Gold-catalyzed [4+3] and [4+4]-annulation reactions of: T -butyl propiolate derivatives with epoxides and oxetanes for the construction of 1,4-dioxepane and 1,5-dioxocane cores

Article abstract of DOI:10.1039/c6cc02124g

Gold-catalyzed [4+n]-annulations (n = 3, 4) of tert-butyl propiolate derivatives with epoxides or oxetanes proceed smoothly to yield seven- or eight-membered oxacyclic products efficiently. In the context of the [4+3]-annulations, product analysis reveals a retention of stereochemistry upon the intramolecular S_N2 attack of an epoxide. We also report the [4+5]-annulation between one tert-butyl propiolate and γ-lactol, to manifest the utility toward medium-sized rings.

Full text of DOI:10.1039/c6cc02124g

ChemComm
View Article Online
View Journal
COMMUNICATION
Gold-catalyzed [4+3] and [4+4]-annulation
reactions of t-butyl propiolate derivatives with
epoxides and oxetanes for the construction of
1,4-dioxepane and 1,5-dioxocane cores†
Cite this:DOI: 10.1039/c6cc02124g
Received 10th March 2016,
Accepted 12th May 2016
Rajkumar Lalji Sahani and Rai-Shung Liu*
DOI: 10.1039/c6cc02124g
www.rsc.org/chemcomm
Gold-catalyzed [4+n]-annulations (n = 3, 4) of tert-butyl propiolate
derivatives with epoxides or oxetanes proceed smoothly to yield
seven- or eight-membered oxacyclic products efficiently. In the
context of the [4+3]-annulations, product analysis reveals a retention of
stereochemistry upon the intramolecular SN2 attack of an epoxide. We
also report the [4+5]-annulation between one tert-butyl propiolate and
c-lactol, to manifest the utility toward medium-sized rings.
(1)
The advent of gold catalysis has inspired a rapid development
in the catalytic cycloadditions or annulations of alkynes with
weakly nucleophilic p-bond motifs.1 Considerable interest has
been focused on electron-rich ynamides because their gold
p-alkyne intermediates possess a highly electrophilic keteniminium
character, rendering their reactivity greater than that of normal
alkynes.2,3 Electron-deficient alkynes are somewhat ignored
because their coordination with Au(I) complexes is thought to
be too weak to enhance the electrophilicity. By the efforts of
Shin and our laboratory,4,5 gold complexes could enable hetero-
(4)
Diels–Alder reactions of propiolate derivatives with alkenes,4a The annulations of ynamides or common alkynes with epoxides3c
alkynes,4b nitriles5a and carbonyl species,5b yielding [4+2]-annulation or oxetanes3g were reported to afford [2+n]-annulation products
products efficiently (eqn (1)). These annulation products are [n = 3 or 4, eqn (2)], or to yield [4+3]-cycloadducts for aryl-
strictly limited to six-membered heterocycles. To explore new substituted ynamides and epoxides (eqn (3)).3c Herein, we
cycloadditions, we sought a new utility of propiolate derivatives report distinct [4+3]- and [4+4]-annulation reactions of propiolate
toward the synthesis of seven- and eight-membered oxacyclic derivatives (1) with epoxides (2) and oxetanes (4) respectively, yielding
rings. The success of these [4+n]-annulations (n = 3, 4) relies on 2H-1,4-dioxepin-5(3H)-ones (3) and 7,8-dihydro-1,5-dioxocin-2(6H)-
a tert-butyl ester on propiolate substrates to impede competitive ones (5) respectively (eqn (4)). In contrast to special ynamides
[2+n]-annulations because ethyl propiolates can serve as two (eqn (3)), a broad range of propiolates, epoxides and oxetanes of
carbon-building units.5c
substrates are applicable.
Although there is considerable interest in gold-catalyzed
Seven- and eight-membered lactone ethers of medium sizes
cyclizations of epoxy/alkyne functionalities,6 gold-catalyzed are found in many naturally occurring compounds.7–9 The
annulation of these two functionalities has only one literature particular frameworks such as 1,4-dioxepan-5-ones (I)7 and
reference.3c
1,5-dioxocan-2-ones (II),8 resulting from our cycloaddition
reactions (eqn (4)), provide access to representative naturally
occurring products such as salazinic acid,9a luisol B,9b leiocarpin A,9c
luisol A,9b (+)-penicillide A9d and withametelinol A.9d Although
several synthetic methods7,8 have been developed for the
construction of 1,4-dioxepan-5-one (I) and 1,5-dioxocan-2-one
(II) cores, our [4+3]- and [4+4]-annulation methods employ
Department of Chemistry, National Tsing-Hua University, 101, Sec. 2, Kuang-Fu Rd.
Hsinchu, 30013, Taiwan, Republic of China. E-mail: rsliu@mx.nthu.edu.tw
† Electronic supplementary information (ESI) available. CCDC 1455406 and
1459007. For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c6cc02124g
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun.

View Article Online
Communication
ChemComm
only 25% and 35% yields respectively. Other gold catalysts10
including L1AuCl/AgSbF6 [L1 = P(o-biphenyl)(t-Bu)2], PPh3AuCl/
AgSbF6 and AuCl, were ineffective for this transformation with
the initial propiolate 1a recovered in 70–95% (entries 5–7).
A switch to other gold catalysts (2.5 mol%) including L2AuCl/
AgSbF6 [L2 = tris(2,4-di-tert-butylphenyl)phosphite], IMesAuCl/
AgSbF6 (IMes = 1,3-bis(2,4,6-trimethylphenyl)-1,3-dihydro-2H-
imidazol-2-ylidene), XPhosAuCl (XPhos = 2-dicyclohexylphos-
phino-20,40,60-triisopropylbiphenyl) did not activate the reactions,
giving the initial propiolate 1a in 82–95% recovery (entries 8–10).
We prepared Nolan’s digold complex [(IPrAu)2OH]SbF6 that tends
to form IPrAuSbF6 and IPrAuOH. The reaction of this complex in
DCE (entry 11) gave the desired 3a in 45% yield together with
unreacted 1a in 41% recovery. In this instance, only IPrAuSbF6
catalyzed the reaction whereas IPrAuOH was inactive. In the case
of AgSbF6 (2.5 mol%) alone, initial propiolate 1a was recovered
in 82% yield under the same conditions (entry 12). The use of
HOTf (1.0 mol%) led to a degradation of initial propiolate 1a to
propiolate acid 1a0 together with byproducts 3a0/3a00 in 45%
yields (3a0/3a00 = 2.5/1, entry 13). We envisage that the high
catalytic efficiency of IPrAuCl/AgSbF6 is attributed to its weak
acidity, to avoid catalytic rearrangement of 2-phenyloxirane to
1-phenylethanal.
We assess the generality of these formal [4+3] cycloadditions
with various propiolate derivatives 1b–1g and epoxides 2a–2h;
herein, IPrAuCl/AgSbF6 (2.5 mol%) was used in DCE (35 1C, 5–8 h)
for all cases except entry 12. We tested the reactions of phenyl-
substituted propiolates 1b–1d bearing distinct para-substituents
(X = Cl, Br, OMe) that reacted with epoxide 2a to afford products
3b–3d in good yields (68–75%, entries 1–3). X-ray diffraction of
compound 3d was performed to confirm its molecular frame-
work.11 The annulation of 3-thienyl-substituted propiolate 1e
with epoxide 2a yielded expected 3e in 68% yield (entry 4). The
reactions were also extensible to aliphatic-substituted propiolates
1f–1g (R = cyclopropyl and n-butyl), delivering compounds 3f–3g in
55–70% yields. Herein, the side products 3f0 and 3g0 were produced
in 12% and 25% yields respectively because of the presence of
1-phenylethanal from the rearrangement of 2-phenyloxirane 2a
(entries 5 and 6). Entries 7–15 manifest successful applications
of this strategy with various mono-, 1,1- and 1,2-disubstituted
epoxides 2b–2h. The annulations of styrene-derived epoxides
2b–2d bearing para-substituents (X = F, Br and Me) with tert-
butyl propiolate 1a afforded expected oxacycles 3h–3j in reason-
able yields (55–70%, entries 7–9). These [4+3]-cycloadditions
worked well with isobutylene oxide 2e and phenyl- or n-butyl-
substituted propiolates 1a and 1g, affording the corresponding
compounds 3k and 3l in 60–64% yields respectively (entries 10
and 11). For methylenecyclohexane oxide 2f, IPrAuCl/AgSbF6
failed to give the desired product, but a switch to IPrAuCl/
AgNTf2 (5 mol%) yielded the annulation compound 3m in 60%
yield (entry 12).
Fig. 1 Representative naturally occurring products.
cheap and readily available propiolates, epoxides and oxetanes
(Fig. 1).
Table 1 optimizes catalytic [4+3]-annulation reactions of tert-
butyl-3-phenylpropiolate 1a with 2-phenyloxirane 2a over various
gold complexes. In a typical reaction, a 1,2-dichloroethane (DCE)
solution of species 1a (1.0 equiv.) and 2a (3.0 equiv.) was slowly
added to a DCE solution of the catalyst and MS 4 Å at 35 1C. The
use of IPrAuCl/AgSbF6 (5 mol%, IPr = 1,3-bis(diisopropylphenyl)-
imidazole-2-ylidene) in DCE (35 1C, 3 h) gave complicated mixtures
of products, from which a major component, identified as
the [4+3]-annulation product 3a, was isolated in 45% yield.
(Table 1, entry 1). As 2-phenyloxirane was easily rearranged to
1-phenylethanal catalyzed by Lewis acid catalysts; we employed
a low catalyst loading (2.5 mol%) to increase the yield of desired
3a up to 72% (entry 2). Variations of counter-anions as in
IPrAuCl/AgX (X = NTf2 and OTf, entries 3 and 4) resulted in an
incomplete reaction even after 12 h, producing compound 3a in
Table 1 Catalytic activity over various metal catalysts
Entry Catalysta (mol%)
Time (h) 1a 3ab 1a0 3a0/3a00
1
2
3
4
5
6
7
8
[IPrAuCI]/AgSbF6 (5)
[IPrAuCI]/AgSbF6 (2.5)
[IPrAuCI]/AgNTf2 (2.5)
[IPrAuCI]/AgOTf (2.5)
[L1AuCI]/AgSbF6 (2.5)
[PPh3AuCI]/AgSbF6 (2.5) 12
AuCI (2.5)
L2AuCI/AgSbF6 (2.5)
IMesAuCI/AgSbF6 (2.5)
XPhosAuCI/AgSbF6 (2.5) 24
[(IPrAu)2OH]SbF6 (2.5)
AgSbF6 (2.5)
HOTf (1.0)
3
5
12
12
12
—
—
40 25
25 35
75
70
95
95
82
88
45
72
—
—
—
—
—
—
—
—
—
—
—
—
41
—
—
—
—
—
—
—
—
—
—
—
—
45c
—
—
—
—
—
—
24
24
24
9
10
11
12
13
24
24
24
41 45
82
—
—
—
We prepared cis- and trans-1,2-disubstituted epoxides 2g and
2h to examine the stereospecificity of the reactions (Table 2,
entries 13–15); this stereochemical course provides insight into the
reaction mechanism. The reactions of cis-2-methylstyrene oxide 2g
with tert-butyl propiolates 1a and 1d afforded annulation products
a
b
[1a] = 0.02 M, L1 = P(t-Bu)2(o-biphenyl). Product yields are calculated
c
after purification from a silica column. 3a0/3a00 = 2.5 : 1.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2016

View Article Online
ChemComm
Communication
Table 2 [4+3]-annulation with various propiolates and epoxides
Table 3 [4+4]-annulation reactions with various propiolates and oxetanes
a
b
[1] = 0.20 M. L = IPr for entries 1–6; L = P(t-Bu)2(o-biphenyl) for
c
entries 7–13. Product yields are calculated after purification from a
silica column.
cycloadduct 5f in only 45% yield (entry 6) whereas isoalkyl-
substituted propiolates 1f and 1i–1j (R = cyclopropyl, cyclohexyl
and isopropyl) delivered products 5g–5i in enhanced yields
(72–74%, entries 7–9). We also tested the reactions on various
2-phenyloxetane 4b–4e bearing para-substituents (X = Me, Cl,
Br and F), affording the desired products 5j–5m in 74–82%
a
[1] = 0.20 M. Product yields are calculated after purification from a
b
silica column. 5.0 mol% of IPrAuCI/AgNTf2 was used in entry 12.
3n and 3o in one diastereomeric form; their respective yields yields (entries 10–13).
were 68% and 55% (entries 13 and 14). Similarly, the annulation
of trans-2-methylstyrene oxide 2h with propiolate 1d yielded a
distinct isomer 3p stereoselectively in 55% yield (entry 15).
Compounds 3n and 3o assumed a cis-configuration as their
two vicinal protons have the coupling parameter J = 0–1 Hz,
resembling those ( J = 0–1 Hz) of cis-configured seven-membered
(5)
oxacycles3c in eqn (3); this proposed cis-configuration was
confirmed via X-ray diffraction of compound 3o.11 Likewise,
compound 3p and the trans-configured analogue in eqn (3)
have similar parameters J = 4.3–5.0 Hz. Accordingly, these
annulations proceeded through a retention of stereochemistry
with respect to initial epoxides.
New [4+4]-annulation reactions of these propiolate derivatives
with various oxetane species were also investigated to provide access
to eight-membered dioxocinones.12 Acidic P(o-biphenyl)(t-Bu)2AuCl/
AgSbF6 (2.5 mol%) was used to enhance the reaction efficiency
because oxetanes are more chemically stable than oxiranes in the
presence of acid catalysts. To our delight, the treatment of tert-butyl
3-phenylpropiolate 1a with freshly prepared 2-phenyloxetane 4a and
this catalyst in DCE (35 1C) delivered compound 5a in 67% yield
(Table 3, entry 1); its molecular structure was confirmed using X-ray
tert-Butyl propiolate 1a can undergo [4+5]-cycloaddition with
g-lactol 6 (3.0 equiv.) in dry DCE (40 1C, 8 h) to afford a nine-
membered oxacycle 7 in 75% yield; its postulated structure is
assigned by the 13C NMR resonance of the QCPh–O–CH2
methylene carbon at d 66 ppm, very close to that (d 67 ppm)
of compound 3a respectively. This facile annulation reflects the
power of tert-butyl propiolates to serve as four-atom units to
construct various medium-size rings, further highlighting their
synthetic utility.
(6)
diffraction.12 The reactions of 2-phenyloxetane 4a with phenyl- Eqn (6) assesses the role of the tert-butyl group of propiolates 1
substituted propiolates 1b–1d bearing para-substituents (X = Cl, Br, on these catalytic annulation reactions. We tested gold-catalyzed
OMe) afforded cycloadducts 5b–5d in 65–71% yields (entries 2–4). reactions of propiolate acid 1a0 with 2-phenyloxirane 2a, yielding
3-Thienyl substituted propiolate 1e was also applicable to give the additions of two products 3a0 and 3a00 in a combined 75%
compound 5e in 62% yield (entry 5). For n-alkyl-substituted yield (3a0/3a00 = 2 : 1, eqn (6)). This information suggests that
propiolate 1h, its reaction with 2-phenyloxetane 4a afforded propiolate acid 1a0 is not involved in this [4+3]-annulation
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun.

View Article Online
Communication
ChemComm
Ed., 2010, 49, 2840–2859; (d) G. Evano, C. Theunissen and
M. Lecomte, Aldrichimica Acta, 2015, 48, 59–70.
3 For gold-catalyzed cycloadditions of ynamides with small molecules; see
selected examples: (a) P. W. Davies, A. Cremonesi and L. Damitrescu,
Angew. Chem., Int. Ed., 2011, 50, 8931–8935; (b) R. B. Dateer, B. S. Shaibu
and R.-S. Liu, Angew. Chem., Int. Ed., 2012, 51, 113–117; (c) S. N. Karad,
S. Bhunia and R.-S. Liu, Angew. Chem., Int. Ed., 2012, 51, 8722–8726;
(d) R. B. Dateer, K. Pati and R.-S. Liu, Chem. Commun., 2012, 48,
7200–7202; (e) Z. Xin, S. Kramer, J. Overgaard and T. Skrydstrup,
Chem. – Eur. J., 2014, 20, 7926–7930; ( f ) A.-H. Zhou, Q. He, C. Shu,
Y.-F. Yu, S. Liu, T. Zhao, W. Zhang, X. Lu and L.-W. Ye, Chem. Sci., 2015, 6,
1265–1271; (g) S. K. Pawar, D. Vasu and R.-S. Liu, Adv. Synth. Catal., 2014,
356, 2411–2416; (h) L. Zhu, Y. Yu, Z. Mao and X. Huang, Org. Lett., 2015,
17, 30–33.
4 (a) H.-S. Yeom, J. Koo, H.-S. Park, Y. Wang, Y. Liang, Z.-X. Yu and
S. Shin, J. Am. Chem. Soc., 2012, 134, 208–211; (b) T. Luo, M. Dai,
S.-L. Zheng and S. L. Schreiber, Org. Lett., 2011, 13, 2834–2836.
5 (a) S. N. Karad and R.-S. Liu, Chem. Sci., 2015, 6, 5964–5968; (b) S. N.
Karad, W.-K. Chung and R.-S. Liu, Chem. Commun., 2015, 51,
13004–13007; (c) M. Chen, N. Sun, H. Chen and Y. Liu, Chem. Commun.,
2016, 52, 6324–6327; (d) G. Luo and L. Chen, Tetrahedron Lett., 2015, 56,
6276–6278.
6 For gold-catalyzed reactions of epoxy/alkynes and oxetane/alkyne func-
tionalities, see ref. 3c, 3g and (a) L.-Z. Dai and M. Shi, Chem. – Eur. J., 2008,
14, 7011–7018; (b) X.-Z. Shu, X.-Y. Liu, H.-Q. Xiao, K.-G. Ji, L.-N. Guo, C.-Z.
Qi and Y.-M. Liang, Adv. Synth. Catal., 2007, 349, 2493–2498; (c) G.-Y. Lin,
C.-W. Li, S.-H. Hung and R.-S. Liu, Org. Lett., 2008, 10, 5059–5062.
7 (a) Z. Shen, P. K. Dornan, H. A. Khan, T. K. Woo and V. M. Dong,
J. Am. Chem. Soc., 2009, 131, 1077–1091; (b) Z. Shen, H. A. Khan and
V. M. Dong, J. Am. Chem. Soc., 2008, 130, 2916–2917; (c) K. A. Rose
and K. Zeitler, Org. Lett., 2010, 12, 4552–4555; (d) Y. A. Cheng,
T. Chen, C. K. Tan, J. J. Heng and Y.-Y. Yeung, J. Am. Chem. Soc.,
2012, 134, 16492–16495.
8 (a) L. Liu, S. Xu and H. Zhou, Tetrahedron, 2013, 69, 8386–8391;
(b) W. Zhao, Z. Li and J. Sun, J. Am. Chem. Soc., 2013, 135, 4680–4683;
(c) W. Zhao, Z. Wang and J. Sun, Angew. Chem., Int. Ed., 2012, 51,
6209–6213; (d) M. O. Erhunmwunse and P. G. Steel, Tetrahedron
Lett., 2009, 50, 3568–3570; (e) A. Deagostino, J. Maddaluno,
C. Prandi and P. Venturello, J. Org. Chem., 1996, 61, 7597–7599.
9 (a) X. C. Cheng, P. R. Jensen and W. Fenical, J. Nat. Prod., 1999, 62,
608–610; (b) Q. Mu, W. Tang, C. Li, Y. Lu, H. Sun, H. Zheng, X. Hao,
Q. Zheng, N. Wu, L. Lou and B. Xu, Heterocycles, 1999, 51,
2969–2976; (c) Q. Zhang, C. Deng, L. Fang, W. Xu, Q. Zhao,
J. Zhang, Y. Wang and X. Lei, Chin. J. Chem., 2013, 31, 355–370;
(d) B. S. Siddiqui, I. A. Hashmi and S. Begum, Heterocycles, 2002, 57,
715–721.
Scheme 1 Mechanism for retention of stereochemistry.
because two isomeric products will form for species 3a through
gold-catalyzed cyclizations of species 3a0 and 3a00. We also
examined the same reaction with ethyl propiolate 1a00 that was
catalytically inactive (eqn (6)).
Scheme 1 rationalizes the stereospecificity of the [4+3]-annulation
reactions, in which cis-methylstyrene oxide was cleaved with a
retention of stereochemistry. We postulate that the addition of this
epoxide at gold-p-alkyne species A forms epoxide B of an oxonium
type, of which the epoxy ring is perpendicular to the plane defined
by alkenylgold. This orientation is favorable for an attack of the
ester group at the epoxy CHPh carbon as depicted in state C. The
retention of stereochemistry in an intramolecular reaction of the
SN2 reaction was noted for two gold catalytic reactions.3c,13
In summary, we have demonstrated the feasibility of the [4+3]-
cycloadditions of various tert-butyl propiolates with mono-, 1,2-
and 1,1-disubsituted epoxides, yielding 2H-1,4-dioxepin-5(3H)-
ones (3) efficiently. These annulations proceed through high
stereospecificity that cis-2-methylstyrene oxide gave cis-configured
annulation products, indicative of a retention of stereochemistry to
cleave the epoxide ring. The same propiolates can undergo smooth
[4+4]-annulations with several oxetanes to yield 7,8-dihydro-1,5-
dioxocin-2(6H)-ones. We also present one instance for the [4+5]-
cycloaddition of one tert-butyl propiolate with g-lactol to yield a
nine-membered oxacycle, thus highlighting the synthetic utility.
Notes and references
¨
1 Selected reviews for gold-catalyzed cycloadditions of alkynes or allenes: 10 (a) A. Zhdanko, M. Strobele and M. E. Maier, Chem. – Eur. J., 2012,
(a) M. E. Muratore, A. Homs, C. Obradors and A. M. Echavarren, Chem. –
18, 14732–14744; (b) S. Dupuy, D. Gasperini and S. P. Nolan, ACS
Catal., 2015, 5, 6918–6921; (c) A. Gomez-Suarez, Y. Oonishi,
S. Meiries and S. P. Nolan, Organometallics, 2013, 32, 1106–1111;
´
˜
Asian J., 2014, 9, 3066–3082; (b) F. Lopez and J. L. Mascarenas, Chem. Soc.
Rev., 2014, 43, 2904–2914; (c) D. Qian and J. Zhang, Chem. Rec., 2014, 14,
280–302; (d) A. Acardi, Chem. Rev., 2008, 108, 3266–3325; (e) R. Dorel and
A. M. Echavarren, Chem. Rev., 2015, 115, 9028–9072.
´
(d) S. Gaillard, J. Bosson, R. S. Ramon, P. Num, A. M. Z. Slawin and
S. P. Nolan, Chem. – Eur. J., 2010, 16, 13729–13740.
11 CCDC 1455406 (3d), 1459007 (3o) and 1062772 (5a).
2 Recent reviews for ynamides, see (a) X.-N. Wang, H.-S. Yeom, L.-C.
Fang, S. He, Z.-X. Ma, B. L. Kedrowski and R. P. Hsung, Acc. Chem. 12 The synthesis of [4+4]-annulation product 5a was mentioned in an
Res., 2014, 47, 560–578; (b) K. A. DeKorver, H. Li, A. G. Lohse, early report; see ref. 5a.
R. Hayashi, Z. Lu, Y. Zhang and R. P. Hsung, Chem. Rev., 2010, 110, 13 S. K. Pawar, C.-D. Wang, S. Bhunia, A. M. Jadhav and R.-S. Liu,
5064–5106; (c) G. Evano, A. Coste and K. Jouvin, Angew. Chem., Int.
Angew. Chem., Int. Ed., 2013, 52, 7559–7563.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2016