Reaction of arynes with vinylogous amides: Nucleophilic addition to the ortho-quinodimethide intermediate

Article abstract of DOI:10.1021/ol4018968

The reaction of arynes with vinylogous amides containing no free N-H bonds proceeds in a [2 + 2] cycloaddition fashion at ambient temperature. The electronic properties of the vinylogous amides allow for the cycloadducts undergoing a facile ring-opening p

Full text of DOI:10.1021/ol4018968

ORGANIC
LETTERS
2013
Vol. 15, No. 17
4366–4369
Reaction of Arynes with Vinylogous
Amides: Nucleophilic Addition to the
ortho-Quinodimethide Intermediate
Ran Li, Xuemei Wang, Zhibin Wei, Chunrui Wu, and Feng Shi*
Key Laboratory of Natural Medicine and Immuno-Engineering of Henan Province,
Henan University, Jinming Campus, Kaifeng 475004, China
fshi@henu.edu.cn
Received July 6, 2013
ABSTRACT
The reaction of arynes with vinylogous amides containing no free NꢀH bonds proceeds in a [2 þ 2] cycloaddition fashion at ambient temperature.
The electronic properties of the vinylogous amides allow for the cycloadducts undergoing a facile ring-opening process, leading to electronically
biased ortho-quinodimethide intermediates. Subsequent nucleophilic addition with alcohols affords 2-substituted benzaldehydes or ketones.
In the past decade, aryne chemistry has become an
emerging area in organic synthesis.1 Among the hot topics,
those involving nucleophilic addition and pericyclic reac-
tions have been heavily studied, many of which exhibit
considerable theoretical value and/or complement conven-
tional approaches in a synthetic point of view.
Since aryne is electrophilic, efforts to seek the nucleo-
philic partner are continuing. Recently, several marked
reports have disclosed that enamides can nucleophilically
react with arynes. Depending on the structure of the
enamides, [2 þ 2] cycloaddition,2 [3 þ 2] annulation,3
[4 þ 2] annulation,4 as well as β-arylation5 can take place.
However, closely related vinylogous amides (1, Scheme 1)
have not been studied much in aryne chemistry.6 Vinylo-
gous amides have been traditionally considered a subtype
of enamides, but their CdC double bonds are more
electronically biased. We considered that the pushꢀpull
nature of this double bond should feature an easier [2 þ 2]
cycloaddition with arynes (for vinylogous amides without
free NꢀH bonds, Scheme 1, first step). Thus, unlike typical
enamides, whose [2 þ 2] cycloaddition requires 110 °C,2a,b
vinylogous amides may react with arynes under much
milder conditions. Moreover, it has been known that
the stability of the [2 þ 2] cycloadduct 3 is also heavily
dependent upon its electronic properties: the more
electron-rich the nitrogen, the easier it will undergo a
ring-opening process to afford an ortho-quinodimethide
(oQDM) (Scheme 1, second step).7,8 Mechanistically, this
can be explained as the ring-opening for the vinylogous
amide cycloadducts, compared to that for the typical
enamide cycloadducts, deviates from a strict retro-4π
(1) For recent reviews: (a) Dubrovskiy, A. V.; Markina, N. A.;
Larock, R. C. Org. Biomol. Chem. 2013, 11, 191. (b) Bhunia, A.; Yetra,
S. R.; Biju, A. T. Chem. Soc. Rev. 2012, 41, 3140. (c) Yoshida, H.;
Takaki, K. Synlett 2012, 23, 1725. (d) Tadross, P. M.; Stoltz, B. M.
Chem. Rev. 2012, 112, 3550. (e) Wu, C.; Shi, F. Asian J. Org. Chem. 2013,
2, 116.
(2) (a) Feltenberger, J. B.; Hayashi, R.; Tang, Y.; Babiash, E. S. C.;
Hsung, R. P. Org. Lett. 2009, 11, 3666. (b) Ma, Z.-X.; Feltenberger, J. B.;
Hsung, R. P. Org. Lett. 2012, 14, 2742. (c) Blackburn, T.; Ramtohul,
Y. K. Synlett 2008, 1159.
(3) Gilmore, C. D.; Allan, K. M.; Stoltz, B. M. J. Am. Chem. Soc.
2008, 130, 1558.
(4) Zhao, M.-N.; Ren, Z.-H.; Wang, Y.-Y.; Guan, Z.-H. Chem.
Commun. 2012, 8105. See also refs 2c and 3.
(5) The β-arylation was reported in ref 2c with a single example. It
was not thoroughly studied.
(6) There is a report where aryne reacts with vinylogous amides
containing free NH bonds in a β-arylation fashion. See: Ramtohul,
Y. K.; Chartrand, A. Org. Lett. 2007, 9, 1029.
(7) Oppolzer, W. Synthesis 1978, 793.
(8) For reviews of oQDMs, see: (a) Segura, J. L.; Martin, N. Chem.
Rev. 1999, 99, 3199. (b) Collier, S. J.; Storr, R. C. Heterocyclic ortho-
Quinodimethanes. In Progress in Heterocyclic Chemistry; Gribble, G. W.,
Gilchrist, T. L., Eds.; Elsevier Science: Oxford, UK, 1998; Vol. 10, pp
25ꢀ48. (c) McCullough, J. J. Acc. Chem. Res. 1980, 13, 270. (d) Fishwick,
C. W. G.; Jones, D. W. ortho-Quinonoid Compounds. In The Quinonoid
Compounds; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, UK, 1988; Vol. 1,
pp 403ꢀ453.
r
10.1021/ol4018968
Published on Web 08/19/2013
2013 American Chemical Society

Table 1. Reaction Optimizationa
Scheme 1. Proposed Reaction Mechanism and Outcome
MeOH/
entry
NR2 (1a)
equiv
conditions yieldb (%)
1
2
3
4
5
6
7
NBn2 (1a-Bn)
NBn2
0
MeCN, rt, 36 h
THF, 70 °C, 36 h
MeCN, rt, 24 h
MeCN, rt, 36 h
MeCN, rt, 24 h
MeCN, rt, 24 h
MeCN, rt, 24 h
39
28
83
52
82
66
23
0
NBn2
1.5
3
NBn2
NiPr2 (1a-Pr)
morpholino (1a-mor)
pyrrolidinyl (1a-pyr)
1.5
1.5
1.5
electrocyclic process to a lone-pair-assisted cleavage of the
C(sp3)ꢀC(sp3) bond of the four-membered ring.9 Thus the
resultant oQDM exists rather as an iminium inner salt 4.
This speaks for two things. First, the ring-opening process
for the vinylogous amide cycloadducts (R2 ¼ H or EWG)
should be much more facile and might occur at ambient
temperature (compared to 110 °C for that of enamides2a,b).
Second, the reactivity of 4 can be exploited, for example,
in a nucleophilic addition reaction leading to product 5,
thereby featuring an overall ortho-difunctionalization of
arynes. This concept has been previously attempted on
ortho-quinomethide (oQM) intermediates generated from
aryne cycloaddition/ring-opening with DMF,10 but the
scope was limited and has not yet been applied to oQDMs.
One should notice that the direct reaction ofthisexternal
nucleophile with arynes must be minimized. Thus, in
addition to the reported nucleophiles used in the DMF
system,10 we envisioned that simple aliphatic alcohols
appear a general, ideal candidate here, thanks to its low
reactivity with arynes.11 Thus, 4 may be readily quenched
by an alcohol, leading to an aminal 5, which should
eventually afford a carbonyl compound 6. Intermediate 5
is expected to be stable enough and release the highly
“arynophilic” amine only slowly so that its reactivity with
arynes was also minimized.
a All reactions were carried out on 0.3 mmol of 1a with 1.5 equiv of 2a
and 3 equiv of CsF in 4.5 mL of solvent. b Isolated yield.
we elected the morpholino group due to its broader
compatibility than the dibenzylamino group.12
With these conditions in hand, we tested the scope and
limitation of this chemistry. The reaction tolerated several
symmetrical or unsymmetrical arynes (entries 1ꢀ3, Table 2).
Aryne 2c afforded 6c as the only isolable isomer (entry 2).
The scope of vinylogous amides was reasonable. Of note
was that those substrates with aliphatic R1 groups gave
inferior yields (compare entries 4 and 5, entries 10 and 11),
presumably due to the competing deprotonation or internal
proton transfer process at the stage of 4.13 The electronic
properties of R1 proved fairly irrelevant (entries 6ꢀ8).
The electron-withdrawing group (EWG) could success-
fully cover ester (entries 1ꢀ8), tertiary amide (entry 9),
ketone (entries 10 and 11), and cyano (entry 12) groups.
Substrates with an EWG being secondary amides (with a
free NꢀH bond) did not work well (not shown).
As such, substrate 1a was first studied against different
reaction conditions (Table 1). Despite a background reac-
tion presumably caused by moisture (entries 1 and 2),
addition of 1.5 equiv of MeOH significantly improved
the yield (entry 3). Too much MeOH proved detrimental
(entry 4). The best amine moiety was dibenzylamine or
diisopropylamine (entries 3 and 5), but a morpholino
group also afforded acceptable result (entry 6). Pyrrolidinyl
wasnot a groupofchoice(entry7). Forsubsequent studies,
Scheme 2. Cyclic Substrate
Exceptions exist. For example, vinylogous amides de-
rived from acetylenedicarboxylates failed to afford pro-
duct 6 (not shown).14 The cyclic substrate 1k led to a
smooth β-arylation event in a low yield (Scheme 2).
As stated before, a reasonably high yield of 6 must mean
a slow release of the amine from adduct 5, so that the side
reactionofamine with aryne issuppressed. However, given
(9) This is suggested by a reviewer, and we thank him/her for this
comment.
(10) (a) Yoshioka, E.; Kohtani, S.; Miyabe, H. Angew. Chem., Int.
Ed. 2011, 50, 6638. (b) Yoshioka, E.; Kohtani, S.; Miyabe, H. Org. Lett.
2010, 12, 1956. (c) Yoshioka, E.; Miyabe, H. Tetrahedron 2012, 68, 179.
(d) Yoshioka, E.; Tanaka, H.; Kohtani, S.; Miyabe, H. Org. Lett. 2013,
15, 3938.
(11) Liu, Z.; Larock, R. C. J. Org. Chem. 2006, 71, 3198.
(12) For unclear reasons, most morpholino-substituted substrates
with R1 ¼ H (1, cf. Scheme 1) gave much better yields than those Bn2N-
substituted counterparts. This would also mean that the yields reported
in entries 1ꢀ3 in Table 2 may be underestimated.
(13) Unforunately, to date, we have been unsuccessful in preparing
the substrate with R1 being a tert-butyl group.
(14) The identity of the product is under investigation and will be
disclosed in due course.
Org. Lett., Vol. 15, No. 17, 2013
4367

Table 2. Reaction Scopea
a Reactions conditions: 1 (0.3 mmol), 2 (0.45 mmol), CsF (0.9 mmol), and MeOH (0.45 mmol) in 4.5 mL of MeCN. Reactions may not be individually
optimized. b Isolated yield. c Reaction took 12 h.
the low concentration of the aryne, we considered that in
the presence of a second strong electrophile, the released
amine would have less chance to react with arynes. Since a
vinylogous amide can be generated in situ from a propio-
late and an amine, we reasoned that a propiolateꢀamineꢀ
aryne system may constitute a three-component variant of
this chemistry, where amine may be used in catalytic
amount (Scheme 3).
Ideally, in this scenario, amine would first react with
propiolate to form the vinylogous amide 1, which should
subsequently react with aryne, as described in Scheme 1, to
generate 4. Nucleophilic addition then affords 5, which
slowly releases the desired product 6 while regenerating the
amine. We tested this possibility (Table 3) using a model
reaction containing ethyl propiolate, benzyne precursor
2a, and an alcohol as the nucleophile in the presence of a
secondary amine in catalytic quantity. We incorporated 2
equivofpropiolate toprevent the amine from reacting with
benzyne and used 2a as the limiting reagent. Although a
synthetically useful yield has not been realized, the concept
proved amendable. Up to 4 turnovers of amine could be
(15) Increasing the quantity of amine did not significantly improve
the overall yield. For additional results, see the Supporting Information.
4368
Org. Lett., Vol. 15, No. 17, 2013

Table 3. Catalytic Protocola
Scheme 3. Proposed Catalytic Variant
entry
HNR2
ROH
yieldb (%)
1
2
3
4
5
6
7
HNBn2
morpholine
TMPc
HNiPr2
HNiPr2
HNiPr2
HNiPr2
MeOH
MeOH
MeOH
MeOH
EtOH
26
38
trace
40
36
iPrOH
tBuOH
40
realized.15 The nature of alcohol was not critical, as
trace
i
MeOH, EtOH, or PrOH afforded the same yield and
a All reactions were carried out on 0.3 mmol of 2a in 4.5 mL of
MeCN. Amine and propiolate were mixed first in solvent, and all other
reagents were added after 1 h. b Isolated yield. c TMP = 2,2,6,6-
tetramethylpiperidine.
t
turnover. Only tertiary alcohols such as BuOH were
problematic, likely due to its lowered nucleophilicity.
Currently, we tend to believe that the low turnover number
might result from the undesirable kinetics of some steps in
this catalytic cycle,16 as well as the loss of the amine for side
reactions.
In summary, our preliminary results have demonstrated
that vinylogous amides react with arynes in a [2 þ 2]
cycloaddition/ring-opening fashion to furnish an electro-
nically biased oQDM intermediate (rather as an iminium
inner salt). The electronic properties of vinylogous amides
not only allow for this process taking place at ambient
temperature but also render this oQDM intermediate the
opportunity for being trapped by a nucleophile to achieve
ortho-difunctionalization of arynes.17 Although the syn-
theticutilities of the oQDM intermediate have to be further
exploited,18,19 we believe that this preliminary study
complements the aryne reactivity with enamides and sup-
ports that fine-tuning of a nucleophile may result in
significantly altered mechanism, reaction conditions, and
final product in aryne chemistry. Further studies to expand
the scope are underway.
Acknowledgment. This project was financially supported
by the Program for Science & Technology Innovation Talents
in Universities of Henan Province (No. 13HASTIT010)
and National Natural Science Foundation of China (No.
21002021). We thank Dr. Jiang Zhou (Peking University)
for his help in the spectroscopic analysis.
Supporting Information Available. Additional results
for the cataytic protocol, experimental details, and char-
(16) The narrow scope for the catalytic protocol also comes from the
undesirable kinetics of the first step. For example, many electron-poor
alkynes react with secondary amines only at elevated temperature, which
is not tolerated for the subsequent reaction with arynes.
1
acterization of the products, including full H and 13C
NMR spectra. This material is available free of charge via
the Internet at http://pubs.acs.org.
(17) A related process offering a similar outcome but a completely
different mechanism is the aryne reaction with β-dicarbonyl compounds.
See: (a) Tambar, U. K.; Stoltz, B. M. J. Am. Chem. Soc. 2005, 127, 5340.
(b) Yoshida, H.; Watanabe, M.; Ohshita, J.; Kunai, A. Chem. Commun.
2005, 3292. Since some vinylogous amides can be prepared from
β-dicarbonyl compounds, some of our products (e.g., 6e, 6f, 6l) are
accessible that way. However, (i) such an event has not been known to
deliver products with aldehyde moieties (e.g., 6a) or been compatible
with β-ketoamides (e.g., 6j); (ii) the regioselectivity with unsymmetrical
β-diketones (e.g., for 6k) has not been resolved; (iii) vinylogous amides
can also be prepared from other substrates (e.g., substituted propiolates),
and therefore, our reaction has broader product scope.
(18) Our preliminary studies to react the oQDM intermediate with
dienophiles in an intermolecular [4 þ 2] cycloaddition fashion have been
unsuccessful.
(19) Our preliminary studies to react the iminium functionality of 4
with other possible nucleophiles, including NaBH4 and aryl boronic
acid, have also been unsuccessful.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 17, 2013
4369