Silver(II) oxide-mediated synthesis of 2,4-diarylquinazolines

Article abstract of DOI:10.1016/j.tetlet.2018.05.019

A single-pot procedure for the synthesis of 2,4-diarylquinazolines is described which involves a silver oxide-mediated C–H activation/C–N bond formation process. The generality of this method with respect to substituent effects is presented along with studies leading to process optimization. Mechanistic investigations provide support for the involvement of radical intermediates in the reaction process.

Full text of DOI:10.1016/j.tetlet.2018.05.019

Accepted Manuscript
Silver(II) Oxide-Mediated Synthesis of 2,4-Diarylquinazolines
Wing Cheung, Raymond J. Patch, Mark R. Player
PII:
S0040-4039(18)30614-2
https://doi.org/10.1016/j.tetlet.2018.05.019
TETL 49966
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
13 April 2018
8 May 2018
10 May 2018
Please cite this article as: Cheung, W., Patch, R.J., Player, M.R., Silver(II) Oxide-Mediated Synthesis of 2,4-
Diarylquinazolines, Tetrahedron Letters (2018), doi: https://doi.org/10.1016/j.tetlet.2018.05.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
Silver(II) Oxide-Mediated Synthesis of 2,4-Diarylquinazolines
Wing Cheung, Raymond J. Patch and Mark R. Player*
Janssen Research & Development, Cardiovascular and Metabolic TA, Welch & McKean Roads,
Spring House, PA 19477, U.S.A.
Corresponding Author: Tel.: 215-628-7860
E-mail Address: mplayer@its.jnj.com
ABSTRACT: A single-pot procedure for the synthesis of 2,4-diarylquinazolines is described
which involves a silver oxide-mediated C-H activation/C-N bond formation process. The
generality of this method with respect to substituent effects is presented along with studies
leading to process optimization. Mechanistic investigations provide support for the involvement
of radical intermediates in the reaction process.
KEYWORDS: Diarylquinazoline, N-arylbenzamidine, silver oxide, radical oxidation
In support of a drug discovery research program, we were interested in developing a
focused compound library of 1,2-diarylbenzimidazoles, as exemplified by compound 1. As a
prelude to the development of this library, a retrosynthetic analysis was performed, in which a
double disconnection of the imidazole ring system was considered (Figure 1). In this case,
imidoyl chloride 3 was envisioned to undergo a nucleophilic substitution reaction with aniline 4
to generate intermediate 2, which would subsequently be cyclized under a palladium-mediated

C-H activation/C-N bond formation reaction to produce the target benzimidazole 1.1 It also
seemed reasonable to anticipate that this reaction sequence could be carried out in a single pot
operation. To test this hypothesis, anilide 5 was first converted to the corresponding imidoyl
chloride 3 by heating with thionyl chloride (Scheme 1). After removal of excess thionyl
chloride, the residue was dissolved in DMPU, treated with 4 (1.5 eq), palladium acetate (10 mol
%), potassium phosphate (3 eq), and silver (I) oxide (4 eq) as an oxidant, and heated to 120 °C
for 12 h. Following reaction work-up, benzimidazole 1 was isolated in 44% yield.
Figure 1. Retrosynthetic analysis of benzimidazole synthesis.
In an attempt to expand the substrate scope, aniline 4 was replaced with p-
methylbenzylamine 6 and subjected to the same reaction conditions. However, in this case, the
corresponding N-1-benzyl-substituted benzimidazole was not obtained; rather, 2,4-
diarylquinazoline 7 was isolated in 48% yield along with amidine adduct 8, in a ratio of 3 to 1. 2, 3
At the time this work was under investigation, there were few methodologies described for
delivering 2,4-diarylquinazolines based upon this type of bond disconnection strategy (Scheme
1).4 Movassaghi and Hill reported a closely related conversion, in which arylnitriles were reacted
with activated N-arylbenzamides to produce the corresponding substituted quinazolines in a
single-step procedure.4c More recently, Lin described a methodology in which arylamidines
were converted to quinazolines using hypervalent iodine-mediated oxidative annulation.4d Most
recently, Lv and colleagues have reported on the preparation of quinazolines from N,N’-
disubstituted amidines via I2/KI-mediated oxidative C-C bond formation under basic

conditions.4e Herein we report on a complementary one-pot approach to convert readily
available N-arylbenzamides to 2,4-diarylquinazolines via the corresponding N-arylbenzamidines
by way of an oxidative reaction with silver(II) oxide under neutral conditions.
Scheme 1. Initial syntheses of benzimidazole 1 and quinazoline 7.
With the above result in hand, subsequent experiments were conducted to optimize
reaction conditions and to examine the substrate scope. During exploratory studies, it was found
that 2,4-diarylquinazoline 7 formed even in the absence of palladium acetate, which we had
initially presumed would be necessary for inducing a C-H-activated ring closure. Moreover,
under thermal conditions in the absence of both the palladium catalyst and the silver oxidant,
only substitution adduct 8 was obtained. Potassium phosphate was similarly found to be
nonessential; thus, silver(I) oxide alone in DMPU was sufficient for effecting cyclization.
Elevation of the reaction temperature (ca. to 170 °C) did not affect product distribution, nor did it
enhance cyclization. Since it was noted that the reaction is accompanied by the formation of
water along with reduced silver (mirrored walls of the flask), the addition of molecular sieves
was investigated as a means of driving the removal of water. Indeed, the incorporation of
molecular sieves (powdered, 4Å) during the reaction (3:1 w/w relative to amide) resulted in an
increased conversion to 7 (from 48% to 76%). The use of greater amounts of molecular sieves
proved deleterious to the process, leading to reduced product recovery.
Reasoning that the cyclization of 8 may have proceeded through a radical pathway,
several oxidants known to operate via radical mechanisms were thus investigated as potential
replacements for silver(I) oxide. However, none of the inorganic metal salts studied, e.g.

Cu(OAc)2,5 CuO,6 Mn(OAc)2,7 and MnO,8 effected cyclization; in each case, only amidine adduct
8 was formed. Other oxidants such as iodobenzene diacetate, 1-fluoro-2,4,6-trimethylpyridinium
tetrafluoroborate9 and DDQ also failed to provide cyclized product 7. In addition, several silver
salts including Ag(OAc)2, AgBF4, AgNO3, Ag(OAc), PhCO2Ag, MeSO3Ag, Ag2PO4, and AgSO4
were found to be ineffective at inducing ring closure under these reaction conditions. And while
AgF and Ag(CO3)2 afforded modest yields of cyclized product 7 (21 and 52% respectively), AgO
was identified as an equally efficacious substitute for Ag2O in this reaction. Furthermore, by
increasing the stoichiometry of either silver oxide to 6 eq, complete consumption of adduct 8 was
realized, and improved isolated yields of 7 (~ 84%) were obtained.
Finally, the effects of various solvents on this process were investigated (Table 1).
Amongst the polar aprotic solvents evaluated, only dioxane provided results comparable to those
obtained in DMPU. In DMF, only marginal conversion to product 7 was obtained, whereas
DMSO proved to be incompatible with this reaction, as neither quinazoline 7 nor amine adduct 8
were detected. In other polar aprotic solvents (DME, ethyl acetate, NMP, HMPA) yields of the
target quinazoline were moderate, whereas in the protic solvent, tert-butanol, the reaction
progressed poorly. In non-polar solvents such as benzene and toluene, moderate product
conversion was realized, whereas in fluorinated solvents, product yields were improved. Thus,
in benzotrifluoride, a 68% yield of quinazoline product was obtained; in hexafluorobenzene, the
conversion was almost quantitative. This dramatically improved conversion in
hexafluorobenzene could reflect the favorable stabilization of a putative radical intermediate(s)
resulting from – stacking interactions with this solvent.10c In order to overcome potential
solubility limitations of various substrate reagents in this solvent, a co-solvent mixture with
DMPU (1:1) was used for investigating the scope of this reaction. It was determined that in this
solvent mixture, a reaction temperature of 100 °C was sufficient for effecting complete
cyclization. Thus, in a typical experiment, a mixture of the imidoyl chloride (100 mg; prepared
as described above), molecular sieves (300 mg) and DMPU (2 mL) was treated with the benzylic
amine (1.5 eq) and the resulting solution was heated to 80 °C for 0.5 h to allow for adduct
formation. After cooling the solution to rt, hexafluorobenzene (2 mL) and silver(II) oxide (6 eq)
were subsequently introduced and the resulting mixture was heated at 100 °C for 12 h. After
filtration and concentration of the filtrate, the product quinazoline was purified by silica gel
chromatography. Using these optimized reaction conditions, the substrate scope was then

explored to examine the general versatility of this approach. Because adducts derived from
meta-substituted anilides were not expected to cyclize regiospecifically, these were not included
in the anilide test set (Table 2).
Table 1. Optimization of Reaction Conditions: Solvent Screen.
Solvent
DMPU
DMF
Yield (%)
Solvent
EtOAc
DCE
Yield (%)
84
57
28
44
93
55
68
26
9
43 (incomplete)
31
NMP
Benzene
C6F6
DME
Dioxane
DMSO
HMPA
85
Toluene
F3CPh
t-BuOH
No adduct or product
73 (incomplete)
Investigation of substituent effects of the amide N-aryl ring revealed that the reaction
accommodates a range of substituents in this region of the molecule; success of the reaction is
more dependent on the steric nature of a substituent than on its electronic effects. Thus, at the
para position, halo (10-12), alkoxy (13) and alkyl (14) moieties are well tolerated (yields of
biarylquinazolines > 90%). However, with ortho-substituted anilides (entries 7 and 8), modest
decreases in product yields were obtained (15 and 16; 85% and 72%, respectively). The reaction
can also extend to heteroarylamines, as exemplified by the anilide derived from 5-amino-1,3-
dimethylpyrazole (entry 9). Reaction with benzylamine under the above conditions provided the
corresponding pyrazolopyrimidine 17 in good yield (80%).
Table 2. Substrate Scope: Aniline Ring Substituents.

Entry
Ar
Ph
Product
9
Yield (%)
1
2
3
4
5
6
7
8
9
93
86
93
91
89
92
85
72
80
4-F-Ph
10
4-Cl-Ph
11
4-Br-Ph
12
4-MeO-Ph
4-Me-Ph
13
14
2-Me-Ph
15
2-MeO-Ph
1,3-Dimethylpyrazol-5-yl
16
17
Substituent effects of the aroyl ring were next examined (Table 3). Again, while para-
substituents were well tolerated regardless of their electronic nature (18-23), ortho-substituents
negatively impacted the overall process (entries 7 and 8; 62% and 44%, respectively). In the
case of an isonicotinoyl amide (entry 9), the significant reduction in yield is the result of partial
degradation of the pyridyl ring that occurred under the reaction conditions employed for imidoyl
chloride formation. Nonetheless, a moderate yield of the corresponding quinazoline 26 (42%)
could still be obtained from this process.
Table 3. Substrate Scope: Aroyl Ring and Benzylic Amine Substituents.
Entry
R
Ar2
Ar3
Product
Yield (%)
1
2
3
H
H
H
4-MePh
4-F-Ph
4-Cl-Ph
Ph
Ph
Ph
18
19
20
91
88
93

4
H
H
4-Br-Ph
4-MeO-Ph
4-F3C-Ph
2-Me-Ph
2-Cl-Ph
Pyrid-4-yl
Ph
Ph
Ph
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
90
83
87
62
44
42
88
94
95
92
91
65
47
87
92
94
67
85
82
84
5
6
H
Ph
7
H
Ph
8
H
Ph
9
H
Ph
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Me
Me
Me
Me
Me
Me
Me
H
4-F-Ph
Ph
4-Cl-Ph
4-F3C-Ph
4-MeO-Ph
4-Me2N-Ph
4-CN-Ph
4-AcHN-Ph
4-MePh
2-F-Ph
Ph
Ph
Ph
Ph
Ph
Ph
H
Ph
H
Ph
2-Cl-Ph
2,6-di-MeO-Ph
2,5-di-Me-Ph
Naphth-1-yl
Pyrazin-2-yl
H
Ph
H
Ph
H
Ph
H
Ph
Finally, the benzylamine component could also be varied with respect to steric and
electronic effects without adversely affecting the overall process, albeit with a few notable
exceptions (Table 3). Whereas electron-donating substituents such as methoxy and
dimethylamino were well tolerated (entries 13 and 14, respectively), as were some electron-
withdrawing substituents (halo; entries 10, 11, 18, 19 and trifluoromethyl; entry 12), the reaction
proved to be less suited for the 4-cyano and 4-acetamido benzylamines (entries 15 and 16,
respectively). And while this process accommodated a single ortho substituent (entries 18, 19, 21
and 22), the presence of two (2, 6-dimethoxy; entry 20) resulted in a modest reduction in overall
yield (67%). Heterocyclic benzylic amines may also be compatible reaction partners, as
exemplified by 2-pyrazinemethanamine (entry 23), from which quinazoline 40 was obtained in
good yield (84%). It is important to point out that even in those cases where the presence of
certain substituents adversely affected the overall yield of the reaction, the process still provided
an expeditious means of generating the corresponding diarylquinazolines in practical amounts.
In order to determine if a radical mechanism was operational during this silver(II) oxide-
mediated transformation, the optimized two-step reaction sequence was carried out with

benzamidine 41 and p-methylbenzylamine (6) in the presence of radical scavengers (2 eq.), e.g.
2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO),11 1,1-diphenylethylene,12a or 2,6-di-tert-butyl-4-
methylphenol (BHT).12 The first two inhibitors significantly suppressed the formation of the
diarylquinazoline 34 (42% and 39% yields, respectively (vs. 87% in the absence of the
scavenger), while the presence of BHT completely inhibited the transformation.13 These results
are consistent with the involvement of a radical-induced oxidative cyclization mechanism, as
depicted in Figure 3. In this proposed mechanism, oxidation of benzamidine 41 could proceed
through a benzyl amidinium radical (42a),14 which in turn would be further oxidized to the
corresponding diimine (42b) with concomitant loss of water and reduction of the silver(II) to
silver(0). Ring formation could ensue by way of a 6π-electrocyclization rearrangement to afford
the dihydroquinazoline and related tautomers (42c-e), which would then undergo rapid oxidative
aromatization to provide the quinazoline product, 34.15
Figure 3. Proposed Mechanistic Pathway.
In conclusion, we have demonstrated a facile one-pot synthesis 2,4-diarylquinazolines,
starting from readily available aryl anilides and benzylic amines. Under mild oxidative reaction
conditions employing silver(II) oxide as oxidant, the reaction was found to likely proceed
through a radical mechanism. The reaction tolerates a broad range of substituents, though

sterically demanding ortho-substituents in the aroyl and benzylamine reaction components may
compromise the cyclization efficiency. Notwithstanding this limitation, the method reported
herein appears to have otherwise broad application, and provides mild, neutral reaction
conditions for accessing the 2,4-diarylquinazoline ring system. This complementary approach
may be particularly useful in cases where alternative methods may not be tolerated.
Acknowledgement
This research did not receive any specific grant from funding agencies in the public, commercial,
or not-for-profit sectors.
Supplementary data
1
Experimental details; characterization data (HRMS summary data and H NMR) of final
compounds.
References:
(1) (a) Kumar, R.K. and Punniyamurthy, T. RSC Adv. 2012, 2, 4616. (b) Sun, M.; Chen, C.;
Bao, W. RSC Adv. 2014, 4, 47373.
(2) For recent reviews of synthesis of quinazoline derivatives, see: (a) Connolly, D. J.; Cusack,
D.; O’Sullivan, T. P.; Guiry, P. J. Tetrahedron 2005, 61, 10153. (b) Michael, J. P. Nat. Prod.
Rep. 2008, 25, 166. (c) Michael, J. P. Nat. Prod. Rep. 2007, 24, 223. (c) Witt, A.; Bergman,
J. Curr. Org. Lett. 2003, 7, 659.
(3) For selected examples with alternative bond disconnections, see: (a) Zhang, Z-H.; Zhang, X-
N.; Mo, L-P.; Li, Y-X.; Ma, F-P. Green Chem. 2012, 14, 1502. (b) Yan, Y.; Wang, Z. Chem.
Comm. 2011, 47, 9513. (c) Bandaru, M.; Sabbavarapu, N. M.; Pavan, A. K. B. S.; Akula A.
K.; Durga, N. Y. V. Eur. J. Chem. 2012, 3(2), 252. (d) Zhang, J.; Zhu, D.; Yu, C.; Wan, C.;
Wang, Z. Org. Lett. 2010, 12(12), 2841. (e) Rachakonda, S.; Pratap, P. S.; Rao, M. V. B.
Synthesis 2012, 44, 2065. (f) Han, B.; Wang, C.; Han, R-F.; Yu, W.; Duan, X-Y.; Fang, R.;
Yang, X-L. Chem. Comm. 2011, 47, 7818. (g) Karnakar, K.; Kumar, A. V.; Murthy, S. N.;
Ramesh, K.; Nageswar, Y. V. D. Tetrahedron Lett. 2012, 53, 4613. (h) McGowan, M. A.;
McAvoy, C. Z.; Buchwald, S. L. Org. Lett. 2012, 14(14), 3800. (i) Flemming, J. P.; Berry,

M. B.; Brown, J. M. Org. Biomol. Chem. 2008, 6, 1215. (j) Kumar, V.; Mohan, C.; Gupta,
M.; Mahajan, M. P. Tetrahedron 2005, 61, 3533. (k) Szczepankiewicz, W.; Wagner, P.;
Danicki, M.; Suwinski, J. Tetrahedron Lett. 2003, 44, 2015.
(4) For selected examples with the same bond disconnections, see: (a) Katritzky, A. R.; Zhang,
G.; Jiang, J. J. Org. Chem. 1995, 60, 7625. (b) Bedi, P. M. S.; Kumar, V.; Mahajan, M. P.
Bioorg. Med. Chem. Lett. 2004, 14, 5211. (c) Movassaghi, M.; Hill, M. D. J. Am. Chem. Soc.
2006, 128, 14254. (d) Lin, J-P.; Zhang, F-H.; Long, Y-Q. Org. Lett. 2014, 16, 2822-2825.
(e) Lv, Z.; Wang, B.; Hu, Z.; Zhao, Y.; Chang, J. J. Org. Chem. 2016, 81, 9924-9930.
(5) Jenkins, C. L.; Kochi, J. K. J. Am. Chem. Soc. 1972, 94(3), 843.
(6) Sadana, A.; Katzer, J. R. J. Catal. 1974, 35(1), 140.
(7) Cossy, J.; Bouzide, A.; Leblanc, C. Synlett 1993, 3, 202.
(8) Ghosh, R.; Son, Y. C; Makwana, V. D.; Suib, S. L. J. Catal. 2004, 224(2), 288.
(9) Kiselyov, A. S.; Strekowski, L.; Semenov, V. V. J. Heterocycl. Chem. 1993, 30(2), 329.
(10) For stabilization of radicals by fluorinated solvents, see: (a) Grignard, B.; Jerome, C.;
Calberg, C.; Jerome, R.; Detrembleur, C. Eur. Poly J. 2008, 44(3), 861. (b) Ovalioglu, H.;
Peksoz, A.; Kirimli, H. E.; Yalciner, A. J. Dispersion Sci. and Technol. 2010, 31(3), 332.
(c) Pickett, P.D.; Tillman, E.S.; Voter, A.F. Polymer 2011, 52(1), 55.
(11) (a) Beckwith, A.L.J.; Bowry, V.W.; Ingold, K.U. J. Am. Chem. Soc. 1992, 114, 4983. (b)
Wang, X.; Ye, Y.; Zhang, S.; Feng, J.; Xu, Y.; Zhang, Y.; Wang, J. J. Am. Chem. Soc.
2011, 133, 16410.
(12) (a) Cho, S. H.; Yoon, J.; Chang, S. J. Am. Chem. Soc. 2011, 133, 5996. (b) Lifchits, O.;
Demoulin, N.; List, B. Angew. Chem., Int. Ed. 2011, 50, 9680.
(13) Though not structurally characterized, a putative BHT-benzamidine 41 adduct was detected
in the reaction mixture by LC/MS (m/e = 519.5+), along with mostly unreacted
benzamidine 41. Similarly, a putative diphenylethylene-benzamidine 41 adduct (minor
peak) was detected in its reaction mixture by LC/MS (m/e = 480.4+), along with unreacted
benzamidine 41 and quinazoline 34. However, a TEMPO-benzamidine 41 adduct could
not be detected in its reaction mixture.
(14) An analogous N-arylbenzimidamidyl radical species has recently been postulated as a
plausible reaction intermediate, generated by a silver(I) carbonate-mediated oxidation of an
N-arylbenzamidine: He, C-L; Liu, R.; Li, D-D.; Zhu, S-E.; Wang, G-W. Org. Lett. 2013,
15, 1532.

(15) The same 6-electrocyclization mechanism has been invoked in the thermal conversion of
an aza-Wittig-derived 3-benzylidinyl-1,2-diaryl-1,3-diazadiene (such as 42b) to the
corresponding dihydroquinazoline (42e). Rossi, E.; Calabrese, D.; Farma, F. Tetrahedron.
1991, 47, 5819-5834.

HIGHLIGHTS
 One-pot synthesis of 2,4-diarylquinazolines from N-arylbenzamides and benzylamines
 Proceeds under neutral reaction conditions
 Silver oxide-mediated 6-electrocyclization/oxidation process
 Broad scope with respect to N-arylbenzamides and benzylamine components