Potassium isopropoxide: For sulfination it is the only base you need!

Article abstract of DOI:10.1002/chem.201303756

PEPPSI max: KOiPr has been identified as the key ingredient for Pd-catalyzed sulfination (see scheme). Potassium is essential to keep the thiol concentration low, and isopropoxide is necessary for precatalyst activation and to break up Pd-sufide-based resting states. Together with the reactive Pd-PEPPSI-IPent^Cl o-picoline catalyst, this system couples profoundly deactivated partners at RT that other catalysts cannot accomplish in refluxing toluene. Copyright

Full text of DOI:10.1002/chem.201303756

DOI: 10.1002/chem.201303756
Potassium Isopropoxide: For Sulfination It is the Only Base You Need!
Mahmoud Sayah and Michael G. Organ*[a]
Recently we reported a promising sulfinating system
using the Pd-PEPPSI-IPent precatalyst (6) to couple chal-
lenging substrates under moderate reaction conditions.[1,2]
However, unlike cross-coupling with nucleophilic organome-
tallics that readily reduce the PdII precatalyst, sulfinating
with arylthiols using tert-butoxide base does not provide an
obvious/immediate source of reductant,[3] assuming for now
that it is not an electron transfer process. Consequently we
showed that reduction could be achieved by pretreatment of
4 or 6 with dibutylmagnesium, morpholine, or LiOiPr, of
which the latter is most pragmatic. That said, precatalyst ac-
tivation with isopropoxide had to be heated to approximate-
ly 808C, even if the sulfination itself proceeded smoothly at
RT, which added inconvenience and an additional step to
monitor in the process. This encouraged us to look at the
precatalyst reduction step itself in more detail in an attempt
to better understand it and devise a simpler, if not invisible
precatalyst activation process to streamline sulfination using
the highly active Pd-NHC catalyst system (Figure 1). This
has resulted in much greater understanding of the sulfina-
tion mechanism, a vastly easier protocol to follow, and the
movement of the process to conditions so mild they would
have been considered unattainable even a year ago.[4]
Our initial studies revealed that both isopropoxide
(entry 3 vs. 4, and entry 5 vs. 7) and elevated temperatures
(entry 1 vs. 2, and entry 5 vs. 6) greatly impact precatalyst
activation (Table 1). To track activation in the PEPPSI
Table 1. Effect of temperature and isopropoxide base on Pd-PEPPSI pre-
catalyst activation and ensuing sulfination.
Entry
Precatalyst
Additive
T [8C]
Yield [%][d]
1
2
3
4
5
6
7
8
4
4
6
6
7
7
7
10
–
–
–
70
80
70
0
90
0
99
0
99
99
99
LiOiPr (20%)[a]
80, then 23
23
–
–
70
LiOiPr (10%)
23[b]
40
KOiPr (200%)[c]
[a] A mixture containing 1, the Pd-PEPPSI precatalyst, and LiOiPr was
heated to 808C for 30 min, cooled to RT, KOtBu and 2 were added, and
the reaction was stirred at RT for 24 h. [b] When the reaction was per-
formed using degassed solvent, 90% conv. to 3 was obtained in just
5 min. [c] No KOtBu was used at all. [d] Percent yield is reported on iso-
lated products following silica gel column chromatography; reactions
were performed in duplicate.
system, precatalyst 4 (in benzene) was treated with LiOiPr
1
with warming, and progress was followed by H NMR spec-
troscopy (Figure 2). When the base was added (spectrum b),
we observed approximately 30% dissociation of 3-chloro-
pyridine (downfield region, not shown) and the formation of
dimer 9, the structure of which was confirmed by preparing
it independently (see the Supporting Information). At this
stage there was no evidence of the formation of acetone
(1.6 ppm) or isopropanol (1.05 and 3.8 ppm), suggesting that
reduction had yet to take place. Upon heating to 508C
(spectrum c) the first sign of acetone formation took place.
When heated to 808C (spectrum d), the spectrum had radi-
cally changed. The LiOiPr was fully consumed and the ace-
tone peak was now quite prominent with an equimolar
amount of iPrOH, while the intermediate dimer (9) was
gone.
Figure 1. Pd-NHC complexes used in this study.
[a] M. Sayah, Prof. M. G. Organ
Department of Chemistry, York University
4700 Keele Street, Toronto, Ontario, M3J 1P3 (Canada)
E-mail: organ@yorku.ca
At this stage we expected to have the Pd0-NHC mono-
mer, likely coordinated to 3-chloropyridine, which would
then enter the sulfination catalytic cycle directly, or form
a new complex with the thiol(s) present (vide infra). Howev-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201303756.
16196
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16196 – 16199

COMMUNICATION
and discovered that in fact the simple pyridine derivatives
activate much more readily. This result is highly unanticipat-
ed as one would have expected the more electron-poor 3-
chloropyridine to assist in precatalyst reduction/activation.[5]
When the arylthiol is added to the reaction mixture con-
taining the PdII precatalyst (i.e., 5), we believe that thiol/
chloride exchange is the first thing to happen that leads to
the formation of 10. One question that arises is whether the
reductive elimination of diaryldithiol is the way in which Pd0
is produced or whether isopropoxide is involved in the
direct reduction of 10? From an enthalpy point of view, re-
ductive elimination of diaryldithiol is not favoured and in
fact we have shown that addition of diphenyldithiol to
a Pd0-NHC complex will move the equilibrium fully to the
PdII species even at 08C (i.e., Pd0 oxidatively adds fully to
the dithiol).[1a] To examine the potential role of isopropoxide
in dithiol precatalyst activation (i.e., 10) we again turned to
1H NMR spectroscopy to follow the activation process
(Table 2). PdII complex 10 was treated with various basic
Table 2. Base-activation study of dithiol complex 10.
Figure 2. Activation of precatalyst 4 using LiOiPr as a function of tem-
perature. a) Precatalyst 4 in [D6]benzene at RT; b) sample in a) plus
2.0 equiv of LiOiPr at RT; c) sample in b) warmed to 508C; d) sample in
c) warmed to 808C.
Entry
Base additive (equiv)
Conversion [%] to 13b[a]
er, we were surprised to isolate the PdII complex 5a in ap-
proximately 90% yield, which revealed that the 3-chloropyr-
idine ligand had in fact been reduced. To account for this,
we propose that Pd0 is in fact formed concomitant with the
oxidation of isopropoxide to acetone and that the activated
Pd0-NHC complex undergoes oxidative addition to the 3-
chloropyridine. The equivalent of iPrOH that is produced
from the above oxidation of LiOiPr then serves as the
source of hydride to complete the chloride reduction. To
confirm this LiOiPr(D7) was used with 4 and the 3-deuterio-
pyridine Pd-NHC complex 5b was isolated exclusively (see
the Supporting Information). Although the source of the hy-
dride is not in question, it is unclear exactly how the reduc-
tive steps that are necessary to provide 5a yield a PdII com-
plex at the end? We think it is most likely that only a small
amount of the Pd0-NHC complex actually forms, which
enters into a catalytic cycle of its own with LiOiPr to reduce
all of the 3-chloropyridine. This reduction could happen to
intact molecules of 4 directly, or to the 3-chloropyridine that
dissociates from it and the pyridine thus produced simply re-
ligates to the PdII-NHC-Cl2 complex. This would account for
the near quantitative yield of 5a.
1
2
3
4
5
KOtBu (5), LiOiPr (3)
KOtBu (3)
LiOtBu (3)
LiOiPr (3)
KOiPr (3)
100
20
0
0
100
[a] Conversion is the percentage of 10 that is transformed into 13b and
was followed and determined by H NMR spectroscopy over time.
1
salts, or salt combinations, and then trapped with 11 to con-
firm that reduction had taken place. When KOtBu and
LiOiPr were used together (entry 1) full conversion to 13b
(via 13a) was observed, whereas KOtBu (entry 2), LiOtBu
(entry 3), or LiOiPr (entry 4) by themselves yielded low or
zero conversion. Surprisingly, substitution of KOiPr for
LiOiPr (entry 5) led to complete reduction of 10, which is
indicative of both a significant base and counter ion effect.
Whereas the reaction mixtures in entries 3 and 4 were ho-
mogeneous, those of entries 1 and 5 were heterogeneous.
Removal of thiol from solution as the mostly insoluble po-
tassium salt would lower its concentration and help to pull
the equilibrium toward 12, whereas the lithium salts are
soluble, which discourages the reduction of 10. It seems
most likely that the small amount of 13b that is produced
(entry 2) does occur via reductive elimination of ditolyldi-
thiol. Although this is not favoured, it would be assisted by
cleavage of the dithiol as it forms by KOtBu, a strong reduc-
tant, to the potassium thiol salt that precipitates out. To con-
On the surface, it would appear that a remarkable
amount of chemistry has taken place just to produce a re-
duced analogue of 4 or 6. To see if there was any advantage
in this “precatalyst pre-activation”, we investigated sulfina-
tion using 7 in place of 6 (Table 1, entry 7 vs. 4, respectively)
Chem. Eur. J. 2013, 19, 16196 – 16199
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16197

M. G. Organ and M. Sayah
firm this reductant role for KOtBu under the sulfination
conditions, we treated 13c with ditolyldithiol and KOtBu
and this led to the quantitative formation of 13b (see the
Supporting Information).
elimination of thiolate, which again as the K salt would pre-
cipitate. The Pd0-NHC complex (19) thus produced is able
to undergo oxidative addition with the aryl halide present
and catalysis ensues. We also discovered that up to 50% of
dithiol 16 is drawn further off cycle into resting state 17 that
cannot be rescued by KOtBu.[1a] Another vital role of KOiPr
that we have uncovered in this study is the cleavage of this
tri-Pd complex by isopropoxide, which is just small enough
(relative to tert-butoxide) to attack Pd, which upon complex-
ation with pyridine would produce 18, thus 17 is not in equi-
librium with 16.
To confirm the necessity of both potassium and isoprop-
oxide for facile sulfination we performed the control reac-
tion with dithiol precatalyst 10 with just KOiPr as base
(Table 1, entry 8) and the reaction proceeded smoothly. This
result would also suggest that heating to 808C with Pd-
PEPPSI-IPr (4) is only necessary to reduce the 3-chloropyri-
dine ligand as the dithiol PdII complex activated readily and
completed the coupling under mild conditions.
With a thorough understanding of all of the events associ-
ated with precatalyst activation in hand, we wanted to press
forward to see if a general and highly reactive sulfination
system could be created to make sulfination a mild, robust,
and widely applicable operation. The most reactive catalysts
and associated protocols in the literature require very high
temperatures (e.g., 1108C),[3,4] and even then some moder-
ately deactivated coupling partners (e.g., sterically and/or
electronically aryl haldies and thiols) simply do not couple.[3]
Precatalyst 8 combines a 2-methylpyridine ligand, which as-
sists in precatalyst activation, and the dichloro NHC carbene
core that we have shown to be highly effective in the Ne-
gishi coupling of secondary alkyls—another notoriously dif-
ficult transformation.[6] Indeed, we were delighted to see
that a wide variety of difficult sulfinations could now be car-
ried out routinely at room temperature following one simple
protocol in which everything is simply mixed together.[7] Bis
di-ortho-substituted aryl thiols were readily prepared (e.g.,
21, 23) indicating that even the most sterically congested
coupling partners can be tolerated. Electron-poor heterocy-
clic thiols posed no obstacles (e.g., 24, 26, 27, 29). Even the
most stercially and electronically deactivated oxidative addi-
tion partner could be routinely coupled in excellent yield
(e.g., 22, 25), even if a highly electron-poor thiol is used
(e.g., 28). Finally, the proficiency of precatalyst 8 was dem-
onstrated as a loading as low as 0.1 mol% still led to com-
plete conversion to product (e.g., 20).
Given that the conversion to 13 with isopropoxide is
strongly favoured (vide infra), reduction with this base must
be taking place by another pathway. The first clue as to the
mechanism at work was the appearance of acetone and iso-
1
propanol in the H NMR spectrum of entry 5 in Table 2 (see
the Supporting Information). If 12 were formed by global
reductive elimination of ditolydithiol there would be no free
PdII present to oxidize the isopropoxide. We considered that
the isopropoxide-arylthiol adduct (i.e., ArS-OiPr) formed by
the cleavage of dithiol in the presence of this base could be
susceptible to base-induced elimination of thiolate concomi-
tant with the formation of acetone. However, treatment of
ditolyldithiol with KOiPr in the absence of any Pd complex
did not yield acetone (see the Supporting Information), thus
this method of precatalyst reduction seems unlikely. Taken
together, the precatalyst activation mechanism that we pro-
pose is shown in Figure 3. Substitution of one thiolate on 16
Figure 3. Mechanism of PEPPSI precatalyst activation in sulfination.
In conclusion, we have dissected out the details of activa-
tion of Pd-PEPPSI precatalysts in sulfination reactions. The
signature 3-chloropyridine ligand of the PEPPSI family,
which was chosen originally for the general belief of en-
hanced precatalyst activation capability relative to simple
pyridine,[5] actually is fully reduced (i.e., Cl to H) under the
reductive conditions of the sulfination protocol using iso-
propoxide base. So, whether one starts with 4 or 6, or with
the simple corresponding pyridine analogues 5 or 7, the
same PdII complex then undergoes immediate ligand ex-
change of thiolate for chloride producing the penultimate
with isopropoxide will produce the mixed PdII salt 18 that
can undergo reduction in one of two possible ways. b-Hy-
À
dride elimination would produce the corresponding ArS
À
Pd H species that would yield ArSH upon reductive elimi-
nation, which would be driven by formation of insoluble
ArSK by isopropoxide (M=K) and pull the equilibrium
over. Conversely, deprotonation of the Pd-coordinated iso-
propoxide by another molecule of KOiPr would produce
Pd0 directly concurrent with the production of acetone and
16198
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16196 – 16199

Metal-Catalysed Sulfination
COMMUNICATION
precatalyst (i.e., 16). A significant portion of this advanced
precatalyst is then disproportionated into a tri-Pd complex
(i.e., 17). This resting state cannot be rescued by KOtBu,
but KOiPr can, which is presumably a function of the subtle
difference in size of the base and the great stability of this
resting state species (Table 3). Isopropoxide then reduces di-
Acknowledgements
The authors thank the NSERC Canada for funding this work.
Keywords: catalysis · cross-couple · PEPPSI · sulfination
[1] a) M. Sayah, A. Lough, M. G. Organ, Chem. Eur. J. 2013, 19, 2749–
2756; b) M. Sayah, M. G. Organ, Chem. Eur. J. 2011, 17, 11719–
11722.
Table 3. Sulfination of sterically and electronically deactivated oxidative
addition and thiol partners at room temperature.[a]
[2] For other reports on the high reactivity of the Pd-PEPPSI-IPent pre-
catalyst, see: a) K. H. Hoi, S. C¸ alimsiz, R. D. J. Froese, A. C. Hopkin-
son, M. G. Organ, Chem. Eur. J. 2012, 18, 145–151; b) C. Valente, S.
C¸ alimsiz, K. H. Hoi, D. Mallik, M. Sayah, M. G. Organ, Angew.
Chem. 2012, 124, 3370–3388; Angew. Chem. Int. Ed. 2012, 51, 3314–
3332; c) K. H. Hoi, S. C¸ alimsiz, R. D. J. Froese, A. C. Hopkinson,
M. G. Organ, Chem. Eur. J. 2011, 17, 3086–3090; d) S. C¸ alimsiz,
M. G. Organ, Chem. Commun. 2011, 47, 5181–5183; e) S. C¸ alimsiz,
M. Sayah, D. Mallik, M. G. Organ, Angew. Chem. 2010, 122, 2058–
2061; Angew. Chem. Int. Ed. 2010, 49, 2014–2017; f) M. Dowlut, D.
Mallik, M. G. Organ, Chem. Eur. J. 2010, 16, 4279–4283; g) M. G.
Organ, S. C¸ alimsiz, M. Sayah, K. H. Hoi, A. J. Lough, Angew. Chem.
2009, 121, 2419–2423; Angew. Chem. Int. Ed. 2009, 48, 2383–2387.
[3] E. Alvaro, J. F. Hartwig, J. Am. Chem. Soc. 2009, 131, 7858–7868.
[4] Arylthiol couplings in the literature using different metals and gener-
al use, highly active ligand systems all require approximately 1008C
(or higher) in order to achieve any sign of coupling unless the oxida-
tive addition partner is strongly activated (e.g., aryl iodide). For ex-
amples, see: a) ref. [3]; b) S. Jammi, P. Barua, L. Rout, P. Saha, T.
Punniyamurthy, Tetrahedron Lett. 2008, 49, 1484–1487; c) M. A.
Fernꢁndez-Rodrꢂguez, Q. Shen, J. F. Hartwig, Chem. Eur. J. 2006, 12,
7782–7796; d) L. Cai, J. Cuevas, Y.-Y. Pengb, V. W. Pike, Tetrahedron
Lett. 2006, 47, 4449–4452; e) M. Murata, S. L. Buchwald, Tetrahedron
2004, 60, 7397–7403; f) G.-Y. Gao, A. J. Colvin, Y. Chen, X. P.
Zhang, J. Org. Chem. 2004, 69, 8886–8892; for a recent review see:
g) C. C. Eichman, J. P. Stambuli, Molecules 2011, 16, 590–608.
[5] J. Nasielski, E. A. B. Kantchev, C. J. OꢃBrien, M. G. Organ, Chem.
Eur. J. 2010, 16, 10844–10853.
[a] Reaction components were simply mixed together and allowed to stir
for 24 h under argon. [b] The same yield was obtained when the reaction
was performed with 0.1% of 8. [c] The same reaction with JosiPhos and
PdACHTUNGTRENNUNG(dba)2 gave 0%, whereas BrettPhos-Pd-G3 gave 18%.
[6] M. Pompeo, N. Hadei, R. D. J. Froese, M. G. Organ, Angew. Chem.
2012, 124, 11516–11519; Angew. Chem. Int. Ed. 2012, 51, 11354–
11357.
[7] Although the reactions work equally well with KOiPr, the base is not
stable for prolonged periods of time, as are KOtBu, and LiOiPr. So,
it is easier to allow the transmetallation to take place in situ using
these two bases than it is to prepare fresh KOiPr every few weeks to
ensure the integrity of the base.
[8] Pd-PEPPSI-IPr and Pd-PEPPSI-IPent are commercially available
through Sigma–Aldrich. Pd-PEPPSI-IPentCl o-picoline precatalyst is
available through Total Synthesis, Ltd. (www.totalsynthesis.ca).
[9] When the reactions in Table 3 were performed using highly reactive
phosphine catalysts, low (or no) conversion was observed (see the
Supporting Information).
thiol resting state 16 to produce the catalytically active spe-
cies 19 and sulfination ensues. The presence of potassium,
which can come from KOiPr or a mixed base system (i.e.,
KOtBu, and LiOiPr)[7] is essential for any turnover whatso-
ever, and this is attributed to the insolubility of KSAr in tol-
uene, which keeps the concentration of the thiolate poison
low and thus avoids pushing the active catalyst back into un-
productive resting states (i.e., 16 or 17). Finally, a highly re-
active precatalyst (Pd-PEPPSI-IPentCl o-picoline, 8) has
been engineered specifically for sulfination that has a readily
dissociatable 2-methylpyridine and the dichloro-NHC core
that rapidly drives the catalytic cycle.[6,8] This procedure is
operationally simple and works with substrates that will not
work with other catalysts currently being used that are con-
sidered highly active for sulfination.[9]
Received: October 7, 2013
Published online: October 31, 2013
Chem. Eur. J. 2013, 19, 16196 – 16199
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16199