Light-mediated deoxygenation of alcohols with a dimeric gold catalyst

Article abstract of DOI:10.1002/ejoc.201403351

A new protocol for the reductive deoxygenation of primary alcohols was explored. This photo-mediated method combines a novel approach to bromination of alcohols merged with the powerful reducing capability of [Au₂(dppm)₂]Cl₂ [dppm = 1,1-bis(diphenylphosphino)methane] as a photoredox catalyst. The highly efficient methods discussed are marked by the use of UVA light-emitting diodes, which have significantly reduced reaction times and lowered setup cost.

Full text of DOI:10.1002/ejoc.201403351

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201403351
Light-Mediated Deoxygenation of Alcohols with a Dimeric Gold Catalyst
Terry McCallum,[a] Ekaterina Slavko,[a] Mathieu Morin,[a] and Louis Barriault*[a]
Keywords: Radical reactions / Photoredox catalysis / Gold / Deoxygenation / Bromination
A new protocol for the reductive deoxygenation of primary
[dppm = 1,1-bis(diphenylphosphino)methane] as a photo-
alcohols was explored. This photomediated method com- redox catalyst. The highly efficient methods discussed are
bines a novel approach to bromination of alcohols merged
with the powerful reducing capability of [Au2(dppm)2]Cl2
marked by the use of UVA light-emitting diodes, which have
significantly reduced reaction times and lowered setup cost.
Introduction
Radical reductive deoxygenation of alcohols has been of
great interest in organic chemistry since the pioneering
work of Barton and McCombie.[1] Since then, many func-
tional groups have been utilized in radical deoxygenation
reactions (Scheme 1), including various xanthates, phos-
phites, and benzoyl esters. However, the removal of these
groups requires the use of hazardous radical initiators (2,2Ј-
azobisisobutyronitrile, peroxides, or trialkylboroane) and
hydrogen-atom donors (organostannane reagents) or air-
unstable reagents (SmI2).[2] In search of milder conditions,
catalytic variations have been developed but still contain
potentially toxic reagents or involve multiple steps.[3] Sel-
dom are these alcohol-protection and radical-reduction re-
actions compatible with one-pot procedures.[4] Increasing
efforts have been put forward in the development of robust
waste-minimizing protocols that abate the need for stoichio-
metric toxic and hazardous reagents. Moreover, the afore-
mentioned protecting groups in Scheme 1 are restrictive by
limiting substrate precursors to secondary and tertiary
alcohols.[5] Competitive fragmentation pathways between
the desired reduction product and radical intermediates
lead to varying ratios of the desired product and byproducts
unless specific xanthates are used.[6] For this reason, most
deoxygenation protocols are limited to alcoholic precursors
that result in more stable radical intermediates, which
underscores the need for deoxygenation protocols of pri-
mary alcohols. Merging photomediated bromination with
photoredox catalysis in the reduction of the resulting alkyl
bromide eliminates competitive pathways. Herein, we report
a photomediated one-pot reductive deoxygenation protocol
that combines photochemical bromination with photoredox
catalysis.
[a] Centre for Catalysis, Research and Innovation, Department of
Chemistry, University of Ottawa,
Scheme 1. Photomediated one-pot reductive deoxygenation proto-
col. NHC = N-heterocyclic carbene, HMPA = hexamethylphos-
phoramide, MCZ = methylcarbazole, DTBP = 1,1-di-tert-butyl
peroxide, ppy = 2-phenylpyridine, Ts = para-tolylsulfonyl, LED =
light-emitting diode.
10 Marie Curie, Ottawa, ON, K1N 6N5, Canada
E-mail: lbarriau@uottawa.ca
http://mysite.science.uottawa.ca/lbarriau/index.php
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201403351.
Eur. J. Org. Chem. 2015, 81–85
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
81

T. McCallum, E. Slavko, M. Morin, L. Barriault
SHORT COMMUNICATION
Attempts at using DMF stoichiometrically or catalyti-
cally in acetonitrile showed sluggish reactivity. Dimethyl-
acetamide (DMA) was also shown to be inactive as a sol-
vent. Upon identification of the optimized conditions, a
wide range of primary alcohols was surveyed (Table 1). Ali-
phatic primary bromides 6a and 6b were obtained in excel-
lent yields. As expected, a wide variety of functionalities
were tolerated, including alkenes (see compounds 6c–f),
alkynes (see compound 6g), and the protecting groups O-
benzyl (see compounds 6h–j) and N-carbobenzyloxy (see
compound 6k). To achieve bromination with acid-sensitive
functional groups such as N-tert-butoxycarbonyl (Boc) and
O-tert-butyldimethylsilyl (TBS), the photomediated process
was performed in the presence of 2,6-lutidine. Correspond-
ing bromides 6l and 6m were produced in yields of 76 and
50%, respectively. Diols were also compatible with this
method by adding 2 equiv. of CBr4 (see compounds 6n and
6o). Interestingly, alkyl alcohol 5q containing a pyridine
was compatible with the reaction conditions to produce 6q,
albeit with a much longer irradiation time and the addition
of NaBr (3 equiv.). The photomediated halogenation
method was also extended to the formation of alkyl iodides
by using CHI3 and UVA irradiation (400 nm).[15] Reactions
with secondary alcohols were not efficient; they most often
led to the formation of formyl ester byproducts in varying
ratios.
Results and Discussion
Although most organic compounds do not absorb visible
light, chemists have developed photosensitizers that mimic
the bioinorganic capabilities of photosystem II that induce
chemical reactions.[7] Even with a multitude of photosensi-
tizers, limited examples utilize photochemical sensitization
in reductive deoxygenation reactions.[8] Seminal contri-
butions from the groups of MacMillan, Yoon, and Stephen-
son have demonstrated the effectiveness of Ru- and Ir-based
photocatalysts in the efficient formation of C–C, C–N, and
C–O bonds.[9]
The conversion of alcohols into the corresponding hal-
ides is a process commonly utilized in organic synthesis.
Although there exists several halogenation methods, the
Appel reaction (CX4 and PPh3) is among the most efficient
and reliable.[10] However, stoichiometric amounts of tri-
phenylphosphine oxide are generated as a byproduct, which
might render this transformation incompatible with metal-
catalyzed photoredox processes. Recently, Stephenson and
co-workers reported a phosphine-free visible-light-mediated
halogenation reaction of alcohols by using [Ru(bpy)3]Cl2
(bpy = 2,2Ј-bipyridyl) and polyhalomethanes in DMF.[11]
The proposed mechanism involves the formation of a
Vilsmeier–Haack reagent through the addition of electron-
·
deficient CBr3 to DMF, which activates the alcohol for a
nucleophilic displacement with a halide. In 2013, we re-
ported the reduction of unactivated alkyl and aryl bromides
by using [Au2(dppm)2]Cl2 [1, dppm = 1,1-bis(diphenylphos-
phino)methane] as a photocatalyst.[12] While exploring or-
thogonal reactivity of this catalyst, we imagined that a light-
mediated deoxygenation of alcohols could be achieved
through a one-pot radical halogenation/reduction process.
Initial attempts to convert phenylpropanol (2) directly
into propylbenzene (3) by using [Au2(dppm)2]Cl2 (1, 5 mol-
%), N,N-diisopropylethylamine (DIPEA), and CBr4 with a
UVA LED in DMF were fruitless. This led to dismal yields,
as side reactions seemed to be taking place possibly from
radical interactions with the amine base (Scheme 2).[13]
However, in a control experiment, irradiation of 3-phenyl-
propanol in the absence of the gold catalyst and DIPEA
gave the corresponding bromide in 92% yield.[14] This excit-
ing result indicated that a Vilsmeier–Haack reagent could
be generated without the use of a photocatalyst. It
prompted us to investigate a light-mediated halogenation
reaction of general applicability that could be combined
with photoredox processes.
Table 1. Photomediated bromination of primary alcohols.
[a] Yield was determined by 1H NMR spectroscopy with mesitylene
as a standard. [b] NaBr (2–3 equiv.) was added after irradiation,
and the mixture was stirred overnight in the absence of light.
[c] 2,6-Lutidine (3 equiv.) was added, and the mixture was irradi-
ated for 6 h. [d] Irradiated overnight with NaBr (3 equiv.).
Encouraged by these results, we examined the one-pot
deoxygenation reaction (Table 2). Initial experiments re-
Scheme 2. Photomediated bromination.
82
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 81–85

Light-Mediated Deoxygenation of Alcohols
vealed that the subsequent addition of the dimeric gold cat- readily converted into corresponding homoalanine 7s in
alyst to the reaction mixture containing the bromoalkane 85% yield (Table 2, entry 4). The deoxygenation of alcohols
gave the corresponding reduced product in low and variable such as 5t containing an acid-sensitive group was achieved
yield. After extensive solvent screening, we found that the in the presence of 2,6-lutidine (3 equiv.) to produce 7t in
addition of iPrOH as a co-solvent and a hydrogen donor, 50% yield (Table 2, entry 6). Tetra-O-acetylated glucose 5u
after the bromination step, was necessary to obtain high was also compatible, and desired product 5u was obtained
conversions. With the optimal reaction conditions in hand, in 70% yield (Table 2, entry 7). Aryl compounds 5v and 5w
we explored the scope of the reaction by using primary were transformed into corresponding alkanes 7v and 7w in
alcohols. 1-Heptadecanol (5a) was reduced to heptadecane yields of 70 and 84%, respectively. Interestingly, compound
(7a) in excellent yield (Table 2, entry 1). Alcohols 5j and 5r 7w led to a 3:1 mixture of an E/Z olefin. Noteworthy, the
also showed that O-benzyl (Bn) and N-benzyloxycarbonyl bromoalkyl intermediate (i.e., 6w) was isolated as a single
(Cbz) protecting groups were well tolerated in this system E isomer. As shown in Scheme 3, this method is not limited
(Table 2, entries 2 and 3). Homoserine derivative 5s was to deoxygenation processes; radical cyclizations of alcohols
5e–g containing allyl [Eqs. (1) and (2)] and propargyl
groups [Eq. (3)] afforded the corresponding cyclized N-tos-
ylpyrrolidine and piperidines 8e–g in yields ranging from 76
to 84%.
Table 2. One-pot reductive deoxygenation reaction.[a]
Scheme 3. One-pot deoxygenation/cyclization. [a] NaBr (2–
3 equiv.) was added after irradiation, and the mixture was stirred
overnight in the absence of light.
As in Nishina’s study,[16] irradiation of CBr4 promotes
the homolytic cleavage of a C–Br bond to produce an elec-
trophilic radical, ·CBr3. As proposed by Stephenson,[11] this
electrophilic radical adds to DMF to generate intermediate
9, which can be readily oxidized with CBr4 or O2 to afford
Vilsmier–Haack intermediate 10 (Scheme 4). The latter in
the presence of primary alcohol 5 produces intermediate 11,
which undergoes nucleophilic displacement with a bromide
anion to deliver corresponding bromoalkyl 6. At this point,
photoexcited dimeric gold complex 1* undergoes oxidative
[a] Reductive deoxygenation protocol: A mixture of alcohol 5 quenching by single-electron transfer (SET) with organic
(1 equiv.), CBr4 (1 equiv.), and DMF was irradiated for 0.25–1.0 h
with a LED (365 nm). Upon completion of the bromination reac-
or trialkylamines, primary radical intermediate 12 is con-
tion, as indicated by TLC, iPrOH, DIPEA (10 equiv.), and
quencher 6. In the presence of H donors [H] such as iPrOH
verted into reduced product 7 and [Au2(dppm)2]3+ is re-
[Au2(dppm)2]Cl2 (5.0 mol-%) were added, and the mixture was de-
duced to its original oxidation state by using a sacrificial
electron donor such as R3N. A control experiment in which
trialkylamine was removed for the second irradiation
showed no formation of dehalogenated product; this dem-
onstrated that the iPrOH/DMF solvent mixture did not
participate in background reactions.
gassed with argon and irradiated for 2–3 h with a LED (365 nm).
[b] Yield of isolated product. Yields given in parentheses are those
that were determined by H NMR spectroscopy with mesitylene as
a standard. [c] 2,6-Lutidine (3 equiv.) was added/NaBr (2–3 equiv.)
was added after UVA irradiation, and the mixture was stirred over-
night in the absence of light. [d] Irradiated overnight for the first
step.
1
Eur. J. Org. Chem. 2015, 81–85
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
83

T. McCallum, E. Slavko, M. Morin, L. Barriault
SHORT COMMUNICATION
[1]
[2]
D. H. R. Barton, S. W. McCombie, J. Chem. Soc. Perkin Trans.
1 1975, 1574–1585.
a) P. Girard, J. L. Namy, H. B. Kagan, J. Am. Chem. Soc. 1980,
102, 2693–2698; b) S. C. Dolan, J. MacMillan, J. Chem. Soc.,
Chem. Commun. 1985, 1588–1589; c) H. Sano, M. Ogata, T.
Migita, Chem. Lett. 1986, 15, 77–80; d) D. H. R. Barton, P.
Blundell, J. Dorchak, D. O. Jang, J. C. Jaszberenyi, Tetrahedron
1991, 47, 8969–8984; e) D. H. R. Barton, D. O. Jang, J. C. Jasz-
berenyi, Tetrahedron Lett. 1992, 33, 5709–5712; f) D. H. R.
Barton, D. O. Jang, J. C. Jaszberenyi, Tetrahedron Lett. 1992,
33, 2311–2314; g) D. H. R. Barton, D. O. Jang, J. C. Jaszber-
enyi, J. Org. Chem. 1993, 58, 6838–6842; h) K. Nishiyama, M.
Oba, M. Oshimi, T. Sugawara, R. Ueno, Tetrahedron Lett.
1993, 34, 3745–3748; i) M. Oba, K. Nishiyama, Synthesis 1994,
624–628; j) L. Zhang, M. Koreeda, J. Am. Chem. Soc. 2004,
126, 13190–13191; k) K. Lam, I. E. Markó, Org. Lett. 2008,
10, 2773–2776; l) K. Lam, I. E. Markó, Chem. Commun. 2009,
95–97; m) S. Ueng, L. Fensterbank, E. Lacîte, M. Malacria,
D. P. Curran, Org. Lett. 2010, 12, 3002–3005.
[3]
[4]
R. Lopez, D. Hays, G. Fu, J. Am. Chem. Soc. 1997, 119, 6949–
6950.
a) A. Myers, M. Movassaghi, B. Zheng, J. Am. Chem. Soc.
1997, 119, 8572–8573; b) J. D. Nguyen, B. Reiß, C. Dai, C. R. J.
Stephenson, Chem. Commun. 2013, 49, 4352–4354.
[5]
S. W. McCombie, W. B. Motherwell, M. J. Tozer, Organic Reac-
tions (Ed.: S. E. Denmark), Wiley, Hoboken, 2012, vol. 77, pp.
161–432, online.
Scheme 4. Proposed mechanism.
[6]
[7]
[8]
D. H. R. Barton, W. B. Motherwell, A. Stange, Synthesis 1981,
743–745.
V. Balzani, A. Credi, M. Venturi, ChemSusChem 2008, 1, 26–
58.
a) I. Saito, H. Ikehira, R. Kasatani, M. Watanabe, T. Matsu-
ura, J. Am. Chem. Soc. 1986, 108, 3115–3117; b) D. R. Prud-
homme, Z. Wang, C. J. Rizzo, J. Org. Chem. 1997, 62, 8257–
8260; c) A. Bordoni, R. M. de Lederkremer, C. Marino,
Carbohydr. Res. 2006, 341, 1788–1795.
Conclusions
In conclusion, we developed a photomediated reductive
deoxygenation reaction of unactivated primary alcohols. As
shown, this procedure for the bromination and deoxygen-
ation of primary alcohols is simple and functional-group
tolerant. This method was enabled by the unexpected dis-
covery of a bromination protocol for primary alcohols by
using only light, CBr4, and DMF. It was demonstrated that
CBr4 as a source of bromine was mild and compatible with
photoredox processes involving the powerful reducing abil-
ity of [Au2(dppm)2]Cl2. The high synthetic value of this
transformation was shown through the genesis of alkyl radi-
cals that yield reduction and cyclization products from
alcohol precursors. The bromination of alcohols seems to
operate through initial photolysis of a C–Br bond in CBr4
and proceeds through SN-type displacement of an activated
intermediate. Application of the method in the context of
natural product synthesis will be reported in due course.
[9]
For reviews on photoredox catalysis in organic Synthesis see:
a) J. M. R. Narayanam, C. R. J. Stephenson, Chem. Soc. Rev.
2011, 40, 102–113; b) C. K. Prier, D. A. Rankic, D. W. C. Mac-
Millan, Chem. Rev. 2013, 113, 5322–5363; c) D. M. Schultz,
T. P. Yoon, Science 2014, 343, 985-1239176-8; for dehalogena-
tion of activated bromoalkanes, see: d) J. M. R. Narayanam,
J. W. Tucker, C. R. J. Stephenson, J. Am. Chem. Soc. 2009, 131,
8756–8757; for recent examples of selected topics in photo-
redox catalysis, see: e) R. S. Andrews, J. J. Becker, M. R.
Gagné, Angew. Chem. Int. Ed. 2012, 51, 4140–4143; Angew.
Chem. 2012, 124, 4216–4219; f) M. Nakajima, Q. Lefebvre, M.
Rueping, Chem. Commun. 2014, 50, 3619–3622; g) S. P. Pitre,
C. D. McTiernan, H. Ismaili, J. C. Scaiano, ACS Catal. 2014,
4, 2530–2535; h) D. J. Wilger, J. M. Grandjean, T. R. Lammert,
D. A. Nicewicz, Nature Chem. 2014, 6, 720–726; i) G. Bergon-
zini, C. S. Schindler, C. Wallentin, E. N. Jacobsen, C. R. J. Ste-
phenson, Chem. Sci. 2014, 5, 112–116; j) L. J. Rono, H. G.
Yayla, D. Y. Wang, M. F. Armstrong, R. R. Knowles, J. Am.
Chem. Soc. 2013, 135, 17735–17738; k) J. Du, K. L. Skubi,
D. M. Schultz, T. P. Yoon, Science 2014, 344, 392–396; l) C.
Cassani, G. Bergonzini, C. Wallentin, Org. Lett. 2014, 16,
4228–4231; m) B. Sahoo, M. N. Hopkinson, F. Glorius, J. Am.
Chem. Soc. 2013, 135, 5505–5508; n) J. C. Tellis, D. N. Primer,
G. A. Molander, Science 2014, 345, 433–436; o) Z. Zuo, D. T.
Ahneman, L. Chu, J. A. Terrett, A. G. Doyle, D. W. C. Mac-
Millan, Science 2014, 345, 437–440; p) J. W. Beatty, C. R. J. Ste-
phenson, J. Am. Chem. Soc. 2014, 136, 10270–10273; q) E. Ar-
ceo, I. D. Jurberg, A. Álvarez-Fernández, P. Melchiorre, Nature
Chem. 2013, 5, 750–756.
Acknowledgments
The authors thank the Natural Sciences and Engineering Research
Council (NSERC) (for Accelerator, Discovery, and CREATE
grants to L. B.) and the University of Ottawa (for a University
Research Chair to L. B.) for support of this research. Acknowledg-
ment is also made to the donor of the American Chemical Society
Petroleum Research Fund for support of this research. T. M. and
M. M. thank NSERC (CREATE on medicinal chemistry and
biopharmaceutical development). We also thank Michel Grenier
and Prof. Tito Scaiano for insightful discussions.
[10]
[11]
R. Appel, Angew. Chem. Int. Ed. Engl. 1975, 14, 801–811; An-
gew. Chem. 1975, 87, 863–874.
C. Dai, J. M. R. Narayanam, C. R. J. Stephenson, Nature
Chem. 2011, 3, 140–145.
84
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 81–85

Light-Mediated Deoxygenation of Alcohols
[12] G. Revol, T. McCallum, M. Morin, F. Gagosz, L. Barriault,
Angew. Chem. Int. Ed. 2013, 52, 13342–13345; Angew. Chem.
2013, 125, 13584–13587.
[15]
See the Supporting Information. The reactions proceeded in
good to excellent yields. However, a lower yield was observed
for N-Boc-protected substrate 5l in spite of the presence of 2,6-
lutidine.
[13] J. C. Scaiano, M. Barra, R. Sinta, Chem. Mater. 1996, 8, 161–
166.
[16] Y. Nishina, B. Ohtani, K. Kikushima, Beilstein J. Org. Chem.
2013, 9, 1663–1667.
[14] Reaction was run on a 2 g scale, which furnished 1-bromo-3-
phenylpropane in 92% yield. Over the course of reaction opti-
mization, our group moved to a UVA LED system. Low-cost
LEDs offer consistent light coverage and small heat excess.
Received: October 15, 2014
Published Online: November 12, 2014
Eur. J. Org. Chem. 2015, 81–85
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
85