Catalytic ketyl-olefin cyclizations enabled by proton-coupled electron transfer

Article abstract of DOI:10.1021/ja404342j

Concerted proton-coupled electron transfer is a key mechanism of substrate activation in biological redox catalysis. However, its applications in organic synthesis remain largely unexplored. Herein, we report the development of a new catalytic protocol fo

Full text of DOI:10.1021/ja404342j

Communication
pubs.acs.org/JACS
Catalytic Ketyl-Olefin Cyclizations Enabled by Proton-Coupled
Electron Transfer
Kyle T. Tarantino, Peng Liu, and Robert R. Knowles*
Department of Chemistry, Princeton University, Princeton, New Jersey 08544, United States
S
* Supporting Information
ABSTRACT: Concerted proton-coupled electron transfer
is a key mechanism of substrate activation in biological
redox catalysis. However, its applications in organic
synthesis remain largely unexplored. Herein, we report
the development of a new catalytic protocol for ketyl-
olefin coupling and present evidence to support concerted
proton-coupled electron transfer being the operative
mechanism of ketyl formation. Notably, reaction outcomes
were correctly predicted by a simple thermodynamic
formalism relating the oxidation potentials and pKa values
of specific Brønsted acid/reductant combinations to their
capacity to act jointly as a formal hydrogen atom donor.
roton-exchange reactions can exert a profound influence on
Pthe rates and thermodynamics of associated electron
transfers.1 The significance of these effects is evident in the
critical roles that proton-coupled electron transfers (PCET) play
in processes as diverse and essential as photosynthetic water
Figure 1. Energetic advantages of concerted PCET and application to
oxidation, ribonucleotide reduction, aerobic respiration, and
enzymatic C−H bond oxidation.2 Much current research in
PCET reactivity is focused on the advantages afforded by
concerted transfer pathways, wherein independent electrons and
protons are simultaneously exchanged in a single elementary
step. The energetic coupling enforced by a shared transition state
allows a favorable driving force associated with one exchange
event to compensate for unfavorable energetics in the other
without requiring the generation of an intermediate (Figure 1).3
The more favorable driving forces that result are often
complemented by diminished activation barriers,4 allowing
concerted PCET to proceed more rapidly than either competing
sequential transfer pathway.
While these kinetic advantages are now widely recognized,
concerted PCET5 has rarely been invoked as a mechanism of
substrate activation in organic synthesis.6 Yet, its potential to
function in this capacity is uncommonly broad. Many organic
functionalities exhibit large changes in pKa upon undergoing one-
electron oxidation or reduction.7 As such, coordinating a
favorable proton transfer to occur in concert with electron
transfer can enable redox reagents with potentials far less
energetic than those of their substrates to be successfully
employed. This attribute makes concerted PCET activation
particularly attractive for applications in photoredox catalysis.8
Many synthetically valuable radical classes remain inaccessible
using these technologies, as the potentials required for their
generation are unattainable using established catalyst systems, or
the necessary electron-transfer events are too slow to be
catalytic ketyl formation.
operative within the lifetime of a given catalyst excited state. In
principle, the rate accelerations afforded by concerted PCET may
provide a general solution to both of these limitations and
provide opportunities to significantly expand the scope of
substrates amenable to use in these platforms.9
Herein we demonstrate the feasibility of concerted PCET
activation in organic synthesis in the context of a new catalytic
protocol for ketyl-olefin coupling (Figure 1). Ketyls are versatile
synthetic intermediates derived from the one-electron reduction
of carbonyl compounds that play a key role in numerous
important bond-forming and bond-breaking processes.10 How-
ever, the strongly negative potentials required to generate ketyls
by electron transfer places practical limits on the scope of viable
redox partners, and to date only a handful of catalytic ketyl-olefin
coupling chemistries have been developed.11 We anticipated that
the large change in basicity (>17 pKa units in H2O)12,13 attendant
to one electron reduction would make ketones attractive
candidates for concerted PCET activation and presents an
opportunity to develop catalytic ketyl chemistries that employ
comparatively mild outer sphere reductants in combination with
a catalytic proton donor.
Received: May 1, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja404342j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Research Design. The success of these efforts was
predicated on correctly evaluating the feasibility of ketyl
formation under the action of specific combinations of one-
electron redox catalysts and Brønsted acids. We expected that the
reduction potentials of typical ketones (E1/2red = −2.48 V vs Fc
for acetophenone)14 should render them inert to all but the most
strongly reducing photocatalysts. Similarly, ketones are weakly
basic (pKa in MeCN = −0.1 for acetophenone), requiring strong
acids to generate any meaningful concentration of protio-
oxocarbenium ions.15 However, we anticipated that the joint
action of acids and reductants with pKas and potentials far
removed from these values could facilitate efficient ketyl
formation through a concerted PCET manifold.
Scheme 1. Proposed Catalytic Cycle
The selection of which acids and reductants to employ could
be assessed in quantitative terms by considering an energetic
analogy between PCET and related hydrogen atom transfer
(HAT) processes. As in HAT, the thermodynamics of a PCET
reaction can be described by the difference in bond dissociation
free energies (BDFEs) between two bonds undergoing exchange.
While no bond in the acid-reductant pair is cleaved homolyti-
cally, Mayer and co-workers recently proposed that an energy
thermodynamically equivalent to a BDFE may be calculated for
any acid/reductant pair from the pKas and redox potentials of its
constituents as well as a constant term that accounts for proton
reduction (Figure 2).7,16 This formalism, which finds its basis in
vs Fc) and the radical cation HEH+•. Concurrently, the Brønsted
acid catalyst could be expected to reversibly form a hydrogen-
bonded complex with the ketone substrate. Concerted PCET
would follow, with electron transfer from the redox catalyst
occurring concomitantly with proton transfer to the ketone
oxygen along the hydrogen-bond coordinate to generate a
neutral ketyl intermediate. This radical would add conjugately to
the pendant acrylate to form a new carbocyclic ring and an α-
carbonyl radical. Hydrogen atom transfer from HEH to this
intermediate would generate the desired closed shell product.
The oxidized HEH radical would then regenerate the active form
of the acid/reductant donor pair through electron- and proton-
transfer events with the excited state of the redox catalyst and the
conjugate base of the Brønsted acid.
Optimization and Results. In accord with our thermody-
namic analysis, visible light irradiation of a THF solution
containing ketone 1 in the presence of 2 mol % Ru(bpy)3(BArF)2
and HEH resulted in no conversion. (Table 1, entry 1). Similarly,
inclusion of Brønsted acids whose pKa values, in combination
with the Ru(I) reductant, did not furnish formal BDFEs
approaching those of the ketyl O−H bond was similarly
ineffective (Table 1, entries 2−4). However, addition of 5 mol
% diphenyl phosphoric acid (pKa in MeCN ∼ 13, ‘BDFE’ = 33
kcal/mol)21 resulted in full conversion of the ketone starting
material and produced the desired cyclization products 2 and 3 as
a 4.6:1 mixture in 78% overall yield in 4 h at room temperature
(Table 1, entry 5). Other combinations utilizing more strongly
reducing redox agents or stronger Brønsted acids were uniformly
successful (Table 1, entries 6−8). Notably, when more strongly
reducing catalysts were employed, Brønsted acids that previously
failed to activate 1 toward PCET were found to be effective, in
accord with formal BDFEs commensurate with that of the O−H
bond of the ketyl intermediate (Table 1, entry 9).22 An
evaluation of alternative stoichiometric hydrogen atom donors
established that the selectivity for the lactone product 2 could be
improved to 10:1 when 2-phenyl-dihydrobenzothaizoline (BT)
was employed in place of HEH (Table 1, entry 10). This suggests
that C−C bond formation may be reversible in these reactions
Figure 2. Thermodynamic cycle for determination of formal BDFE
values and application to ketyl PCET.
the classical bond strength calculations popularized by
Bordwell,17 allows the thermodynamics of any proposed PCET
event to be readily evaluated by comparing the formal BDFE
(‘BDFE’) of a given acid/reductant pair to the strength of the
new bond formed in the transfer event. For ketyl formation,
CBS-QB3 calculations provided a BDFE for the O−H bond in
the acetophenone ketyl of 26 kcal/mol. We reasoned that at
formal BDFE values approaching the strength of the ketyl O−H
bond, concerted PCET may become kinetically feasible, and
subsequent ketyl reactivity might be observed.18
To evaluate the correspondence between formal bond
strengths and reaction outcomes, we studied the intramolecular
cyclization of ketone 1 using the photoredox catalyst Ru-
(bpy)3(BArF)2 in the presence of various Brønsted acids. We
envisioned a catalytic cycle (Scheme 1) initiated by excitation of
the ruthenium photocatalyst with 450 nm light. The resulting
excited state (*E1/2 = 0.39 V vs Fc)19 would be reduced by
red
20
ox
Hantzsch dihydropyridine (HEH) (E1/2 = 0.51 V vs Fc) to
generate the more strongly reducing RuI(bpy)3 (E1/2ox = −1.71 V
B
dx.doi.org/10.1021/ja404342j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
c
Table 1. Optimization of the Ketyl-Olefin Coupling
Table 2. Substrate Scope of the Ketyl-Olefin Coupling
entry
acid catalyst
redox catalyst
‘BDFE’ % yield
2:3
1
2
3
4
5
6
7
8
9
none
Ru(bpy)3(BArF)2
Ru(bpy)3(BArF)2
Ru(bpy)3(BArF)2
Ru(bpy)3(BArF)2
Ru(bpy)3(BArF)2
Ru(bpy)3(BArF)2
Ir(ppy)2(dtbpy)PF6
fac-Ir(ppy)3
−
0
0
−
−
−
−
4.6:1
4.3:1
4.8:1
4.8:1
4.9:1
10:1
BzOH
45
41
35
33
27
29
24
31
33
NEt3·HBF4
lutidine·HBF4
(PhO)2PO2H
pTSA
0
0
78
74
93
92
74
89
(PhO)2PO2H
(PhO)2PO2H
lutidine·HBF4
(PhO)2PO2H
Ir(ppy)2(dtbpy)PF6
Ru(bpy)3(BArF)2
b
10
a
Yields and isomeric ratios were determined by GC analysis of crude
reaction mixtures relative to calibrated internal standards. Visible light
irradiation was provided by 26 W fluorescent lamps. Formal BDFE
values (‘BDFE’) calculated using the thermodynamic cycle presented
in Figure 2 from pKa and potential data in MeCN. For details, see
b
Supporting Information. BT used in place of HEH.
and that the diastereoselectivity is determined, at least in part, by
the relative rates of the HAT steps.
Having established effective conditions for the cyclization of 1,
we next investigated the scope and generality of this protocol. On
a preparative scale, model substrate 1 cyclized to produce 2 and 3
in a combined 73% isolated yield as an 11:1 mixture of
diastereomers (Table 2, entry 1). Incorporation of α-oxygen and
α-nitrogen substituents was well-tolerated, providing 2,3-
disubstituted tetrahydrofuran (5 and 6) and pyrrolidine products
(8 and 9) in good yields (Table 2, entries 2 and 3). Ketones 10
and 13, whose reduction potentials are nearly 900 mV more
negative than the oxidation potential of the active Ru(I) catalyst,
were also found to be viable substrates (Table 2, entries 4 and
5).13 Alkyl branching α to the carbonyl was tolerated, as
demonstrated by the successful cyclization of a tetralone-derived
substrate to produce fused polycyclic product 17 in high
diastereoselectivity (Table 2, entry 6). An ortho-substituted
acetophenone derivative 19 provided cyclized products in 96%
yield as a 2:1 mixture of diastereomers favoring the lactone
product 20 (Table 2, entry 7). The six-membered analogue 22
was also found to be a viable substrate, though it cyclized with
poor diastereoselectivity (Table 2, entry 8). Acrylonitrile 25 was
also cyclized under these conditions, furnishing γ-hydroxy nitrile
product 26 (Table 2, entry 9). Lastly, it was found that these
conditions could enable ketyl additions to styrenyl acceptors,
such as 28, resulting in the formation of dehydrated product 29
in good yield (Table 2, entry 10). Notably, for the acrylate
substrates, this protocol provides selective access to the cis-fused
bicyclic lactone products, which are generally the minor
diastereomeric products observed when these and similar
substrates are cyclized under the action of SmI2.23
a
b
1
Isolated yield of 17. Determined by H NMR analysis of the crude
c
reaction mixture. Reactions run on 0.5 mmol scale. Yields and
isomeric ratios are for isolated material following chromatography.
Visible light irradiation was provided by 26 W fluorescent lamps.
concentrations studied. Variation of the acid and ketone
concentrations in these assays revealed that the quenching
process exhibits a first-order dependence on each component.
Additionally, an isotope effect of 1.22 0.02 was observed in
quenching studies conducted with protiated and deuterated
diphenyl phosphoric acid. Collectively, these results preclude
direct electron transfer from being the mechanism of excited-
state quenching.
In principle, the observed rate law is consistent with either a
stepwise pathway involving rate-limiting proton transfer
followed by fast electron transfer to a protio-oxocarbenium ion
or a concerted proton−electron transfer to a hydrogen-bonded
Mechanism of ketyl formation. We studied the mecha-
nism of ketyl formation using fluorescence quenching techniques
and acetophenone as a model substrate.24,25 A Stern−Volmer
red
analysis revealed that acetophenone (E1/2 = −2.48 V vs Fc)
does not quench the excited state of Ir(ppy)3 (*E1/2ox = −2.11 V
vs Fc) in acetonitrile at 25 °C.26 However, inclusion of diphenyl
phosphoric acid resulted in a large decrease in the measured
fluorescence. Control experiments confirmed that the phos-
phoric acid itself does not quench the Ir(III) excited state at the
C
dx.doi.org/10.1021/ja404342j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(2) (a) Meyer, T. J.; Huynh, M. H. V.; Thorp, H. H. Angew. Chem., Int.
Ed. 2007, 46, 5284. (b) Stubbe, J.; Nocera, D. G.; Yee, C.; Chang, M.
Chem. Rev. 2003, 103, 2167. (c) Kaila, V. R. I.; Verkhovsky, M. I.;
ketone·acid complex. The former possibility can be discounted
by considering the large pKa difference in MeCN between
protonated acetophenone (pKa = −0.1)15 and the diphenyl
phosphoric acid (pKa ∼ 13).21 Taking this free energy difference
of +17.9 kcal/mol as the minimal kinetic barrier to proton
transfer necessitates a rate constant <4.6 × 10−1 M−1s−1. As such,
this stepwise mechanism is too slow to be operative within the
1.9 μs lifetime of the Ir(III) excited state.26
Wikstrom, M. Chem. Rev. 2010, 110, 7062. (d) Hatcher, E.; Soudachov,
̈
A.; Hammes-Schiffer, S. J. Am. Chem. Soc. 2004, 126, 5763.
(3) Mayer, J. M.; Rhile, I. J. Biochim. Biophys. Acta, Bioenerg. 2004, 51,
1655.
(4) Schrauben, J. N.; Cattaneo, M.; Day, T. C.; Tenderholt, A. L.;
Mayer, J. M. J. Am. Chem. Soc. 2012, 134, 16635.
As both sequential transfer mechanisms can be discounted, the
observed rate law suggests that concerted mechanisms are likely
operative in ketyl formation. Assuming a BDFE of 26 kcal/mol
for the acetophenone ketyl O−H bond, the driving force for
concerted transfer with these reagents is −1.9 kcal/mol.
Importantly, while these experiments were carried out in
MeCN to allow comparisons to established pKa and potential
values, catalytic reactions of 1 and BT under the action of
Ir(ppy)3 and diphenyl phosphoric acid in MeCN were viable,
delivering the cyclization products in 83% yield as a 5.3:1 mixture
of diastereomers.
Analogous fluorescence quenching studies are not possible in
the ruthenium-catalyzed reactions due to the more modest
reducing abilities of the RuII(bpy)3 excited state (*E1/2ox = −1.19
V vs Fc, ‘BDFE’ = 45 kcal/mol with diphenyl phosphoric acid).
However, analysis of the relevant energies required by the
stepwise pathways would suggest that concerted PCET is also
operative in the catalytic reactions. While direct electron transfer
between RuI(bpy)3 and 1 is endergonic by ∼18 kcal/mol, the
driving force for ketyl formation in the analogous PCET
mechanism is only +7.0 kcal/mol. This assumption is bolstered
by the lack of consumption of 1 under conditions known to
generate Ru(I), both in the presence and absence of Brønsted
acids of sufficient acidity to quantitatively protonate any
transiently generated ketyl radical anion (pKa ∼20 in MeCN)
(Table 1, entries 1−4).
(5) In this work, ‘concerted PCET’ is used to imply the simultaneous
transfer of electrons and protons from independent donors to a single
acceptor or the microscopic reverse of this process.
(6) (a) Rosenthal, J.; Luckett, T. D.; Hodgkiss, J. M.; Nocera, D. G. J.
Am. Chem. Soc. 2006, 128, 6546. (b) Yang, J. Y.; Nocera, D. G. J. Am.
Chem. Soc. 2007, 129, 8192.
(7) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010, 110,
6961.
(8) (a) Ischay, M. A.; Anzovino, M. E.; Du, J.; Yoon, T. P. J. Am. Chem.
Soc. 2008, 130, 12886. (b) Du, J.; Yoon, T. P. J. Am. Chem. Soc. 2009,
131, 14604. (c) Nicewicz, D. A.; MacMillan, D. W. C. Science 2008, 322,
77. (d) McNally, A.; Prier, C. K.; MacMillan, D. W. C. Science 2011, 334,
1114. (e) Narayanam, J. M. R.; Tucker, J. W.; Stephenson, C. R. J. J. Am.
Chem. Soc. 2009, 131, 8756. (f) Tucker, J. W.; Stephenson, C. R. J. J. Org.
Chem. 2012, 77, 1617. (g) Hamilton, D. S.; Nicewicz, D. A. J. Am. Chem.
Soc. 2012, 134, 18577.
(9) Yoon has reported that formic acid activates enones for use in
photoredox catalysis. In this work, a stepwise mechanism that proceeds
through an oxocarbenium intermediate was proposed: Du, J.; Espelt, L.
R.; Guzei, I. A.; Yoon, T. P. Chem. Sci. 2011, 2, 2115.
(10) (a) Edmonds, D. J.; Johnston, D.; Procter, D. J. Chem. Rev. 2004,
104, 3371. (b) Nicolaou, K. C.; Ellery, S. P.; Chen, J. S. Angew. Chem., Int.
Ed. 2009, 48, 7140.
(11) (a) Corey, E. J.; Zheng, G. Z. Tetrahedron Lett. 1997, 38, 2045.
(b) Hays, D.; Fu, G. C. J. Org. Chem. 1996, 61, 4. (c) Estev
́
ez, R. E.;
Oller-Lopez, J. L.; Robles, R.; Melgarejo, C. R.; Gansauer, A.; Cuerva, J.
́
̈
M.; Oltra, J. E. Org. Lett. 2006, 8, 5433. (d) For a recent review, see:
Streuff, J. Synthesis 2013, 45, 281.
(12) Arnett, E. M. Prog. Phys. Org. Chem. 1963, 1, 223.
(13) Hayon, E.; Ibata, T.; Lichtin, N. N.; Simic, M. J. Phys. Chem. 1972,
76, 2072.
(14) Fukuzumi, S.; Ishikawa, K.; Hironaka, K.; Tanaka, T. J. Chem. Soc.,
Perkin Trans. 1987, 2, 751.
(15) Kolthoff, I. M.; Chantooni, M. K. J. Am. Chem. Soc. 1973, 95, 8539.
(16) Waidmann, C. R.; Miller, A. J. M.; Ng, C. A.; Scheuermann, M. L.;
Porter, T. R.; Tronic, T. A.; Mayer, J. A. Energy Environ. Sci. 2012, 5,
7771.
(17) Bordwell, F. G.; Cheng, J. P.; Harrelson, J. A. J. Am. Chem. Soc.
1988, 110, 1229.
(18) Similar thermochemical approaches have been used in the
development of Ni-based catalysts for hydrogen production: DuBois, D.
L.; Bullock, R. M. Eur. J. Org. Chem. 2011, 1017.
In conclusion, we have developed a new protocol for catalytic
ketyl radical chemistry enabled by concerted proton-coupled
electron transfer. We anticipate that concerted PCET will prove
to be a general mode of catalytic activation and the elements of
reaction design described herein will prove successful in their
application to other substrate classes and transformations as well.
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures and characterization data. This material
is available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
rknowles@princeton.edu
(19) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; Von
Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85.
(20) (a) Zhu, X.; Li, H.; Li, Q.; Ai, T.; Lu, J.; Yang, Y.; Cheng, J.
Chem.Eur. J. 2003, 9, 871. (b) Fukuzumi, S.; Koumitsu, S.; Hironaka,
K.; Tanaka, T. J. Am. Chem. Soc. 1987, 109, 305.
■
Notes
The authors declare no competing financial interest.
(21) Rueping, M.; Nachtsheim, B. J.; Ieawsuwan, W.; Atodiresei, I.
Angew. Chem., Int. Ed. 2011, 50, 6706.
ACKNOWLEDGMENTS
(22) Reactions with Ir(ppy)2(dtbpy)PF6 and Ir(ppy)3 alone provide
conversion in the model reaction, but reactions with acid are
considerably faster; see SI.
(23) Enholm, E. J.; Trivellas, A. Tetrahedron Lett. 1989, 30, 1063.
(24) Fukuzumi has studied a system wherein ketones activated by
HClO4 quench the excited state of RuII(bpy)3. See refs 14 and 23.
(25) Fukuzumi, S.; Chiba, M.; Tanaka, T. Chem. Lett. 1989, 18, 31.
(26) Dixon, I. M.; Collin, J.; Sauvage, J.; Flamigni, L.; Encinas, S.;
Barigelletti, F. Chem. Soc. Rev. 2000, 29, 385.
■
Financial support was provided by Princeton University and the
ACS-PRF. We acknowledge Mark Watson for computational
assistance.
REFERENCES
■
(1) (a) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.;
Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D. G.;
Meyer, T. J. Chem. Rev. 2012, 112, 4016. (b) Mayer, J. M. Annu. Rev.
Phys. Chem. 2004, 55, 363. (c) Reece, S. Y.; Nocera, D. G. Annu. Rev.
Biochem. 2009, 78, 673.
D
dx.doi.org/10.1021/ja404342j | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX