Mechanistic studies of an unusual epoxide-forming elimination of a β-hydroxyalkyl rhodium porphyrin

Article abstract of DOI:10.1039/b511974j

A new and remarkably facile sp³-C-O bond forming reaction of β-hydroxyalkyl Rh porphyrins to form epoxides has been discovered and its mechanism investigated. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b511974j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Mechanistic studies of an unusual epoxide-forming elimination of a
b-hydroxyalkyl rhodium porphyrin
Yuan-Zhang Han,a Melanie S. Sanford,b Michael D. Englanda and John T. Groves*a
Received (in Berkeley, CA, USA) 22nd August 2005, Accepted 16th November 2005
First published as an Advance Article on the web 15th December 2005
DOI: 10.1039/b511974j
A new and remarkably facile sp3-C–O bond forming reaction
30 min at 25 uC, and afforded complexes 1a–c in 50–60% isolated
yield.{ The products were characterised by 1H NMR spectroscopy
of b-hydroxyalkyl Rh porphyrins to form epoxides has been
based on the dramatic upfield shifts of the alkyl proton and OH
resonances (resulting from the diamagnetic anisotropy associated
with the porphyrin ligand). For example, the a-protons of complex
1b appear at d 24.57, while the hydroxyl proton resonates at
d 23.80 in C6D6.12
discovered and its mechanism investigated.
Carbon–oxygen bond formation at transition metal centers is a
fundamental organometallic transformation that serves as the
functionalization step in many important catalytic processes. For
example, Pd-catalysed aryl etherification1 and alkene alkoxycar-
bonylation,2 Pt-3 and Pd-catalysed4 alkane oxidation, and Rh-
mediated olefin hydrofunctionalization5 all involve the formation
of carbon–oxygen bonds as a key step in the catalytic cycle.
However, despite the significance of this transformation, relatively
few C–O coupling reactions at transition metal centers have been
directly observed and studied, and most still require elevated
temperatures, extended reaction times, and/or electronically/
sterically constrained substrates.4a,6–10 Such limitations hinder the
development and scope of processes in which C–O bond formation
is critical to product release and catalyst turnover. Furthermore, a
significant and potentially important class of C–O coupling
reactions—those involving the formation of sp3-carbon–oxygen
bonds—remain exceedingly rare.8–10
Complexes 1a–c showed no propensity toward inter- or
intramolecular C–O bond formation under neutral conditions.
However, the addition of several equivalents of base resulted in
rapid intramolecular sp3-C–O coupling to yield epoxides
(Scheme 2). Under optimized reaction conditions (in C6D6 with
3 equiv. 18-Crown-6), complex 1b reacted with KOtBu to produce
1,2-epoxyhexane and K[(TPP)RhI] in quantitative yield, as
1
measured by H NMR spectroscopy.13 This reaction proceeded
under unprecedentedly mild conditions,4a,6–10 and was complete
within 5 min at room temperature.
To determine the mechanism of this transformation, the
stereospecifically deuterated substrate 1b–d was synthesized via
reaction of Na[(TPP)RhI] with trans-1,2-oxido[1-D]hexane. The
1H NMR spectra of 1b–d and 1b are compared in Fig. 1. Trace (A)
is the spectrum of complex 1b and trace (B) is the decoupled
spectrum of complex 1b obtained upon irradiation of Hc at
d 22.43. The very high-field positions of Hc and the –OH proton
(d 23.8) with respect to the highest field CH2 protons (d 21.4 and
20.9) and the gradually descending shifts of the other alkyl
resonances indicate that the conformation of 1b is as shown, with
the larger n-butyl group anti to the rhodium, similar to 1a.12 Very
similar chemical shifts have been reported for analogous gauche
b-hydroxy-N-alkyl porphyrins.11e Accordingly, the following
coupling constants in 1b can be obtained from spectra (A) and
We report herein our preliminary investigations of a series of
porphyrin RhIII b-hydroxyalkyl complexes, which undergo clean
and remarkably facile sp3-C–O bond forming reactions upon the
addition of base. Furthermore, we describe mechanistic investiga-
tions that implicate an SN2 pathway for this unusual transformation.
The nucleophilicity of square planar RhI porphyrins is well-
documented, and these complexes react cleanly with many
sterically unencumbered electrophiles via an SN2 mechanism.11
As such, a series of (TPP)RhIII [TPP = tetraphenylporphyrin]
b-hydroxyalkyl adducts was readily prepared by reaction of
Na[(TPP)RhI] with unsubstituted or mono-substituted epoxide
derivatives (Scheme 1).11a,b These reactions were complete within
2
3
3
2
(B): Jab = 9.34 Hz, Jac = 8.45 Hz, Jbc = 1.47 Hz, JRh–H
=
1
2.93 Hz. The H NMR spectrum of compound 1b–d is shown in
spectrum (C), indicating that Ha is the deuterated position in 1b–d.
This identification confirms that the formation of
b-hydroxyalkylrhodium(III) porphyrin complex 1b occurred with
inversion of stereochemistry at Ca,11 presumably via SN2 attack of
Na[(TPP)RhI] on the epoxide.
Scheme 1 Synthesis of (TPP)RhIII b-hydroxyalkyl complexes.
aDepartment of Chemistry, Princeton University, Princeton, NJ 08544
E-mail: jtgroves@princeton.edu; Fax: +1 609-258-0348;
Tel: +1 609-258-3593
bCurrent address, Department of Chemistry, University of Michigan,
Ann Arbor, MI 48109
Scheme 2 C–O bond forming reactions of complexes 1a–c.
Chem. Commun., 2006, 549–551 | 549
This journal is ß The Royal Society of Chemistry 2006

View Article Online
the spectrum of an authentic sample of 1-hexene oxide, while
spectrum B shows that of trans-1,2-oxido[1-D]hexane (containing
24% unlabelled 1-hexene oxide). Finally, spectrum C shows the
spectrum of the volatile components of the reaction mixture from
1b–d following trap-to-trap distillation.
The 1H NMR spectra of the epoxide product (C) and of trans-
1,2-oxido[1-D]hexane (B) are clearly identical, indicating that the
epoxide ring opening/ring closing sequence led to overall retention
of stereochemistry. Based on this data, it is clear that C–O
coupling proceeded with inversion of configuration at Ca,
implicating an SN2 mechanism (Scheme 3). Alternate mechanisms,
including pre-coordination of the alkoxide6,7 or radical chain
pathways11d can be ruled out, as they would involve retention or
scrambling of stereochemistry, respectively. Notably, the proposed
mechanism is consistent with conclusions of Collman et al. that a
closely related RhI/RhIII alkyl exchange reaction proceeds by an
SN2 pathway.14,15 Similar mechanisms have also been observed in
C–O bond forming reductive elimination from PtIV centers, which
serve as the product forming step in the Shilov reaction.3c By
extension, it can be inferred that the anti-Markovnikov hydro-
functionalization of olefins we have recently described for oxygen,
nitrogen, and carbon nucleophiles5 all proceed via such an SN2
displacement of
a
rhodium(I) porphyrin leaving group.
Accordingly, efforts to make this transformation truly catalytic
will need to address (i) the factors that affect the leaving group
ability of the RhI as well as (ii) the compatability of other
elementary steps of the catalytic cycle with the basic reaction
medium required for the C–O coupling process.
Fig. 1 1H NMR spectra of alkyl-rhodium(III) complex 1b. Integration of
trace C indicates ca. 60% deuteration.
The carbon–oxygen bond-forming reaction described herein is
particularly notable in light of the diverse and unique reactivity of
Rh porphyrins. For example, [Por]Rh complexes are well-known
to promote the C–H activation of methane and toluene16a as well
as to mediate highly regioselective alkene/CO insertion reactions16b
under extremely mild conditions. Developing methods to couple
such transformations (which produce [Por]Rh s-alkyl and
s-formyl products) with subsequent functionalization steps could
provide novel insights and solutions to many challenges in
catalysis, including alkane oxidation, regioselective olefin hydro-
functionalization, and the Fisher–Tropsch synthesis of hydro-
carbons. Efforts towards these applications are currently underway
in our laboratories.5
The subsequent addition of KOtBu/18-crown-6 to complex 1b
led to rapid C–O bond formation to produce the corresponding
epoxide. The cyclized product was analysed via 1H NMR
spectroscopy (Fig. 2, Spectra A, B, and C). Spectrum A shows
In conclusion, this report describes a new, base-promoted sp3-
C–O coupling reaction of b-hydroxyalkyl RhIII porphyrins.
Mechanistic investigations reveal that this transformation proceeds
via deprotonation of the b-hydroxyalkyl ligand followed by
nucleophilic displacement of K[(TPP)RhI] by an SN2 pathway.
Fig. 2 1H NMR (300 MHz) spectrum (C6D6) of (A) authentic 1-hexene
oxide, (B) authentic trans-1,2-oxido[1-D]hexane (containing 24% unla-
belled 1-hexene oxide), and (C) product obtained from treatment of
complex 1b–d with 3 equiv. KOtBu/ 3 equiv. 18-crown-6.
Scheme 3 Mechanism of reductive C–O bond formation for 1b–d.
This journal is ß The Royal Society of Chemistry 2006
550 | Chem. Commun., 2006, 549–551

View Article Online
2 J. Wolowska, G. R. Eastham, B. T. Heaton, J. A. Iggo, C. Jacob and
R. Whyman, Chem. Commun., 2002, 2084 and references therein.
3 For examples, see: (a) A. E. Shilov and G. B. Shul’pin, Activation and
Catalytic Reactions of Saturated Hydrocarbons in the Presence of Metal
Complexes, Kluwer, Dordecht, 2000; (b) R. A. Periana, D. J. Taube,
S. Gamble, H. Taube, T. Satoh and H. Fujii, Science, 1998, 280, 560; (c)
S. S. Stahl, J. A. Labinger and J. E. Bercaw, Angew. Chem., Int. Ed.,
1998, 37, 2181; (d) G. A. Luinstra, L. Wang, S. S. Stahl, J. A. Labinger
and J. E. Bercaw, J. Organomet. Chem., 1995, 504, 75.
4 (a) A. R. Dick, J. W. Kampf and M. S. Sanford, J. Am. Chem. Soc.,
2005, 127, 12790; (b) L. V. Desai, K. L. Hull and M. S. Sanford, J. Am.
Chem. Soc., 2004, 126, 9542; (c) A. R. Dick, K. L. Hull and
M. S. Sanford, J. Am. Chem. Soc., 2004, 126, 2300.
5 M. S. Sanford and J. T. Groves, Angew. Chem., 2004, 43, 588.
6 (a) Q. Shelby, N. Kataoka, G. Mann and J. Hartwig, J. Am. Chem.
Soc., 2000, 122, 10718; G. Mann, C. Incarvito, A. L. Rheingold and
J. F. Hartwig, J. Am. Chem. Soc., 1999, 121, 3224; G. Mann and
J. F. Hartwig, J. Am. Chem. Soc., 1996, 118, 13109; (b)
R. A. Widenhoefer and S. L. Buchwald, J. Am. Chem. Soc., 1998,
120, 6504; R. A. Widenhoefer, H. A. Zhong and S. L. Buchwald, J. Am.
Chem. Soc., 1997, 119, 6787.
7 Y.-S. Lin and A. Yamamoto, Organometallics, 1998, 17, 3466.
8 (a) B. S. Williams and K. I. Goldberg, J. Am. Chem. Soc., 2001, 123,
2576; B. S. Williams, A. W. Holland and K. I. Goldberg, J. Am. Chem.
Soc., 1999, 121, 252; (b) A. J. Canty and H. Jin, J. Organomet. Chem.,
1998, 565, 135; A. J. Canty, H. Jin, B. W. Skelton and A. H. White,
Inorg. Chem., 1998, 37, 3975.
9 Y.-S. Lin, S. Takeda and K. Matsumoto, Organometallics, 1999, 18,
4897.
10 Oxidatively induced C–O coupling is more common. For examples, see:
(a) K. Koo and G. L. Hillhouse, Organometallics, 1998, 17, 2924; (b)
J. E. Ba˚ckvall, Acc. Chem. Res., 1983, 16, 335.
11 (a) H. Ogoshi, J.-I. Setsune and Z.-I. Yoshida, J. Organomet. Chem.,
1980, 185, 95; (b) B. B. Wayland, S. L. Van Voorhees and K. J.
Del Rossi, J. Am. Chem. Soc., 1987, 109, 6513; (c) R. Guilard and
K. M. Kadish, Chem. Rev., 1988, 88, 1121; (d) J. P. Collman,
J. I. Brauman and A. M. Madonik, Organometallics, 1986, 5, 310; (e)
J. P. Collman, P. D. Hampton and J. I. Brauman, J. Am. Chem. Soc.,
1986, 108, 7862.
The authors acknowledge the National Science Foundation
(CHE-316301) for support of this research. Additionally, MSS
thanks the National Institutes of Health for a post-doctoral
fellowship (GM64158-02).
Notes and references
{ Preparation of b-hydroxy-alkyl(tetraphenylporphyrinato)-rhodium(III) com-
plexes (1a–c). To a solution of 10 mg RhTPPI in 10 mL of degassed
ethanol, 0.5 mL of 0.5 N aqueous NaOH solution containing 5 mg NaBH4
was added. The mixture was stirred under nitrogen at 55–65 uC for 1 h and
then cooled to 25 uC. The appropriate epoxide (2 mL) was added and the
mixture was stirred for 1 h under nitrogen. 1a–c were isolated by
preparative TLC ( silica/CH2Cl2) in 50–60% yield. 1a NMR (CDCl3): d
8.76 (s, pyr-H, 8H); 8.2 (m, o-H, 8H); 7.8 (m, m,p-H, 12H); 21.43 (d, 3H);
22.31 (m, 1H); 24.11 (br.s, -OH, 1H); 24.88 (m, 2H). FAB MS
(m-nitrobenzyl alcohol matrix; M+): m/z Found: 774 (100), 775(63),
776(22), 777(8). Calcd: 774 (100), 775(55), 776(15), 777(4); +HR FAB MS
(3-NBA/Li matrix),17 M + Li m/z Found: 781.2019, calcd: 781.2026 (288);
1b NMR (C6D6): d 8.85 (s, 8H); 8.24–8.11 (m, 8H); 7.50 (m, 12H); 0.16 (t,
3H); 0.05 (m, 2H); 20.45 (m, 2H); 20.90 (m, 1H); 21.4 (m, 1H); 22.43 (m,
2
1Hc); 23.80 (s, 1H); 24.55 (td, 1Ha) 24.60 (dm, 1Hb); Jab = 9.34 Hz,
3Jac = 8.45 Hz, 3Jbc = 1.47 Hz, 2JRh–H = 2.93 Hz. FAB MS (m-nitrobenzyl
alcohol matrix; M+): m/z Found: 816(100), 817(59), 818(42), 819(12). Calcd:
816(100), 817(59), 818(17), 819(5); +HR FAB MS (3-NBA/Li matrix), M +
H m/z Found: 817.2400, calcd: 817.2414 (21.66); 1c NMR (C6D6): d 8.79
(s, pyr-H, 8H); 8.16 (m, o-H, 4H); 8.00 (d, o9-H, 4H); 7.47–7.32 (m, m,p-H,
12H); 22.00 (q, b-H, 2H); 22.94 (t, -OH, 1H); 24.78 (dt, a-H, 2H). FAB
MS (m-nitrobenzyl alcohol matrix; M+): m/z Found: 760 (100), 761(70),
762(25), 763(10). Calcd: 760 (100), 761(53), 762(12), 763(5); +HR FAB MS
(3-NBA/Li matrix), M + Li m/z Found: 767.1847, calcd: 767.1869 (22.92).
b-Hydroxy-1-deuterio-n-hexyl(tetraphenylporphyrinato)-rhodium(III)
TPPRhCHDCH(OH)C4H9 (1b–d) was prepared in a similar fashion as 1b
from trans-1,2-oxido[1-D]hexane. The high field region of the 1H NMR
spectrum of 1b–d displayed the expected J-coupling collapse and a
characteristic upfield shift of Hb due to deuterium substitution of Ha,11d
which partially obscures the residual Hb. Integration of this region
indicated approximately 60% deuteration of the Ha site and small amounts
of impurities in the sample. NMR (C6D6): d 8.85 (s, 8H); 8.24–8.1 (m, 8H);
7.50 (m, 12H); 0.16 (t, 3H); 0.05 (m, 2H); 20.45 (m, 2H); 20.90 (m, 1H);
21.4 (m, 1H); 22.43 (m, 1H); 23.80 (s, 1H); 24.58 (m, Ha, 0.4H; Hb, 1H).
FAB MS (m-nitrobenzyl alcohol matrix; M+): m/z Found: 816(35),
817(100), 818(65), 819(26), 820(12). Calcd: 816(26), 817(100), 818(54),
819(16), 820(4). The reaction of b-hydroxylalkyl rhodium (III) porphyrins
with potassium t-butoxide. 1a–c and 1b–d (4 mg) were dissolved in 0.7 mL of
C6D6 in a dry box, 2–3 equiv. of 18-crown-6 and excess potassium
t-butoxide were added. The reactions were followed by 1H NMR. The
reaction mixtures were then transferred to 50 mL flasks, the volatile
components were isolated by bulb-to-bulb distillations (liquid N2 trap) and
12 The connectivity of complex 1a was also verified by X-ray crystal-
lography. However, the structure showed considerable disorder,
particularly, in the b-hydroxyalkyl ligand, Y.-Z. Han, PhD Thesis,
Princeton University, 1992.
1
13 The identity of the products was confirmed by comparison of their H
NMR spectra with those of authentic samples.
14 J. P. Collman, J. I. Brauman and A. M. Madonik, Organometallics,
1986, 5, 215.
15 Solvent and steric effects in related C–O coupling reactions of RhIII
porphyrins are also consistent with an SN2 mechanism (ref. 5).
16 (a) K. J. Delrossi, X. X. Xhang and B. B. Wayland, J. Organomet.
Chem., 1995, 504, 47; (b) B. B. Wayland, S. J. Ba and A. E. Sherry,
J. Am. Chem. Soc., 1991, 113, 5305; (c) R. S. Paonessa, N. C. Thomas
and J. Halpern, J. Am. Chem. Soc., 1985, 107, 4333.
1
analyzed by H NMR.
1 For recent reviews, see: (a) D. Prim, J. M. Campagne, D. Joseph and
B. Andrioletti, Tetrahedron, 2002, 58, 2041; (b) J. F. Hartwig, Acc.
Chem. Res., 1998, 31, 852.
17 Mass spectrometry was provided by the Washington University Mass
Spectrometry Resource, an NIH Research Resource (Grant No.
P41RR0954).
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 549–551 | 551