Optimizing ligand structure for low-loading and fast catalysis for alkynyl-alcohol and-amine cyclization

Article abstract of DOI:10.1039/c9dt01870k

A series of [Ru(Cp/Cp?)(P^R₂N^R′₂)(MeCN)]PF₆ complexes were prepared, in which the steric and electronic properties of the primary coordination sphere were varied (R = Ph, t-Bu, Bn; and Cp vs. Cp?). These complexes were catalytically active in the cyclization of alkyne substrates with an intramolecular nucleophile (amine or alcohol) to produce 5-and 6-membered heterocycles. The effect of the 1° coordination sphere structure on catalyst performance was evaluated. Steric bulk around the metal centre was a key feature to achieve rapid catalysis at low temperatures. The catalyst [Ru(Cp)(P^t-Bu₂N^Ph₂)(MeCN)]PF₆ gave a turnover number that was >1 order of magnitude more active than previous catalysts in the cyclization of the benchmark substrate 2-ethynylaniline. This catalyst is tolerant of a diversity of functional groups and is competent at the formation of various substituted indoles.

Full text of DOI:10.1039/c9dt01870k

Dalton
Transactions
View Article Online
View Journal
PAPER
Optimizing ligand structure for low-loading and
fast catalysis for alkynyl-alcohol and -amine
cyclization†
Cite this: DOI: 10.1039/c9dt01870k
James M. Stubbs, Benjamin J. Bridge and Johanna M. Blacquiere
*
A series of [Ru(Cp/Cp*)(PR2NR’2)(MeCN)]PF6 complexes were prepared, in which the steric and electronic
properties of the primary coordination sphere were varied (R = Ph, t-Bu, Bn; and Cp vs. Cp*). These com-
plexes were catalytically active in the cyclization of alkyne substrates with an intramolecular nucleophile
(amine or alcohol) to produce 5- and 6-membered heterocycles. The effect of the 1° coordination
sphere structure on catalyst performance was evaluated. Steric bulk around the metal centre was a key
feature to achieve rapid catalysis at low temperatures. The catalyst [Ru(Cp)(Pt-Bu2NPh2)(MeCN)]PF6 gave a
turnover number that was >1 order of magnitude more active than previous catalysts in the cyclization of
the benchmark substrate 2-ethynylaniline. This catalyst is tolerant of a diversity of functional groups and is
competent at the formation of various substituted indoles.
Received 5th May 2019,
Accepted 8th May 2019
DOI: 10.1039/c9dt01870k
rsc.li/dalton
nium complexes, [Ru(Cp)(PR2NR′2)(MeCN)]PF6 (1), for the cycli-
Introduction
zation of alkynyl alcohols and amines (Fig. 1).6 Complexes of
Oxygen- and nitrogen-containing heterocycles of varying ring type 1 selectively cyclize 2-ethynylbenzyl alcohol (EBA) to give
sizes are important motifs in many classes of molecules, the 6-membered ring iso-chromene (IC) product, which indi-
including: natural products, pharmaceuticals and conjugated cates this catalyst family follows a vinylidene mechanism.6b
polymers.1 A relatively mild and atom-economic route into Attempted high temperature cyclization of EBA revealed a de-
such structures is the catalytic cyclization of an alkyne with an activation product, B (R = t-Bu, R′ = Bn), in which the pendent
alcohol or amine (1° or 2°).2 Transition metal catalysts that tertiary amine of the catalyst attacked the alpha carbon of the
mediate this reaction either activate the alkyne by vinylidene.6b In an effort to increase catalyst performance and
π-coordination or through formation of a vinylidene intermedi- circumvent deactivation, a series of catalysts were prepared in
ate (i.e. III, Scheme 1). Nucleophilic attack at Cα of a vinylidene which the nature of the pendent amine was varied.6a
selectively gives endo cyclized products, in contrast to Cyclization of 2-ethynylaniline (EA) derivatives revealed that
π-coordination that gives endo or exo products.2h,3 Ruthenium the highest catalyst performance was achieved with complexes
complexes predominantly follow a vinylidene route, but other in which the Brønsted basicity of the PR2NR′2 ligand amine was
mechanisms have been identified for select catalysts.4
similar to, or exceeded, that of the substrate amine. In the
Since deprotonation and protonation steps occur through-
out the cyclization mechanism, a Brønsted base is an essential
component of the catalytic system (Scheme 1).5 The base can
be exogenous (i.e. pyridine)3c or it can be incorporated into the
ligand framework,3d,6 a common structural feature of metal–
ligand cooperative (MLC) catalysts.5a,7 We have recently
employed the tunable8 PR2NR′ (1,5-R′-3,7-R-1,5-diaza-3,7-
2
diphosphacyclooctane) ligand family, A, in piano-stool ruthe-
Department of Chemistry, University of Western Ontario, London, Ontario, N6A 5B7,
Canada. E-mail: johanna.blacquiere@uwo.ca
†Electronic supplementary information (ESI) available: General, synthetic and
catalytic procedures; IR and NMR spectra; and additional catalysis data; crystal-
lographic information. CCDC 1891141. For ESI and crystallographic data in CIF
or other electronic format see DOI: 10.1039/c9dt01870k
Scheme 1 Cyclization of alkynyl amine or alcohol substrates mediated
by a metal catalyst following a vinylidine (i.e. III) mechanism.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
PF6 to give Cp (1a–d) and Cp* (2a and 2d) complexes, respect-
ively (Scheme 2). Complexes 1c, 1d, 2a and 2d are new entries
into the [Ru(Cp/Cp*)(PR2NR′2)(MeCN)]PF6 family of catalysts
and each was obtained in excellent yield (1c: 87%, 1d: 90%, 2a:
99%, 2d: 95%).
1
Complexes 1c, 1d, 2a and 2d were characterized by H, 31P
{1H}, and 13C{1H} NMR spectroscopy, MALDI mass spec-
trometry, and IR spectroscopy. Successful ligand coordination
was confirmed in each case by MALDI MS, which revealed a
major signal for a [M − MeCN − PF6]+ fragment that is consist-
ent with previously synthesized complexes of this type.6 One
singlet is observed in the 31P{1H} NMR spectra of 1d, 2a and 2d
at room temperature. In contrast, the R = Bn derivative 1c has
several signals at room temperature. Coalescence to one major
signal at 36.6 ppm is observed on cooling to −90 °C in CH2Cl2.
These observations are consistent with a dynamic structure for
1c, due to changes in conformation of the Ru–PBn2NBn2 metallo-
cyclic rings and the benzyl substituents.9 The 31P chemical shift
for the Cp* complexes are ca. 8–15 ppm upfield from the corres-
ponding signal for the Cp analogue (1a: 38.4 ppm; 2a:
30.7 ppm; 1d: 55.5 ppm; 2d: 40.7 ppm in CD2Cl2). This differ-
ence is due to the greater donor properties of Cp* vs. Cp, which
attenuates the donation of the phosphines to the Ru centre.
Single crystals of 2d were obtained and the X-ray structure
confirmed the expected connectivity (Fig. 2). The Ru–P(1)/P(2)
bond lengths (both 2.306(1) Å) and P(1)–Ru–P(2) angle
(77.75(4)°) are very similar to those found in the closely related
chloro complex RuCl(Cp*)(Pt-Bu2NPh2) (cf. Ru–P = 2.3077(4) and
2.3001(4) Å; P–Ru–P = 78.272(15)°).10 However, the solid-state
structure of 2d reveals two unusual features as compared to
related complexes. First, both of the R′ = phenyl substituents
Fig. 1 General structure for: the PR2NR’2 ligand family A and the catalyst
family [Ru(Cp)(PR2NR’2)(MeCN)]PF6, 1. Complex B was an identified cata-
lysts deactivation product for catalyst 1 (R = t-Bu; R’ = Bn) in the cycliza-
tion of 2-ethynylbenzyl alcohol.
present study we continue the evaluation of optimal structure
for the Ru-(PR2NR′2) class of catalysts through modification of
the primary coordination sphere.
Results and discussion
Synthesis and characterization of [Ru(Cp/Cp*)(PR2NR′
(MeCN)]PF6 catalysts
)
2
A group of [Ru(Cp/Cp*)(PR2NR′2)(MeCN)]PF6 complexes was
prepared that differed in the nature of the ancillary ligand (1 =
Cp; 2 = Cp*; Cp* = pentamethylcyclopentadienyl) and in the
structure of the PR2NR′2 ligand (Scheme 2). The phosphine sub-
stituents were varied (R: a = Ph; b = t-Bu; c = Bn), while the
pendent amine substituent (R′ = Bn) was held constant. The
initial catalyst screen (vide infra) was conducted in parallel
with our previous evaluation of substituent effects of the sec-
ondary coordination sphere,6a which revealed that pendent
amine substituents R′ = benzyl and phenyl give similar cycliza-
tion performance. Given the easier synthetic accessibility of
ligands containing R′ = phenyl versus benzyl substituents, two
additional catalysts were prepared herein with a phenyl substi-
tuent on the pendent amine (1d and 2d). Each complex (1a–d,
2a and 2d) was prepared by coordination of a PR2NR′2 ligand to
the metal precursors [Ru(Cp)(MeCN)3]PF6 or [Ru(Cp*)(MeCN)3]
are co-planar with the lone pair of the amine of the Pt-Bu2NPh
2
ligand, which suggests the lone pair is delocalized into the
π-system. This is supported by the planar geometry found for
N3 (sum of bonding angles = 359.05°). Second, the two six-
membered metallocycles comprised of Ru and the Pt-Bu2NPh
2
ligand are both in a boat conformation, positioning one
Fig. 2 Thermal displacement plot of 2d with ellipsoids at 50% prob-
ability. t-Butyl groups on P1 and P2, hydrogen atoms (except those on
C2, C8 and C9), along with the [PF6]− counterion were removed for
clarity. Bond lengths (Å): P1–Ru1 = 2.306(1); P2–Ru1 = 2.306(1); N1–
Ru1 = 2.052(4). Bond angles (°): P1–Ru1–P2 = 77.75(4); C13–N2–C14 =
108.5(3); C13–N2–C25 = 119.3(3); C14–N2–C25 = 119.3(3); C15–N3–
C16 = 108.2(3); C15–N3–C31 = 124.9(3); C16–N3–C31 = 124.9(3).
Scheme 2 Synthesis of complexes 2a–e and 3a. (i) 1.05 eq. PR2NR’
MeCN, 70 °C, 4 h.
,
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
pendent amine (N3) close to the bound acetonitrile ligand (i.e. at 55 °C in THF with catalytic loadings of 0.1, 0.5, 1, and
the active site of the catalyst). This conformation is uncommon 3 mol%. Incomplete conversion to IC was observed with all
for solid-state structures of [Ru(Cp/Cp*)(PR2NR′2)(L/X)]+ com- catalysts and loadings. Despite the moderate catalyst perform-
plexes (L = MeCN with PF6−; X = Cl), which generally crystallize ance, trends with respect to ligand structure are evident. The
with this proximal metallocycle in a chair conformation, with relative activity of the four catalysts, 1a–c and 2a, is consistent
the tertiary amine positioned away from the active site. A at each loading, but the distinctions between the catalysts is
similar boat conformation has previously only been observed most evident from the 1 mol% data (Fig. 3b, red bars).
in cases where the amine is engaged in hydrogen bonding Consistent with a previous study,6b the R = t-Bu derivative 1b
with a metal-bound substrate. In the case of 2d, a stabilizing gave approximately twice the yield of IC as compared to the
C–H/π hydrogen-bonding interaction11 between the aceto- R = Ph complex 1a. The benzyl derivative 1c has the lowest per-
nitrile methyl group and the aryl of the amine substituent is formance giving only 12% conversion. The performance trend
proposed. The distance between the plane of the aryl ring and of t-Bu > Ph > Bn reveals that the R substituent electronic pro-
H2A is 2.74(1) Å, which is within the 3.05 Å distance that is the perties are not the dominant factor in catalyst activity under
typical maximum for such interactions.11a Additionally, the these conditions since IC yield does not track with the
angle between the centre of the aryl ring, H2A and C2 is expected phosphine donor strength. Rather, increased sterics
124.0(2)°, which is also within the expected range of about the metal centre appears to be critical for high catalyst
112–168°.11b The relatively close distances between the distal performance. This is further supported by a comparison of the
pendent aryl amine and the C–H groups of the Cp* ligand two catalyst derivatives with different ancillary ligands Cp (1a)
suggest that two weaker CH/π interactions may also exist.
and Cp* (2a). The catalyst with the larger Cp* ligand, 2a, gave
ca. twice the yield of IC as compared to 1a.
Catalytic studies
Together, the above screen of the primary-coordination
sphere ligand effects and our prior study of the secondary
coordination sphere, indicate that the optimal ligand set
Catalyst screen for the cyclization of 2-ethynylbenzyl alcohol
under mild conditions. The performance of Cp (1a–c) and
Cp* (2a) Ru-(PR2NR′2) catalysts was first evaluated in the cycli-
zation of 2-ethynylbenzyl alcohol (EBA) to give the 6-membered
heterocycle isochromene (IC; Fig. 3a). Catalysis was conducted
should include Cp*, and a PR2NR′ ligand with R = t-Bu and R′
2
= Ph or Bn. These R′ substituents are expected to give similar
results based on our prior study and, due to the easier syn-
thetic accessibility of the PtBu2NPh ligand, we elected to use
2
complex 2d (i.e. [Ru(Cp*)(PtBu2NPh2)(MeCN)]PF6). The perform-
ance of this catalyst was monitored over time during the cycli-
zation of EBA (Fig. 3c). Additionally, 2d was compared to direct
analogues with either a Cp ligand (1d) or R = Ph substituent
(2a). Notably, catalyst 2d reached a conversion maximum of
62% isochromene within only 1 h. In contrast, the Cp catalyst
1d did not reach a plateau in conversion within 6 h, and the IC
yield at this point is only 31%. This suggests that the Cp* ana-
logue promotes catalyst initiation at a lower temperature than
the Cp derivative. Unsurprisingly, the weak C–H/π interaction
observed in the solid-state structure of 2d does not hinder
MeCN lability. The Cp* and R = Ph catalyst 2a likewise
approaches a conversion plateau within 1 h, but the max con-
version (47%) is lower than that of 2d. The rapid catalysis with
2d at 55 °C suggests that higher loadings could afford IC in
high yield and short reaction times. Indeed, an increase in the
loading of 2d to 2 mol% afforded IC in ca. 90% yield within
10 min (Fig. 3c). This is an outcome that was previously unat-
tainable with other Ru-(PR2NR′2) catalysts due to a combination
of slow catalysis at low temperatures and competitive catalyst
deactivation at higher temperatures.6 The 10 min reaction
time is also significantly faster than reported catalysts for EBA
cyclization that have reaction times of 1–24 h.3b,12
Fig. 3 (a) Cyclization of 2-ethynylbenzyl alcohol (EBA) to give iso-chro-
mene (IC). (b) Yields of IC after 24 h in THF at 55 °C using 1a–c and 2a at
0.1 (purple), 0.5 (orange), 1 (red), 3 (blue) mol%. Conversion data with
3 mol% 1c was not achieved in this screen due to inaccurate catalyst
addition due to low solubility of 1c in the stock solution. (c) Time trace
of cyclization of EBA in THF at 55 °C by 1 mol% (solid) 1d (blue), 2a
(purple), 2d (red) and 2 mol% (dashed) 2d (red).
Self-exchange study to examine catalyst initiation. The faster
catalysis with the Cp* catalyst 2d relative to the Cp analogue
1d could be due to faster catalyst initiation for 2d (i.e. MeCN
dissociation; I → II in Scheme 1) at low temperatures. To gain
direct evidence of the differences in MeCN lability with Cp*
versus Cp catalysts, the rate constants for self-exchange were
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
determined through magnetization transfer experiments.13
Catalysts 1a and 2a were selected for this study due to the
acceptable chemical shift difference between Ru-bound and
free MeCN (Scheme 3). The rate constant for 2a is greater than
that of 1a by a factor of 1.3. While the difference is small, it
does confirm that the barrier to acetonitrile dissociation is
lower for 2a and this is likely due to the larger steric profile of
Cp* vs. Cp.
Catalyst screen in the cyclization of 2-ethynyaniline. The
catalytic activity of complexes 1a–c and 2a was screened in the
cyclization of 2-ethynylaniline (EA) to give the 5-membered
heterocycle indole (Ind; Fig. 4). Catalysis was conducted at
55 °C in THF with catalytic loadings of 0.1, 0.5, 1, and 3 mol%.
Catalysts 1b and 2a achieved quantitative conversion at loadings
≥0.5 mol%, which shows that cyclization to give the 5-mem-
Fig. 5 Cyclization of 2-ethynylaniline (EA) in Me-THF with 0.5 mol% Cp
catalyst 1d (blue) and Cp* catalyst 2d (red) at 40 (solid), 55 (dashed) and
70 °C (dotted).
bered Ind is much easier than the 6-membered IC. A distinction elevated temperatures (70 °C) with 1a was possible.6a Thus, in
in catalyst performance is evident at 0.1 mol% and the trend contrast to EBA cyclization, catalyst deactivation via complex B
(2a > 1b > 1a > 1c) is the same as that observed with EBA.
does not inhibit high temperature cyclization of EA. The temp-
Informed by the reaction profiles for EBA cyclization erature dependence of Cp and Cp* catalysts 1d and 2d was
(Fig. 3c), we hypothesized that the higher yield of Ind with the evaluated in the cyclization of EA at 40, 55 and 70 °C (Fig. 5).
Cp* catalyst 2a as compared to the Cp analogue 1a was a con- At 40 °C the Cp catalyst 1d exhibits very poor performance
sequence of the different initiation rates of the catalysts at reaching only 9% conv. by 6 h. In contrast, 2d reaches 55%
55 °C. We have previously identified that cyclization of EA at Ind, but the reaction profile shows catalysis is quite slow. At
55 °C the trend in activity is very similar to that found for cycli-
zation of EBA, including that both catalysts reach about the
same product yield by 24 h (>95% Ind for 1d and 2d, ESI†).
Consistent with our previous study of the 2° coordination
sphere, the 24 h conversion for 1d is the same as catalyst 1b
(cf. Fig. 4b), which has the same structure except with a benzyl
substituent on the pendent amine. But, the 6 h %Ind values of
22 and 84% for 1d and 2d, respectively, indicate that the Cp
catalyst 1d is significantly slower. Increasing the temperature
to 70 °C affords complete conversion to Ind with both catalysts
within 10 min. This data reinforces that the Cp* catalyst has
more facile entry into the catalytic cycle at lower temperatures
due to higher lability of the labile placeholder ligand.
Scheme 3 Self-exchange of acetonitrile and kexchange values for 1a and
2a as determined by magnetization-transfer experiments at 25 °C.
The cyclization of EA was conducted at 70 °C with catalysts
1a–c to compare phosphine substituent effects under con-
ditions where MeCN lability is expected to be high for all cata-
lysts. With 0.1 mol% catalyst, Ind yields of 30, 7 and 3% were
obtained for 1a, 1b and 1c, respectively, after a 24 h reaction
time. At this higher temperature the trend in performance
with respect to R substituent is Ph > t-Bu > Bn, which is
different than the trend at lower temperature (cf. 55 °C: t-Bu >
Ph > Bn). While further mechanistic analysis is needed, this
70 °C trend may be indicative of optimal phosphine donor
strength for steps within the catalytic cycle (II → III → IV → II
in Scheme 1), rather than those needed to promote initiation
(I → II in Scheme 1).
The rapid and high yield catalysis with 0.5 mol% of the Cp
and Cp* catalysts 1d and 2d at 70 °C (Fig. 5, dotted lines)
suggest that the catalyst loading could be lowered even further.
Indeed, with 0.2 mol% 1d and 2d ca. 95% and 70% Ind was
formed, respectively, within 2 h (Fig. 6). The 0.1 mol% data
Fig. 4 (a) Cyclization of 2-ethynylaniline (EA) to give indole (Ind). (b)
Yields of Ind in THF at 55 °C using 1a–c and 2a at 0.1 (purple), 0.5
(orange), 1 (red), 3 (blue) mol%. Conversion data with 3 mol% 1c was not
revealed that the TONs for 1d and 2d are 800 and 330, respect-
achieved in this screen due to inaccurate catalyst addition due to low
solubility of 1c in the stock solution.
ively, which indicate that the lifetime of the Cp catalyst is
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
Table 1 Tolerance screen using additives for the cyclization of 2-ethy-
nylaniline using complex 1da
Entry
Additive
Ind (%)
%changeb (%)
1
2
3
4
5
6
7
8
None
40
40
40
41
35
43
37
36
37
5
6
4
5
7
13
31
20
30
31
28
—
0
0
+3
−13
+8
−8
−10
−8
−88
−85
−90
−88
−83
−68
−23
−50
−25
−23
−30
Fig. 6 Cyclization of 2-ethynylaniline (EA) in Me-THF at 70 °C with Cp
catalyst 1d (blue) and Cp* catalyst 2d (red) at loadings of 0.2 (solid) and
0.1 (dashed) mol%.
Dimethyl terephthalate
PhF, PhCl, PhBr, PhI
Ph2CHCOOH
Benzophenone
Benzyl alcohol
DMF
Styrene
approximately double that of the Cp* catalyst. Given the pro-
posed higher MeCN lability with the latter, this could be a con-
sequence of the instability of the low-coordinate active catalyst
(i.e. II in Scheme 1). The turnover number for 1d of 800 is ca.
100 and 20 times greater than known non-MLC RuCl(Cp)
9
Diphenylacetylene
Benzyl amine
Benzonitrile
NaI
Phenyl acetylene
1,2-Ethandithiol
K2CO3
10
11
12
13
14
15
16
17
18
19
20
3c
(PPh3)2/AgBF4 and MLC [RuCl(Cp)(L)2(MeCN)]PF6 (L = P(2-
P-6-tBu-py)Ph2)3d catalysts, respectively.
NEt3
Benzaldehyde
Trifluoromethylbenzene
Nitrobenzene
p-Tolylboronic acid
Catalyst robustness and scope studies. The robustness14 of
the long-lifetime catalyst 1d toward a variety of functional
groups was evaluated by conducting the cyclization of EA in
the presence of a group of additives (Table 1). In all cases,
cyclization was conducted with 0.5 mol% 1d in the presence of
1 equivalent of additive relative to substrate and the conver-
sion was evaluated after 15 h. A control reaction without addi-
tive afforded 40% Ind (Table 1, entry 1), which is an ideal
value to observe either a significant decrease or increase in
yield in the presence of an additive. A negligible impact on
a Conditions: 2-Ethynylaniline (150 mM), additive (150 mM), and tetralin
(50 mM), as an internal standard, in THF at 55 °C. Samples were analysed
after 15 h and in situ %conv. values were determined by calibrated
GC-FID. b %change = ((%IndAdditive − %IndControl)/%IndControl) × 100.
The scope of catalyst 1d was evaluated in the cyclization of
yield (<13% change relative to control, which is within 5% a variety of ortho-alkyne aniline substrates (Table 2). For each
Ind yield) was observed with: dimethyl terephthalate, PhX substrate, cyclization was conducted at 70 °C with both 0.5
(X = F, Cl, Br, I), diphenylacetic acid, benzophenone, benzyl and 1 mol% 1d. At the 0.5 mol% loading, the conversion of all
alcohol, dimethylformamide, styrene and diphenyl acetylene the substrates was lower than for the benchmark substrate EA
(entries 2–9). This indicates that catalyst 1d is tolerant of: (entry 1) indicating that any substitution on the parent sub-
esters, aryl halides, carboxylic acids, ketones, alcohols, strate results in a decrease in both conversion and rate.
amides, alkenes and internal alkynes, all groups that are Electron withdrawing (CF3) and donating (OMe) groups para to
useful for downstream functionalization. A number of addi- the alkyne gave 42 and 58% product yields, respectively
tives were significantly detrimental to catalyst activity (>50% (entries 2 and 4). The non-linear relationship with alkyne elec-
change in %Ind relative to control), including: benzyl amine, tronics (CF3 vs. H vs. OMe) may be a function of a change in
benzonitrile, sodium iodide, phenyl acetylene, 1,2-ethane- rate-determining step or a consequence of competing catalyst
dithiol and potassium carbonate (entries 10–15). When deactivation with the CF3 functional group (see Table 1,
employed for cyclization, catalyst 1d is not compatible with entry 18). Effective cyclization of both substrates was achieved
functional groups that have a propensity to bind competitively by increasing the loading to 1 mol% giving the CF3 and OMe
to the metal centre, such as: primary amines, nitriles, halide products in 74 and 80% yield, respectively (entries 3 and 5).
salts, terminal alkynes and thiols. A smaller, but still negative Catalyst performance is less sensitive to withdrawing (F and
impact (15–50% change relative to control) on catalysis is CF3) and donating (OMe) groups para to the amine giving
observed for the additives: potassium carbonate, triethyl- yields of 85, 42, 69%, respectively with 0.5 mol% 1d (entries 6,
amine, benzaldehyde, trifluoromethylbenzene, nitrobenzene 8 and 10). The different values for the two electron-withdraw-
and p-tolylboronic acid (entries 15–20). Unfortunately, 1d has ing substituents is likely a consequence of the incompatibility
limited compatibility with aldehyde, nitro, tri-fluoromethyl of the catalyst with the CF3 functional group (Table 1,
and boronic acid functional groups. The poor compatibility of entry 18). The ca. 15% lower yield with the OMe- vs.
1d toward the bases K2CO3 and NEt3 is notable given that F-substituted derivative is expected based on our prior study of
these are commonly employed in synthesis and could be catalyst secondary-coordination sphere effects.6a The electron
present as contaminants.
donating OMe group increases the basicity of the substrate
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 2 Substrate screen to examine different electronic and steric pro-
perties for the intramolecular cyclization of alkynes using complex 1da
with the Ac-protected substrate are both due to the attenuated
nucleophilicity of the substrate amine. Attempted cyclization
of 2-(prop-2-ynyl)-aniline with 0.5 or 1 mol% 1d resulted in
minor substrate consumption and trace amounts of 6-mem-
bered heterocycles dihydroisoquinoline and isoquinoline
(entries 18 and 19). The latter is likely formed following dehy-
drogenation15 of the expected cyclization product dihydroiso-
quinoline. While the PR2NR′2 catalyst 1d reported here operates
at a lower loading than previously reported catalysts,3c,d the
poor performance of 1d toward the 6-membered N-heterocycle
product is a current limitation of this catalyst family.
Conv. (%)
Entry
1
Substrate
Loading 1d (mol%)
1 h
6 h
0.5
100
100
Experimental
Reagents and general procedures
2
3
0.5
1
7
—
42
74
All air- and water-sensitive reactions were manipulated under
Ar or N2 using standard Schlenk or glovebox techniques,
respectively, unless otherwise stated. All glassware was oven
dried prior to use. Benzylamine (>98%), and triphenyl-
phosphine oxide (99%) was obtained from Alfa Aesar.
Triethylamine (99%) was obtained from Caledon Laboratory
Chemicals. Pyrene (98%), benzyl chloride (99%), 2-ethynylani-
line (98%), 2-ethynylbenzyl alcohol (95%), tetralin (1,2,3,4-tet-
rahydronaphthalene; 99%), and 2-methyltetrahydrofuran (Me-
THF; >99% anhydrous) were obtained from Sigma-Aldrich.
Chloroform-d1 (99.8%), acetonitrile-d3 (99.8%), toluene-d8
(99.5%), benzene-d6 (99.6%), and dichloromethane-d2 (99.8%)
were obtained from Cambridge Isotope Laboratories. 4-Fluoro-
2-iodoaniline (97%), 4-amino-3-bromobenzotrifluoride (95%),
3-amino-4-bromobenzotrifluoride (95%), 2-bromo-4-methoxy-
aniline (95%) were obtained from Oakwood Chemicals.
4
5
0.5
1
39
_
58
80
6
7
0.5
1
57
—
85
96
8
9
0.5
1
27
—
42
93
10
11
0.5
1
42
—
69
99
12
13
14
0.5
1
5
3
9
20 (24 h)
78
[Ru(Cp)(MeCN)3]PF6,16 [Ru(Cp)(PPh2NBn2)(MeCN)]PF6 (1a),6b
18
[Ru(Cp)(PtBu2NBn2)(MeCN)]PF6 (1b),17 PtBu2NPh
,
2-iodo-m-
15
16
17
0.5
1
5
13
—
—
17
30
97
2
anisidine,19 2-(prop-2-ynyl)-aniline,20 2-iodo-N-tosylaniline,21
and 2-iodo-N-acetylaniline,22 were synthesized following litera-
ture procedures. Dry and degassed tetrahydrofuran (THF),
diethyl ether, toluene, dichloromethane (DCM), hexanes,
pentane and acetonitrile (MeCN) were obtained from an
18
19
0.5
1
0
—
3b
6b
a Conditions: Substrate (150 mM) and trimethoxybenzene (25 mM), as Innovative Technology 400–5 Solvent Purification System and
an internal standard, with 1d in Me-THF at 70 °C. Samples were moni-
stored under N2. These dry and degassed solvents, except for
tored by 1H NMR spectroscopy and in situ %conv. values are listed.
b Mixture of dihydroisoquinoline and isoquinoline in a 2 : 1 and 1 : 1
ratio at 0.5 and 1 mol% respectively.
MeCN, were stored over 4 Å molecular sieves (Fluka and acti-
vated at 150 °C for over 12 h). Acetone was dried with Cs2CO3
and degassed by bubbling with N2. THF was further dried with
CaH2 and distilled under Ar. Triethylamine and ethanol were
amine relative to the pendent amine of the catalyst, which dried with 4 Å molecular sieves and degassed by bubbling with
negatively impacts proton-transfer steps. The moderate yields N2. Chloroform-d1 was dried with 4 Å molecular sieves and
with each of these three substrates were improved to >95% by degassed by bubbling with N2. Benzylamine was dried with
increasing the catalyst loading to only 1 mol% (Table 2, entries NaOH, distilled under vacuum and stored under N2. All other
7, 9 and 11). Mono-protection of the amine with acetyl or tosyl chemicals were used as received.
groups resulted in poor yields with both 0.5 (Ac: 9%; Ts: 17%)
Charge-transfer Matrix Assisted Laser Desorption/
or 1 (Ac: 20%; Ts: 30%) mol% 1d (entries 12, 13, 15 and 16, Ionization (MALDI) mass spectra were collected on an AB Sciex
respectively). Increasing the loading to 5 mol% 1d afforded the 5800 TOF/TOF mass spectrometer using pyrene as the matrix
Ac- and Ts-protected substrates in excellent yields of 78 and in a 20 : 1 molar ratio to metal complex. Samples were spotted
97%, respectively (entries 14 and 17). We expect that the need on the target plate as solutions in DCM. All NMR spectra were
for higher catalyst loadings and the comparatively lower yield recorded on either an Inova 400 or 600 MHz, or Mercury/
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
1
Bruker 400 MHz instrument. H and 13C{1H} spectra acquired signal), −144.3 (sept, 1JP–F = 711.2 Hz, PF6). MALDI MS (pyrene
in CDCl3 were referenced internally against the residual matrix): Calc. m/z 677.2 [Ru(Cp)(PBn2NBn2)]+, Obs. m/z 677.2.
solvent signal (CHCl3) to TMS at 0 ppm. 31P{1H} spectra were
[Ru(Cp)(PtBu2NPh2)(MeCN)]PF6 (1d). Yield: 90%. 1H
referenced externally to 85% phosphoric acid at 0.00 ppm. (400 MHz, CD2Cl2) δ: 7.35–7.26 (m, Ph–H, 4H), 7.01–6.85 (m,
Infrared spectra were collected on solid samples using a Ph–H, 6H), 4.87 (s, Cp–H, 5H), 3.93–3.85 (m, CH2, 2H),
PerkinElmer UATR TWO FTIR spectrometer. Quantification of 3.74–3.59 (m, CH2, 4H), 3.58–3.51 (m, CH2, 2H), 2.03 (s,
catalytic reactivity was achieved using an Agilent 7890a gas NCCH3, 3H), 1.46–1.29 (m, PC(CH3)3, 18H). 31P{1H} (162 MHz,
1
chromatography with a flame ionization detector (GC-FID), CD2Cl2) δ: 55.3 (s, PtBu), −144.3 (sept, JP–F = 711.2 Hz, PF6).
3
fitted with a HP-5 column. Calibration curves for 2-ethynylben- 13C{1H} (101 MHz, CD2Cl2) δ: 152.8 (t, JC–P = 6.1 Hz, N–CAr),
3
zyl alcohol, isochromene, 2-ethynylaniline, indole, and were 151.7 (t, JC–P = 5.1 Hz, N–CAr), 130.4 (s, CAr), 130.3 (s, CAr),
prepared to determine the response factors relative to tetralin. 128.7 (s, Ru–CNCH3), 121.6 (s, CAr), 121.0 (s, CAr), 117.4 (s, CAr),
1
The amount of each species was quantified, relative to the 116.9 (s, CAr), 80.4 (s, Cp), 48.8 (d, JC–P = 17.2 Hz, P–CH2–N),
1
1
internal standard (tetralin), using area counts corrected with 48.6 (d, JC–P = 16.2 Hz, P–CH2–N), 47.0 (d, JC–P = 14.1 Hz,
the response factors.
1
P–CH2–N), 46.8 (d, JC–P = 14.1 Hz, P–CH2–N), 35.7 (ABX,
PC(CH3)3), 27.2 (s, PC(CH3)3), 4.3 (s, CH3CN). Anal. Calc. for
C31H44F6N3P3Ru: C, 48.56; H, 5.78; N, 5.48. Found: C, 48.89;
H, 6.05; N, 5.48. MALDI MS (pyrene matrix): Calc. m/z 581.2
Synthetic procedures
Synthesis of PBn2NBn2. This procedure was based on the syn- [Ru(Cp)(PtBu2NPh2)]+, Obs. m/z 581.2.
thesis of PMe2NPh
.
Trihydroxymethylphosphine (THP)
[Ru(Cp*)(PPh2NBn2)(MeCN)]PF6 (2a). Yield: 99%. 1H
23
2
(0.976 g, 7.87 mmol, 2 equiv.) was added to a 100 mL Schlenk (600 MHz, CD2Cl2) δ: 7.57–7.37 (m, Ph–H, 16H), 7.24–7.18 (m,
flask with a stir bar, and THF (ca. 5 mL) was added by Ph–H, 2H), 6.94–6.90 (m, Ph–H, 2H), 4.07–4.05 (m, Ph–CH2–N,
cannula. The solution was cooled to −40 °C and BnCl 2H), 3.53–3.47 (m, Ph–CH2–N and P–CH2–N, 4H), 3.18–3.12
(8.84 mmol, 2.2 equiv.) was added dropwise by syringe whilst (m, P–CH2–N, 2H), 2.83–2.74 (m, P–CH2–N, 2H), 2.46 (s,
4
stirring. The reaction was left to warm to room temperature NCCH3, 3H), 1.35 (t, JH–P = 1.92 Hz, Cp–CH3, 15H). 31P{1H}
1
overnight while stirring. The solvent was removed under (243 MHz, CD2Cl2) δ: 30.7 (s, PPh), −144.4 (sept, JP–F = 712.0
vacuum and NEt3 (∼20 mL) was added by cannula to the trans- Hz, PF6). 13C{1H} (151.5 MHz, CD2Cl2) δ: 136.9 (s, CAr–CH2N),
1
3
parent oil. The solution was left to stir at room temperature for 135.6 (s, CAr–CH2N), 132.3 (t, JC–P = 19.6 Hz, JC–P = 19.6 Hz,
2
72 h. The reaction was filtered via cannula to remove N–CAr), 131.1 (s, CAr), 130.6 (d, JC–P = 5.2 Hz, CAr), 130.6 (d,
NEt3·HBr. The remaining NEt3 was removed from the filtrate 2JC–P = 5.2 Hz, CAr), 129.9–129.7 (m, CAr), 129.1 (s, CAr), 129.0 (s,
under vacuum. To the resulting transparent oil was added CAr), 128.7 (s, CAr), 128.2 (s, CAr), 126.5 (s, Ru–CNCH3), 93.1 (s,
3
3
EtOH (50 mL) and BnNH2 (8.26 mmol, 2.1 equiv.) by cannula Cp–CH3), 66.8 (t, JC–P = 12.7 Hz, Ph–CH2–N), 66.5 (t, JC–P
=
1
and syringe, respectively. The reaction was heated at reflux and 11.0 Hz, Ph–CH2–N), 54.8 (d, JC–P = 17.5 Hz, P–CH2–N), 54.7
stirred for 24 h, at which point the reaction was cooled to (d, JC–P = 18.8 Hz, P–CH2–N), 47.4 (d, JC–P = 22.4 Hz, P–CH2–
room temperature and left to stir for an additional 48 h N), 47.2 (d, JC–P = 25.4 Hz, P–CH2–N), 10.0 (s, Cp–CH3), 4.9 (s,
1
1
1
without precipitation occurring. EtOH was removed under CH3CN). MALDI MS (pyrene matrix): Calc. m/z 719.2 [Ru(Cp)
vacuum to produce a white residue. The residue was dissolved (PPh2NPh2)]+, Obs. m/z 719.2.
in acetonitrile and cooled to −35 °C to force precipitation after
[Ru(Cp*)(PtBu2NPh2)(MeCN)]PF6 (2d). Yield: 95%. 1H
one week. The solution was decanted to isolate the white pre- (400 MHz, CD2Cl2) δ: 7.35–7.24 (m, Ph–H, 4H), 7.04–6.95 (m,
cipitate which was dried under vacuum for 24 h. Crude yield: Ph–H, 3H), 6.93–6.84 (m, Ph–H, 3H), 3.82–3.73 (m, CH2, 4H),
7%. 31P {1H} (400 MHz, CDCl3) δ: 60.0 (broad, P).
3.65–3.55 (m, CH2, 2H), 3.28–3.21 (m, CH2, 2H), 1.95 (s,
NCCH3, 3H), 1.77 (t, JH–P = 1.52 Hz, Cp–CH3, 15H), 1.41–1.35
4
General procedure for the synthesis of [Ru(Cp/Cp*)(PR2NR′
)
2
(MeCN)]PF6 complexes. To a 100 mL Schlenk flask with a stir (m, PC(CH3)3, 18H). 31P{1H} (162 MHz, CD2Cl2) δ: 40.7 (s, PtBu),
bar, [Ru(Cp)(MeCN)3]PF6 or [Ru(Cp*)(MeCN)3]PF6 (1 equiv.), −144.5 (sept, 1JP–F = 709.6 Hz, PF6). 13C{1H} (101 MHz, CD2Cl2)
ligand PR2NR′ (1.05 equiv.) and acetonitrile (20 mL) was δ: 172.6 (s, Ru–CNCH3), 154.1 (t, JC–P = 7.6 Hz, N–CAr), 151.8
3
2
3
added. The flask was heated to 65 °C for 4 hours with stirring. (t, JC–P = 6.1 Hz, N–CAr), 130.4 (s, CAr), 130.2 (s, CAr), 121.5 (s,
The solvent was removed under vacuum and the remaining CAr), 120.4 (s, CAr), 118.8 (s, CAr), 115.9 (s, CAr), 90.1 (s, Cp–
1
1
solid was triturated with pentane (3 × 2 mL). Acetonitrile CH3), 49.1 (d, JC–P = 17.2 Hz, P–CH2–N), 48.9 (d, JC–P = 17.2
(2 mL) was added and the resulting suspension was filtered. Hz, P–CH2–N), 48.7 (d, 1JC–P = 14.1 Hz, P–CH2–N), 48.5 (d, 1JC–P
2
The solid was washed with acetonitrile until the washings were = 13.1 Hz, P–CH2–N), 35.6 (ABX, PC(CH3)3), 27.4 (t, JC–P = 2.0
colourless. The solvent volume of the filtrate was reduced Hz, PC(CH3)3), 12.5 (s, CH3–Cp), 4.3 (s, CH3CN). Anal. Calc. for
under vacuum to ca. 0.5 mL and diethyl ether (5 mL) was C36H54F6N3P3Ru: C, 51.67; H, 6.50; N, 5.02. Found: C, 51.37;
added to precipitate the product. The solvent was decanted off H, 6.86; N, 5.39. MALDI MS (pyrene matrix): Calc. m/z 651.3
and the product was dried under vacuum.
[Ru(Cp*)(PtBu2NPh2)]+, Obs. m/z 651.3.
General procedure A for Sonogashira (X = I). The synthesis
[Ru(Cp)(PBn2NBn2)(MeCN)]PF6 (1c). Yield: 87%. 1H NMR
complicated by the presence of multiple conformers (see ESI† followed a modified procedure based on a literature method.24
for spectrum) 31P{1H} (162 MHz, CD2Cl2) δ: 36.6 (s, PBn, major To a 200 mL Schlenk flask, Pd(PPh3)2Cl2 (equiv.), CuI (0.05
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
equiv.), 2-iodoaniline derivative (500 mM,
1
equiv.) in matched literature values.28 2-Ethynyl-5-trifluoromethylaniline:
1
THF : NEt3 (1 : 1). Trimethylsilyl acetylene (1.25 equiv.) was Sonogashira B; Deprotection A; the H NMR spectrum matched
added to the reaction by syringe at room temperature while literature values.29 2-Ethynyl-N-tosylaniline: Sonogashira A,
stirring resulting in almost immediate colour change. The Deprotection B; the 1H NMR spectrum matched literature
reaction was left for 4 h and an aliquot was checked by 1H values.21 2-Ethynyl-N-acetylaniline: Sonogashira A, Deprotection A;
NMR spectroscopy to ensure reaction had proceeded to com- the 1H NMR spectrum matched literature values.28
pletion. Upon completion, the reaction was filter through a
Catalytic procedures
plug of silica under air and the filtrate was evaporated to
dryness. The residue was dissolved in ethyl acetate (20 mL)
General procedure for the catalytic cyclization of substrates.
and washed with brine (3 × 40 mL). The product was dried In a glovebox, the following stock solutions were prepared:
with Na2SO4, filtered, and then dried under vacuum. The 2-ethynylaniline (246 mg, 2.10 mmol, 0.300 M) and tetralin
product was then subsequently used in the next step without (185 mg, 1.4 mmol, 0.2 M) in THF (14.00 mL); 1a (10 mg,
further purification.
0.012 mmol, 6 mM) in THF (2.00 mL); 1b (10 mg, 0.013 mmol,
General procedure B for Sonogashira (X = Br). The synthesis 6 mM) in THF (2.10 mL); 1c (10 mg, 0.012 mmol, 6 mM) in
followed a modified procedure based on a literature method.24 THF (1.93 mL); 1d (10 mg, 0.013 mmol, 6 mM) in THF
To a 200 mL Schlenk flask, Pd(PPh3)2Cl2 (equiv.), CuI (0.05 (2.17 mL); 2a (10 mg, 0.011 mmol, 6 mM) in THF (1.84 mL);
equiv.), 2-iodoaniline derivative (500 mM,
1 equiv.) in 2d (10 mg, 0.012 mmol, 6 mM) in THF (1.99 mL). Five sets
THF : NEt3 (1 : 1). Trimethylsilyl acetylene (1.25 equiv.) was (A–F) of five 4 mL vials (30 vials total) containing stir bars were
added to the reaction by syringe at room temperature while charged with the 2-ethynylaniline/tetralin stock solution
stirring resulting in almost immediate colour change. The (250 μL) and additional THF (125 µL). To each vial was added
reaction was refluxed at 90 °C for 24 h when an aliquot was catalyst stock solution (125 µL, set A = 1a, B = 1b, C = 1c, D =
1
checked by H NMR spectroscopy to ensure reaction had pro- 1d, E = 2a, F = 2d) giving a final volume of 500 μL. The final
ceeded to completion. Upon completion, the reaction was concentrations for all vials were 0.150 M in substrate and
filter through a plug of silica under air and the filtrate was 0.75 mM in catalyst. A final vial was charged with substrate/
evaporated to dryness. The residue was dissolved in ethyl internal standard stock solution (100 μL) for use as the time =
acetate (20 mL) and washed with brine (3 × 40 mL). The 0 sample, required for accurate quantification of substrate and
product was dried with Na2SO4, filtered, and then dried under product. The vials were capped and removed from the glove
vacuum. The product was then subsequently used in the next box and heated to 55 °C (sets A–F) with stirring. After 0.167,
step without further purification.
0.5, 1, 2, 6, and 24 hours one vial from each of the sets was
Deprotection procedure A. In a 200 mL round bottom under removed from heat, cooled, and exposed to air to quench. A
air, the TMS-protected substrate was deprotected in MeOH 20 μL aliquot was diluted to 3 mM (0.980 μL) in acetonitrile
(50 mL) using K2CO3 (1 equiv.). The reaction was stirred for and analyzed by GC-FID. A 10 μL aliquot of the T0 sample was
1 h. The solvent was evaporated to dryness. The residue was diluted with acetonitrile (990 μL) and analyzed by GC-FID.
dissolved in ethyl acetate (20 mL) and washed with water (3 ×
High-throughput catalytic procedure. A representative pro-
30 mL). The substrate was evaporated to dryness. Further puri- cedure is given for 2-ethynylaniline. In a glovebox, the follow-
fication was performed using a silica gel flash column using a ing stock solutions were prepared: 2-ethynylaniline (435 mg,
gradient ethyl acetate–hexane solvent system increasing from a 3.72 mmol, 0.300 M) and tetralin (328 mg, 2.48 mmol,
ratio of 1 : 99.
0.200 M) in THF (12.390 mL). Stock solutions of catalysts
Deprotection procedure B. In a 200 mL Schlenk flask, the (9 mM and 1.5 mM) were prepared as above. Reaction com-
TMS-protected substrate was dissolved in THF (30 mL). The ponents were added to a cooled (0 °C) 8 × 12 reaction plate in
reaction mixture was cooled to −25 °C and TBAF (1 M, the following order: catalyst, solvent, then substrate. Stock
1 equiv.) was added dropwise whilst stirred. The reaction was solutions of catalysts were robotically dispensed to their appro-
left to warm to room temperature and then quenched with priate concentration amounts: 0.15, 0.75, 1.50, and 3.00 mM
water (1 mL). The solvent was evaporated to dryness. The (0.1, 0.5, 1, 3 mol%). Solvent and substrate were added by
residue was dissolved in ethyl acetate (20 mL) and washed with Eppendorf pipette to the well plate and to a T0 sample. Final
water (3 × 30 mL). The substrate was evaporated to dryness. conditions: 150 mM Substrate, 0.1/0.5/1/3 mol% catalyst,
Further purification was performed using a silica gel flash 100 μL reaction volume in THF. The 96 well plate was sealed
column using a gradient ethyl acetate–hexane solvent system with a Teflon sheet, a rubber sheet and an aluminium cover,
increasing from a ratio of 1 : 99.
Substrates. 2-Ethynyl-4-fluoroaniline:
to minimize evaporation, and the plate was heated to 55 °C for
A; 24 h. After the plate had cooled, the solutions were daughtered
Sonogashira
Deprotection A; the 1H NMR spectrum matched literature into a second plate and diluted to 2.5 mM (based on the start-
values.25 2-Ethynyl-4-methoxyaniline: Sonogashira B; Deprotection ing concentration of 2-ethynylaniline) in acetonitrile for
A; the 1H NMR spectrum matched literature values.26 2-Ethynyl-4- GC-FID analysis. A 10 μL aliquot of the T0 sample was diluted
trifluoromethylaniline: Sonogashira B; Deprotection A; the with acetonitrile (990 μL) and analyzed by GC-FID.
1H NMR spectrum matched literature values.27 2-Ethynyl-5-meth-
General procedure for the catalytic cyclization of scope sub-
oxyaniline: Sonogashira A; Deprotection B; the 1H NMR spectrum strates. A representative procedure is provided for 2-ethynyl-N-
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Dalton Transactions
Paper
tosylaniline. In a glovebox, the following stock solutions were to either the alkyne or amine functionality of the substrate.
prepared: 2-ethynyl-N-tosylaniline (41 mg, 0.15 mmol, Additionally, secondary amine substrates can be cyclized in
0.300 M) and dimethyl terephthalate (5 mg, 0.03 mmol, excellent yield using somewhat higher catalyst loadings. The
0.05 M) in Me-THF (0.50 mL); and 1d (10 mg, 0.013 mmol, elucidated trends in catalyst performance with respect to struc-
6 mM) in Me-THF (2.17 mL). Two 4 mL vials containing stir ture are currently guiding strategies to achieve cyclization of
bars were charged with the 2-ethynyl-N-tosylaniline/tetralin larger ring O- and N-heterocycles.
stock solution (150 μL) and additional Me-THF (75 µL). To
each vial was added catalyst stock solution (75 μL, 1d) giving a
final reaction volume of 300 μL. The final concentrations for
both vials were 0.150 M in substrate and 0.75 mM in catalyst.
Conflicts of interest
A final vial was charged with substrate/internal standard stock The authors declare no competing financial interest.
solution (100 μL) for use as the time = 0 sample, required for
accurate quantification of substrate and product. The vials
were capped and removed from the glove box and heated to
70 °C with stirring on a heating block. After 1 and 6 h, one vial
Acknowledgements
was removed from the heat, cooled, and exposed to air to Roxanne Clément of the Centre for Catalysis Research and
quench. The reaction vials was evaporated to dryness. The Innovation (CCRI) at the University of Ottawa is thanked for
samples were dissolved in CDCl3 and analyzed by 1H NMR her assistance with high-throughput experiments. This work
spectroscopy to quantify substrate and product relative to was financially supported by
a Natural Sciences and
internal standard.
Engineering Research Council (NSERC) of Canada Discovery
Grant and The University of Western Ontario. JMS thanks the
Ontario Graduate Scholarship program for funding. BJB
thanks NSERC for a CGS-M.
Conclusions
A group of [Ru(Cp/Cp*)(PR2NR′2)(MeCN)]PF6 catalysts were pre-
pared in which the nature of the primary coordination sphere
References
– Cp vs. Cp* and PR2NR′ phosphine substituent R – was
2
varied. Catalyst performance was assessed in metal–ligand
cooperative cyclization of alkynyl-alcohol and -amine sub-
strates to give O- and N-heterocycles. At lower temperatures,
catalyst derivatives with a Cp* ancillary ligand and R = t-Bu
phosphine substituent gave the fastest conversion and highest
yields. These trends suggest that the increased steric bulk at
the metal centre increases the lability of the placeholder
MeCN ligand and formation of the active low-coordinate cata-
lyst (i.e. II in Scheme 1). At higher temperatures, both Cp and
Cp* derivatives give very rapid catalysis (i.e. quantitative con-
version of 2-ethynylaniline within 10 min at 70 °C), which
suggests initiation is facile in both cases. The Cp derivative 1d
achieves a turnover number at 70 °C of 800 which is ca. double
that of the Cp* analogue 2d. This is also ca. 20 fold higher
than the previous highest TON for reported catalysts.
Additionally, the trends in R substituent at this temperature
are Ph > t-Bu > Bn. Both of these trends indicated that the less
sterically hindered catalyst has a longer lifetime, which could
be a consequence of suppressed deactivation of the active cata-
lyst due to more facile coordination of a labile ligand to give a
protected adduct (i.e. the MeCN-bound pre-catalyst). Thus, a
more sterically hindered ligand set affords faster catalysis at
low temperatures, but lower lifetimes at high temperatures. A
robustness screen of the longer-lifetime Cp derivative 1d
revealed that the catalyst is not compatible with additives that
can competitively coordinate to the metal, but it is tolerant of
a wide range of functional groups including: alcohols, car-
boxylic acids and aryl halides. A scope analysis indicated that
1d is tolerant of electron donating or withdrawing groups para
1 (a) M. Ishikura, T. Abe, T. Choshi and S. Hibino, Nat. Prod.
Rep., 2013, 30, 694–752; (b) A. J. Kochanowska-Karamyan
and M. T. Hamann, Chem. Rev., 2010, 110, 4489–4497;
(c) N. Majumdar, N. D. Paul, S. Mandal, B. de Bruin and
W. D. Wulff, ACS Catal., 2015, 5, 2329–2366; (d) L. D. Quin
and J. A. Tyrell, Fundamentals of Heterocyclic Chemistry:
Importance in Nature and in the Synthesis of Pharmaceuticals,
John Wiley & Sons Inc., Hoboken, New Jersey, 2010;
(e) O. Gidron and M. Bendikov, Angew. Chem., Int. Ed.,
2014, 53, 2546–2555.
2 (a) S. W. Youn and T. Y. Ko, Asian J. Org. Chem., 2018, 7,
1467–1487; (b) A. Blanc, V. Bénéteau, J.-M. Weibel and
P. Pale, Org. Biomol. Chem., 2016, 14, 9184–9205;
(c) A. V. Gulevich, A. S. Dudnik, N. Chernyak and
V. Gevorgyan, Chem. Rev., 2013, 113, 3084–3213;
(d) N. T. Patil and Y. Yamamoto, Chem. Rev., 2008, 108,
3395–3442; (e) A. S. K. Hashmi, Chem. Rev., 2007, 107,
3180–3211; (f) C. Praveen and P. T. Perumal, Synthesis,
2016, 48, 855–864; (g) C. Bruneau and P. H. Dixneuf, Acc.
Chem. Res., 1999, 32, 311–323; (h) J. A. Varela and C. Saá,
Synthesis, 2016, 48, 3470–3478.
3 (a) B. M. Trost and Y. H. Rhee, J. Am. Chem. Soc., 2002, 124,
2528–2533; (b) P. N. Liu, F. H. Su, T. B. Wen, H. H. Y. Sung,
I. D. Williams and G. Jia, Chem. – Eur. J., 2010, 16, 7889–
7897; (c) A. Varela-Fernández, A. Varela Jesús and C. Saá,
Adv. Synth. Catal., 2011, 353, 1933–1937; (d) R. N. Nair,
P. J. Lee, A. L. Rheingold and D. B. Grotjahn, Chem. – Eur.
J., 2010, 16, 7992–7995; (e) S. W. Roh, K. Choi and C. Lee,
Chem. Rev., 2019, 119, 4293–4356.
This journal is © The Royal Society of Chemistry 2019
Dalton Trans.

View Article Online
Paper
Dalton Transactions
4 (a) L.-H. Chung and C.-Y. Wong, Chem. – Eur. J., 2019, 25, 12 (a) T. Cai, Y. Yang, W.-W. Li, W.-B. Qin and T.-B. Wen,
2889–2897; (b) C.-F. Yeung, L.-H. Chung, H.-S. Lo,
C.-H. Chiu, J. Cai and C.-Y. Wong, Organometallics, 2015,
34, 1963–1968.
Chem. – Eur. J., 2018, 24, 1606–1618; (b) A. Varela-
Fernández, C. González-Rodríguez, J. A. Varela, L. Castedo
and C. Saá, Org. Lett., 2009, 11, 5350–5353.
5 (a) D. B. Grotjahn, Top. Catal., 2010, 53, 1009–1014; 13 The kexchange rates are not strictly quantitative due to errors
(b) A. J. Arita, J. Cantada, D. B. Grotjahn and A. L. Cooksy,
Organometallics, 2013, 32, 6867–6870; (c) A. A. Ogunlana,
introduced by NOEs: B. E. Mann, J. Magn. Reson., 1976, 21,
17–23.
J. Zou and X. Bao, J. Organomet. Chem., 2018, 864, 160–168; 14 K. D. Collins and F. Glorius, Nat. Chem., 2013, 5, 597–601.
(d) Y. Wakatsuki, J. Organomet. Chem., 2004, 689, 4092– 15 J. M. Stubbs, R. J. Hazlehurst, P. D. Boyle and
4109.
6 (a) J. M. Stubbs, D. E. Chapple, P. D. Boyle and 16 E. P. Kündig and F. R. Monnier, Adv. Synth. Catal., 2004,
J. M. Blacquiere, ChemCatChem, 2018, 10, 3694–3702; 346, 901–904.
(b) J. M. Stubbs, J. P. J. Bow, R. J. Hazlehurst and 17 J.-P. J. Bow, P. D. Boyle and J. M. Blacquiere, Eur. J. Inorg.
J. M. Blacquiere, Dalton Trans., 2016, 45, 17100–17103. Chem., 2015, 2015, 4162–4166.
7 (a) J. R. Khusnutdinova and D. Milstein, Angew. Chem., Int. 18 E. S. Wiedner, J. Y. Yang, W. G. Dougherty, W. S. Kassel,
J. M. Blacquiere, Organometallics, 2017, 36, 1692–1698.
Ed., 2015, 54, 12236–12273; (b) S. Werkmeister,
J. Neumann, K. Junge and M. Beller, Chem. – Eur. J., 2015,
R. M. Bullock, M. R. DuBois and D. L. DuBois,
Organometallics, 2010, 29, 5390–5401.
21, 12226–12250; (c) V. T. Annibale and D. Song, RSC Adv., 19 A. Wetzel and F. Gagosz, Angew. Chem., Int. Ed., 2011, 50,
2013, 3, 11432–11449; (d) R. H. Crabtree, New J. Chem., 7354–7358.
2011, 35, 18–23; (e) H. Grützmacher, Angew. Chem., Int. Ed., 20 B. Das, P. Kundu and C. Chowdhury, Org. Biomol. Chem.,
2008, 47, 1814–1818; (f) H. Li, B. Zheng and K.-W. Huang, 2014, 12, 741–748.
Coord. Chem. Rev., 2015, 293–294, 116–138; (g) J. I. van der 21 A. Varela-Fernandez, J. A. Varela and C. Saá, Synthesis,
Vlugt, Eur. J. Inorg. Chem., 2012, 2012, 363–375. 2012, 44, 3285–3295.
8 (a) R. M. Bullock and M. L. Helm, Acc. Chem. Res., 2015, 48, 22 (a) H. Shen, J. Fu, H. Yuan, J. Gong and Z. Yang, J. Org.
2017–2026; (b) R. M. Bullock, A. M. Appel and M. L. Helm,
Chem. Commun., 2014, 50, 3125–3143; (c) U. J. Kilgore,
M. P. Stewart, M. L. Helm, W. G. Dougherty, W. S. Kassel,
Chem., 2016, 81, 10180–10192; (b) N. Chaisan, W. Kaewsri,
C. Thongsornkleeb, J. Tummatorn and S. Ruchirawat,
Tetrahedron Lett., 2018, 59, 675–680.
M. R. DuBois, D. L. DuBois and R. M. Bullock, 23 M. D. Doud, K. A. Grice, A. M. Lilio, C. S. Seu and
Inorg. Chem., 2011, 50, 10908–10918; (d) U. J. Kilgore, C. P. Kubiak, Organometallics, 2012, 31, 779–782.
J. A. S. Roberts, D. H. Pool, A. M. Appel, M. P. Stewart, 24 Z.-Y. Liao, P.-Y. Liao and T.-C. Chien, Chem. Commun.,
M. R. DuBois, W. G. Dougherty, W. S. Kassel, R. M. Bullock 2016, 52, 14404–14407.
and D. L. DuBois, J. Am. Chem. Soc., 2011, 133, 5861– 25 N. Sakai, K. Annaka and T. Konakahara, J. Org. Chem.,
5872. 2006, 71, 3653–3655.
9 S. Latypov, A. Strelnik, A. Balueva, Y. Spiridonova, 26 A. Carpita, A. Ribecai and P. Stabile, Tetrahedron, 2010, 66,
A. Karasik and O. Sinyashin, Eur. J. Inorg. Chem., 2016,
2016, 1068–1084.
10 T. A. Tronic, W. Kaminsky, M. K. Coggins and J. M. Mayer,
Inorg. Chem., 2012, 51, 10916–10928.
7169–7178.
27 J. H. Lee, B. S. Lee, H. Shin, D. H. Nam and D. Y. Chi,
Synlett, 2006, 0065–0068.
28 N. Sakai, K. Annaka, A. Fujita, A. Sato and T. Konakahara,
J. Org. Chem., 2008, 73, 4160–4165.
11 (a) M. Nishio, CrystEngComm, 2004, 6, 130–158;
(b) H. Suezawa, S. Ishihara, Y. Umezawa, S. Tsuboyama and 29 N. Sakai, K. Tamura, K. Shimamura, R. Ikeda and
M. Nishio, Eur. J. Org. Chem., 2004, 4816–4822. T. Konakahara, Org. Lett., 2012, 14, 836–839.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2019