experience electrophilic substitution, a series of oxida-
tions were conducted using varied blends of AcOH and
CHCl3 as the reaction solvent (entries 2ꢀ6). Under the
chosen test conditions, an acceptable level of conver-
sion was charted using a 1:7 mixture of AcOH and
CHCl3 (entry 5). It is likely that an acid cosolvent
assists with catalyst turnover but, at a high concentra-
tion, inhibits both initial binding of Pd(II) to the
biquinolyl and subsequent intramolecular electrophilic
palladation.
With an optimal solvent composition so identified, the
oxidation was investigated on larger scales (entries 7ꢀ10).
Little was gained by increasing the reagent loading (entry
8), and monoacetate 12 could be deliberately targeted by
appropriate adjustment to the stoichiometry of PhI(OAc)2
(entry 9). Reactions on g10 mmol scale benefitted from a
longer reacton time (entry 10). Significantly, it was found
that diester 13 (a highly crystalline material) could be
isolated in pure form by direct evaporation of the reaction
solvent followed by trituration of the residue with EtOAc.
Saponification of 13 with methanolic KOH (2 h, rt) gave
azaBINOL (1) in high yield (>85%).21
azaBINOLs in good yield directly from 10 on a multigram
scale (eq 3).
In summary, a concise synthesis of azaBINOL (1) from
2-chloroaniline (in 35% overall yield) has been developed
that utilizes a Pd(II)-catalyzed directed double CꢀH func-
tionalization. The key step is notable for several reasons:
(1) an achiral (tropos) biaryl (10) is converted to a chiral
(atropos) biaryl ligand system (13/20), and as such the
reaction provides a viable platform for the study of enan-
tioselective Pd(II)-catalyzed oxidations;22 (2) the reaction
progresses via putative 6-membered palladacycles rather
than the more commonly encountered 5-membered exam-
ples; and (3) in this case, electrophilic metalation of an
electron-deficient azine occurs without resort to an activa-
tion tactic (e.g., N-oxide formation).23 In comparing the
original 7/8 step synthesis of azaBINOL (1) with the new
route described herein the improvement is marked. The
CꢀH functionalization based approach is shorter and
higher yielding overall; none of the steps involved require
special precautions or awkward experimental procedures,
and chromatography can be avoided.
It was previously demonstrated that racemic divalerate
20 is effectively resolved into (þ)-(R)-20 and (ꢀ)-(S)-1 by a
practical enzymatic hydrolysis using bovine pancreas ace-
tone powder.7c Substitution of valeric acidfor aceticacid in
the solvent blend used for Sanford oxidation was the only
change required to access this valuable precursor to scalemic
The ready availability of azaBINOL (1) (and its easily
resolved diester derivative 20) made possible by this work
will facilitate further investigation of biquinolyl templates
in a wide variety of contexts, including: organocatalysis,
metal mediated synthesis, and materials chemistry. Work
along these lines will be reported in due course.
(15) Regioselective rhodation of the C8 position of quinoline may
offer an alternative means to directly access 10 from 14, see: Kwak, J.;
Kim, M.; Chang, S. J. Am. Chem. Soc. 2011, 133, 3780–3783.
(16) For the Ni(0) mediated reductive coupling of 8-bromoquinoline
to give 8,80-biquinolyl (10), see: (a) Benito, Y.; Canoira, L.; Rodriguez,
J. G. Appl. Organomet. Chem. 1987, 1, 535–540. Biquinolyl 10 has also
been prepared via pyrrolytic deauration of 8-quinolylgold(I) complexes:
(b) Vaughan, L. G. J. Organomet. Chem. 1980, 190, C56–C58.
(17) Colon, I.; Kelsey J. Org. Chem. 1986, 51, 2627–2637.
(18) If residual NiCl2/ZnCl2 were not removed from the precipitated
10, a significant quantity of 7-chloro-8,80-biquinolyl was generated in
the Sanford oxidation step (established by XRD). Nonproductive
reduction of 19 to quinoline (14) occurs to a limited extent during its
conversion to 10. The Pd(II) catalyzed acetoxylation of samples of 10
contaminated by traces of quinoline proceeded sluggishly and failed to
give an acceptable yield of 13. Inorganic salts were removed from 10 by
its redissolution in EtOAcꢀCH2Cl2 followed by filtration/concentra-
tion; trace quinoline was removed by trituration with t-BuOMe. See
Supporting Information.
Acknowledgment. The National Science Foundation
(CHE-0722319) and Murdock Charitable Trust (2005265)
are thanked for support of the OSU NMR facility.
Supporting Information Available. All experimental
1
procedures, characterization data, and H and 13C NMR
spectra. CIF file for 7-chloro-8,80-biquinolyl. This material is
(19) A six-membered palladacycle somewhat related to 11 and
generated by cyclometallation of 2-(diphenylphosphino)-1,10-binaphthyl is
known; however, see: Huang, X.-J.; Mo, D.-L.; Ding, C.-H.; Hou, X.-K.
Synlett 2011, 943–946.
(20) Simple pyridines and quinolines are often used as directing
groups for palladation but are seldom metalated themselves (e.g., see
ref 2a). For a recent review of cyclopalladation, see: Dupont, J.;
Consorti, C. S.; Spencer, J. Chem. Rev. 2005, 105, 2527–2571.
(21) Monoacetate 12 was similarly hydrolyzed to yield 7-hydroxy-
8,80-biquinolyl, a potentially useful new tropos N,O-ligand. See Sup-
porting Information for details of both saponification reactions.
(22) Enantioselective Pd(II)-catalyzed CꢀH functionalization reac-
tions are in their infancy; for a particularly significant example by Yu
andco-workers, seeref4b. Othertypes ofPd(II)-catalyzed enantioselective
transformations have been reported, for an example, see: Jensen, D. R.;
Pugsley, J. S.; Sigman, M. S. J. Am. Chem. Soc. 2001, 123, 7475–7476.
(23) For use of N-oxide derivatives to activate azines toward metal
catalyzed CꢀH functionalization, see: (a) Lapointe, D.; Markiewicz, T.;
Whipp, C. J.; Toderian, A.; Fagnou, K. J. Org. Chem. 2011, 76, 749–759.
(b) Sun, H.-Y.; Gorelsky, S. I.; Stuart, D. R.; Campeau, L.-C.; Fagnou,
K. J. Org. Chem. 2010, 75, 8180–8189. (c) Fagnou, K. Top. Curr. Chem.
2010, 292, 35–56.
Org. Lett., Vol. 13, No. 15, 2011
4027