Direct generation of oxygen-stabilized radicals by H? transfer from transition metal hydrides

Article abstract of DOI:10.1021/ja511883b

Transition-metal hydrides generate α-alkoxy radicals by H? transfer to enol ethers. We have measured the rate constant for transfer from CpCr-(CO)₃H to n-butyl vinyl ether and have examined the chemistry of radicals generated by such transfers.

Full text of DOI:10.1021/ja511883b

Communication
pubs.acs.org/JACS
Direct Generation of Oxygen-Stabilized Radicals by H• Transfer from
Transition Metal Hydrides
Jonathan L. Kuo,† John Hartung,‡ Arthur Han,‡ and Jack R. Norton*,†
†Department of Chemistry, Columbia University, 3000 Broadway, New York, New York 10027, United States
‡Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States
S
* Supporting Information
α-Alkoxy radicals are primarily generated by indirect
methods.7 Curran, for example, has generated the radical in eq
ABSTRACT: Transition-metal hydrides generate α-al-
2 by a 1,5 H• transfer.8 Reiser has achieved a similar result by a
1,6 H• transfer.9 Sammis and co-workers have used oxygen-
centered radicals to accelerate both of these reactions, as part of
relay cyclizations.10
koxy radicals by H• transfer to enol ethers. We have
measured the rate constant for transfer from CpCr-
(CO)3H to n-butyl vinyl ether and have examined the
chemistry of radicals generated by such transfers. Radicals
from appropriate substrates undergo 5-exo cyclization,
with higher diastereoselectivity than the analogous all-
carbon radicals. From such radicals it is straightforward to
make substituted tetrahydrofurans.
It is easier to generate α-alkoxy radicals with special
stabilization. For example, anomeric radicals have been reported
by Giese and Rychnovsky.11 α-Alkoxy radicals can be
captodatively stabilized by an electron-accepting substituent
(such as CO2R);12 however the additional stabilization tempers
their nucleophilicity.2,12
Parsons and co-workers13 have produced α-alkoxy radicals by
the addition of thiophosphonyl radicals to enol ethers, which are
increasingly available as synthetic intermediates (eq 3).14
he stabilizing effect of π-donor substituents on carbon
T
radicals is well established. Recent theoretical work shows
that a methoxy substituent stabilizes a methyl radical by 7.43
kcal/mol, more than a carbomethoxy substituent (4.88 kcal/
mol).1 α-Alkoxy substituents also increase the nucleophilicities of
carbon-centered radicals. Such substituents have been shown to
theoretically accelerate the addition of radicals to olefins with
electron-withdrawing substituents and should thus increase the
rate of appropriate radical cyclizations.2 Furthermore, replacing
carbons with oxygens in the precursors can accelerate radical
cyclizations by 2−3 orders of magnitude.3,4
However, the literature contains few α-alkoxy radicals,
presumably because of the challenges associated with their
generation. For example, it is difficult to abstract a halogen atom
from an α-haloether because of competitive ionization.5 As a
result, only poor leaving groups are suitable for generating α-
alkoxy radicals by abstraction. Indeed, Rawal and co-workers
have shown that an α-alkoxy radical can only be generated from
an alkoxyselenide such as 1 (eq 1).6
However, this reaction requires several equiv of a phosphorus
hydride and obliges the introduction of a phosphonothioate ester
into the product.13 An alternative route to such radicals involves
the transfer of H• from a transition-metal hydride (M−H). Such
transfers are known to generate radicals from alkenes15 and
alkynes.16
We have shown that transfer of H• to α-substituted acrylate
esters can initiate radical cyclizations.17 The resulting radicals
can, in addition to cyclization, (1) transfer the H• back to M• to
regenerate 3, (2) transfer a different H• to M• to achieve an
isomerization, or (3) accept a second H• to achieve hydro-
genation; these possibilities are shown in Scheme 1. The rates of
H• transfer to phenyl-, alkyl-, and carbomethoxy-substituted
olefins have been studied in detail for CpCr(CO)3H.15g
Generally, such rates are affected by the stability of the radical
formed, though substituents on the carbon receiving the H•
retard the rate by several orders of magnitude. Transfer to the c
double bond in the acrylate substrate 3 has been shown to
compete successfully with transfer to the d double bond.
Cyclization of an enol ether such as 2 requires analogous relative
rates; i.e., transfer to a must be faster than to b.
Received: November 19, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/ja511883b
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
kH for HAT to methyl methacrylate (details on the kinetic
corrections and isotope effects can be found in the Supporting
Information).18 The rate constant for HAT suggests that enol
ethers can be used as we have previously used acrylates, as
acceptors for H•.
Scheme 1. Pathways for Radical Cyclization by M−H’s
We thus prepared enol ether 5 and treated it with the metal
hydrides our group has used in radical cyclizations: catalytic
21
CpCr(CO)3H16,17,20 or Co(dmgBF2)2(THF)2 under H2, or
stoichiometric HV(CO)4(dppe).15e We were surprised to find
that each system produced a different product from 5. The
cobaloxime system removed the vinyl substituent to give the
alcohol 7, the chromium system gave almost entirely recovered
starting material, but the vanadium reagent afforded the cyclized
product 8 in 71% yield (Scheme 2).
We have shown elsewhere that Co(dmgBF2)2(THF)2
generates an acidic species under H2,22 and that species
presumably catalyzes hydrolysis of 5 to form 7. The recovery
only of starting material with CpCr(CO)3H is less easily
explained, given that HV(CO)4dppe cyclizes 5 to 8; CpCr-
(CO)3H is generally an effective stoichiometric and catalytic
cyclization reagent.
Treating an excess of 5 with CpCr(CO)3H (7 mol %) at
elevated temperatures left most of the substrate unchanged after
several days, although we observed some hydrogenation of the f
double bond (see Scheme 3); we did not observe any
We thus needed a value for kH, the rate constant for transfer of
H• from CpCr(CO)3H to an enol ether. We selected n-butyl
vinyl ether (4) as a model substrate. When 4 is mixed with an
excess of CpCr(CO)3H in benzene at 50 °C, we do not observe
any hydrogenation. This suggests that the reverse reaction, with
rate constant ktr, is considerably faster than the hydrogenation
reaction, with rate constant khyd
.
In the presence of CpCr(CO)3D, excess 4 readily undergoes
exchange. Terminal vinyl signals appear in the 2D NMR, and the
1
Scheme 3. Reaction of 5 with Metal-Hydride Systems
hydride signal from CpCr(CO)3H grows in the H NMR. By
monitoring these changes over time we obtain the pseudo-first-
order rate constant kobs for H/D exchange.15g
As is shown in Figure 1, kobs is linear in [4], confirming that
exchange is second order overall. The slope of this line reflects
hydrogenation of the e double bond. These results were
surprising because H• transfer to olefins is generally slowed
when there are substituents on both carbons.15g (Alkyl
substituents on the carbon that receives the H• generally reduce
the rate constant for transfer by 3 orders of magnitude.) Indeed, a
crude measurement of the rate constant for H• transfer to trans-
β-methylstyrene (a model for the f double bond in 5) is only
about a tenth of our rate constant for H• transfer to n-butyl vinyl
ether (a model for the e double bond in 5).23 The rate constants
for transfer to e and f double bonds can be compared directly by
treating a substrate such as 5 with CpCr(CO)3D. Replacing the
Ph substituent with a naphthyl substituent provided better
Figure 1. Pseudo-first-order kinetics, plot of kobs vs [4], for H• transfer
at 323 K in C6D6.
only those cases in which a D• is transferred and an H• is
abstracted. Thus, it must be corrected for D• reabstraction (see
Scheme 2).15g,18 The result is kD, the rate constant for the
transfer of 2D to butyl vinyl ether. Correction by the established
19
kH/kD suggests that kH is 6.84 × 10−3 M−1 s−1, roughly half the
Scheme 2. Pathways for the Reaction of 4 with M−D
1
resolution in the H NMR. After 9 was treated with 2 equiv of
1
2
CpCr(CO)3D at 60 °C for 3 days, both H and D NMR
indicated almost three times more D incorporation into the g
double bond than into the h double bond (Figure 2).
In combination, these observations show that CpCr(CO)3H is
not effective in the cyclization of 5 because of a fast reverse
reaction, not an inability to transfer H• (Scheme 4). HAT from
CpCr(CO)3H to the f double bond of 5 (giving 5B) is slow but
leads mostly to hydrogenation of that bond, yielding 6. The
B
DOI: 10.1021/ja511883b
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
Table 1. Substrate Scope of Enol Ether Cyclization
Figure 2. Percentage deuterium incorporation after reaction with 2
equiv of CpCr(CO)3D (measured by 1H NMR).
effectiveness of HV(CO)4(dppe) lies in the weakness of its V−H
bond (57.5 kcal/mol).15e The equilibrium for the initial H•
transfer (to 5A) is more favorable for cyclization (a slower
reaction from 5A back to 5).24 Thus, the rate of H• reabstraction,
which we have previously neglected, does matter.25
Scheme 4. Reaction Pathways for 5 and M−H (M =
CpCr(CO)3 or V(CO)4dppe)
To investigate the generality of our approach we treated other
enol ethers with HV(CO)4dppe. After 3 days at 50 °C we
obtained the tetrahydrofuran derivatives (and a 2,3-dihydrofur-
an) in Table 1 in good yields. As in our previous reports,26 the
rate constant for hydrogenation, khyd, can be considered constant,
so the yields reflect the values of kcyc for the various cyclizations.26
Electron-rich aromatic substituents on C6 (e.g., 2,5-dimethox-
yphenyl, entry 3) give lower yields, an observation consistent
with the idea that alkoxy-substituted radicals are nucleophilic.
Entries 1−5 in Table 1 cyclize with high diastereoselectivity.
Similar results have been observed by both Parsons (eq 3)13 and
Sammis10 in similar cyclizations. The preference for a cis product
can be explained via the transition state proposed by Beckwith
and Houk (Scheme 5).27 The conformer with the methyl group
pseudoequatorial prevails, resulting in the cis product. An oxygen
in the backbone shortens the bond lengths (C−O bonds are
about 0.15 Å shorter than C−C bonds), further improving the
selectivity over what we obtained with acrylate substrates,
presumably due to enhanced 1,3-diaxial-type interactions.28 The
2,3-dihydrofuran prepared from a phenolic enol ether (Table 1,
entry 6) displays reduced diastereoselectivity, presumably
because the rigid backbone restricts access to the transition state.
Thus, enol ethers are an effective starting point for generating
α-alkoxy radicals, which carry out radical cyclizations with high
diastereoselectivity. However, the necessary H• transfers are
readily reversible, so such cyclizations are best done with
hydrides (such as HV(CO)4(dppe)) that contain weak M−H
bonds.
a
Yields are isolated yields upon treatment with 2.1 equiv of
HV(CO)4dppe at 50 °C for 3 days. Ratio refers to cis:trans
diastereoselectivities.
Scheme 5. Conformers of the α-Alkoxy Radical from 5
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthetic details, compound characterization, kinetic traces,
kinetic analysis, and spectra. This material is available free of
charge via the Internet at http://pubs.acs.org
C
DOI: 10.1021/ja511883b
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(b) Tang, L.; Papish, E. T.; Abramo, G. P.; Norton, J. R.; Baik, M.-H.;
AUTHOR INFORMATION
Corresponding Author
*E-mail: jrn11@columbia.edu.
Notes
■
́
Friesner, R. A.; Rappe, A. J. Am. Chem. Soc. 2006, 128, 11314.
(19) Bullock, R. M.; Samsel, E. G. J. Am. Chem. Soc. 1990, 112, 6886.
(20) Hartung, J.; Pulling, M. E.; Smith, D. M.; Yang, D. X.; Norton, J. R.
Tetrahedron 2008, 64, 11822.
(21) (a) Li, G.; Han, A.; Pulling, M. E.; Estes, D. P.; Norton, J. R. J. Am.
Chem. Soc. 2012, 134, 14662. (b) Li, G.; Estes, D. P.; Norton, J. R.;
Ruccolo, S.; Sattler, A.; Sattler, W. Inorg. Chem. 2014, 53, 10743.
(22) Estes, D. P.; Grills, D. C.; Norton, J. R. J. Am. Chem. Soc. 2014,
136, 17362.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This material is based upon work supported by the National
Science Foundation under Grant No. CHE-1401049 (IUPAC
Novel Molecular and Supramolecular Theory and Synthesis
Approaches for Sustainable Catalysis). J.L.K. acknowledges the
National Science Foundation Graduate Research Fellowship
Program. The authors are grateful to Scott A. Snyder, Deven P.
Estes, and Gang Li for helpful discussions. Kevin D. Vo is
acknowledged for the synthesis of starting materials.
(23) β-Methyl styrene gives both exchange and hydrogenation, making
a measurement of kH impractical.
(24) We have not observed H/D exchange of 4 with DV(CO)4dppe
except over tens of hours.
(25) It is unclear what factors influence the rate of H/D exchange and
the rate of hydrogenation, although α-alkoxy radicals have been shown
to hydrogenate exceptionally slowly by SnBu3H.6
(26) Han, A.; Spataru, T.; Hartung, J.; Li, G.; Norton, J. R. J. Org. Chem.
2014, 79, 1938.
(27) (a) Beckwith, A. L. J.; Schiesser, C. H. Tetrahedron 1985, 41, 3925.
(b) Spellmeyer, D. C.; Houk, K. N. J. Org. Chem. 1987, 52, 959.
(28) Sammis has studied the diastereoselectivity of these cyclizations
extensively. See ref 10 for more examples.
REFERENCES
(1) Zipse, H. Top. Curr. Chem. 2006, 263, 163.
■
(2) Fischer, H.; Radom, L. Angew. Chem., Int. Ed. 2001, 40, 1340.
(3) Beckwith, A. L. J.; Glover, S. A. Aust. J. Chem. 1987, 40, 157.
(4) Beckwith showed that the 2-oxahex-5-enyl cyclization is ca. 10
times slower than the standard hex-5-enyl cyclization. This reduction in
rate was rationalized as conformational effects derived from partial
double-bond character from spin delocalization onto oxygen.
(5) Arpin, P.; Hill, B.; Larouche-Gauthier, R.; Spino, C. Can. J. Chem.
2013, 91, 1193.
(6) (a) Rawal, V. H.; Singh, S. P.; Dufour, C.; Michoud, C. J. Org. Chem.
1993, 58, 7718. (b) Rawal, V. H.; Singh, S. P.; Dufour, C.; Michoud, C. J.
Org. Chem. 1991, 56, 5245.
(7) Lopez, J. C.; Fraser-Reid, B. J. Am. Chem. Soc. 1989, 111, 3450.
(8) (a) Curran, D. P.; Kim, D.; Liu, H. T.; Shen, W. J. Am. Chem. Soc.
1988, 110, 5900. (b) Curran, D. P.; Shen, W. J. Am. Chem. Soc. 1993,
115, 6051.
(9) Lin, H.; Schall, A.; Reiser, O. Synlett 2005, 2603.
(10) Zhu, H.; Leung, J. C. T.; Sammis, G. M. J. Org. Chem. 2014,
DOI: 10.1021/jo502499a.
(11) (a) Buckmelter, A. J.; Kim, A. I.; Rychnovsky, S. D. J. Am. Chem.
Soc. 2000, 122, 9386. (b) Rychnovsky, S. D.; Powers, J. P.; Lepage, T. J. J.
Am. Chem. Soc. 1992, 114, 8375. (c) Giese, B.; Dupuis, J. Tetrahedron
Lett. 1984, 25, 1349.
(12) (a) Burke, S. D.; Rancourt, J. J. Am. Chem. Soc. 1991, 113, 2335.
(b) Lolkema, L. D. M.; Hiemstra, H.; Alghouch, A. A.; Speckamp, W. N.
Tetrahedron Lett. 1991, 32, 1491.
(13) Jessop, C. M.; Parsons, A. F.; Routledge, A.; Irvine, D. J.
Tetrahedron Lett. 2004, 45, 5095.
(14) (a) Shade, R. E.; Hyde, A. M.; Olsen, J. C.; Merlic, C. A. J. Am.
Chem. Soc. 2010, 132, 1202. (b) Winternheimer, D. J.; Shade, R. E.;
Merlic, C. A. Synthesis 2010, 2497. (c) Okimoto, Y.; Sakaguchi, S.; Ishii,
Y. J. Am. Chem. Soc. 2002, 124, 1590. (d) Handerson, S.; Schlaf, M. Org.
Lett. 2002, 4, 407. (e) Nakamura, A.; Tokunaga, M. Tetrahedron Lett.
2008, 49, 3729. (f) Bosch, M.; Schlaf, M. J. Org. Chem. 2003, 68, 5225.
(15) (a) Sweany, R. L.; Comberrel, D. S.; Dombourian, M. F.; Peters,
N. A. J. Organomet. Chem. 1981, 216, 37. (b) Sweany, R. L.; Halpern, J. J.
Am. Chem. Soc. 1977, 99, 8335. (c) Iwasaki, K.; Wan, K. K.; Oppedisano,
A.; Crossley, S. W. M.; Shenvi, R. A. J. Am. Chem. Soc. 2014, 136, 1300.
(d) King, S. M.; Ma, X.; Herzon, S. B. J. Am. Chem. Soc. 2014, 136, 6884.
(e) Choi, J.; Pulling, M. E.; Smith, D. M.; Norton, J. R. J. Am. Chem. Soc.
2008, 130, 4250. (f) Choi, J. W.; Norton, J. R. Inorg. Chim. Acta 2008,
361, 3089. (g) Choi, J.; Tang, L.; Norton, J. R. J. Am. Chem. Soc. 2007,
129, 234. (h) Crossley, S. W.; Barabe,
2014, 136, 16788.
́
F.; Shenvi, R. A. J. Am. Chem. Soc.
(16) Estes, D. P.; Norton, J. R.; Jockusch, S.; Sattler, W. J. Am. Chem.
Soc. 2012, 134, 15512.
(17) Smith, D. M.; Pulling, M. E.; Norton, J. R. J. Am. Chem. Soc. 2007,
129, 770.
(18) (a) Tang, L.; Papish, E. T.; Abramo, G. P.; Norton, J. R.; Baik, M.-
́
H.; Friesner, R. A.; Rappe, A. J. Am. Chem. Soc. 2003, 125, 10093.
D
DOI: 10.1021/ja511883b
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX