N- versus O-arylation of aminoalcohols: Orthogonal selectivity in copper-based catalysts

Article abstract of DOI:10.1021/ja068926f

Two complementary protocols for copper-catalyzed arylation of aminoalcohols were developed. Selective N-arylation was accomplished at room temperature using 2-isobutyrylcyclohexanone (a β-diketone) as supporting ligand, while selective O-arylation require

Full text of DOI:10.1021/ja068926f

Published on Web 03/07/2007
N- versus O-Arylation of Aminoalcohols: Orthogonal Selectivity in
Copper-Based Catalysts
Alexandr Shafir, Phillip A. Lichtor, and Stephen L. Buchwald*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
Received December 13, 2006; E-mail: sbuchwal@mit.edu
Table 1. Effect of Spacer Length of N- and O-Arylation
In 2002, we disclosed a set of complementary procedures for
the copper-catalyzed N- and O-arylation of â-aminoalcohols.1 While
no added ligand was required for these Ullmann-type couplings,
attempts to expand the scope to other types of aminoalcohols
resulted in low levels of chemoselectivity. We reasoned that since
a 1,2-aminoalcohol itself can assist in the reaction by acting as a
supporting ligand an additional ligand might be required for
nonchelating substrates. Therefore, ligand-assisted Ullmann cou-
pling of aminoalcohols was investigated; the preliminary findings
of this study are disclosed in this report.
Reactionsa,b
Recently, a number of ligands were shown to expedite Ullmann
arylation of aliphatic amines.2-4 We began by examining the
coupling of 5-amino-1-pentanol with 3-iodobromobenzene using
5% CuI and 20% of diketone L1 (Scheme 1).4 After only 7 h at
room temperature, the desired N-arylated product 1a was isolated
in 97% yield, with none of the O-arylated regioisomer detectable.
Our previous work on copper-catalyzed C-O coupling using
phenanthroline or the more soluble tetramethylphenanthroline (L2)5
prompted us to see if a switch to this type of ligand might alter the
observed chemoselectivity. Indeed, the coupling of 3-iodobro-
mobenzene with 5-amino-1-pentanol in the presence of 5% CuI
and 10% L2 at 90 °C was complete after 16 h, affording the
O-arylated product 1b in 86% yield (Scheme 2).6
a Using 1.5-2.0 equiv of aminoalcohol. b Isolated yields, average of two
runs. c GC yield. d,e Ligand-free conditions; see Supporting Information.
Scheme 1. N-Arylation of 5-Amino-1-pentanol
Together, the protocols shown in Schemes 1 and 2 represent a
complementary set of conditions for copper-catalyzed N- and
O-arylation of 5-amino-1-pentanol. To test the generality of these
processes, arylation of a series of linear (C2-C6) aminoalcohols
was performed. As shown in Table 1, the catalyst based on L1
was highly effective for substrates with n g 3, affording the
N-arylated products 3a-6a in excellent yields and chemoselectivity.
Only the reaction of ethanolamine led to a sluggish reaction and a
sizable fraction of the O-arylated product. It is likely that the
formation by ethanolamine of stable fiVe-membered chelate rings7
might interfere with the normal functioning of the catalyst.
Interestingly, in the absence of added ligand, the N-arylated
ethanolamine (2a) was obtained selectively in 92% yield (40:1)
after only 14 h at room temperature.
Scheme 2. O-Arylation of 5-Amino-1-pentanol
Scheme 3. N- and O-Arylation of 4- and 3-Piperidinols
In a related series of experiments, an OH-NH2 separation of at
least four methylene units was found to be necessary to ensure selec-
tive (>15:1) O-arylation using ligand L2. Although only traces of
the N-arylated isomer were detected for these substrates, some (2-
4%) of the N-, O-diarylated byproduct was observed, along with some
reduced ArH. The reasons for N,O-diarylation in the presence of
excess aminoalcohol are currently under investigation. The coupling
of aminoalcohols with fewer methylene groups (n ) 2, 3) gave
rise to a mixture of N-, O-, and double-arylation products. Although
some success with O-arylation was achieved under ligand-free condi-
tions, an efficient O-selective method for these is yet to be realized.
The data presented in Table 1 suggest that the selectivity in
ligand-assisted reactions is only achieved for nonchelating ami-
noalcohols and that 1-amino-3-propanol represents a “borderline”
case. This hypothesis was put to the test in the arylation of 3- and
4-piperidinols. Despite being structurally related, the two differ
significantly in their metal-chelating abilities. As expected, the
coupling of the nonchelating 4-piperidinol employing ligands L1
and L2 led to the selective formation of both the N-arylated 7a
(80%, 30:1) and the O-arylated 7b (81%, 16:1) (Scheme 3A). Under
the same reaction conditions, the potentially chelating 3-piperidinol
gave rise to a mixture of products. On the other hand, submitting
a mixture of 3-piperidinol and 4-iodotoluene to the “ligand-free”
protocol developed for N-arylation of ethanolamine afforded the
9
3490
J. AM. CHEM. SOC. 2007, 129, 3490-3491
10.1021/ja068926f CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Copper-Catalyzed N- and O-Arylation of Aminoalcoholsa
a Isolated yields, average of two runs. b With L1. c Selectivity: %CN:%CO. d With L2. e Selectivity: %CO:(%CN + % double). f Balance: ArH (from
ArI) and Ar2O.
alcohol using L2 afforded the expected 17b in 88% yield (Scheme
4), attempts to perform N-arylation using L1 were not successful.
In this case, switching to a palladium catalyst afforded the
N-arylated 17a in 91% yield.9
In summary, we have disclosed two complementary copper-based
methods for the selective N- and O-arylation of aminoalcohols.
Figure 1. A working hypothesis of the mechanistic cycle.
Scheme 4. N- and O-Arylation of 4-Aminophenethyl Alcohol (Ar )
Work is currently underway to understand the exact role of ligands
and to expand the scope of these methods.
p-tolyl)a
Acknowledgment. We thank the NIH (GM058160) and the NCI
(NIH/NCI 5-T32-CA09112-30) for funding, Merck, Amgen, and
Boehringer Ingelheim for unrestricted support, and Chemetall for
the gift of Cs2CO3. We thank Timothy E. Barder for DFT
calculations.
a See Supporting Information for details.
N-arylated product 8a smoothly in 83% yield (25:1, Scheme 3B).
It was also possible to achieve a moderately selective ligand-free
O-arylation of 3-piperidinol.
Supporting Information Available: Detailed experimental pro-
cedures and characterization of products. The material is available free
of charge via the Internet at http://pubs.acs.org.
With this set of two orthogonal catalysts, the targeting of either
the N- or the O-terminus in several molecules was investigated.
As shown in Table 2, electron-rich and electron-deficient as well
as heteroaryl iodides underwent both N- and O-arylation with good
to excellent chemoselectivities.
Our current hypothesis is that the two initial steps, coordination
(1) and deprotonation (2) (Figure 1), are responsible for the observed
selectivities. These events are interdependent since the pKa of Cu-
bound nucleophiles is significantly lower than that of the free
species. For anionic L1, the lowered electrophilicity of the Cu(I)
center might disfavor the binding of alcohol and allow the high
affinity of amine to Cu(I) to dictate the outcome.8 The more Lewis
acidic (L2)Cu(I) species may lead to non-negligible concentration
of the copper-bound alcohol; facile deprotonation of such species
would favor O-selectivity. Alternatively, the (L2)Cu(I) catalyst may
require a pre-deprotonated nucleophile (3), favoring once again the
more acidic alcohol.
References
(1) Job, G. E.; Buchwald, L. Org. Lett. 2002, 4, 3703-3706.
(2) For reviews of copper-catalyzed cross-coupling, see: (a) Ley, S. V.;
Thomas, A. W. Angew. Chem., Int. Ed. 2003, 42, 5400-5449. (b) Kunz,
K.; Scholz, U.; Ganzer, D. Synlett 2003, 2428-2439. (c) Beletskaya, I.
P.; Cheprakov, A. V. Coord. Chem. ReV. 2004, 248, 2337-2364.
(3) (a) Kwong, F. Y.; Buchwald, S. L. Org. Lett. 2003, 5, 793-796. (b) Ma,
D. W.; Cai, Q.; Zhang, H. Org. Lett. 2003, 5, 2453-2455. (c) Lu, Z. K.;
Twieg, R. J.; Huang, S. P. D. Tetrahedron Lett. 2003, 44, 6289-6292.
(4) Shafir, A.; Buchwald, S. L. J. Am. Chem. Soc. 2006, 128, 8742-8743.
(5) (a) Wolter, M.; Nordmann, G.; Job, G. E.; Buchwald, S. L. Org. Lett.
2002, 4, 973-976. (b) Also see: Nordmann, G.; Buchwald, S. L. J. Am.
Chem. Soc. 2003, 125, 4978-4979.
(6) The use of activated 3 Å MS suppressed the formation of biaryl ethers.
(7) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry, 5th ed.; John
Wiley & Sons: New York, 1988.
(8) Preliminary calculations (DFT) revealed that, for the neutral â-diketonate-
Cu(I) complex, coordination of ethylamine is favored by a factor of 1.7
× 109 over the coordination of ethanol.
(9) For examples of Pd-catalyzed C-N coupling of aminoalcohols, see: (a)
Harris, M. C.; Huang, X. H.; Buchwald, S. L. Org. Lett. 2002, 4, 2885-
2888. (b) Shen, Q. L.; Shekhar, S.; Stambuli, J. P.; Hartwig, J. F. Angew.
Chem., Int. Ed. 2005, 44, 1371-1375.
Unlike the efficient coupling of aliphatic amines, N-arylation of
the less acidic aromatic amines proved to be more challenging.
Thus, while copper-catalyzed O-arylation of 4-aminophenethyl
JA068926F
9
J. AM. CHEM. SOC. VOL. 129, NO. 12, 2007 3491