Generation and trapping of radicals derived from cyclobutadieneiron tricarbonyl

Article abstract of DOI:10.1021/om060263n

Trapping products consistent with the formation of radical intermediates a to and on the ring carbon of cyclo-butadieneiron tricarbonyl have been synthesized. These examples constitute the first strong experimental evidence for the formation of radicals derived from cyclobutadieneiron tricarbonyl.

Full text of DOI:10.1021/om060263n

Organometallics 2006, 25, 3787-3790
3787
Generation and Trapping of Radicals Derived from
Cyclobutadieneiron Tricarbonyl
Jeffrey H. Byers,*,† Stephen F. Sontum,† Tina S. Dimitrova,† Sumaya Huque,†
Benjamin M. Zegarelli,† Yong Zhang,† Jerry P. Jasinski,*,‡ and Raymond J. Butcher§
Department of Chemistry and Biochemistry, Middlebury College, Middlebury, Vermont 05753,
Department of Chemistry, Keene State College, Keene, New Hampshire 03435, and
Department of Chemistry, Howard UniVersity, Washington, D.C. 20059
ReceiVed March 23, 2006
Summary: Trapping products consistent with the formation of
radical intermediates R to and on the ring carbon of cyclo-
butadieneiron tricarbonyl haVe been synthesized. These exam-
ples constitute the first strong experimental eVidence for the
formation of radicals deriVed from cyclobutadieneiron tricar-
bonyl.
for the syntheses of a wide variety of novel polycyclic systems
through intramolecular Diels-Alder reactions, which in turn
have proven to be useful precursors to interesting ring systems
found in natural products.5e
Introduction
Cyclobutadiene has long been a molecule of interest to
organic chemists,1 but has been directly observed only under
the most stringent conditions.2 Due to its exceptional instability,
it has usually been incorporated into more complex organic
structures as its Fe(CO)3 derivative, cyclobutadieneiron tricar-
bonyl3 (henceforth, CBIT), which can undergo further trans-
formations under a wide variety of conditions. Free cyclobuta-
diene or cyclobutadiene derivatives for subsequent trapping
experiments are commonly generated upon oxidation of these
Fe(CO)3 complexes, most commonly with ceric ammonium
nitrate.3
Recently, there has been a modest, but significant resurgence
of interest in the reactivity of derivatives of CBIT. CBIT-
stabilized cation 14 has recently been studied extensively by
the Snapper group,5 who have demonstrated the synthetic utility
of this interesting intermediate, which reacts as an electrophilic
partner in a variety of transformations. These reactions have
been used to append olefinic substituents to CBIT, allowing
Given our general interest in radical reactions, and more
specifically in the radical reactions of transition metal polyene
complexes, we became interested in the generation and trapping
of the R-methylene CBIT radical (2). The ways in which a
modest number of common transition metal complexes interact
with adjacent radicals have been studied. For example, experi-
mental6 and theoretical7 studies have indicated a significant
radical stabilizing effect by ferrocene. Radicals stabilized by
iron-diene complexes, formally pentadienyl radicals, have been
implicated in electrochemical studies,8 as well as in radical
additions to the uncomplexed alkene in iron-triene complexes.9
10
Cyclizations of radicals mediated by (alkyne)Co2(CO)6 and
reductive dimerizations of radicals derived from (alkyne)Co2-
(CO)6-stabilized cations11 have also been studied. Given the
paucity of data on radicals stabilized by transition metal-olefin
complexes, we felt that examining the reactivity of radical 2
would be intrinsically interesting and might lead to useful
methodology.
Very little published work has appeared concerning reactive
intermediate 2. Most notably, Creary7 has carried out calcula-
tions at the B3LYP/LANL2DZ level on the isodesmic reaction
between radical 2 and toluene, generating the benzylic radical
and methyl CBIT, showing a ∆E of +7.7 kcal/mol, indicating
that radical 2 should possess exceptional stability. This com-
putational study also predicted a short exocyclic C-C bond
length (1.361 Å), indicative of a double bond and implying a
η3-complex with significant spin delocalization onto the Fe. On
the basis of this study, we reasoned that formation of radical 2
should be straightforward, but the question loomed whether it
could be rendered reactive enough to add to olefins. There has
* To whom correspondence should be addressed. E-mail: byers@
middlebury.edu; jjasinsk@keene.edu.
† Middlebury College.
‡ Keene State College.
§ Howard University.
(1) (a) Deniz, A. A.; Peters, K. S.; Snyder, G. J. Science 1999, 286,
1119. (b) Maier, G. Angew. Chem., Int. Ed. Engl. 1988, 27, 309. (c) Bally,
T.; Masamune, S. Tetrahedron 1980, 36, 343. (d) Maier, G. Angew. Chem.
1974, 86, 491.
(2) (a) Cram, D. J. Nature 1992, 356, 29. (b) Lin, C. Y.; Krantz, A. J.
Chem. Soc., Chem. Commun. 1972, 1111. (c) Chapman, O. L.; McIntosh,
C. L.; Pacansky, J. J. Am. Chem. Soc. 1973, 95, 2744.
(3) (a) Seyferth, D. Organometallics 2003, 22, 2. (b) Watts, L.;
Fitzpatrick, J. D.; Pettit, R. J. Am. Chem. Soc. 1965, 87, 3253. (c) Grubbs,
R. H.; Pancoast, T. A.; Grey, R. A. Tetrahedron Lett. 1974, 28, 2425. (d)
Grubbs, R. H.; Pancoast, T. A. J. Am. Chem. Soc. 1977, 99, 2382.
(4) (a) Davis, R. E.; Simpson, H. D.; Grice, N.; Pettit, R. J. Am. Chem.
Soc. 1971, 93, 6688. (b) Eschbach, C. S.; Seyferth, D.; Reeves, P. C. J.
Organomet. Chem. 1976, 104, 363.
(6) Creary, X.; Mehrsheikh-Mohammadi, M. E.; McDonald, S. J. Org.
Chem. 1989, 54, 2904.
(7) Creary, X. Org. Lett. 2000, 2, 2069.
(5) (a) J. A.; Randall, M. L.; Snapper, M. L. J. Am. Chem. Soc. 1996,
118, 9196. (b) Snapper, M. L.; Tallarico, J. A.; Randall, M. L. J. Am. Chem.
Soc. 1997, 119, 1478. (c) Limanto, J.; Snapper, M. L. J. Org. Chem. 1998,
63, 6440. (d) Porter, J. R.; Snapper, M. L. Synthesis 1999, 1407. (e) Limanto,
J.; Snapper, M. L. J. Am. Chem. Soc. 2000, 122, 8071. (f) Deak, H. D.;
Stokes, S. S.; Snapper, M. L. J. Am. Chem. Soc. 2001, 123, 5152. (g)
Limanto, J.; Tallarico, J. A.; Porter, J. R.; Khuong, K.; Houk, K. H.; Snapper,
M. L. J. Am. Chem. Soc. 2002, 124, 14748.
(8) Hwang, W.-S.; Liao, R.-L.; Ong, C. W. Polyhedron 1989, 8, 479.
(9) Wei, L. L.; Lai, M. C.; Ong, C. W. Tetrahedron Lett. 2000, 41, 3407.
(10) (a) Salazar, K. L.; Nicholas, K. M. Tetrahedron 2000, 56, 2211.
(b) Salazar, K. L.; Khan, M. A.; Nicholas, K. M. J. Am. Chem. Soc. 1997,
119, 9053.
(11) (a) Melikyan, G. G.; Khan, M. A.; Nicholas, K. M. Organometallics
1995, 14, 2170. (b) Melikyan, G. G.; Combs, R. C.; Lamirand, J.; Khan,
M.; Nicholas, K. M. Tetrahedron Lett. 1994, 35, 363.
10.1021/om060263n CCC: $33.50 © 2006 American Chemical Society
Publication on Web 06/21/2006

3788 Organometallics, Vol. 25, No. 15, 2006
Notes
Scheme 1
Scheme 2
been scant experimental evidence for radicals of this type.
Reductive McMurry radical coupling of the sodium salt of
hydroxymethyl CBIT has been shown to form dimer 3.12 The
reductive pinacol coupling of acetyl CBIT by aluminum
amalgam, possibly proceeding through a ketyl intermediate, has
also been reported.13
Scheme 3
Results and Discussion
reaction in refluxing benzene in the presence of AIBN. This
was not surprising, in light of the fact that this reagent will
not react with styrene due to slow rates of phenylseleno trans-
fer to well-stabilized benzylic radicals.22 The more reactive
reagent, methylphenylselenomalononitrile,23 which is known to
add to styrene, could be added to these CBIT-substituted ole-
fins 6 and 7 upon radical initiation, in 64% and 25% yields,
respectively (Scheme 2). These proved to be very sluggish
reactions, however, probably due to the intermediacy of CBIT-
stabilized radicals, which would be expected to undergo the
phenylselenide-transfer step slowly due to their low reactivity.
NMR analysis of the crude reaction mixture yielding 9 showed
only one singlet in the normal olefinic region (5.9 ppm), leading
us to conclude that only one olefin stereoisomer was gen-
erated. Phase-sensitive NOESY NMR spectroscopy of purified
9 gave a significant negative off-diagonal peak showing
correlation between this vinylic resonance and the ortho-proton
on the CBIT ring, supporting the Z-stereochemical assignment
shown.
Our initial studies focused on the direct generation and
trapping of radical 2, with the assumption that this radical would
prove highly nucleophilic, and relatively unreactive. Since
phenyl selenides have proven to be useful radical precursors in
substrates where solvolysis might be otherwise problematic,14
we chose to generate phenylselenide 4 as a precursor to radical
2. This was accomplished by treatment of hydroxymethyl
CBIT15 with PhSeH and catalytic H2SO4, leading to formation
of 4 in 87% yield. The attempted trapping of radical 2 by
allyltributylstannane,16 generated from 4 under standard AIBN
conditions, yielded only minute traces of the desired allylated
product, as evidenced by a minor component in the GC/MS
exhibiting an ion at 245 m/z (M+ - 1). The only isolable product
of this reaction (Scheme 1) was dimer 3, as evidenced by its
molecular ion at 410 m/z, and NMR data, which matched
literature values.12 The dimer presumably arose from radical-
radical coupling, characteristic of radicals too unreactive to add
to olefins at synthetically useful rates.17
We reasoned that the rate of radical-olefin reactions arising
from 2 might be enhanced if a more electron-deficient allyl-
stannane was employed. When 4 was allowed to react with
2-carboethoxyallyltributylstannane18 in the presence of AIBN,
the desired allylated product was obtained in 42% yield (Scheme
1). The successful trapping of this electron-deficient olefin not
only confirmed our thesis that 2 should be a highly nucleophilic
radical but also diminished any outstanding concerns that
allylation might arise through allylstannane nucleophilicity,
rather than through a radical process. Thus, we consider this
reaction strong evidence for the formation of radical 2.
With the above evidence in hand, we turned our attention to
radical reactions arising from the SmI2 ketyl derived from formyl
CBIT (10). The Sm(III) ketyl generated from benzaldehyde-
Cr(CO)3 has been successfully trapped with acrylate esters,
leading to butyrolactones.24 Treatment of 10 with SmI2 in the
presence of methyl acrylate or ethyl 2-cyanoacrylate yielded
only diol 11. Apparently, the ketyl intermediate is too unreactive
to add to even highly reactive olefins. Optimizing this reaction
by omitting the electron-deficient olefin, we obtained diol 11
as one diastereomer in 94% yield (Scheme 3). While we were
unable to generate crystals suitable for X-ray crystallographic
analysis from diol 11, its di-p-nitrobenzoate ester (12) did prove
suitable for this purpose and indicated that the stereochemistry
was indeed the syn stereochemistry shown. High syn diastereo-
selectivity has also been observed in the SmI2-promoted di-
merizations of other similar organometallic aldehydes, such as
ferrocenecarboxaldehyde,25,26 (benzaldehyde)Cr(CO)3,25 (di-
We also hoped to demonstrate the intermediacy of radicals
similar to 2 through successful atom-transfer addition reactions
to vinyl19 (6) and ethynyl-CBIT20 (7). Dimethyl methylphen-
ylselenomalonate21 did not react with either of these olefins upon
(12) Adams, C. M.; Holt, E. M. Organometallics 1990, 9, 980.
(13) Marcinal, P.; Hannoir-Guisez, N.; Cuignet, E. TraV. Soc. Pharm.
Montpellier 1973, 33, 281.
(14) (a) Byers, J. H.; Whitehead, C. C.; Duff, M. E. Tetrahedron Lett.
1996, 37, 2743. (b) Rawal, V. H.; Singh, S. P.; Dufour, C.; Michoud, C. J.
Org. Chem. 1993, 58, 7718. (c) Beckwith, A. L. J.; Pigou, P. E. Aust. J.
Chem. 1986, 39, 77.
(15) Fitzpatrick, J. D.; Watts, L.; Emerson, G. F.; Pettit, R. J. Am. Chem.
Soc. 1965, 87, 3254.
(19) (a) Biehl, E. R.; Reeves, P. C. Synthesis 1974, 883. (b) Nunn, E. E.
Aust. J. Chem. 1976, 29, 2549.
(20) Bunz, U. H. F.; Wiegelmann-Kreiter, J. E. C. Chem. Ber. 1996,
129, 785.
(21) Byers, J. H.; O’Leary, B. M.; Zraly, M. A. Synth. Commun. 1996,
26, 4039.
(16) Keck, G. E.; Enholm, E. J.; Yates, J. B.; Wiley, M. R. Tetrahedron
1985, 41, 4079.
(17) Viehe, H. G.; Merenyi, R.; Stella, L.; Janousek, Z. Angew. Chem.,
Int. Ed. Engl. 1979, 18, 917.
(22) Byers, J. H.; Lane, G. C. J. Org. Chem. 1993, 58, 3355.
(23) (a) Curran, D. P.; Thomas, G. J. Am. Chem. Soc. 1992, 114, 4436.
(b) Curran, D. P.; Eichenberger, E.; Collis, M.; Roepel, M. G.; Thoma, G.
J. Am. Chem. Soc. 1994, 116, 4279.
(18) (a) Byun, H.; Reddy, K. C.; Bittman, R. Tetrahedron Lett. 1994,
35, 1371. (b) Villieras, J.; Rambaud, M. Synthesis 1982, 925. (c) Tanaka,
H.; Hai, A. K. M.; Ogawa, H.; Torii, S. Synlett 1993, 11, 835.
(24) Merlic, C. A.; Walsh, J. C. Tetrahedron Lett. 1998, 39, 2083. (b)
Taniguchi, N.; Uemura, M. Tetrahedron Lett. 1997, 38, 7199.
(25) Taniguchi, N.; Uemura, M. Tetrahedron 1998, 54, 12775.

Notes
Organometallics, Vol. 25, No. 15, 2006 3789
Scheme 4
Scheme 5
potassium/benzophenone prior to use. Dry benzene was distilled
from CaH2 prior to use. Nuclear magnetic resonance (NMR)
analysis was conducted on a 400-MHz Bru¨ker Avance nuclear
magnetic resonance spectrometer. Gas chromatography and mass
spectrometry (GC/MS) were performed with a 15 m × 0.25 µm
Agilent cross-linked methyl silicone HP-5 capillary column on an
Agilent 6890/5973 GC/MS. Zero-point energies were calculated
at the B3LYP/LANL2DZ level using Gaussian 2003 on a Compaq
ES40 cluster. Spin density was determined at the B3LYP/6-31G*
level using Spartan 2002 on a Dell Latitude D610 personal
computer. Vibrational analyses were carried out to ensure energy
minima.
[1,2-Di-η4-cyclobutadienylethane]bis(tricarbonyliron) (3). Phen-
ylselenide 4 (120 mg, 0.33 mmol), allyltributylstannane (331 mg,
1.0 mmol), and AIBN (55 mg, 0.33 mmol) were dissolved in 2
mL of benzene. The reaction mixture was deoxygenated with
bubbling N2 for 10 min and heated to reflux for 18 h. Analysis of
the crude reaction mixture by TLC showed unreacted limiting
reagent 4, so an additional 25 mg (0.15 mmol) of AIBN was added.
The mixture was deoxygenated and heated to reflux for an additional
4 h. Analysis by TLC and GC/MS showed complete consumption
of 4. Trace quantities of 3-butenyl CBIT could be detected in the
GC/MS: (m/z) 245 (M+ - 1), 217, (M+ - 1 - 28), 189 (M+ - 1
- 56), 163 (M+ - 1 - 84). MPLC with 97% hexane/3% EtOAC
enal)Fe(CO)3 complexes,25 and (propargyl aldehyde)Co(CO)6
complexes.27
Stereoselectivity in SmI2-promoted pinacolization has been
described as arising from binding of the SmIII ketyl to the
unreduced second aldehyde prior to C-C bond formation and
the subsequent second reduction.25
Aldehyde 10 could be converted to its benzyl imine (13),
which underwent reductive dimerization with SmI2 to form
diamine 14 (Scheme 4), albeit in diminished yield (44%) and
with poor (1.1:1) diastereoselectivity, again, as expected from
the literature.28 Given the poor stereoselectivity, no attempt was
made to assign stereochemistry to either of the otherwise fully
characterized isolated dimers. The N-stabilized radical generated
in this reaction should be less reactive than the previously
studied O-stabilized radicals.29 This diminished reactivity could
preclude reaction with a formally nonradical species, somewhat
akin to the behavior observed in other well-stabilized radicals,
such a capto-dative radicals.17 Thus, the dimerization may occur
between two fully reduced, discrete Sm-bound species, leading
to an intermolecular coupling through an open, and less
stereoselective, transition state.
Finally, we were interested in studying the formation of
radical 16, in which the radical is formed directly on one of the
ring carbons. The isodesmic reaction of 16 with benzene to form
CBIT and a phenyl radical gave a ∆Ε of -3.7 kcal/mol at the
B3LYP/LANL2DZ level. The spin density of radical 16 also
showed that the radical was mostly localized on the ring carbon,
orthogonal to the π system, similar to a phenyl radical. Reaction
of iodide 15 with 2-carboethoxyallyltributylstannane (Scheme
5) led to ester 17 in 39% yield, indicative of a reaction
proceeding through radical 16.
1
(v/v) gave 34 mg of 3 (50%), homogeneous by TLC. H NMR
matched that reported in the literature.7 MS (m/z): 410 (M+).
Tricarbonyl(η4-phenylselenomethylcyclobutadiene)iron (4).
Hydroxymethyl CBIT (270 mg, 1.21 mmol) was dissolved in 10
mL of benzene, and 288 mg (1.8 mmol) of PhSeH was added. The
mixture was cooled to 5 °C, and 3 drops (∼50 mg) of concentrated
H2SO4 was added. The mixture was stirred for 2 h, gradually
warming to room temperature. The crude reaction mixture was
washed with two 10 mL portions of 10% NaOH and 20 mL of
water, dried over MgSO4, and filtered, and solvents were removed
by rotary evaporation. MPLC with hexane gave 380 mg (87%) of
1
4, homogeneous by TLC, as a yellow oil. H NMR (CDCl3): δ
3.40 (s, 2H); 3.92 (s, 2H); 3.97 (s, 1H); 7.35 (m, 3H), 7.61 (m,
2H). 13C NMR (CDCl3): δ 25.9, 61.8, 64.6, 84.3, 128.3, 129.6,
129.9, 134.8, 214.8. IR (neat): 2042, 1958 cm-1 (CO). MS (m/z):
334 (M+ -28). Anal. Calcd for C14H10O3SeFe: C, 46.56, H, 2.79.
Found: C, 46.76, H, 2.88
Tricarbonyl(η4-(3-carboethoxy-3-butenyl)cyclobutadiene)-
iron (5). Phenylselenide 4 (200 mg, 0.55 mmol), 2-carboethoxy-
lallyltributylstannane (0.688 mg, 1.7 mmol), and AIBN (18 mg,
0.11 mmol) were dissolved in 2 mL of benzene. The solution was
deoxygenated with bubbling N2, and the mixture was heated to
reflux under N2 for 18 h. The reaction mixture was cooled to room
temperature, and TLC analysis showed complete consumption of
limiting reagent 4. The reaction mixture was purified by MPLC
with hexane (to elute PhSeSePh), followed by 99% hexane/1%
EtOAc (v/v) to remove 2-carboethoxyallyltributylstannane, and 95%
hexane/5% EtOAc (v/v) to elute 146 mg of an oil, which while
homogeneous by TLC, appeared to be a mixture of 5 and a second,
unidentified component by 1H NMR. Kugelrohr distillation (3
mmHg, 175 °C) yielded 96 mg (55%) of 5, pure by TLC and NMR,
Conclusion
In conclusion, we have provided the first strong evidence for
the formation and trapping of a variety of radicals derived from
CBIT. Radicals generated R to the cyclobutadiene ring appear
to be quite stable, as evidenced by their sluggish reactivity, and
highly nucleophilic, given that they only appear to react with
electron-deficient olefins. The radical intermediate generated
directly on the ring carbon of CBIT behaves rather similarly to
the phenyl radical.
1
as an oil. H NMR (CDCl3): δ 1.07 (t, 3H, J ) 7.0 Hz), 2.00 (t,
2H, J ) 7.5 Hz), 2.33 (t, 2H, J ) 7.5 Hz), 3.37 (s, 1H), 3.49 (s,
2H), 4.08 (q, 2H, J ) 7.0 Hz), 5.27 (d, 1H, J ) 2 Hz), 6.23 (d, 1H,
J ) 2 Hz). 13C NMR (CDCl3): δ 14.3, 27.0, 32.6, 59.9, 60.8, 64.2,
88.6, 125.3, 140.2, 166.5, 215.6. IR (neat): 2042, 1961, 1716 cm-1
(CO). MS: m/z 290 (M+ - 28). Anal. Calcd for C14H14O5Fe: C,
52.86, H, 4.44. Found: C, 52.55, H, 4.34.
Tricarbonyl(η4-2, 2-dicyano-4-phenylselenobutylcyclobuta-
diene)iron (8). A 10 mm NMR tube fitted with a resealable valve
was charged with 0.225 g (1.03 mmol) of vinyl CBIT (6) and 0.470
g (2.00 mmol) of methylphenylselenomalononitrile in 2 mL of
CDCl3. A catalytic amount of AIBN (5 mg, 0.03 mmol) was also
Experimental Section
General Methods. Reagent grade hexane and ethyl acetate were
distilled prior to use. Tetrahydrofuran (THF) was distilled from
(26) Taniguchi, N.; Uemura, M. Tetrahedron Lett. 1998, 39, 5385.
(27) Lake, K.; Dorrell, M.; Blackman, N.; Khan, M. A.; Nicholas, K.
M. Organometallics 2003, 22, 4260.
(28) Kim, M.; Knettle, B. W.; Dahlen, A.; Hilmersson, G.; Flowers, R.
A. Tetrahedron 2003, 59, 10379.
(29) Wu, Y.-D.; Wong, C.-L.; Chan, K. W. K. J. Org. Chem. 1996, 61,
746.

3790 Organometallics, Vol. 25, No. 15, 2006
Notes
added to the tube and was dissolved in the CDCl3. The tube was
then sealed, and the solution was deoxygenated by three freeze-
pump-thaw cycles on a 3 Torr vacuum manifold. The tube was
heated to 68 °C in an oil bath for 68 h. Progress of the reaction
was monitored by NMR and GC/MS after the first 68 h and every
24 h after that until the reaction was complete. Every 24 h time
the tube was opened, an additional 5 mg portion of AIBN was
added, and the reaction mixture was redeoxygenated. After 240 h,
the reaction was complete. The crude product was purified by flash
chromatography on silica gel with 10% EtOAc/90% hexane (v/v)
to obtain 0.302 g (65%) of an amber oil, homogeneous by TLC.
1H NMR (CDCl3): δ 1.89 (s, 3H); 2.18 (m, 2H); 3.89 (s, 1H); 4.0
(m, 1H); 4.18 (s, 2H); 7.3-7.85 (m, 5H). 13C NMR (CDCl3): δ
26.3, 30.8, 36.8, 42.8, 62.8, 64.1, 86.0, 117.4, 117.8, 127.4 129.5,
213.9. MS: m/z 218 (M+), 190 (M+ - CO), 162 (M+ - 2CO),
134 (M+ - 3CO). Anal. Calcd for C19H14FeN2O3Se: C, 50.36, H,
3.11, N, 6.18. Found: C, 50.69, H, 2.73, N, 6.28.
(34%) of 12 as a pale yellow crystalline solid, homogeneous by
TLC. H NMR (CDCl3): δ 8.30 (m, 4H), 8.18 (m, 4H), 5.71 (s,
2H), 4.31 (s, 2H), 4.24 (s, 4H). IR (KBr): 2048, 1974, 1736 (CO)
cm-1. Crystals suitable for X-ray analysis were grown by slow
diffusion of hexane into a THF solution of 12.
1
(1,2-Di-η4-cyclobutadien-yl-1,2-dibenzylamino)bis(tricarbo-
nyliron) (14). Formyl CBIT (10) (579 mg, 2.6 mmol) was dissolved
in 5 mL of toluene. Benzylamine (0.31 g, 2.9 mmol) was added,
and the toluene was removed by rotary evaporation, yielding imine
13, which was used without further purification. 1H NMR
(CDCl3): δ 7.62 (s, 1H), 7.20-7.45 (m, 5H), 4.59 (s, 1H), 4.52 (s,
2H), 4.24 (s, 2H). Imine 13 was dissolved in 10 mL of THF, and
the resulting solution was cooled to -78 °C. A 50 mL portion of
SmI2 (0.1 M in THF) was added via syringe, and the resulting
solution was stirred for 1 h, gradually warming to room temperature.
Methanol (2 mL) was added in order to quench the reaction, and
50 mL of ether was added. The resulting solution was washed with
two 50 mL portions of water, the remaining organic phase was
dried over MgSO4 and filtered, and solvents were removed by rotary
evaporation. MPLC with 10% EtOAc/90% hexane (v/v) gave 181
mg (23%) of a higher rf diamine and 167 mg (21%) of a lower rf
diamine, each homogeneous by TLC. High rf diamine: 1H NMR
(CDCl3) δ 7.2-7.4 (m, 10H), 4.19 (d, J ) 9.1 Hz, 2H), 4.15 (s,
2H), 3.90, (d, J ) 9.1 Hz, 2H), 3.90 (d, J ) 12.7 Hz, 2H), 3.71 (d,
J ) 12.7 Hz, 2H), 3.13 (s, 2H), 1.4 (bs, 2H); 13CNMR (CDCl3) δ
214.7, 139.9, 128.6, 128.3, 127.3, 87.1, 64.1, 63.7, 62.1, 59.0, 52.4.
Anal. Calcd for C30H24Fe2N2O6: C, 58.08, H, 3.90, N, 4.54.
Found: C, 58.08, H, 3.98, N, 4.49. Low rf diamine: 1H NMR
(CDCl3) δ 7.2-7.4 (m, 10H), 4.40 (d, J ) 9.1 Hz, 2H), 4.16 (s,
2H), 3.92 (d, J ) 9.1 Hz, 2H), 3.80 (d, J ) 13.1 Hz, 2H), 3.64 (d,
J ) 13.1 Hz, 2H), 3.15 (s, 2H), 2.22 (s, 2H); 13C NMR (CDCl3) δ
214.4, 140.0, 128.5, 128.2, 127.2, 86.9, 66.1, 61.3 (2 peaks), 58.0,
52.1. IR (neat): 2040, 1965 Anal. Calcd for C30H24Fe2N2O6: C,
58.08, H, 3.90, N, 4.54. Found: C, 58.08, H, 3.93, N, 4.52.
Tricarbonyl(η4-(2-carboethoxy-2-propenyl)cyclobutadiene)-
iron (17). Iodo CBIT30 (0.268 g, 0.840 mmol) and 2-carboethoxy-
lallyltributylstannane (0.682 g, 1.68 mmol) were dissolved in 1 mL
of C6D6 and added to a 10 mm NMR tube fitted with a resealable
valve. A catalytic amount of AIBN (5 mg, 0.03 mmol) was then
added to the tube, and the solution was deoxygenated by three
freeze-pump-thaw cycles, sealed, and then heated to 80 °C for
50 h. The crude product was purified by MPLC on silica gel with
3% EtOAc/97% hexane (v/v) to yield 100 mg (39%) of 17,
Tricarbonyl(η4-2,2-dicyano-4-phenylseleno-3-butenylcyclo-
butadiene)iron (9). A 10 mm NMR tube fitted with a resealable
valve was charged with 0.234 g (1.08 mmol) of ethynyl CBIT (7)
and 0.51 g (2.16 mmol) of methylphenylselenomalononitrile in 2
mL of C6D6. A catalytic portion of AIBN (5 mg, 0.03 mmol) was
also added to the tube. The tube was then sealed, and the solution
was deoxygenated by three freeze-pump-thaw cycles. The tube
was heated to 85 °C in an oil bath and left for 24 h. Progress of
the reaction was monitored by NMR in 24 h increments. An
additional 5 mg portion of AIBN was added after each 24 h
increment. After 72 h, the reaction was complete, as evidenced by
the disappearance of limiting reagent 7. The crude product was
purified by MPLC with 20% EtOAc/80% hexane (v/v), to give 122
mg (25%) of addition product 9, homogeneous by TLC. 1H NMR
(CDCl3): δ 2.08 (s, 3H); 4.06 (s, 2,H); 4.12 (s, 1H); 5.87 (s, 1H);
7.20-7.70 (m, 5H). 13C NMR (CDCl3): δ 27.0, 31.8, 64.1, 64.6,
80.2, 114.9, 122.9, 128.2, 128.9, 129.8, 132.4, 171.0, 213.0. IR
(neat): 2048, 1977 cm-1 (CO). Anal. Calcd for C19H12O3N2SeFe:
C, 50.59, H, 2.68, N, 6.21. Found: C, 50.61, H, 2.62, N, 6.17.
[syn-1,2-Di-η4-cyclobutadien-yl-1,2-dihydroxyethane]bis(tri-
carbonyliron) (11). Formyl CBIT (10) (176 mg, 0.8 mmol) was
dissolved in 10 mL of dry THF. After cooling the contents of the
flask to -78 °C, 17.6 mL (1.76 mmol) of SmI2 solution (0.1 M in
THF) was slowly added to the stirred solution via syringe. The
blue solution was stirred for an additional 40 min under N2 and
then quenched with 2 mL of methanol. Saturated NaHCO3 was
added, and the resulting solution was extracted twice with 50 mL
portions of EtOAc. The combined organic layers were dried over
MgSO4 and concentrated on a rotary evaporator. The residue was
purified by column chromatography on silica gel with 50% EtOAc/
50% hexane (v/v) to give 166 mg of diol 11 as pale orange crystals
1
homogeneous by TLC. H NMR (CDCl3): δ 0.89 (t, J ) 7.1 Hz,
3H); 1.30 (q, J ) 7.1 Hz, 2H); 3.00 (s, 2H); 4.12 (s, 2H); 4.26 (s,
1H); 5.20 (s, 1H); 6.24 (s, 1H). 13C NMR (CDCl3): δ 14.2, 30.6,
60.6, 61.0, 64.5, 84.8, 126.3, 138.4, 167.3, 214.7. MS: m/z 304
(M+), 276 (M+ - CO), 248 (M+ - 2CO), 220 (M+ - 3CO). Anal.
Calcd for C13H12O2Fe: C, 51.34, H, 3.98. Found: C, 51.71, H,
4.25.
1
(94%), homogeneous by TLC. Mp: 115-119 °C (dec). H NMR
(CD3COCD3): δ 4.45 (s, 2H), 4.40 (s, 6H), 3.95 (s, 2H). 13C NMR
(CD3COCD3): δ 215.2, 86.8, 71.6, 64.6, 64.2, 62.7. IR (KBr): 3430
(OH), 2047, 1971, 1957 cm-1 (CO). Anal. Calcd for C16H10O8Fe2:
C, 43.48, H, 2.28. Found: C, 43.83, H, 2.31.
Acknowledgment. Financial support is acknowledged from
the NSF (CHE-0315745) and Vermont EPSCOR. Summer
research fellowships for Y.Z. and B.M.Z. were provided through
Pfizer Summer Undergraduate Research Fellowships. A summer
research fellowship for T.S.D. was provided by the Vermont
Genetics Network. We thank Prof. Thomas Cheatham, Depart-
ment of Medicinal Chemistry, University of Utah, for use of
his computational resources.
[syn-1,2-Di-η4-cyclobutadien-yl-1,2-di-p-nitrobenzoylethane]-
bis(tricarbonyliron) (12). A 104 mg (0.24-mmol) portion of diol
11 was dissolved in 10 mL of THF under N2. The resulting solution
was cooled to 0 °C, and 47 mg (1.18 mmol, 60% by wt in mineral
oil) of NaH was added. After evolution of H2 was complete, 139
mg (0.875 mmol) of p-nitrobenzoyl chloride was added. The
reaction was stirred for 72 h, warming to room temperature. TLC
analysis showed the presence of unreacted diol, so an additional
24 mg (0.59 mmol) of NaH was added, followed by 50 mg (0.27
mmol) of p-nitrobenzoyl chloride. The reaction mixture was allowed
to stir overnight. The crude reaction mixture was then added to 50
mL of EtOAc, washed with 50 mL of water, dried over MgSO4,
and filtered, and solvents were removed by rotary evaporation. Flash
chromatography with 80% hexane/20% EtOAc (v/v) gave 51 mg
Supporting Information Available: Crystallographic infoma-
tion files (CIF) and experimental parameters for 12, results of
calculations. This material is available free of charge via the Internet
at http://pubs.acs.org.
OM060263N
(30) Bunz, U. Organometallics 1993, 12, 3594.