Synthesis of gold allyloxysulfonium complexes and elimination to form an α,β-unsaturated aldehyde

Article abstract of DOI:10.1039/c9cc06589j

Treatment of the gold vinyl carbene/allylic cation complex (E)-[(IPr)AuC(H)C(H)C(4-C₆H₄OMe)₂]⁺ OTf^- with sulfoxides at -95 °C formed the corresponding gold allyloxysulfonium complexes [(IPr)AuC(H)(OSR

Full text of DOI:10.1039/c9cc06589j

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Synthesis of gold allyloxysulfonium complexes
and elimination to form an a,b-unsaturated
aldehyde†
Cite this: Chem. Commun., 2019,
55, 13745
Received 24th August 2019,
Accepted 17th October 2019
Nana Kim and Ross A. Widenhoefer
*
DOI: 10.1039/c9cc06589j
rsc.li/chemcomm
Treatment of the gold vinyl carbene/allylic cation complex
(E)-[(IPr)AuC(H)C(H)C(4-C6H4OMe)2]+ OTfÀ with sulfoxides at À95 8C
(2)
formed the corresponding gold allyloxysulfonium complexes
+
[(IPr)AuC(H)(OSR2)C(H) C(4-C6H4OMe)2] OTfÀ [R = Me, –(CH2)4–,
Q
Ar] in Z95 Æ 5% NMR yield. Allyloxysulfonium gold complexes under-
went elimination at or below room temperature to form 3,3-bis(4-
methoxyphenyl)acrylaldehyde in Z67% yield.
We have recently reported the synthesis of cationic, two
coordinate gold(I) allylic cation/vinyl carbene complex 1, gene-
rated via the Lewis acid-mediated ionization of
a gold
Sulfoxides, heterocyclic N-oxides, and related nucleophilic oxidants
containing an oxygen atom bound to a leaving group (O–LG) have
been widely employed as oxygen transfer agents in a variety of
gold(I)-catalyzed transformations.1–5 Notable examples include the
intramolecular oxidation/rearrangement of sulfinyl alkynes to
form b-dicarbonyl compounds (eqn (1)) and the oxidative
cycloaddition of enynes to form cyclopropyl carboxaldehydes
(eqn (2)).3,4 Two general mechanisms have been invoked for
oxygen atom transfer in these transformations involving attack
of the nucleophilic oxidant at either the C1 position of a gold
carbene complex to generate a gold alkoxysulfonium/pyridinium
intermediate (A) or at the alkyne carbon atom of a gold p-alkyne
complex to generate a gold alkenyloxysulfonium/pyridinium inter-
mediate (B) (Scheme 1). Heterolytic cleavage of the O–LG bond of
A or B then completes oxygen atom transfer to the substrate, in
the latter case to generate a reactive a-oxo carbene complex.1,6
However, despite the widespread employment of these oxygen
atom transfer reagents in gold(I) catalysis, the key gold oxysulfonium/
pyridinium intermediates have never been directly observed.
As such, significant gaps remain in our understanding of
the structure and reactivity of these intermediates and in the
mechanisms of oxygen atom transfer in gold(I)-catalysis.
(g-methoxy) vinyl complex (Scheme 2).7 Complex 1 reacted rapidly
at À95 1C with a range of nucleophiles, preferentially at the C1
carbon atom. Included in this family of transformations was
the reaction of 1 with 4-picoline-N-oxide to form 3,3-bis(4-
methoxyphenyl)acrylaldehyde (2), presumably via the unob-
served gold allyloxypyridinium intermediate I (Scheme 2).8,9
Despite the instability of intermediate I, we reasoned that
reaction of vinyl carbene complex 1 with sulfoxides might lead
to formation of detectable gold allyloxysulfonium complexes
owing to the stronger S–O bond relative to an N–O bond.10
Indeed, here we report the direct observation and thermal
elimination of gold allyloxysulfonium complexes.
Treatment of a freshly prepared solution of 1 in CD2Cl2 with
dimethylsulfoxide (DMSO; 1.3 equiv.) at À95 1C led to immediate
disappearance of the bright red/orange color of 1 with formation
of the gold allyloxydimethylsulfonium complex 3a in 96 Æ 5%
(1)
Scheme 1 Proposed mechanisms for oxygen atom transfer from nucleo-
philic oxidants (O–LG) to cationic gold carbene (top equation) and
p-alkyne complexes (bottom equation).
French Family Science Center, Duke University, Durham, North Carolina, USA.
E-mail: rwidenho@chem.duke.edu
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9cc06589j
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 13745--13748 | 13745

View Article Online
Communication
ChemComm
1
CD2Cl2 at 33, À16, and À28 1C, respectively, employing H NMR
spectroscopy. In each case, disappearance of 3 obeyed first-order
kinetics (Table 1, entries 1–5) with activation energies that
decreased in the order 3a (DG‡306K = 22.0 kcal molÀ1) 4 3b
(DG‡257K = 19.2 kcal molÀ1) 4 3c (DG2‡34K = 17.6 kcal molÀ1).
To quantify the effect of the electron donor ability of the R2S
moiety on the rate of elimination from complexes 3, we analysed
the kinetics of the disappearance of complexes 3d–3f at À27 1C
(Table 1, entries 5–8). A plot of log k versus Ss gave acceptable fit
(R2 = 0.90) with a slope of r = 1.0 Æ 0.2 (Fig. 1),11 which established
that the electron density on sulfur increased in the transition state
for elimination relative to the ground state.
The relative stabilities of complexes 3a–3c with respect to
elimination tracked with the relative stabilities of these
complexes with respect to sulfoxide displacement. For example,
treatment of the allyloxydiphenylsulfonium complex 3c with
THTSO (1 equiv.) at À32 1C led to rapid (15 min) and quanti-
tative displacement of DPSO to form the allyloxytetramethylene-
sulfonium complex 3b (Scheme 4). Similarly, treatment of 3b
with DMSO (1 equiv.) at À24 1C for 1 h led to quantitative
displacement of THTO to form the allyloxydimethylsulfonium
complex 3a.
Additional kinetic experiments were performed to gain
insight into the mechanism of the conversion of complexes 3
to 2. To determine the activation parameters for elimination,
the rate of conversion of 3b to 2 was determined as a function
of temperature from À28 to À3 1C (Table 1, entries 2–4 and
9–15); an Eyring plot of these data provided the activation
parameters DH‡ = 20.4 Æ 0.6 kcal molÀ1 and DS‡ = 1.1 Æ 0.5 eu
(Fig. 1). We likewise evaluated the effect of exogenous sulfoxide
on the rate of elimination. For example, the rate of elimination of
3b (17 mM) decreased nominally (o10%) in the presence of three
equivalents of THTSO (51 mM; Table 1, entries 2–4 and 16).
In comparison, the rate of elimination of 3a in CD2Cl2 decreased
Scheme 2 Synthesis of gold vinyl carbene/allylic cation complex 1 [IPr =
1,3-bis(2,6-bis-(diisopropylphenyl)imidazol-2-ylidene)] and reaction with
4-picoline-N-oxide to form aldehyde 2.
1
yield by H NMR spectroscopy (Scheme 3). In a similar manner,
treatment of 1 with tetrahydrothiophene sulfoxide (THTSO),
diphenylsulfoxide (DPSO), or p-substituted diarylsulfoxides
(4-C6H4R)2SQO (R = Me, Cl, OMe) at À95 1C formed the corres-
ponding gold allyloxysulfonium complexes 3b–3f in Z95 Æ 5%
yield (Scheme 3). Complexes 3 were thermally unstable and were
characterized in solution by NMR spectroscopy at or below 0 1C.
For example, selective attack of DMSO at the C1 vinyl carbene
carbon atom of 1 was established by a pair of mutually-coupled
doublets at d 3.60 and 5.65 ( J = 11.3 Hz) in the 1H NMR spectrum
of 3a assigned to the aliphatic C1 and vinylic C2 protons,
respectively.‡ Similarly, the 13C NMR spectrum of the
13C-labelled isotopomer [(IPr)Au13C(H)(OSMe2)13C(H)QC(4-
C6H4OMe)2]+ OTfÀ (3a-13C2) displayed a pair of mutually-
1
1
coupled multiplets at d 61.6 (dd, JCH = 139.8, JCC = 41.3 Hz)
1
1
and 113.6 (dd, JCH = 158.0, JCC = 42.3 Hz) assigned to the
aliphatic C1 and vinylic C2 carbon atoms, respectively.
When a solution of 3a in CD2Cl2 was warmed at 20 1C for
B5 h, 3a was completely consumed to form aldehyde 2 in 72%
yield along with [(IPr)Au(SMe2)]+ OTfÀ (Scheme 3).§ In a similar
manner, complex 3b underwent elimination at 0 1C over
the course of 30 min to form 2 in quantitative yield while
complexes 3c–3f underwent elimination at or below 0 1C to
form 2 in 67–88% yield (Scheme 3). To quantify the reactivity
differences between complexes 3a–3c with respect to elimination,
we analysed the kinetics of the disappearance of 3a, 3b, and 3c in
Table 1 First-order rate constants for the conversion of gold allyloxy-
sulfonium complexes 3 (17 mM) to aldehyde 2 in CD2Cl2
Entry
Cmpd
Temp (1C)
k (104 sÀ1 a
)
1
2
3
4
5
6
7
8
3a
3b
3b
3b
3c
3d
3e
3f
3b
3b
3b
3b
3b
3b
3b
3b
3a
3a
33
À16
À16
À16
À28
À27
À27
À27
À9
11.7 Æ 0.6
2.14 Æ 0.02
2.33 Æ 0.10
2.41 Æ 0.06
9.7 Æ 0.22
3.71 Æ 0.05
15.8 Æ 0.4
1.31 Æ 0.02
7.62 Æ 0.11
7.97 Æ 0.12
12.5 Æ 0.2
15.2 Æ 0.3
1.33 Æ 0.04
1.11 Æ 0.01
0.315 Æ 0.022
2.12 Æ 0.05
9.6 Æ 0.4
9
10
11
12
13
14
15
16b
17c
18d
À9
À3
À3
À20
À21
À28
À16
33
33
1.48 Æ 0.09
a
b
First-order decay was observed in all cases. Reaction mixture contained
c
Scheme 3 Synthesis of gold allyloxysulfonium complexes 3 and thermal
elimination to form 2.
THTSO (51 mM). Reaction mixture contained DMSO-d6 (0.45 M).
d
Reaction mixture contained DMSO-d6 (3.0 M).
13746 | Chem. Commun., 2019, 55, 13745--13748
This journal is ©The Royal Society of Chemistry 2019

View Article Online
ChemComm
Communication
Fig. 1 Hammett plot of the elimination of complexes 3c–3f, where
r = 1.0 Æ 0.2 (left plot). Eyring plot of the elimination of 3b over the
temperature range from À28 to À3 1C, where ln(k/T) = (À10 300 Æ 280)/T +
(26 Æ 1) (right plot).
Scheme 6 Potential dissociative (top path) and non-dissociative (bottom
path) elimination pathways for the conversion of complexes 3 to aldehyde 2.
increasing solvent polarity, which points to increased delocali-
zation of positive charge in the transition state for elimination
relative to the ground state. However, the small positive entropy
of activation (DS‡ = 1.1) determined for the elimination of 3b
appears inconsistent with a dissociative elimination pathway
and points to a non-dissociative pathway for elimination of 3
(Scheme 6, bottom pathway). Indeed, the entropy of activa-
tion determined for the elimination of 3b is similar to values
determined for b-hydride and b-alkyl elimination from transi-
tion metal alkyl complexes.13 Likewise, the Hammett reaction
constant determined for the elimination of gold allyloxyldiaryl-
sulfonium complexes 3c–3f (r = 1.0 Æ 0.2) appears too small
to account for the release of free sulfide in the transition state
for elimination, pointing to a non-dissociative pathway for
elimination where sulfide remains tricoordinated throughout
elimination. As points of comparison, Hammett reaction con-
stants for the oxidation of alkyl aryl sulfides with electrophilic
halogen sources involving rate-limiting formation of RArSX+
range from r = À3.20 to À4.25.14
Scheme 4 Sulfoxide exchange experiments involving complexes 3a–3c.
by a factor of eight as the DMSO-d6 concentration increased from
0 to 3.0 M (Table 1, entries 1, 17 and 18), which we attribute to
increasing solvent polarity.¶ Evaluation of the effect of free THT
on the rate of elimination of 3b was precluded by the rapid
substitution of THTSO with THT to form the a-sulfonium carbe-
noid complex 4 (Scheme 5). Specifically, conversion of an equi-
molar mixture of 3b and THT to 4 at À24 1C (k = 7.22 Æ 0.01 Â
10À4 sÀ1) was an order of magnitude faster than was decomposition
of 3b and displayed first-order kinetics to Z3 half-lives, which
points to a dissociative mechanism involving rate-limiting for-
mation of free 1 followed by rapid trapping by THT (Scheme 5).
We considered two mechanisms for the conversion of gold
allyloxysulfonium complexes 3 to aldehyde 2 involving either
(1) dissociative elimination via an anti-periplanar transition
state (TS-II) to form p-aldehyde intermediate II and free sulfide
followed by ligand exchange to form 2 and [(L)AuSR2]+
(Scheme 6, top) or (2) non-dissociative elimination via a syn-
periplanar transition state (TS-III) to form the three-coordinate
gold p-aldehyde sufide intermediate III followed by dissociation
of 2 (Scheme 6, bottom).12
Although b-hydride elimination from gold(I) complexes is
rare,15 the syn-migratory insertion of alkynes into Au(I)–hetero-
atom bonds is known.16–18 Notable among these examples is the
syn-insertion of dimethyl acetylenedicarboxylate into the Au–P
bond of gold phosphine complexes of the form Au(SAr)(PAr3) to
form gold alkenylphosphonium complexes.18
In summary, we have generated thermally unstable gold
allyloxysulfonium complexes 3 via the reaction of gold vinyl
carbene/allylic cation complex 1 with sulfoxides and we have
demonstrated the thermal elimination of complexes 3 to form
aldehyde 2 and a gold sulfide complex. Kinetic analysis of the
elimination of complexes 3, in particular the small positive
entropy of activation and modest Hammett reaction constant
determined for the elimination of gold allyloxyldiarylsulfonium
complexes 3c–3f, points to a mechanism involving non-
dissociative b-elimination through a syn-periplanar transition
state. Unfortunately, experiments that would unambiguously
distinguish between dissociative and non-dissociative mechanisms
for elimination from complexes 3, such as sulfide labelling
experiments, were precluded by rapid exchange of free sulphide
with the C1 sulfoxide group of complexes 3.
Both pathways depicted in Scheme 6 are consistent with the
observed first-order dependence of the rate of elimination on
[3], the zero-order dependence of the rate of elimination of 3b
on [THTSO], and the decreasing rate of elimination of 3a with
Scheme 5 Reaction of 3b with THT to form 4.
We thank the NSF (CHE-1800273) for support of this research.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 13745--13748 | 13747

View Article Online
Communication
ChemComm
8 For additional examples of oxygen atom transfer to well defined
gold carbene complexes see: (a) R. J. Harris and R. A. Widenhoefer,
Conflicts of interest
´
Angew. Chem., Int. Ed., 2014, 53, 9369–9371; (b) C. Garcıa-Morales,
There are no conflicts to declare.
X.-L. Pei, J. M. Sarria Toro and A. M. Echavarren, Angew. Chem., Int. Ed.,
2019, 58, 3957–3961.
9 For examples of oxidative trapping of carbenium ions with sulfoxides
see: (a) Y. Ashikari, T. Nokami and J.-I. Yoshida, Org. Lett., 2012, 14,
938–941; (b) Y. Ashikari, T. Nokami and J.-I. Yoshida, Org. Biomol.
Chem., 2013, 11, 3322–3331; (c) Y. Ashikari, T. Nokami and J.-I. Yoshida,
J. Am. Chem. Soc., 2011, 133, 11840–11843.
10 (a) R. T. Sanderson, Polar Covalence, Elsevier, 1983; (b) R. T.
Sanderson, Chemical Bonds and Bond Energy, Academic Press, 1976.
Notes and references
‡ The large H1–H2 vicinal coupling constants observed for complexes 3
was likewise observed for the analogous a-ammonium carbenoid
complex [(IPr)AuC(H)(py)C(H)QC(4-C6H4OMe)2]+ OTfÀ and was attri-
buted to the large H1–C1–C2–H2 dihedral angle of 1601 estimated from
X-ray analysis.7
§
1H NMR analysis of the reaction mixture upon complete consumption
11 For validation of the additivity of Hammett
s values see:
(a) P. Beltrame, P. L. Beltrame and L. Bellotti, J. Chem. Soc. B,
1969, 932–934; (b) P. Beltrame, P. L. Beltrame, M. L. Cereda and
G. Lazzerini, J. Chem. Soc. B, 1969, 1100–1102; (c) J. E. Baldwin and
J. A. Kapecki, J. Am. Chem. Soc., 1970, 92, 4868–4873; (d) D. J.
McLennan and R. J. Wong, J. Chem. Soc., Perkin Trans. 2, 1972,
279–285; (e) P. Beltrame, P. L. Beltrame, O. Sighinolfi and
M. Simonetta, J. Chem. Soc. B, 1967, 1103–1108; ( f ) P. Beltrame,
D. Pitea and M. Simonet, J. Chem. Soc. B, 1967, 1108–1111.
of 3 was fully consistent with the formation of the corresponding gold
sulfide complexes [(IPr)AuSR2]+ OTfÀ. However, given the limited
dispersion of resonances in the 1H NMR spectrum, it was not possible
to distinguish between bound and free sulfide.
¶ Dielectric constants for CH2Cl2 and DMSO-d6 (3.0 M) in CD2Cl2 are
e = 8.93 and 15.3, respectively.19,20
1 For reviews on gold a-oxo carbenoid complexes see: (a) Z. Zheng,
Z. Wang, Y. Wang and L. Zhang, Chem. Soc. Rev., 2016, 45, 12 For computational analysis of the conversion of s- to p-bound
4448–4458; (b) Y. Wang and L. Zhang, Synthesis, 2015, 47, 289–305;
(c) L. Zhang, Acc. Chem. Res., 2014, 47, 877–888; (d) H.-S. Yeom and
S. Shin, Acc. Chem. Res., 2014, 47, 966–977; (e) J. Xiao and X. Li, Angew.
ligands at gold(I) centers see: L. N. dos Santos Comprido, J. E. M. N.
¨
Klein, G. Knizia, J. Kastner and A. S. K. Hashmi, Chem. – Eur. J.,
2017, 23, 10901–10905.
Chem., Int. Ed., 2011, 50, 7226–7236; ( f ) M. C. B. Jaimes and 13 (a) M. Brookhart, A. F. Volpe, D. M. Lincoln, I. T. Hovath and J. M. Millar,
A. S. K. Hashmi, in Modern Gold Catalyzed Synthesis, ed. A. S. K. Hashmi
and F. D. Toste, Wiley-VCH, Weinheim, 2012, pp. 273–296.
2 For the gold-catalyzed ring opening/oxidative trapping of cyclo-
propenes see: (a) J. T. Bauer, M. S. Hadfield and A.-L. Lee, Chem.
Commun., 2008, 6405–6407; (b) M. S. Hadfield, J. T. Bauer, P. E. Glen
and A.-L. Lee, Org. Biomol. Chem., 2010, 8, 4090–4095.
J. Am. Chem. Soc., 1990, 112, 5634–5636; (b) L. Wang and T. C. Flood,
J. Am. Chem. Soc., 1992, 114, 3169–3170; (c) S. P. Ermer, G. E. Struck,
S. P. Bitler, R. Richards, R. Bau and T. C. Flood, Organometallics, 1993, 12,
2634–2643; (d) N. M. Doherty and J. E. Bercaw, J. Am. Chem. Soc., 1985,
107, 2670–2682; (e) B. J. Burger, B. D. Santarsiero, M. S. Trimmer and
J. E. Bercaw, J. Am. Chem. Soc., 1988, 110, 3134–3146.
3 G. Li and L. Zhang, Angew. Chem., Int. Ed., 2007, 46, 5156–5159.
14 (a) D. Thenraja, P. Subramaniam and C. Srinivasan, J. Chem. Soc.,
Perkin Trans. 2, 2002, 2125–2129; (b) U. Miotti, G. Modena and
L. Seda, J. Chem. Soc. B, 1970, 802–805; (c) F. Ruff and A. Kucsman,
J. Chem. Soc., Perkin Trans. 2, 1975, 509–519.
´
4 C. A. Witham, P. Mauleon, N. D. Shapiro, B. D. Sherry and F. D. Toste,
J. Am. Chem. Soc., 2007, 129, 5838–5839.
5 For additional examples of gold(I)-catalyzed oxidative cyclization of
1,n-enynes see: (a) B. P. Taduri, S. M. Abu Sohel, H.-M. Cheng, G.-Y. 15 (a) C.-M. Ting, Y.-L. Hsu and R.-S. Liu, Chem. Commun., 2012, 48,
´
Lin and R.-S. Liu, Chem. Commun., 2007, 2530–2532; (b) D. Vasu,
H.-H. Hung, S. Bhunia, S. A. Gawade, A. Das and R.-S. Liu, Angew.
6577–6579; (b) M. C. Reis, C. S. Lopez, E. Kraka, D. Cremer and
O. N. Faza, Inorg. Chem., 2016, 55, 8636–8645.
Chem., Int. Ed., 2011, 50, 6911–6914; (c) S. Bhunia, S. Ghorpade, 16 M. Joost, L. Estevez, S. Mallet-Ladeira, K. Miqueu, A. Amgoune and
D. B. Huple and R.-S. Liu, Angew. Chem., Int. Ed., 2012, 51, D. Bourissou, J. Am. Chem. Soc., 2014, 136, 10373–10382.
2939–2942; (d) S. Ghorpade, M.-D. Su and R.-S. Liu, Angew. Chem., Int. 17 (a) T. Tamai, K. Fujiwara, S. Higashimae, A. Nomoto and A. Ogawa,
Ed., 2013, 52, 4229–4234; (e) D. Qian and J. Zhang, Chem. Commun.,
2011, 47, 11152–11154; ( f ) W. Wang, J. Yang, F. Wang and M. Shi,
Org. Lett., 2016, 18, 2114–2117; (b) X. Zhang and K. T. Wang, RSC
Adv., 2015, 5, 34439–34446.
Organometallics, 2011, 30, 3859–3869; (g) K. Ji and L. Zhang, Org. Chem. 18 H. Kuniyasu, T. Nakajima, T. Tamaki, T. Iwasaki and N. Kambe,
Front., 2014, 1, 34–38. Organometallics, 2015, 34, 1373–1376.
6 Base-mediated elimination of alkoxysulfonium ions is common in a 19 A. Jouyban and S. Soltanpour, J. Chem. Eng. Data, 2010, 55, 2951–2963.
range of DMSO-based alcohol oxidation processes: (a) T. T. Tidwell, 20 A. A. Maryott and E. R. Smith, Table of dielectric constants of pure
Synthesis, 1990, 857–870; (b) T. T. Tidwell, Org. React., 2004, 39.
7 N. Kim and R. A. Widenhoefer, Chem. Sci., 2019, 10, 6149–6156.
liquids, Circular of the National Bureau of Standards, 514; Washington,
D.C.; U. S. Gov. Print. Office, 1951.
13748 | Chem. Commun., 2019, 55, 13745--13748
This journal is ©The Royal Society of Chemistry 2019