Transition-metal-catalyzed alkylations of amines with alkyl halides: Photoinduced, copper-catalyzed couplings of carbazoles

Article abstract of DOI:10.1002/anie.201301202

N-alkylations of carbazoles with a variety of secondary and hindered primary alkyl iodides can be achieved by using a simple precatalyst (CuI) under mild conditions (0 °C) in the presence of a Br?nsted base; at higher temperature (30 °C), secondary alkyl bromides also serve as suitable coupling partners. A Li[Cu(carbazolide)₂] complex has been crystallographically characterized, and it may serve as an intermediate in the catalytic cycle. Copyright

Full text of DOI:10.1002/anie.201301202

Angewandte
Chemie
DOI: 10.1002/anie.201301202
Homogeneous Catalysis
Transition-Metal-Catalyzed Alkylations of Amines with Alkyl Halides:
Photoinduced, Copper-Catalyzed Couplings of Carbazoles**
Alex C. Bissember, Rylan J. Lundgren, Sidney E. Creutz, Jonas C. Peters,* and Gregory C. Fu*
The development of new methods for the formation of
carbon–nitrogen bonds is a central challenge in organic
chemistry,[1] owing in part to the propensity of nitrogen-
containing compounds to exhibit interesting bioactivity.[2]
A
variety of powerful approaches have been devised, including
venerable processes such as reductive amination[3] and
Ullmann coupling,[4,5] as well as the more recently developed
Buchwald–Hartwig reaction.[6] Despite the utility of such
processes, the discovery of novel methods for the formation of
carbon–nitrogen bonds remains an important objective.[7]
Perhaps the most classic “textbook” approach to carbon–
nitrogen bond construction is the reaction of an amine with an
alkyl halide [Eq. (1)],[8] a transformation that continues to
play a critical role in organic synthesis.[9] Because this process
generally follows an SN2 pathway, elevated reaction temper-
atures are typically employed for hindered primary and
secondary electrophiles that are unactivated.[1] To our knowl-
edge, there are virtually no examples of transition-metal-
catalyzed variants of this method, which would enable
carbon–nitrogen bond formation to proceed under milder
conditions.[10]
We recently reported that Ullmann C–N couplings of
carbazole with aryl halides can be accomplished at room
temperature through photolysis (for example, [Eq. (2)]).[11]
As part of a mechanistic investigation of this process, we
determined that irradiation of a mixture of copper–carbazo-
lide complex 1 and aryl iodide 2 leads to cyclized coupling
product 3 (Scheme 1). The formation of a Csp3 ÀN bond in the
product suggested to us that it might be possible to develop
Scheme 1. Previous mechanistic study of photoinduced Ullmann cou-
plings of aryl halides: Observation of Csp3 ÀN bond formation.
a method whereby a transition metal would catalyze the
coupling of an amine with an alkyl halide. Herein, we describe
the achievement of this objective, specifically, a photoinduced,
copper-catalyzed coupling of carbazoles with alkyl iodides
that occurs under very mild conditions (08C) [Eq. (3)].
[*] Dr. A. C. Bissember, Dr. R. J. Lundgren,[+] S. E. Creutz,[+]
Prof. J. C. Peters, Prof. G. C. Fu
Division of Chemistry and Chemical Engineering
California Institute of Technology
Pasadena, CA 91125 (USA)
E-mail: jpeters@caltech.edu
gcfu@caltech.edu
[+] These authors contributed equally to this work.
[**] This work was supported by the Gordon and Betty Moore
Foundation (grant to J.C.P.), the National Science Foundation
(fellowship support for S.E.C.), and the Natural Sciences and
Engineering Research Council of Canada (fellowship support for
R.J.L.). Assistance with X-ray crystallography from Larry M. Henling
is gratefully acknowledged.
Because we had established that carbazole readily
engages in photoinduced Ullmann C–N couplings with aryl
halides,[11] it represented a natural starting point from which
to explore the proposed alkylation process. Furthermore, the
development of a mild, versatile method for functionalizing
this heterocycle through N-alkylation would represent
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201301202.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
a useful proof-of-concept, given that the carbazole subunit is
found in an array of natural and unnatural products that are of
interest in fields ranging from biology to medicinal chemistry
to materials science.[12,13]
proceeds with diminished efficiency at higher temperature
(Table 1, entry 16) or in a borosilicate reaction vessel
(entry 17). The reaction is modestly air- and moisture-
sensitive (Table 1, entries 18 and 19).
In an initial study, we determined that our reported
conditions for photoinduced, copper-catalyzed Ullmann N-
arylation [Eq. (2)] do indeed also effect N-alkylation of
carbazole by cyclohexyl iodide; however, the yield of the
desired product was modest. Nevertheless, through optimi-
zation of the appropriate reaction parameters, we were able
to develop a method that furnishes N-cyclohexylcarbazole in
good yield at 08C (Table 1, entry 1).
This metal-catalyzed N-alkylation process can be applied
to couplings of a wide range of secondary and hindered
primary alkyl iodides (Table 2), thereby complementing the
Table 2: Photoinduced, copper-catalyzed N-alkylation of carbazole:
Scope with respect to the alkyl iodide (for the reaction conditions, see
[Eq. (3)]).
Entry
Electrophile
Yield [%][a]
1
2
68
62
Table 1: Photoinduced, copper-catalyzed N-alkylation: Coupling of car-
bazole with cyclohexyl iodide at 08C.
3
4
61
70
Entry
Change from “standard” conditions
Yield [%][a]
5
6
69
80
1
2
3
none
no CuI
no hn
76
16
1
4
no CuI and no hn
1
5
6
7
8
no LiOtBu
<1
36
21
28
9
76
66
26
37
65
43
58
65
41
61
NaOtBu instead of LiOtBu
KOtBu instead of LiOtBu
Cs2CO3 instead of LiOtBu
K3PO4 instead of LiOtBu
[{(m-tol)3P}2Cu(carbazolide)] instead of CuI
Cu(OTf)2 instead of CuI
Cu powder instead of CuI
1.2 equiv of CyI and 1.2 equiv of LiOtBu
5 mol% CuI
7
8
64
61
9
10
11
12
13
14
15
16
17
18
19
All data are the average of two experiments. [a] Yield of purified product.
1 mol% CuI
SN2 reaction;[15] for each coupling (entries 2–8), in the absence
308C instead of 08C
borosilicate, instead of quartz, test tube
under air instead of under nitrogen
0.1 equiv of H2O
À
of CuI and/or light, essentially no C N bond formation is
observed (ꢀ 10%). Thus, photoinduced, copper-catalyzed
alkylations of carbazole with cyclic, including heterocyclic,
electrophiles occur in uniformly good yield at 08C (Table 2,
entries 1–4). Furthermore, acyclic secondary iodides are
suitable coupling partners (Table 2, entries 5–7). Neopentyl
iodide is a prototypical example of a poor electrophile for an
SN2 reaction,[16] yet the copper-catalyzed N-alkylation of
carbazole proceeds smoothly with this substrate (61% yield;
Table 2, entry 8).
We have also examined the scope of this metal-catalyzed
N-alkylation process with respect to the carbazole coupling
partner. Although the reaction of 1-fluorocarbazole occurs in
modest yield (Table 3, entry 1), both 2- and 3-substituted
carbazoles undergo photoinduced N-alkylation with cyclo-
heptyl iodide in fairly good yield (Table 3, entries 2–4).
Furthermore, we have explored a stereoselective copper-
catalyzed N-alkylation process. Thus, reaction of carbazole
with a trans-2-substituted cyclohexyl iodide leads to the trans
coupling product with excellent diastereoselectivity (> 20:1;
[Eq. (4)]). This highlights the potential stereochemical com-
plementarity between our photoinduced method versus
[a] The yield was determined by GC analysis versus a calibrated internal
standard (average of two runs).
Under otherwise identical conditions in the absence of
CuI, light, or both CuI and light, little or no carbon–nitrogen
bond formation is observed (Table 1, entries 2–4).[14] Simi-
larly, no coupling occurs if LiOtBu is omitted (Table 1,
entry 5), and the other Brønsted bases that we have inves-
tigated are significantly less effective (including NaOtBu,
KOtBu, Cs2CO3, and K3PO4: Table 1, entries 6–9). Among
the copper sources that we have examined, CuI is the most
useful from the standpoints of convenience, cost, and
efficiency (e.g., Table 1, entries 10–12). If less electrophile
(Table 1, entry 13) or less CuI (entries 14 and 15) is employed,
then a lower yield is obtained; nevertheless, the formation of
the C–N coupling product in 43% yield in the presence of 1%
CuI establishes that a substantial turnover number can be
achieved. This photoinduced, copper-catalyzed N-alkylation
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Table 3: Photoinduced, copper-catalyzed N-alkylation of substituted
carbazoles.
Entry
1
Carbazole
Yield [%][a]
Scheme 2. A possible pathway for photoinduced, copper-catalyzed N-
alkylation of carbazole. SET=single-electron transfer.
45
reaction conditions and may be an intermediate in the
alkylation pathway (for an outline of a possible mechanism,
see Scheme 2).[17] When we examine the electrospray mass
spectra of coupling reactions in progress, we do indeed detect
substantial quantities of ions with molecular weights of 395
and 397 (and the ratio is approximately constant), which
correspond to the masses of [63Cu(carbazolide)2]À and
[65Cu(carbazolide)2]À. We were able to synthesize Li[Cu(car-
bazolide)2] as its [Li(CH3CN)4]+ (4) and [Li([12]crown-4)2]+
(5) adducts, and we obtained an X-ray crystal structure of the
latter (Figure 1a). In the solid state, this two-coordinate
2
73
3
4
X=OMe
Et
59
67
All data are the average of two experiments. [a] Yield of purified product.
a traditional SN2 approach to N-alkylation, which would
proceed with inversion. In practice, however, our attempts to
achieve this particular coupling through an SN2 pathway
(NaH/DMF) led to complete consumption of the electrophile
À
without any appreciable C N bond formation (< 1%).
In a preliminary investigation, we have established that
this copper-catalyzed N-alkylation is not limited to alkyl
iodides as the electrophilic partner: alkyl bromides can also
be employed. For example, upon irradiation at 308C in the
presence of CuI and base, carbazole couples with cyclic and
acyclic bromides to form the desired carbon–nitrogen bond in
moderate to very good yield ([Eq. (5)] and [Eq. (6)]).
Figure 1. [Li([12]crown-4)2][Cu(carbazolide)2] (5): a) X-ray structure
(thermal ellipsoids drawn at 50% probability and H atoms omitted for
clarity). b) Excitation (emission wavelength: 410 nm) and emission
(excitation wavelength: 325 nm) spectra in CH3CN.
copper complex adopts a linear geometry (]N-Cu-N = 1788)
with a dihedral angle of 898 between the two carbazolide
planes. The excitation and emission profiles of complex 5 are
shown in Figure 1b.
[Li(CH3CN)4][Cu(carbazolide)2] (4) catalyzes the N-alky-
lation of carbazole by cyclohexyl iodide with a rate and
efficiency similar to CuI [Eq. (7)]. When copper complex 4 is
treated with cyclohexyl iodide at 08C in the absence of light,
With respect to the mechanism of the present reaction, we
hypothesize that Li[Cu(carbazolide)2] is formed under the
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
loaded onto a silica-gel column using hexanes and purified by
chromatography.
Received: February 11, 2013
Published online: && &&, &&&&
Keywords: alkylations · amine · copper ·
.
homogeneous catalysis · photochemistry
[1] a) E. Schaumann, Sci. Synth. 2008, 40a, 7 – 22; b) H. Butenschçn,
Sci. Synth. 2008, 40a, 157 – 201; c) S. A. Lawrence, Sci. Synth.
2008, 40a, 501 – 577; d) U. Scholz, B. Schlummer, Sci. Synth.
2007, 31b, 1565 – 1678.
no carbon–nitrogen bond formation occurs; on the other
hand, if this mixture is irradiated, N-cyclohexylcarbazole is
produced in 68% yield [Eq. (8)], indicating that complex 4 is
[2] For some leading references, see: The Alkaloids: Chemistry and
Biology, Vol. 70 (Ed.: H.-J. Knçlker), Elsevier, San Diego, 2011.
[3] For leading references, see: P. Margaretha, Sci. Synth, Knowl-
edge Updates 2010/01, 405 – 442.
[4] F. Ullmann, Ber. Dtsch. Chem. Ges. 1903, 36, 2382 – 2384.
[5] For recent reviews and leading references, see: a) J. D. Senra,
L. C. S. Aguiar, A. B. C. Simas, Curr. Org. Synth. 2011, 8, 53 – 78;
b) Y. Jiang, D. Ma in Catalysis without Precious Metals (Ed.:
R. M. Bullock), Wiley-Blackwell, Weinheim, 2010, pp. 213 – 233;
c) L. Penn, D. Gelman in Chemistry of Organocopper Com-
pounds, Part 2 (Ed.: Z. Rappoport, I. Marek), Wiley, Chichester,
2009, pp. 881 – 990.
[6] For example, see: a) D. S. Surry, S. L. Buchwald, Chem. Sci. 2011,
2, 27 – 50; b) J. F. Hartwig, S. Shekhar, Q. Shen, F. Barrios-
Landeros in Chemistry of Anilines, Vol. 1 (Ed.: Z. Rapaport),
Wiley, New York, 2007, pp. 455 – 536.
chemically competent for the catalytic process. Thus, the data
that we have accumulated to date are consistent with the
suggestion that Li[Cu(carbazolide)2] may be an intermediate
in this metal-catalyzed N-alkylation reaction.
[7] For examples of very recent approaches, see: a) R. P. Rucker,
A. M. Whittaker, H. Dang, G. Lalic, J. Am. Chem. Soc. 2012, 134,
6571 – 6574; b) T. J. Barker, E. R. Jarvo, Angew. Chem. 2011,
123, 8475 – 8478; Angew. Chem. Int. Ed. 2011, 50, 8325 – 8328.
[8] In introductory textbooks on organic chemistry, the alkylation of
amines with alkyl halides is often the first reaction of amines that
is described. For example, see: a) M. Loudon in Organic
Chemistry, 5th ed., Roberts & Company, Greenwood Village,
2009, chap. 23.7A; b) J. McMurry, Organic Chemistry, Brooks/
Cole, Belmont, 2012, chap. 24.7.
[9] In a recent “Analysis of the reactions used for the preparation of
drug candidate molecules” (J. S. Carey, D. Laffan, C. Thomson,
M. T. Williams, Org. Biomol. Chem. 2006, 4, 2337 – 2347),
“Heteroatom alkylation & arylation” was the reaction category
with the highest frequency use (19%; Table 2). Within this
category, “N-Substitution” was the most-often-used process
(57%; Table 10), and, within “N-Substitution”, “N-Akylation
with Alk-X” was the most frequently employed reaction (36%;
Table 11).
In summary, we have developed a transition-metal-
catalyzed variant of a classic process, the N-alkylation of an
amine with an alkyl halide. Specifically, we have established
that, upon photolysis, a range of N-alkylations of carbazoles
can be achieved with secondary and hindered primary alkyl
iodides (thereby complementing SN2 reactions of unhindered
primary electrophiles) under mild conditions (08C) with
a simple precatalyst (CuI) in the presence of a Brønsted base.
Our data are consistent with the formation of Li[Cu(carba-
zolide)2] under the coupling conditions and a possible role for
this complex in the catalytic cycle. Additional studies of
photoinduced, copper-catalyzed processes, including mecha-
À
nistic investigations (e.g., of the C N bond-forming step), are
underway.
[10] There are isolated reports of N-alkylations of amines with
unactivated alkyl halides that proceed in the presence of
a substoichiometric quantity of a transition metal. For example,
see: a) A. Aydın, I. Kaya, Electrochim. Acta 2012, 65, 104 – 114
(1658C; primary alkyl bromide); b) X. Tu, X. Fu, Q. Jiang, Z.
Liu, G. Chen, Dyes Pigm. 2011, 88, 39 – 43 (808C; primary alkyl
bromide); c) M. Kozuka, T. Tsuchida, M. Mitani, Tetrahedron
Lett. 2005, 46, 4527 – 4530 (838C; primary alkyl bromide).
[11] S. E. Creutz, K. J. Lotito, G. C. Fu, J. C. Peters, Science 2012, 338,
647 – 651.
[12] For two recent examples, see: a) T. Hirota, J. W. Lee, P. C. St.
John, M. Sawa, K. Iwaisako, T. Noguchi, P. Y. Pongsawakul, T.
Sonntag, D. K. Welsh, D. A. Brenner, F. J. Doyle III, P. G.
Schultz, S. A. Kay, Science 2012, 337, 1094 – 1097; b) H.
Uoyama, K. Goushi, K. Shizu, H. Nomura, C. Adachi, Nature
2012, 492, 234 – 238.
Experimental Section
Representative procedure: The carbazole (1.00 mmol), LiOtBu
(152 mg, 1.90 mmol), and CuI (19.5 mg, 0.10 mmol) were added to
an oven-dried 10 mL quartz test tube that contained a stir bar. The
test tube was fitted with a rubber septum, the joint was wrapped with
electrical tape, and the test tube was evacuated and backfilled with
nitrogen (3 cycles). Acetonitrile (4.0 mL) and the alkyl iodide
(1.90 mmol) were added in turn by using a syringe, and then the
test tube was detached from the nitrogen line, and the puncture holes
of the septum were covered with vacuum grease. The resulting
mixture was stirred for 5 min, and then the test tube was suspended in
an ice-filled dewar. The stirred mixture was irradiated with a 100 watt
Hg lamp, positioned directly above the dewar, for 10 h. Next, the
mixture was passed through a long plug of silica gel (eluant: hexanes;
monitored by TLC), the solvent was removed, and the residue was
[13] For a few recent reviews, see: a) occurrence, biogenesis, and
synthesis of biologically active carbazoles: A. W. Schmidt, K. R.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Reddy, H.-J. Knçlker, Chem. Rev. 2012, 112, 3193 – 3328;
b) carbazole-based polymers: J. Li, A. C. Grimsdale, Chem.
Soc. Rev. 2010, 39, 2399 – 2410.
bond formation stopped when the mercury lamp was turned off,
and resumed when the lamp was turned back on; b) Cyclohexyl
iodide is stable to photolysis in CH3CN at 08C for 10 h (> 98%
recovery); c) On a gram scale (1.15 g of product), the coupling of
carbazole with cyclohexyl iodide at 308C proceeded in 58%
yield.
À
[14] Our observation of C N bond formation upon irradiation in the
absence of CuI (entry 2 of Table 1) may be due to the photo-
physical properties of lithium carbazolide itself. For example,
see: H. W. Vos, H. H. Blom, N. H. Velthorst, C. MacLean, J.
Chem. Soc. Perkin Trans. 2 1972, 635 – 639. For the excitation and
emission spectra of lithium carbazolide in CH3CN, see the
Supporting Information.
[16] For example, see: T. H. Lowry, K. S. Richardson in Mechanism
and Theory in Organic Chemistry, HarperCollins, New York,
1987, pp. 377 – 378.
[17] We are not aware of previous investigations of [Cu(carbazo-
lide)2]À complexes. However, [Cu(anilide)2]À complexes have
been proposed as intermediates in nonphotochemical arylations
of anilines (e.g.: C.-K. Tseng, C.-R. Lee, C.-C. Han, S.-G. Shyu,
Chem. Eur. J. 2011, 17, 2716 – 2723). See also:R. Giri, J. F.
Hartwig, J. Am. Chem. Soc. 2010, 132, 15860 – 15863.
[15] a) Under our standard conditions: an unactivated tertiary alkyl
iodide and an unactivated secondary alkyl chloride were not
suitable coupling partners; a small amount of hydrodehaloge-
nation (reduction) of the electrophile was observed; CuBr and
À
CuCl furnished slightly lower yields; at partial conversion, C N
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Homogeneous Catalysis
A. C. Bissember, R. J. Lundgren,
S. E. Creutz, J. C. Peters,*
G. C. Fu*
&&&& — &&&&
Transition-Metal-Catalyzed Alkylations of
Amines with Alkyl Halides:
Photoinduced, Copper-Catalyzed
Couplings of Carbazoles
N-alkylations of carbazoles with a variety
of secondary and hindered primary alkyl
iodides can be achieved by using a simple
precatalyst (CuI) under mild conditions
(08C) in the presence of a Brønsted base;
at higher temperature (308C), secondary
alkyl bromides also serve as suitable
coupling partners. A Li[Cu(carbazolide)2]
complex has been crystallographically
characterized, and it may serve as an
intermediate in the catalytic cycle.
6
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!