Alkyl isomerisation in three-coordinate iron(II) complexes

Article abstract of DOI:10.1039/b209389h

The tertiary to iso-butyl isomerisation of three-coordinate iron(II) diketiminate complexes is reported and a hydride intermediate is proposed on the basis of exchange experiments.

Full text of DOI:10.1039/b209389h

Alkyl isomerisation in three-coordinate iron(II) complexes†
Javier Vela, Jeremy M. Smith, Rene J. Lachicotte and Patrick L. Holland*
Department of Chemistry, University of Rochester, Rochester, NY 14627, USA.
E-mail: holland@chem.rochester.edu
Received (in Purdue, IN, USA) 26th September 2002, Accepted 21st October 2002
First published as an Advance Article on the web 4th November 2002
The tertiary to iso-butyl isomerisation of three-coordinate
iron(II) diketiminate complexes is reported and a hydride
intermediate is proposed on the basis of exchange experi-
ments.
The isomerisation of transition metal alkyl complexes
typically leads to more than one isomer at equilibrium.10
Experiment and theory have shown that electronic factors
determine the relative energies of the isomers.11 Thus linear
isomers tend to be favoured for iron and other metals, and
tertiary alkyl iron complexes are rare.12
Complexes of the late transition metals with bulky bidentate
ligands have gained attention because they polymerise ethylene
into highly branched polyethylene.1 The mechanisms of
propagation, chain transfer, and chain walking in group 10
systems have been studied thoroughly.2 More recently Gibson,
Brookhart, and coworkers have reported iron(II) catalysts that
give linear polyethylene with high molecular weight.3 The
The formation of the tert-butyl compound 2a with the more
sterically demanding ligand leads us to think that 2b is initially
formed during the production of 4b.§ A plausible mechanism
(Scheme 1) to account for tertiary to iso-butyl isomerisation
involves b-hydride elimination to form olefin hydride inter-
mediate 3b; after rotation or dissociation, reinsertion gives 4b.
The bulkier ligand with R = tert-butyl7 raises the energy of the
four-coordinate transition state (relative to 2a/2b) more than the
ligand with R = methyl.
active catalyst is thought to be a low-coordinate alkyliron(II
)
complex.4 The differences in polymer structure suggest that
there are differences in the rate of the ‘chain walking’ process
that interconverts linear and branched alkylmetal species.
However, the mechanism of isomerisation in low-coordinate
alkyliron(II) complexes has not been studied experimentally
except by analysis of polymer structure, presumably because of
the paramagnetism of the active complexes.5 A study of
In order to establish the existence of 3b, 4b was heated at 60
°C in the presence of ethylene. The 1H NMR spectrum showed
the clean formation of a new compound, along with isobutylene
(Scheme 1). This compound was identified as ethyl complex 5b
by independent synthesis from ethylmagnesium chloride and
1b, and its structure was determined by X-ray diffraction.‡
The transformation of 4b to 5b implicates 3b as an
undetected intermediate. However, another mechanism has
been proposed to explain chain transfer in a-olefin polymer-
isations by late metal catalysts (Scheme 2).1,5 Mechanism A
starts with alkene binding (i) to form four coordinate alkyl-
olefin intermediate 6, followed by intramolecular hydride
transfer (ii) and loss of the new alkene (iii). Mechanism B
involves a hydride intermediate 3, like the one proposed above
for butyl isomerisation. We have found that the first-order rate
constant for conversion of iso-butyl complex 4b to ethyl
complex 5b is independent of ethylene concentration between
0.1 and 0.9 M. When exchange is monitored with propene or
3,3,3-trifluoropropene instead of ethylene, the rate constants are
nearly identical to that for ethylene.¶ These data are inconsistent
with mechanism A. Moreover, these observations rule out v and
vi as rate-determining steps, allowing us to conclude that iv (b-
hydride elimination) is rate limiting. An Eyring plot of rate
constants in the range 20–74 °C yielded the activation
parameters DH‡ = +77 ± 2 kJ mol21 and DS‡ = 270 ± 8 J
relevant diamagnetic cobalt(
I) complexes appeared very re-
cently.6
This communication describes paramagnetic, low-coordinate
butyl complexes of iron(II) and demonstrates that alkyl
isomerisation and alkene transfer occur through transient
hydride complexes. Bulky b-diketiminate ligands (series a: R =
tBu; series b: R = Me) are used to modify the size of the
binding pocket, which is smaller in 1a than 1b.7 Our initial
synthetic results are shown along the top and right of Scheme 1.
Treatment of iron chloride complex 1a7 with tert-butylmagne-
sium chloride in diethyl ether at room temperature yields
complex 2a. When the same reaction is performed with 1b, only
the iso-butyl isomer 4b is isolated. The structures of 2a and 4b
have been confirmed by X-ray crystallography (Fig. 1)‡ and are
similar to the benzyl complex recently reported by Sciarone et
al.,8 and to our recently reported alkyl complexes.9 The
products of these reactions are independent of the batch of
Grignard reagent or replacement by di-tert-butylmagnesium.
Scheme 1 Preparation and reactions of iron-butyl complexes. Ar
=
2,6-diisopropylphenyl. Series a: R = tBu, X = Cl. Series b: R = Me, X =
Cl2Li(THF)2.
† Electronic supplementary information (ESI) available: general considera-
tions, synthesis of compounds, kinetic studies and crystal data. See http://
www.rsc.org/suppdata/cc/b2/b209389h/
Scheme 2 Olefin exchange mechanisms.
2886
CHEM. COMMUN., 2002, 2886–2887
This journal is © The Royal Society of Chemistry 2002

mm21, 22 803 reflections measured, 8753 unique (Rint = 0.0198), R1
0.0777, wR2 = 0.1879, [I > 2s(I)], GOF = 1.128.
=
For 4b, C33H50FeN2, M = 530.60, monoclinic, space group P21/c, a =
20.8440(14), b = 8.6523(6), c = 17.9567(11) Å, b = 91.9790(10)°, U =
3236.5(4) Å3, T = 193(2) K, Z = 4, m(Mo-Ka) = 0.487 mm21, 19911
reflections measured, 7792 unique (Rint = 0.0285), R1 = 0.0545,
wR2 = 0.1335, [I > 2s(I)], GOF = 1.041.
For 5b, C31H46FeN2, M = 502.55, monoclinic, space group P21/n, a =
15.7646(12), b = 9.3607(7), c = 20.6882(16) Å, b = 93.9060(10)°, U =
3045.8(4) Å3, T = 193(2) K, Z = 4, m(Mo-Ka) = 0.514 mm21, 16927
reflections measured, 4823 unique (Rint = 0.0234), R1 = 0.0530, wR2
0.1559, [I > 2s(I)], GOF = 1.025.
=
2a: Yield 71%; Analysis found(calcd.) C, 76.40(76.19)%, H,
10.03(10.17)%, N, 4.45(4.56)%; meff(Evans) = 5.6(3) mB; 1H NMR (d/ppm,
400 MHz, C6D6, 21 °C): 130 (br s, 1H, b-CH), 128 (br s, 9H, tBu-CH3), 44
(br s, 18H, C(CH3)3), 25 (s, 4H, m-CH), 229 (s, 12H, iPr-CH3), 2110 (s,
i
i
2H, p-CH), 2116 (br s, 12H, Pr-CH3), 2143 (br s, 4H, Pr-CH); Vis
(pentane): 510 nm (510 M21cm21).
4b: Yield 86%; Analysis found(calcd.) C, 74.10(74.70)%, H,
1
9.56(9.50)%, N, 5.24(5.28)%; meff(Evans) = 6.0(3) mB; H NMR (d/ppm,
400 MHz, C6D6, 21 °C): 130 (br s, 1H, b-CH), 106 (br s, 6H, iBu-CH3), 70
(br s, 6H, a-CH3), 212 (s, 4H, m-CH), 218 (s, 12H, iPr-CH3), 274 (s, 2H,
p-CH), 2115 (br s, 12H, iPr-CH3), 2132 (br s, 4H, iPr-CH); Vis (pentane):
463 nm (810 M21cm21), 490 nm (720 M21cm21).
Fig. 1 ORTEP diagrams with thermal ellipsoids at 50% probability.
Hydrogen atoms not shown. Selected bond lengths (Å) and angles (°): for
2a, Fe(1)–N(11) 2.017(2), Fe(1)–C(14) 2.079(4), N(21)–Fe(1)–N(11)
93.10(9), N(21)–Fe(1)–C(14) 133.11(12); for 4b, Fe–N(1) 1.9830(16), Fe–
C(14b) 2.019(6), N(1)–Fe–N(2) 92.77(7), N(2)–Fe–C(14b) 123.43(17); for
5b, Fe(1)–N(11) 1.9828(19), Fe(1)–C(14) 2.033(3), N(11)–Fe(1)–N(21)
93.44(9), N(11)–Fe(1)–C(14) 132.51(10).
5b: Yield 85%; Analysis found(calcd.) C, 73.29(74.09)%, H,
1
8.78(9.23)%, N, 5.45(5.57)%; meff(Evans) = 5.6(3) mB; H NMR (d/ppm,
400 MHz, C6D6, 21 °C): 130 (br s, 1H, b-CH), 69 (br s, 6H, a-CH3), 212
(s, 4H, m-CH), 220 (s, 12H, iPr-CH3), 276 (s, 2H, p-CH), 2123 (br, 16H,
i
iPr-CH3, Pr-CH); Vis (pentane): 461 nm (740 M21cm21), 489 nm (750
M21cm21).
mol21·K21. The transition state for iv is more ordered than 4b,
as expected for b-hydride elimination.
CCDC 194485–194487. See http://www.rsc.org/suppdata/cc/b2/
b209389h/ for crystallographic files in CIF or other electronic format.
§ A transient intermediate can be observed by 1H NMR spectroscopy upon
mixing of 1b and Grignard reagent, but attempts to isolate this species or
unambiguously characterise it by low temperature NMR and electronic
absorption spectroscopies were unsuccessful.
¶ Rate constants at 63 °C (kobs/s21): ethylene 1.82(9) 3 1023; propene
2.2(1) 3 1023; 3,3,3-trifluoropropene 1.56(6) 3 1023. These data are also
consistent with rate-determining alkene dissociation, but the negative value
of DS‡ argues against this possibility.
In order to examine directly the isomerisation of an isolated
tert-butyl complex to an iso-butyl complex, we heated tert-butyl
compound 2a and found that it converts to 4a in a first-order
process (Scheme 1). The activation parameters of DH‡ = 85 ±
5 kJ mol21 and DS‡ = 254 ± 18 J mol21·K21 are similar to
those for alkene exchange, implying that b-hydride elimination
is again rate limiting. The reaction rate is independent of
isobutylene concentration, consistent with the mechanism in
Scheme 1.
1 G. J. P. Britovsek, V. C. Gibson and D. F. Wass, Angew. Chem., Int. Ed.,
1999, 38, 428; S. D. Ittel, L. K. Johnson and M. Brookhart, Chem. Rev.,
2000, 100, 1169.
2 L. K. Johnson, C. M. Killian and M. Brookhart, J. Am. Chem. Soc.,
1995, 117, 6414; S. A. Svejda, L. K. Johnson and M. Brookhart, J. Am.
Chem. Soc., 1999, 121, 10634; L. H. Shultz, D. J. Tempel and M.
Brookhart, J. Am. Chem. Soc., 2001, 123, 11539.
3 G. J. P. Britovsek, V. C. Gibson, B. S. Kimberley, P. J. Maddox, S. J.
McTavish, G. A. Solan and A. J. P. White and D. J. Williams, Chem.
Commun., 1998, 849; B. L. Small and M. Brookhart, J. Am. Chem. Soc.,
1998, 120, 7143; B. L. Small, M. Brookhart and A. M. A. Bennett, J.
Am. Chem. Soc., 1998, 120, 4049.
4 L. Deng, P. Margl and T. Ziegler, J. Am. Chem. Soc., 1999, 121, 6479;
E. A. H. Griffiths, A. J. P. Britovsek, V. C. Gibson and I. R. Gould,
Chem. Commun., 1999, 1333.
5 G. J. P. Britovsek, M. Bruce, V. C. Gibson, B. S. Kimberley, P. J.
Maddox, S. Mastroianni, S. J. McTavish, C. Redshaw, G. A. Solan, S.
Strömberg, A. J. P. White and D. J. Williams, J. Am. Chem. Soc., 1999,
121, 8728.
6 V. C. Gibson, K. P. Tellmann, M. J. Humphries and D. F. Wass, Chem.
Commun., 2002, 2316.
7 J. M. Smith, R. J. Lachicotte and P. L. Holland, Chem. Commun., 2001,
1542.
8 T. J. J. Sciarone, A. Meetsma, B. Hessen and J. H. Teuben, Chem.
Commun., 2002, 1580.
9 J. M. Smith, R. J. Lachicotte and P. L. Holland, Organometallics, 2002,
21, 4808–4814.
10 D. L. Reger and P. J. McElligott, J. Organomet. Chem., 1981, 216, C12;
D. L. Reger, D. G. Garza and J. C. Baxter, Organometallics, 1990, 9,
873; J. A. Pool, C. A. Bradley and P. J. Chirik, Organometallics, 2002,
21, 1271.
11 T. J. Marks, M. R. Gagne, S. P. Nolan, L. E. Schock, A. M. Seyman and
D. Stern, Pure Appl. Chem., 1989, 61, 1665; J. L. Bennett and P. T.
Wolczanski, J. Am. Chem. Soc., 1997, 119, 10696; J. N. Harvey,
Organometallics, 2001, 20, 4887.
12 W. P. Giering and M. Rosenblum, J. Organomet. Chem., 1970, 25, C71;
D. L. Reger and E. C. Culbertson, Inorg. Chem., 1977, 16, 3104; K. H.
Pannell, T. Giasolli and R. N. Kapoor, J. Organomet. Chem., 1986, 316,
315.
Since 3 lies after the rate-determining step, kinetics do not
distinguish whether olefin exchange (v) is associative or
dissociative. However, heating either a mixture of 4a and 5b or
a mixture of 4b and 5a9 gave product solutions containing 4a,
4b, 5a, and 5b. Observation of alkyl crossover suggests that free
olefin is formed, assuming that there is no route for direct alkyl
exchange between metals. Therefore, it is likely that alkene
substitution in 3 follows a dissociative mechanism, and a three-
coordinate iron(II) hydride may be transiently formed.
Despite the low coordination number and obvious electronic
unsaturation, none of the alkyl complexes reported here have
agostic CH…Fe interactions in the solid state. The closest
Fe…H contacts in 2a, 4b and 5b are at least 2.50 Å. This
contrasts with the bisimine iron complexes reported/modelled
by Gibson and Ziegler, for which g-agostic interactions are
proposed to stabilise the active alkyl species.4,5
In summary, we have prepared 12-electron iron(II) C4
hydrocarbyl complexes, including a rare example of a tert-butyl
iron compound that was kinetically trapped using a hindered b-
diketiminate ligand. The presence of hydride intermediates
explains the kinetics of isomerisation and exchange reactions.
The less substituted complexes are favoured thermodynam-
ically, in accordance with predictions based on the polarity of
the metal–carbon bond,11 and this thermodynamic effect is a
possible explanation for the linear polyethylene from iron
catalysts.
Funding was provided by the National Science Foundation
(CHE-0134658) and the University of Rochester. We thank
Richard Eisenberg and William Jones for helpful discussions.
Notes and references
‡
Crystal data For 2a, C39H62FeN2, M = 614.76, monoclinic, space
group P21/n, a = 9.7226(5), b = 17.6389(9), c = 21.6557(10) Å, b =
96.8150(10)°, U = 3687.6(3) Å3, T = 193(2) K, Z = 4, m(Mo-Ka) = 0.436
CHEM. COMMUN., 2002, 2886–2887
2887