZrClCp2 (eq
1).7 However, we uniformly failed to obtain the desired
addition products such as Cp*(dppe)FeCHdCHZrClCp2.
Instead we herein report these reactions as efficient
ways to prepare C2-bridged heterodinuclear complexes.
(2)
These hydrogen-elimination transformations are ap-
parently unprecedented in the reactions of terminal
alkynes with Schwartz’s reagent or catecholborane. In
the hydroalumination of terminal alkynes, 1-alkynyl-
aluminum compounds are observed only as minor prod-
ucts (<10%).10 A related reaction is the higher temper-
(1)
Treatment of the yellow benzene-d6 solution of iron
acetylide 1 with Schwartz’s reagent, [Cp2Zr(H)Cl]n (3),
(8)
(1) Etzenhouser, B. A.; Cavanaugh, M. D.; Spurgeon, H. N.; Spon-
sler, M. B. J . Am Chem. Soc. 1994, 116, 2221-2222. Etzenhouser, B.
A.; Chen, Q.; Sponsler, M. B. Organometallics 1994, 13, 4176-4178.
Sponsler, M. B. Organometallics 1995, 14, 1921-1927.
(2) This complex has been prepared by a different route (Gu, X.
Ph.D. Dissertation, Syracuse University, 1997) and was desired for
comparison to work by Lapinte and co-workers (see ref 3).
(3) Le Narvor, N.; Toupet, L.; Lapinte, C. J . Am. Chem. Soc. 1995,
117, 7129-7138.
To a solution of 1 (20 mg, 0.033 mmol) in benzene (2 mL) was added
50 µL of 1 M diisobutylaluminum hydride (0.050 mmol) in hexane.
Bubbles evolved upon mixing. The solution was stirred at room
temperature overnight, then evaporated under vacuum, and the
residue was washed with two portions of pentane (1 mL each). The
product (20 mg) was obtained as a brown solid in 64% yield (corrected
for purity, estimated at 80% by NMR integration). 1H NMR (C6D6): δ
7.76 (t, J ) 8 Hz, 4H, Ph), 7.40 (t, J ) 8 Hz, 4H, Ph), 6.90-7.25 (m,
12H, Ph), 2.24 (nonet, J ) 7 Hz, 2H, iBu), 2.82 (m, 2H, PCH2), 2.15
(m, 2H, PCH2), 1.39 (s, 15H, Cp*), 0.74 (d, J ) 7 Hz, 12H, iBu), 0.24
(d, J ) 7 Hz, 4H, iBu). 31P NMR: δ 100.7. IR (KBr, cm-1): 3040w,
2939m, 1961m, 1614s, 1463m, 1429s, 1252s. MS(EI): 613 (M+ - Al-
(iBu)2), 640 (M+ - 2iBu), 654 (M+ - iBu - iPr), 668 (M+ - 2iPr).
(4) Temple, J . S.; Schwartz, J . J . Am. Chem. Soc. 1980, 102, 7381-
7382.
(5) Eisch, J . J . In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon: Oxford, 1992; Vol. 8, p 733.
(6) Brown, H. C.; Gupta, S. K. J . Am. Chem. Soc. 1972, 94, 4370-
4371. Suseela, Y.; Periasamy, M. J . Organomet. Chem. 1993, 450, 47-
52.
(7) Lemke, F. R.; Szalda, D. J .; Bullock, R. M. J . Am. Chem. Soc.
1991, 113, 8466-8477.
10.1021/om9800187 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 11/20/1998

Notes
Organometallics, Vol. 17, No. 26, 1998 5921
Sch em e 1
ature alumination of terminal alkynes with trialkyla-
luminum reagents, in which alkanes are extruded.11
Similar reactions are also known for alkyl zirconium
complexes, such as the methane elimination shown in
eq 3.7
(3)
Sch em e 2
The observed reactions may be viewed formally as
acid-base reactions, similar to the formation of metal
acetylides from the reaction of terminal alkynes and
NaH or KH. However, unusually facile deprotonation
does not appear to be a reasonable explanation for the
observed reactivity. Compound 1 is not expected to be
a particularly acidic alkyne, given the electron-rich
character of the iron center. The most striking difference
between 1 and 2, whose reactions with 3 show com-
pletely different chemoselectivity (eqs 1 and 2), is the
degree of steric congestion near the metal center. We
therefore propose that the facile condensation of 1 with
various hydrides is attributable primarily to steric
congestion around the iron center, which effectively
blocks delivery of the hydride to the R-carbon of the iron
acetylide. Lapinte’s extensive studies on complexes of
Cp*(dppe)Fe offer strong evidence for large steric effects,
such as the remarkable persistence observed for various
16-,12 17-,13 and 19-electron14 complexes. Also indicative
of unusual steric effects is the result that Cp*(dppe)-
FeCtCMe15 was found to be unreactive to 3 even at 70
°C, while other internal alkynes such as Cp(PMe3)2-
RuCtCMe,7 (η5-C5H4Me)2ClZrCtCMe,16 and t-Bu-
CtCMe17 are reported to undergo smooth hydrozircona-
tion at ambient temperature.18
A standard hydrometallation mechanism,19 illus-
trated in Scheme 1, requires that the hydride reagent
come significantly closer to the iron center than is
necessary for the observed condensation, illustrated in
Scheme 2. If the R-carbon is sterically inaccessible, the
electrophilic hydride instead may react with the less
electronically favorable but more accessible C-H bond.
The three-center, two-electron intermediate and four-
center transition state in Scheme 2 are offered as a
possible mechanism.
Molecular mechanics calculations offer further sup-
port for steric hinderance as a primary cause of the
differing reactivity. Comparison of space-filling repre-
sentations of 1 and 2 (Figure 1) reveals a marked
difference in the accessibility of the R-carbon. In addi-
tion, the steric repulsion energies that accompany
complexation of either the CtC bond or the C-H bond
to the Zr of 3 (i.e., the intermediates of Schemes 1 and
2) were evaluated. For 1, complexation at the CtC bond
was found to cost 15 kcal/mol more than complexation
at the C-H bond. However, for 2 the difference was only
1 kcal/mol.
Bullock and co-workers also reported a reaction
between 2 and (dimethylamino)trimethyltin (Me2N-
SnMe3), cleanly producing Cp(PMe3)2RuCtCSnMe3.7
We found that this transformation also works well for
1. The reaction between 1 and Me2NSnMe3 (eq 4) was
finished in ca. 30 min at room temperature. The red-
brown crystalline complex Cp*(dppe)FeCtCSnMe3 (5)
was obtained in almost quantitative yield (98%). A
strong IR absorption at 1949 cm-1 resulting from the
asymmetric CtC stretch in complex 5 was comparable
to 1919 cm-1 for Cp(PMe3)2RuCtCSnMe3. Hydrolysis
(9)
To a solution of 1 (20 mg, 0.033 mmol) in benzene (2 mL) was added
50 µL of 1
M catecholborane (0.050 mmol) in THF at ambient
temperature. The solution was stirred for 30 min, then evaporated
under vacuum, and the residue was washed with two portions of
pentane (1 mL each). The product (18 mg) was obtained as a brown
solid in 53% yield (corrected for purity, estimated at 70% by NMR
integration). 1H NMR (C6D6): δ 8.10 (t, J ) 8 Hz, 4H, Ph), 6.80-7.60
(m, 20H, Ph), 2.60 (m, 2H, PCH2), 2.25 (m, 2H, PCH2), 1.45 (s, 15H,
Cp*). 31P NMR: δ 101.5. IR (KBr, cm-1): 3058m, 2957s, 2923s,
2863m, 1994s, 1615s, 1480s, 1227s, 1049s, 732m. MS(EI): 732 (M+).
(10) See, for example: Negishi, E.; Takahashi, T.; Baba, S. Org.
Synth. 1988, 66, 60-66. When an amine complex of a dialkylaluminum
hydride is used, however, the amine-complexed 1-alkynyl aluminum
compound becomes the major product: Binger, P. Angew. Chem., Int.
Ed. Engl. 1963, 2, 686.
(11) Mole, T.; Surtees, J . R. Aust. J . Chem. 1964, 17, 1229-1235.
(12) Hamon, P.; Toupet, L.; Rabaaˆ, H.; Saillard, J .-Y.; Hamon, J .-
R.; Lapinte, C. Organometallics 1996, 15, 10-12.
(13) Roger, C.; Hamon, P.; Toupet, L.; Rabaaˆ, H.; Saillard, J .-Y.;
Hamon, J .-R.; Lapinte, C. Organometallics 1991, 10, 1045-1054. See
also ref 14.
(14) Hamon, P.; Hamon, J .-R.; Lapinte, C. J . Chem. Soc., Chem.
Commun. 1992, 1602-1603.
(16) Erker, G.; Fro¨mberg, W.; Angermund, K.; Schlund, R.; Kru¨ger,
C. J . Chem. Soc., Chem. Commun. 1986, 372-374.
(15) Cp*(dppe)FeCtCMe was made by methylation (methyl triflate,
78% yield) of iron acetylide 1 and subsequent deprotonation (KOt-Bu,
98% yield), following the general procedure of Selegue: Adams, R. D.;
Davison, A.; Selegue, J . P. J . Am. Chem. Soc. 1979, 101, 7232-7238.
1H NMR (C6D6): δ 8.06 (t, J ) 8 Hz, 4H), 7.40-7.10 (m, 16H), 2.65
(m, 2H), 2.14 (s, 3H), 1.84 (m, 2H), 1.55 (s, 15H). 13C NMR (0.05 M
(17) Hart, D. W.; Blackburn, T. F.; Schwartz, J . J . Am. Chem. Soc.
1975, 97, 679-680.
(18) We cannot rule out the possibility that electronic effects might
also play a role in determining the mode of reactivity, but note that 1
and 2 are reasonably comparable in terms of electron-richness of the
metal centers. In addition, the fact that clean hydrozirconation is
observed for both Cp(PMe3)2RuCtCMe and (η5-C5H4Me)2ClZrCtCMe
indicates that this reaction is not very sensitive to electron density.
(19) Yamamoto, A. Organotransition Metal Chemistry; Wiley: New
York, 1986; pp 260-262.
2
Cr(acac)3): δ 141-129 (m, Ph), 138.8 (t, J PC ) 43 Hz, FeC, tentative
assignment due to overlap with Ph), 110.3 (s, tCMe), 87.3 (s, Cp*),
1
31.3 (t, J PC ) 21 Hz, PCH2), 10.8 (s, Cp*), 8.9 (s, tCMe). 31P NMR:
δ 101.3.

5922 Organometallics, Vol. 17, No. 26, 1998
Notes
(THF: sodium/benzophenone; benzene, pentane: calcium hy-
dride). All reagents were obtained from common commercial
sources unless otherwise stated and were used as received.
Compound 1 was prepared from FeCl2 in four steps according
to the literature method.3 Benzene-d6 was vacuum-transferred
from calcium hydride, degassed by the freeze-pump-thaw
method, and stored over 4 Å molecular sieves. All NMR spectra
were recorded in benzene-d6 on a 300 MHz (Bruker-300 or
General Electric QE-300) spectrometer. All chemical shifts are
referenced to TMS by using known shifts of residual proton
(7.15 ppm) or carbon (128.39 ppm) signals in benzene-d6 or to
external 85% aqueous phosphoric acid (31P NMR). Chromium-
(III) acetylacetonate was used as a relaxation agent for 13C
NMR. Electron impact mass spectra were obtained from the
MS facilities at University of Illinois, UrbanasChampaign. IR
spectra were recorded in KBr disks on an Perkin Elmer 1660
series IR spectrometer. Elemental analyses were performed
by E & R Microanalysis Co., Corona, NY.
F igu r e 1. Space-filling representations of molecular me-
chanics-optimized structures of 1 (left) and 2 (right). The
R-carbons of the acetylide ligands are shaded. In each case,
the perspective that allowed the least obstructed view of
the R-carbon was selected.
Cp *(d p p e)F eCtCZr ClCp 2 (4). To a stirred solution of 1
(40 mg, 0.066 mmol) in benzene (2 mL) was added solid [Cp2-
ZrHCI]n (3, 26 mg, 0.1 mmol) at ambient temperature. Over
10 min the solution color turned from yellow-brown to dark
red. After 6 h, excess 3 was removed by filtration through a
Celite pad, and solvent was evaporated under vacuum. Wash-
ing the product with pentane (2 × 1 mL) and drying under
of 5 in wet benzene cleanly formed 1. Solid samples of
5 were stable in the open air, while solution samples
decomposed slowly (2 days) to 1.
1
vacuum afforded 56 mg of 4 in 98% yield. H NMR: δ 7.98 (t,
(4)
J ) 8 Hz, 4H, Ph), 7.38 (t, J ) 8 Hz, 4H, Ph), 7.21 (m, 4H,
Ph), 7.05 (m, 8H, Ph), 6.03 (s, 10H, Cp), 2.79 (m, 2H, PCH2),
2.01 (m, 2H, PCH2), 1.52 (s, 15H, Cp*). 13C NMR (0.06 M Cr-
(acac)3): δ 207.1 (s, ZrC), 192.2 (t, 2J PC ) 40 Hz, FeC), 139.8-
2
128.9 (m, Ph), 111.9 (s, Cp), 89.3 (s, Cp*), 31.4 (t, J PC ) 21
Complexes with C2 ligands σ-bonded to two metal
atoms are not common.20 Such complexes have been
prepared by several methods, including substitution
reactions of metal halide compounds with deprotonated
metal acetylide complexes (transmetalation),21 depro-
tonation of cationic µ-η1-η2-acetylide complexes,22 alkyne
metathesis of metal acetylide complexes,18 and conden-
sation reactions like the ones in eqs 3 and 4. Formation
of C2-bridged complexes such as 2 from the reaction of
terminal metal acetylide complexes with electrophilic
hydrides adds another method to the synthetic list,
although the reaction will probably be limited to highly
hindered acetylides. Since the functional groups
ZrClCp2, Al(iBu)2, Bcat, and SnMe3 are often employed
in coupling reactions,23 the new complexes might be
useful as building blocks for the preparation of substi-
tuted iron acetylides or conjugated dinuclear complexes.
In summary, acetylide 1 reacts anomalously with
Schwartz’s reagent 3 and other electrophilic hydrides
to form C2-bridged heterodinuclear complexes. The
unusual and complete chemoselectivity is most reason-
ably attributed to the steric protection of the R-carbon
of the iron acetylide ligand provided by the Cp* and
dppe ligands.
Hz, PCH2), 10.8 (s, Cp*). 31P NMR: δ 102.2. IR (KBr, cm-1):
3052w, 2901m, 1843vs, 1471m, 1428m, 1084m, 1013m, 719s,
741m, 691s, 670m, 527s, 484m. MS(EI): 868 (M+). Isotope
pattern (% calcd, % found): 866(6.0, 6.4), 867(4.5, 4.3), 868-
(100, 100), 869(76.7, 76.4), 870(91.7, 91.6), 871(49.1, 49.0), 872-
(57.8, 57.8), 873(27.3, 27.0), 874(22.7, 22.7), 875(10.1, 10.1),
876(4.3, 4.4), 877(1.4, 1.2). Anal. Calcd for C48H49ClFeP2Zr: C,
66.24; H, 5.67. Found: C, 66.18; H, 5.89.
Complex 4 was also cleanly formed when the reaction was
performed under the conditions employed by Bullock and co-
workers for the reaction between 2 and 3 (use of THF as
solvent, warming slowly from -78 °C to room temperature).
Cp *(d p p e)F eCtCSn Me3 (5). To a solution of 1 (50 mg,
0.082 mmol) in benzene (2 mL) was added 21 µL of neat
(dimethylamino)trimethyltin (0.122 mmol). The solution was
stirred for 30 min and then evaporated under vacuum. The
residue was washed with two portions of pentane (1 mL each).
Complex 5 (63 mg, 0.080 mmol) was obtained in 98% yield.
1H NMR: δ 8.13 (t, J ) 8 Hz, 4H, Ph), 7.25 (m, 12H, Ph), 7.03
(m, 4H, Ph), 2.86 (m, 2H, PCH2), 1.85 (m, 2H, PCH2), 1.52 (s,
2
15H, Cp*), 0.22 (s, with 117,119Sn satellites, J SnH ) 25, 29 Hz,
2
9H, SnMe3). 13C NMR (0.05 M Cr(acac)3): δ 165.2 (t, J PC
)
36 Hz, FeC), 140.4-129.2 (m, Ph), 117.8 (s, SnC), 87.8 (s, Cp*),
1
31.4 (t, J PC ) 23 Hz, PCH2), 10.7 (s, Cp*), -7.0 (s, SnMe3).
31P NMR: δ 101.5. IR (KBr, cm-1): 3054w, 2918s, 2847w,
1949vs, 1554m, 1425s, 1179w, 1082m, 740m, 688s, 662m, 527s,
489m. Anal. Calcd for C41H48FeP2Sn: C, 63.36; H, 6.22.
Found: C, 63.65; H, 6.35.
Exp er im en ta l Section
All reactions were performed under nitrogen using standard
Schlenk techniques or with an inert atmosphere drybox.
Solvents were distilled under nitrogen from a drying agent
Com p u ta tion a l Meth od s. Molecular mechanics calcula-
tions were performed by using CAChe, version 4.02 (Oxford
Molecular, Ltd.). The augmented MM2 parameter set was
used. The CtC and C-H complexes from Schemes 1 and 2,
respectively (either 1-3 or 2-3), were geometry optimized in
orientations consistent with the subsequent reactions. In each
case, several starting structures were employed in order to find
the best relative orientation of the two components. Steric
repulsion energies due to complexation were calculated by
subtraction of the total energies of optimized components from
the total energy of the optimized complex (eq 5). To prevent
inclusion of binding energy, the components were taken as 1
(20) Koutsantonis, G. A.; Selegue, J . P. J . Am. Chem. Soc. 1991,
113, 2316-2317, and references therein.
(21) Cross, R. J .; Davidson, M. F. J . Chem. Soc., Dalton Trans. 1986,
411-414. Ogawa, H.; Onitsuka, K.; J oh, T.; Takahashi, S.; Yamamoto,
Y.; Yamazaki, H. Organometallics 1988, 7, 2257-2260. Weng, W.;
Bartik, T.; Brady, M.; Bartik, B.; Ramsden, J . A.; Arif, A. M.; Gladysz,
J . A. J . Am. Chem. Soc. 1995, 117, 11922-11931.
(22) Akita, M.; Terada, M.; Oyama, S.; Mor-oda, Y. Organometallics
1990, 9, 816-825.
(23) Heck, R. F. Palladium Reagents in Organic Synthesis; Academic
Press: New York, 1985; Chapter 6.

Notes
Organometallics, Vol. 17, No. 26, 1998 5923
(or 2) and a model complex without steric repulsions (with the
cyclopentadienyl and phosphine ligands on 1 or 2 replaced by
a single chloride ligand). The total energy of the sterically
modified 1 or 2 appeared as a positive term in the steric
repulsion energy in order to compensate for the appearance
of this molecule in the first component.
between components (unwanted because such attractions are
roughly compensated by solvation in the solution-phase reac-
tion). By reporting only the difference of two steric repulsion
energies (for CtC and C-H coordination), the effect of these
attractions is minimized, assuming them to be similar in the
two coordination modes.
Steric repulsion energy (1‚‚‚3) ) E(1‚‚‚3) - E(1) -
E(ClFeCtCH‚‚‚3) + E(ClFeCtCH) (5)
Ack n ow led gm en t is made to the Donors of The
Petroleum Research Fund, administered by the Ameri-
can Chemical Society, for partial support of this research.
The steric repulsion energies calculated in this way do
include significant and unwanted van der Waals attractions
OM9800187