Mechanisms and rates of proton transfer to coordinated carboxydithioates: Studies on [Ni(S2CR){PhP(CH2CH2PPh2)2}]+ (R = Me, Et, Bun or Ph)

Article abstract of DOI:10.1039/c4dt03543g

The complexes [Ni(S₂CR)(triphos)]BPh₄ (R = Me, Et, Buⁿ or Ph; triphos = PhP{CH₂CH₂PPh₂}₂) have been prepared and characterised. X-ray crystallography (for R = Et, Ph, C₆H₄Me-4, C₆H₄OMe-4 and C₆H₄Cl-4) shows that the geometry of the five-coordinate nickel in the cation is best described as distorted trigonal bipyramidal, containing a bidentate carboxydithioate ligand with the two sulfur atoms spanning axial and equatorial sites, the other axial site being occupied by the central phosphorus of triphos. The reactions of [Ni(S₂CR)(triphos)]⁺ with mixtures of HCl and Cl^- in MeCN to form equilibrium solutions containing [Ni(SH(S)CR)(triphos)]²⁺ have been studied using stopped-flow spectrophotometry. The kinetics show that proton transfer is slower than the diffusion-controlled limit and involves at least two coupled equilibria. The first step involves the rapid association between [Ni(S₂CR)(triphos)]⁺ and HCl to form the hydrogen-bonded precursor, {[Ni(S₂CR)(triphos)]⁺...HCl} (KR1) and this is followed by the intramolecular proton transfer (kR2) to produce [Ni(SH(S)CR)(triphos)]²⁺. In the reaction of [Ni(S₂CMe)(triphos)]⁺ the rate law is consistent with the carboxydithioate ligand undergoing chelate ring-opening after protonation. It seems likely that chelate ring-opening occurs for all [Ni(S₂CR)(triphos)]⁺, but only with [Ni(S₂CMe)(triphos)]⁺ is the protonation step sufficiently fast that chelate ring-opening is rate-limiting. With all other systems, proton transfer is rate-limiting. DFT calculations indicate that protonation can occur at either sulfur atom, but only protonation at the equatorial sulfur results in chelate ring-opening. The ways in which protonation of either sulfur atom complicates the analyses and interpretation of the kinetics are discussed. This journal is

Full text of DOI:10.1039/c4dt03543g

Dalton
Transactions
View Article Online
View Journal
PAPER
Mechanisms and rates of proton transfer to
coordinated carboxydithioates: studies on
[Ni(S2CR){PhP(CH2CH2PPh2)2}]+ (R = Me, Et, Bun
or Ph)†
Cite this: DOI: 10.1039/c4dt03543g
Ahmed Alwaaly,a,b William Clegg,a Richard A. Henderson,*a Michael R. Proberta and
Paul G. Waddella
The complexes [Ni(S2CR)(triphos)]BPh4 (R = Me, Et, Bun or Ph; triphos = PhP{CH2CH2PPh2}2) have been
prepared and characterised. X-ray crystallography (for R = Et, Ph, C6H4Me-4, C6H4OMe-4 and C6H4Cl-4)
shows that the geometry of the five-coordinate nickel in the cation is best described as distorted trigonal
bipyramidal, containing a bidentate carboxydithioate ligand with the two sulfur atoms spanning axial and
equatorial sites, the other axial site being occupied by the central phosphorus of triphos. The reactions of
[Ni(S2CR)(triphos)]+ with mixtures of HCl and Cl− in MeCN to form equilibrium solutions containing
[Ni(SH(S)CR)(triphos)]2+ have been studied using stopped-flow spectrophotometry. The kinetics show that
proton transfer is slower than the diffusion-controlled limit and involves at least two coupled equilibria.
The first step involves the rapid association between [Ni(S2CR)(triphos)]+ and HCl to form the hydrogen-
bonded precursor, {[Ni(S2CR)(triphos)]+⋯HCl} (K1R) and this is followed by the intramolecular proton trans-
fer (kR2) to produce [Ni(SH(S)CR)(triphos)]2+. In the reaction of [Ni(S2CMe)(triphos)]+ the rate law is consist-
ent with the carboxydithioate ligand undergoing chelate ring-opening after protonation. It seems likely
that chelate ring-opening occurs for all [Ni(S2CR)(triphos)]+, but only with [Ni(S2CMe)(triphos)]+ is the pro-
tonation step sufficiently fast that chelate ring-opening is rate-limiting. With all other systems, proton
transfer is rate-limiting. DFT calculations indicate that protonation can occur at either sulfur atom, but
only protonation at the equatorial sulfur results in chelate ring-opening. The ways in which protonation of
either sulfur atom complicates the analyses and interpretation of the kinetics are discussed.
Received 18th November 2014,
Accepted 23rd December 2014
DOI: 10.1039/c4dt03543g
www.rsc.org/dalton
the action of certain metalloenzymes (such as nitrogenases
and hydrogenases), where protonation of a bound substrate
Introduction
Proton transfer is one of the most pervasive reactions in chem- occurs at a metal site ligated by cysteinate ligands.4
istry. Understanding the factors that affect the rates of these
There have been some comparative studies on the protona-
reactions and, consequently, how to control the rates of these tion of non-coordinated analogous oxygen and sulfur sites. In
reactions remain important goals. Of particular interest are general, protonation of oxygen sites are diffusion controlled
the rates of proton transfer to sites coordinated to metals.1 but protonation of the analogous sulfur sites are generally
Whilst there have been numerous studies on proton transfer slower (by up to a factor of ca. 30–50). It has been suggested
rates to coordinated oxygen sites, there have been few studies that the origin of this difference in rates is a consequence of
on sulfur sites.2,3 The factors controlling the rates of proton the hydrogen bonding capabilities of the acid and base reac-
transfers to coordinated sulfur is of fundamental relevance to tants. Hydrogen bonding occurs prior to proton transfer and
stabilizes the transition state for proton transfer. Since sulfur
bases are poorer hydrogen bond acceptors than the analogous
oxygen bases (with a similar pKa), proton transfer to sulfur is
aSchool of Chemistry, Newcastle University, Newcastle upon Tyne, NE1 7RU, UK.
E-mail: richard.henderson@ncl.ac.uk
bDepartment of Chemistry, College of Science, University of Basrah, Basrah, Iraq
†Electronic supplementary information (ESI) available: Summary of kinetic data
for the reactions of [Ni(S2CR)(triphos)]+ (R = Me, Et, Bun or Ph) with HCl or mix-
tures of HCl and Cl− in MeCN, crystallography data, selected bond lengths and
angles, together with pictures of optimized calculated structures for protonated
complexes. CCDC 1033172–1033177. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c4dt03543g
slower.
There have been few kinetic studies on the protonation
of coordinated thiolates. Previously, we have shown that
stopped-flow spectrophotometric studies on protonation of
coordinated thiolates in [Ni(thiolate)(triphos)]+ (triphos = PhP-
{CH2CH2PPh2}2) can give information about both the initial
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
(cooling with a water bath may be necessary), and the mixture
was stirred for 1.5 h. About 20 mL of the solvent was removed
using a rotatory evaporator, maintaining the temperature
below 25 °C. The residue was poured into a mixture of ice
(10 g), 10 M hydrochloric acid (5 mL), and hexane (8 mL). The
free carboxydithioic acid was extracted with hexane (2 × 5 mL)
and the collected organic phases were dried over anhydrous
magnesium sulfate. After filtration, [NMe3Bz]OH (3.5 g of a
40 wt% solution in methanol) was added to the carboxy-
dithioic acid solution. The mixture was then evaporated to
dryness in vacuo. The deep purple or violet crystals of
[NMe3Bz]S2CR were washed with cold hexane and dried at
room temperature.
Fig. 1 Protonation and chelate ring-opening of [Ni(S2CR)(triphos)]+
(triphos omitted for clarity).
hydrogen-bonded intermediate and the subsequent intramole-
cular transfer of the proton.5 Furthermore, in a recent study we
have shown for the isostructural series [Ni(XPh)(triphos)]+ (X =
O, S or Se) that the rates of proton transfer from the weak acid,
lutH+ (lut = 2,6-dimethylpyridine; pKa = 14.1 in MeCN),6 and
deprotonation of the complex conjugate acid by lut, are all
slow and essentially independent of the donor atom, X.7 This
behaviour is a consequence of the steric bulk of the triphos
ligand which limits the access of the sterically-demanding
lutH+ to the donor atom.
In this paper we describe the synthesis, structures and pro-
tonation chemistry of the 5-coordinate, trigonal bipyramidal
[Ni(S2CR)(triphos)]+ (R = Me, Et, Bun or Ph), which contain the
bidentate carboxydithioate ligand. Stopped-flow studies show
that the reactions of these complexes with anhydrous HCl in
MeCN are equilibrium reactions, and involve slow proton
transfer. For R = alkyl, the precursor hydrogen-bonded inter-
mediate is detectable. Furthermore, the effect that Cl− has on
the kinetics of the reaction between HCl and [Ni(S2CMe)-
(triphos)]+ indicate that, after protonation, the bidentate car-
boxydithioic acid undergoes chelate ring-opening to form the
monodentate ligand (Fig. 1). It seems likely that such chelate
ring-opening operates in the reactions of all [Ni(S2CR)-
(triphos)]+ species, but when R = Et, Bun or Ph proton transfer
is slower than chelate ring-opening.
Synthesis of [Ni(S2CR)(triphos)]BPh4 (R = Me, Et, Bun,
Ph, C6H4X-4)
To a suspension of [NiCl(triphos)]BPh4 (0.5 g; 0.53 mmol) in
THF (ca. 20 mL) was added [NMe3Bz]S2CR (0.25 g; 0.75 mmol).
The colour changed rapidly from yellow to dark red, and the
mixture became homogeneous. After the solution was stirred
for ca. 2 h, the solution was filtered and the filtrate was reduce
in vacuo to ca. 10 mL. Addition of an excess of hexane pro-
duced a dark red microcrystalline solid. Recrystallization of
the complex was achieved by dissolving the solid in the
minimum amount of tetrahydrofuran and then adding a large
excess of hexane (ca. 4–5 × volume of tetrahydrofuran). Leaving
the solution undisturbed at room temperature for 48 h pro-
duced red crystals which were dried in vacuo (ca. 60%). The
1
product was characterized by elemental analysis, H NMR and
31P{1H} NMR spectroscopy (Table 1). For R = Et, Ph, C6H4Me-4,
C6H4OMe-4 and C6H4Cl-4 the products were also characterized
by X-ray crystallography (see Fig. 2 and 3, Tables 2 and 3, and
ESI†).
Kinetic studies
Experimental
All kinetic studies were performed using an Applied Photo-
physics SX.18 MV stopped-flow spectrophotometer, modified
to handle air-sensitive solutions, connected to a RISC compu-
ter. The temperature was maintained using a Grant LTD6G
thermostat tank with combined recirculating pump. The
experiments were conducted at 25.0 °C. The wavelength used
was λ = 420 nm. All kinetics were studied in MeCN. The MeCN
was dried over CaH2 and distilled immediately prior to use.
The solutions of complex and reactants (HCl and [NEt4]Cl)
were prepared under an atmosphere of dinitrogen and trans-
ferred to the stopped-flow apparatus using gas-tight, all glass
syringes. The kinetics were studied under pseudo-first-order
conditions, with HCl and [NEt4]Cl present in at least a 10-fold
excess over the concentration of the complex. Mixtures of HCl
and [NEt4]Cl were prepared from stock solutions of the two
reagents. All solutions were used within 1 hour of preparation.
The following chemicals were purchased from Sigma-Aldrich
and used as received without any further purfications:
NiCl2·6H2O, benzyltrimethylammonium hydroxide, bis(2-
diphenylphosphinoethyl)phenylphosphine = triphos, carbon
disulfide, sodium tetraphenylborate, methanol-d1 and D2O.
CD3CN was purchased from Goss Scientific and used as
received. [NiCl(triphos)]BPh4 was prepared by the literature
method.8 [NEt4]Cl was dried in vacuo prior to use.
All manipulations in both the synthetic and kinetic aspects
of this work were performed under an atmosphere of dinitro-
gen using Schlenk or syringe techniques, as appropriate.
Synthesis of [NMe3Bz]S2CR (Bz = CH2Ph; R = Me, Et, Bun
or Ph)
All [NMe3Bz]S2CR were prepared by the literature method.9 Stock solutions of HCl were prepared in situ. Thus a solution
Manipulations were performed in a fume cupboard. Carbon of HCl (25 mmol dm−3) in 25 mL of CH3CN was prepared by
disulfide (0.78 g, 0.01 mol) in dry THF (20 mL) was slowly adding methanol (1.0 mL, 25 mmol) followed by Me3SiCl
added to the appropriate Grignard reagent (10 mL, 0.01 mol) (3.175 mL, 25 mmol). Stock solutions of DCl were similarly
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Table 1 Elemental analyses and spectroscopic data for [Ni(S2CR)(triphos)]BPh4 (R = Me, Et, Bun or Ph)
Elemental analysisa
Compound
C
H
1H NMR spectrum
31P{1H} NMR spectrum
[Ni(S2CMe)(triphos)]BPh4
71.1 (71.8)
6.0 (5.6)
8.45–6.75 (m, 45H, Ph)
3.50–2.23 (br, 8H, CH2)
1.36 (s, 3H, CH3)
111.6 (t, JPP = 50 Hz, PPh)
47.7 (d, JPP = 50 Hz, PPh2)
[Ni(S2CEt)(triphos)]BPh4
[Ni(S2CBun)(triphos)]BPh4
71.2 (72.0)
71.9 (72.4)
6.0 (5.7)
6.0 (5.6)
8.20–6.40 (m, 45H, Ph)
113.7 (t, JPP = 32 Hz, PPh)
52.4 (d, JPP = 32 Hz, PPh2)
3.20–2.17 (br, 8H, CH2)
3.73 (q, JHH = 7.3 Hz, 2H, CH2CH3)
0.80 (t, JHH = 7.3 Hz, 3H, CH3)
9.10–6.79 (m, 45H, Ph)
3.40–2.50 (br, 8H, CH2)
3.72–3.61 (m, 4H, SCH2 & CH2)
1.32–1.01 (m, 2H, CH2)
0.88 (t, JHH = 7.12 Hz, 3H, CH3)
8.45–6.60 (m, 50H, Ph)
113.9 (t, JPP = 32 Hz, PPh)
51.7 (d, JPP = 32 Hz, PPh2)
[Ni(S2CPh)(triphos)]BPh4
71.5 (71.9)
5.9 (5.8)
113.0 (t, JPP = 32 Hz, PPh)
52.8 (d, JPP = 32 Hz, PPh2)
3.24–2.10 (br, 8H, CH2)
a Calculated values in parentheses.
Fig. 3 Structure of the [Ni(S2CPh)(triphos)]+ cation.
H atoms are
omitted for clarity.
Fig. 2 Structure of the [Ni(S2CEt)(triphos)]+ cation.
H atoms are
omitted for clarity; unlabelled atoms follow in sequence from those
shown labelled. The inset shows the atom numbering scheme used in
Table 3.
R = C6H4OMe-4, Sol = THF; and R = C6H4Cl-4, Sol = THF) were
examined on various diffractometers at temperatures between
100 and 150 K with either Mo-Kα or synchrotron radiation as
detailed in the ESI.† In all cases absorption corrections were
applied, based on repeated and symmetry-equivalent reflec-
tions, and the structures were solved by direct methods and
refined on all unique F2 values. Disorder was resolved for
some of the solvent molecules. Table 2 gives selected crystallo-
graphic data for the structures with R = Et and Ph, with
selected coordination geometry in Table 3; full information for
all six structures is provided in the ESI.† The other four struc-
tures are all isomorphous, with essentially the same arrange-
ment of cations and anions in the asymmetric unit, the para
substituents and the solvent molecules occupying equivalent
regions in each case, with little variation in the nickel coordi-
prepared in situ by mixing methanol-d1 (1.0 mL, 25 mmol)
with Me3SiCl (3.175 mL, 25 mmol) in 25 mL of MeCN.
Under all conditions, the stopped-flow absorbance-time
trace was an excellent fit to a single exponential, indicating
a first-order dependence on the concentration of complex
(Fig. 4). The dependences on the concentrations of HCl and
Cl− were determined from analysis of the appropriate graphs
as explained in the text in the results and discussion section.
X-ray crystallography
Single crystals of [Ni(S2CR)(triphos)]BPh4·Sol (R = Me, unsol-
vated; R = Ph, Sol = THF; R = C6H4Me-4, Sol = THF or CHCl3;
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 2 X-ray crystallographic data for two of the [Ni(S2CR)(triphos)]-
BPh4 complexes
Complex
R = Et
R = Ph
Formula
C
61H58BNiP3S2
C69H66BNiOP3S2
1137.8
Orthorhombic
Mr/g mol−1
1017.6
Monoclinic
P21/c
Crystal system
Space group
Pbcn
a/Å
b/Å
c/Å
17.685(4)
13.749(3)
22.807(5)
110.150(2)
5206(2)
4
38.8301(16)
16.6153(5)
18.1236(5)
90
11 692.9(7)
8
β/°
V/Å3
Z
Reflections measured
Unique reflections, Rint
Refined parameters
R (F, F2 > 2σ)
Rw (F2, all data)
Goodness of fit on F2
66 315
9261, 0.1269
615
52 089
11747, 0.0487
694
0.0486
0.1000
1.070
0.52, −0.38
Fig. 4 Stopped-flow absorbance-time curve for the reaction of [Ni-
(S2CEt)(triphos)] + (0.5 mmol dm−3) with HCl (40 mmol dm−3) in MeCN
at 25.0 °C (λ = 420 nm). The experimental trace is shown in black and
the exponential curve fit is shown in grey. The curve fit is defined by the
equation At = 0.325–0.042exp(−0.35t).
0.0530
0.1169
1.061
Max., min. electron density/e Å−3 0.45, −0.39
dm3 mol−1 cm−1). In addition, we have characterised the R =
Et, Ph, C6H4Me-4, C6H4OMe-4 and C6H4Cl-4 derivatives by
Table 3 Selected bond lengths and bond angles in [Ni(S2CR)(triphos)]+ X-ray crystallography. All the crystal structures show discrete
(R = Et or Ph)
cations, anions and, for the aryl derivatives, a molecule of
tetrahydrofuran (a chloroform solvate was also characterized
Complex
R = Et
R = Ph
for the C6H4Me-4 derivative). In the cations, the nickel is
5-coordinate with the coordination sphere made up of the three
phosphorus donors of the triphos ligand and the two sulfur
atoms of the carboxydithioate ligand. The structures of [Ni-
(S2CEt)(triphos)]+ and [Ni(S2CPh)(triphos)]+ are shown in Fig. 2
and 3, and the structures of the other derivatives are shown in
the ESI.† The structures of all the cations are best described as
trigonal bipyramidal.11 The trigonal plane comprises the two
terminal phosphorus atoms of the triphos ligand and one
sulfur from the carboxydithioate ligand. The axial positions of
the trigonal bipyramid are occupied by the central phosphorus
of the triphos ligand and the other sulfur of the carboxydithio-
ate ligand. Selected bond lengths and bond angles associated
with the nickel and its ligands are presented in Table 3 and
the ESI.†
Bond lengths/Å
Ni–P1
Ni–P2
Ni–P3
Ni–S1
2.1733(10)
2.1516(11)
2.2174(11)
2.2342(11)
2.3448(10)
2.2227(7)
2.1484(7)
2.2066(7)
2.2321(7)
2.3179(7)
Ni–S2
Bond angles/°
P1–Ni–P2
P2–Ni–P3
P1–Ni–P3
S1–Ni–S2
P1–Ni–S1
P1–Ni–S2
P2–Ni–S1
P2–Ni–S2
P3–Ni–S1
P3–Ni–S2
S1–C–S2
85.88(4)
87.06(4)
131.15(4)
75.60(4)
95.75(4)
125.86(4)
176.27(4)
100.73(4)
94.35(4)
102.95(4)
113.7(2)
87.70(3)
87.39(3)
130.66(3)
76.16(2)
92.70(3)
107.41(3)
176.60(3)
100.50(3)
94.90(3)
121.76(3)
111.88(15)
Previously, we have described the X-ray crystal structure of
[Ni(2-Spy)(triphos)]BPh4 (2-Spy = 2-pyridinethiolate), in which
the nickel is also five-coordinate and the 2-Spy ligand is biden-
tate.12 The structures of [Ni(S2CR)(triphos)]+ and [Ni(2-Spy)-
(triphos)]+ are very similar. Thus, in [Ni(2-Spy)(triphos)]+ the
structure is also best described as a distorted trigonal bipyra-
mid. Furthermore, the trigonal plane of [Ni(2-Spy)(triphos)]+
also comprises the two terminal phosphorus atoms (Pt1, Pt2;
subscript t = terminal) of the triphos ligand {Ni–Pt1 = 2.1637(9)
Å, Ni–Pt2 = 2.1966(9) Å} with the remaining position occupied
by the nitrogen of the 2-Spy ligand {Ni–N = 2.164(2) Å, Pt1–Ni–
Pt2 = 146.34(3)°, Pt1–Ni–N = 119.08(7)°, Pt2–Ni–N = 94.11(7)°}.
Finally, the two axial positions are occupied by the central
phosphorus (Pc) of the triphos ligand and the sulfur of the
2-Spy ligand {Ni–S = 2.2644(8) Å, Ni–Pc = 2.1310(8) Å; Pt1–Ni–S =
92.08(3)°; Pt2–Ni–S = 93.93(3)°; N–Ni–S = 70.81(6)°; Pt1–Ni–Pc =
87.08(3)°, Pt2–Ni–Pc = 87.08(3)°, N–Ni–Pc = 109.12(7)°, Pc–Ni–S
= 178.98(4)°}.
nation geometry. CCDC references: 1033172–1033177. Pro-
grams were standard Agilent Technologies, Bruker and Rigaku
control and data processing software, SHELXTL, SHELXL-2014,
and OLEX2.10
Results and discussion
Synthesis and structures of [Ni(S2CR)(triphos)]BPh4
All [Ni(S2CR)(triphos)]BPh4 were prepared by the reactions of
[NMe3Bz]S2CR9 with [NiCl(triphos)]BPh4 and isolated as dark
red crystalline solids. The spectroscopic and analytical data
associated with all the complexes are consistent with the for-
mulation (Table 1). The UV-visible spectra of all [Ni(S2CR)
(triphos)]+ in MeCN are very similar broad spectra, absorbing
from 350–700 nm (R = Et; λmax (shoulder) ∼510 nm, ε = 420
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Kinetics and mechanism
concentrations of HCl the rate of the reaction is independent
of [HCl]. The data were analysed by the usual plot of 1/kobs
versus 1/[HCl] which is a straight line and allows derivation of
the experimental rate law shown in eqn (2).13 When R = Me:
a = 4.69 × 103 dm3 mol−1 s−1, b = 67.0 dm3 mol−1; R = Et: a =
39.5 dm3 mol−1 s−1, b = 84.0 dm3 mol−1; R = Bun: a = 19.9 dm3
The kinetics of the reactions of [Ni(S2CR)(triphos)]+ (R = Me,
Et, Bun or Ph) with mixtures of HCl and Cl− {eqn (1)} were
studied under pseudo-first-order conditions ([HCl]/[Ni] ≥ 10)
in MeCN using stopped-flow spectrophotometry. When moni-
tored at a single wavelength, the absorbance-time traces are
excellent fits to a single exponential consistent with the reac-
tion exhibiting a first-order dependence on the concentration
of [Ni(S2CR)(triphos)]+ (Fig. 4). The traces exhibit behaviour
characteristic of an equilibrium reaction {eqn (1)}. Thus,
increasing the concentration of HCl (at a constant concen-
tration of Cl−) results in a larger absorbance change and final
absorbance, while increasing the concentration of Cl− (at a
constant concentration of HCl) results in a smaller absorbance
change and final absorbance (see ESI†).
mol−1 s−1, b = 71.0 dm3 mol−1
.
ꢀ
ꢁ
a½HClꢀ NiðS2CRÞðtriphosÞþ
1 þ b½HClꢀ
Rate ¼
ð2Þ
Effect of Cl− on the reaction between [Ni(S2CR)(triphos)]
(R = Me, Et or Bun) and HCl
In further kinetic studies, the effect that Cl− (addition of
[NEt4]Cl) has on the rates of the reactions with HCl was investi-
gated. The effect that Cl− has on the rate of reaction between
anhydrous HCl and [Ni(S2CEt)(triphos)]+ or [Ni(S2CBun)-
(triphos)]+ is different to the effect Cl− has on the reaction
between HCl and [Ni(S2CMe)(triphos)]+. For the reactions of
HCl with the R = Et or Bun derivatives, addition of Cl− accele-
rates the rate, whereas in the reaction with the R = Me deriva-
tive, Cl− inhibits the reaction (see ESI†).
In analysing the kinetic data for the reactions involving mix-
tures of HCl and Cl−, corrections need to be made to the
concentrations of these species because, in MeCN, HCl and
Cl− associate as shown in eqn (3) (homoconjugation
equilibrium).14
½NiðS2CRÞðtriphosÞþꢀ þ HCl Ð ½NiðHS2CRÞðtriphosÞ ꢀ þ Clꢁ
2þ
ð1Þ
In the discussion below, we will first present the kinetic
results of the studies of the R = Me, Et or Bun complexes with
HCl alone and then report the different effects that introdu-
cing Cl− has on the kinetics of these systems. We will then
discuss the possible mechanisms consistent with the kinetics
of these systems and, finally, present the kinetics of the R = Ph
complex with mixtures of HCl and Cl− to show that the kine-
tics of this system are also consistent with the proposed
mechanism.
KH
Reaction between [Ni(S2CR)(triphos)] (R = Me, Et or Bun) and
HCl
HCl þ Clꢁ Ð HCl2
ð3Þ
ꢁ
An important consequence of this equilibrium reaction is
that the concentrations of HCl and Cl− cannot be changed
independently; changing the concentration of one will perturb
the concentration of the other. Using the literature value of
KH = 158.5 dm3 mol−1, the concentrations of free HCl and Cl−
present in the mixtures have been calculated {henceforth
designated [HCl]e and [Cl−]e}.9 In the kinetic analysis that
follows, it is assumed that HCl2− acts as neither acid nor base.
For the reactions of the R = Et or Bun derivatives with HCl
(in the absence of Cl−), the dependence on the concentration
of acid is described by eqn (2). Thus, the observed rate con-
stant (kobs) for the reactions in the presence of Cl− can be cor-
rected for changes to the rate due to changes in the
concentration of HCl using (kobs − k′R), where k′R = a[HCl]e/(1 +
b[HCl]e). A plot of (kobs − k′R) versus [Cl−]e then describes the
effect that Cl− has on the rate of the reaction. Such a plot is a
straight line going through the origin (Fig. 6, top). Analysis of
this graph shows that, for the R = Et or Bun complexes the rate
law is that shown in eqn (4). When R = Et: c = 380 dm3 mol−1
The kinetics of the reactions of [Ni(S2CR)(triphos)]+ (R = Me, Et
or Bun) with anhydrous HCl (in the absence of [NEt4]Cl)
exhibit a complicated dependence on the concentration of
HCl, as shown in Fig. 5. At low concentrations of HCl the reac-
tion exhibits a first-order dependence on [HCl] but at high
s−1; R = Bun: c = 74 dm3 mol−1 s−1
.
ꢂ
ꢃ
ꢀ
ꢁ
Fig. 5 Kinetic data for the reaction of [Ni(S2CEt)(triphos)]+ ([Ni]
=
a½HClꢀe
Rate ¼
þ c½Clꢁꢀe NiðS2CRÞðtriphosÞþ
ð4Þ
0.5 mmol dm−3) with HCl ( and ) or DCl ( ) in MeCN at 25.0 °C.
Curve fits to the data are those defined by eqn (2) and the parameters
presented in the text. The different types of filled symbols show the
results in different experiments and give an indication of the scatter of
the data. The error bars correspond to 10% errors.
●
○
▲
1 þ b½HClꢀe
For the reaction of the R = Me complex with mixtures of
HCl and Cl− the analysis of the kinetics is more complicated
because addition of Cl− inhibits the reaction. Initial inspection
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
(triphos)]+ with HCl in the presence of Cl− is derived {eqn (6)};
d = 250.0 dm3 mol−1 s−1 and e = 18.0 s−1
.
kobs ꢁ 18
1
¼
ð5Þ
ð6Þ
k′′
1 þ 250½Clꢁꢀe
ꢀ
ꢁ
a½HClꢀe NiðS2CMeÞðtriphosÞþ
Rate ¼
þ e
ð1 þ b½HClꢀeÞð1 þ d½ClꢁꢀeÞ
The scatter in the data shown in both plots of Fig. 6 is, we
believe, in part, a consequence of the necessary corrections to
the concentrations of HCl and Cl− to allow for the homoconju-
gation equilibrium {eqn (3)}. These corrections use a single
value for the homoconjugation equilibrium constant (KH) over
the entire concentration range. It seems likely that the value of
KH might vary slightly with the ionic strength of the solution.
Mechanism
For the R = Et or Bun derivatives, the experimental rate law
shown in eqn (4) is consistent with a mechanism involving
two coupled equilibria. Both mechanisms shown in Fig. 7
would give rise to this form of rate law.
Mechanism 1 involves an initial unimolecular step (poss-
ibly ring-opening of the carboxydithioate ligand, kRa) which is
then protonated in the subsequent step (kRb). Assuming that
the ring-opened intermediate (B) is a steady-state species, the
rate law associated with mechanism 1 is that shown in eqn (7)
Fig. 6 TOP. Kinetic data for the reaction of [Ni(S2CEt)(triphos)]+
(0.5 mmol dm−3) with HCl in the presence of Cl− in MeCN at 25.0 °C.
Graph of (kobs − k’Et) versus [Cl−]e where k’Et and subscript e are
explained in the text. Data points correspond to: [HCl] = 25.0 mmol
●
ꢄ
ꢅ
(
)
kaRkbR=kꢁRa ½HClꢀe
Rate ¼
þ kꢁRb½Clꢁꢀe
ꢁ
ꢄ
ꢅ
1 þ kbR=kꢁRa ½HClꢀe
ꢀ
ꢂ NiðS2CRÞðtriphosÞþ
ð7Þ
dm−3, [Cl−] = 1.0–10.0 mmol dm−3
(
); [HCl] = 50.0 mmol dm−3, [Cl−] =
2.5–30.0 mmol dm−3
(
); [HCl]
=
100.0 mmol dm−3
,
[Cl−]
=
■
2.5–30.0 mmol dm−3
Although eqn (7) is a rate law of the same form as that
observed experimentally {eqn (4)}, this is not the favoured
mechanism for the following reasons. (i) There is no evidence
that any [Ni(S2CR)(triphos)]+ is undergoing rapid chelate ring-
opening and closure in the absence of acid. (ii) The inhibiting
effect of Cl− on the reaction between [Ni(S2CMe)(triphos)]+ and
HCl {eqn (6)} cannot be explained using this mechanism. (iii)
Studies with DCl (vide infra) show that both steps in the mech-
anism are associated with isotope effects, but in mechanism 1,
▲
( ). Line drawn is that defined by eqn (4) and the
parameters detailed in the text. BOTTOM. Kinetic data for the reaction of
Ni(S2CMe)(triphos)]+ (0.5 mmol dm−3) with HCl in the presence of Cl− in
MeCN at 25.0 °C. Graph of (kobs − 18)/k’’ versus [Cl−]e where k’’ and sub-
script e are explained in the text. Data points correspond to: [HCl] =
●
50.0 mmol dm−3, [Cl−] = 2.5–50.0 mmol dm−3
( ); [HCl] = 100.0 mmol
dm−3, [Cl−] = 2.5–50.0 mmol dm−3
). Curve drawn is that defined
▲
(
using eqn (5) and the parameters detailed in the text. The error bars cor-
respond to 10% errors.
of the kinetic data (after correcting the concentrations of HCl
and Cl− for the homoconjugation equilibrium but before cor-
recting kobs for the dependence on the concentration of HCl)
showed that: (i) increasing the concentration of Cl− leads to a
decrease in kobs and (ii) at high concentrations of Cl−, the reac-
tion rate is not zero, but levels off at kobs ∼ 18 s−1. To analyse
the data, the values of kobs (for the reaction in the presence of
Cl−) were corrected for both the effect of changes to the con-
centration of HCl, as described by eqn (2) and given by k″ =
4.69 × 103[HCl]e/(1 + 67.0[HCl]e), and the limiting rate con-
stant at high concentrations of Cl− (kobs = 18 s−1). Thus, a plot
of (kobs − 18)/k″ versus [Cl−]e is a curve (Fig. 6, bottom). Fitting
the data to a curve (using an iterative methodology) gives
eqn (5), from which the rate law for the reaction of [Ni(S2CMe)-
Fig. 7 Mechanisms of protonation of [Ni(S2CR)(triphos)]+ (R = Me, Et or
Bun, Ph).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Table 4 Summary of elementary rate and equilibrium constants for the reactions of [Ni(S2CR)(triphos)]+ (R = Me, Et, Bun or Ph) with mixtures of
HCl and Cl− in MeCN at 25.0 °C
R
KR1/dm3 mol−1
kR2/s−1
KR1k2R/dm3 mol−1 s−1
kR−2/dm3 mol−1 s−1
k−R2/kR3/dm3 mol−1
kR−3/s−1
KR1kR2
Me
67
70
4.69 × 103
4.67 × 103 a
39.5
250
18
110a
84
42.5a
0.47
0.26a
0.28
0.21a
≥0.33
Et
Bun
Ph
380
74
0.10
0.27
0.011
191a
71
49.7a
19.9
83.3a
≤6.7
17.5a
2.18
200
a Studies with DCl.
the chelate ring-opening is not dependent on the acid and (triphos)]+ with mixtures of HCl and Cl− are summarized in
thus kRa can not be associated with an isotope effect.
Table 4.
It is pertinent to note that for [Ni(S2CR)(triphos)]+ {R = Et,
Mechanism 2 is consistent with the kinetic data for all the
complexes described in this paper. The first equilibrium Bun or Ph (vide infra)} the observed rate laws only indicate
involves hydrogen-bonding of HCl to the complex (to form C), equilibrium protonation of these complexes and, because in
whilst the second equilibrium is the intramolecular proton these systems proton transfer is rate-limiting, there is no evi-
transfer from Cl to S (to form D), and the final equilibrium dence for the subsequent chelate ring-opening step. Nonethe-
corresponds to the chelate ring-opening step (to form E). less, we suggest that all the complexes are so similar that the
There is nothing unusual about this mechanism for proton chelate ring-opening follows protonation in all [Ni(S2CR)-
transfer. It seems likely that, in general, any proton transfer (triphos)]+. This proposal is supported by DFT calculations
reaction will involve the formation of a hydrogen-bonded pre- (vide infra). Inspection of the data in Table 4 shows that proto-
cursor, it is just that in this system (because formation of the nation of [Ni(S2CMe)(triphos)]+ is significantly faster than for
hydrogen-bonded precursor is rapid but proton transfer is rela- the other derivatives. The possible reasons for this will be dis-
tively slow) the precursor accumulates.5 Similar kinetics have cussed below. Because protonation is fast for [Ni(S2CMe)-
been observed in the protonation reactions of other ligands (triphos)]+ the chelate ring-opening step becomes rate-limiting.
bound to the robust {Ni(triphos)}2+ site.5,7,15
In contrast, for [Ni(S2CEt)(triphos)]+ or [Ni(S2CBun)(triphos)]+
To accommodate the experimental rate laws for R = Et and proton transfer is slower than chelate ring-opening and hence
Bun only the first two equilibria need to be considered (i.e. proton transfer is rate-limiting.
rate-limiting proton transfer). However, to accommodate the
For the equilibrium reaction between [Ni(S2CMe)(triphos)]+
experimental rate law for R = Me all three coupled equilibria (AMe) and HCl to form E and Cl–, the overall equilibrium con-
need to be considered (i.e. rate-limiting chelate ring-opening stant can be calculated from the kinetic data presented in
after protonation). The rate law associated with mechanism 2 Table 4 (KM1 e·K2Me·K3Me = 6.5 × 104). Using this value we can cal-
is derived assuming that formation of the hydrogen-bonded culate that in a solution prepared from [AMe] = 1.0 mmol dm−3
precursor (C) from A and HCl is a rapidly established equili- and [HCl] = 10.0 mmol dm−3, all the nickel complex is con-
brium. The rate law will then depend on whether proton trans- verted to EMe. However, we have been unable to isolate EMe
fer (kR2) or the chelate ring-opening (kR3) is rate-limiting. If from such reaction mixtures in the absence of Cl–. In all cases
proton transfer is rate limiting (kR2 is slow and k3R is fast) the the only Ni-containing product isolated was [Ni(S2CR)-
rate law associated with mechanism 2 is that shown in eqn (8). (triphos)]BPh4 (A). Furthermore, studies using 31P{1H} NMR
However, if chelate ring-opening is rate-limiting (k2R is fast and spectroscopy on solutions containing [AMe] = 1.0 mmol dm−3
kR3 is slow), and D is a steady-state intermediate then the rate and [HCl] = 10.0 mmol dm−3 showed chemical shifts (associ-
law for mechanism 2 is that shown in eqn (9).
ated with the triphos ligand) which were little different from
those of AMe (δ 10 ppm). Similar problems in isolating thiol
products in the protonation reactions of Ni-thiolate complexes
have been reported for a variety of other systems.5,7,12,15
ꢂ
ꢃ
K1Rk2R½HClꢀe
ꢀ
ꢁ
Rate ¼
þ kꢁR2½Clꢁꢀe NiðS2CRÞðtriphosÞþ ð8Þ
1 þ K1R½HClꢀe
ꢂ
ꢃ
K1Rk2R½HClꢀe
Rate ¼
þ kꢁR3
ð1 þ K1R½HCleꢀÞð1 þ fkꢁR2=k3Rg½ClꢁꢀeÞ
Reaction between [Ni(S2CPh)(triphos)]+ and HCl in the
presence of Cl−
ꢀ
ꢁ
ꢂ NiðS2CRÞðtriphosÞþ
ð9Þ
Eqn (8) is the same form as that observed in the reactions The kinetics of the reaction between [Ni(S2CPh)(triphos)]+ and
with [Ni(S2CEt)(triphos)]+ and [Ni(S2CBun)(triphos)]+ {eqn (4)}, HCl in the presence of Cl− exhibit a first-order dependence on
whilst eqn (9) is the same form as that observed with the concentration of complex. A plot of kobs/[Cl−]e versus
[Ni(S2CMe)(triphos)]+ {eqn (6)}. The values of the elementary [HCl]e/[Cl−]e is a straight line with a small intercept as shown
rate and equilibrium constants for the reactions of [Ni(S2CR)- in Fig. 8. Thus, the rate law associated with this reaction is
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 8 Kinetic data for the reaction of Ni(S2CPh)(triphos)]+ (0.5 mmol
dm−3) with HCl in the presence of Cl− in MeCN at 25.0 °C. Graph of
Fig. 9 Consequences of protonation at axial (S1) and equatorial (S2)
sulfur of [Ni(S2CR)(triphos)]+ (R = Me or Ph).
k
obs/[Cl−]e versus [HCl]e/[Cl−]e where subscript e is explained in the text.
Data points correspond to: [HCl]
=
=
100.0 mmol dm−3
150.0 mmol dm−3
,
,
[Cl−]
[Cl−]
=
=
1.0–15.0 mmol dm−3
(
); [HCl]
●
1.0–15.0 mmol dm−3
▲
( ). Line drawn is that defined by eqn (10). The
Table 5 Selected bond lengths and bond angles of Ni(S(SH)CPh)-
(triphos)]2+ from DFT calculations
error bars correspond to 10% errors.
Complex EPh
Complex FPh
shown in eqn (10) and is consistent with an equilibrium proto-
nation reaction.
Bond lengths
Ni–P1
Ni–P2
Ni–P3
Ni–S1
2.2665
2.2176
2.2630
2.2865
3.5122
2.2665
2.2630
2.2176
2.0423
2.4400
Rate ¼ f2:18½HClꢀe þ 200½Clꢁꢀeg½NiðS2CPhÞðtriphosÞþꢀ ð10Þ
This rate law is consistent with mechanism 2 and the
associated rate law shown in eqn (8), where proton transfer is
rate-limiting. If KP1h[HCl]e < 1, eqn (8) simplifies to the rate law
shown in eqn (11). Comparison of eqn (10) and (11) give the
values of the elementary rate constants shown in Table 4.
Ni–S2
Bond angles
P1–Ni–P2
P1–Ni–P3
P1–Ni–S1
P1–Ni–S2
P2–Ni–P3
P2–Ni–S1
P2–Ni–S2
P3–Ni–S1
P3–Ni–S2
85.90
152.31
94.17
85.90
152.31
90.91
104.47
85.25
167.33
123.24
92.04
þ
Ph
Rate ¼ fKP1hk2Ph½HClꢀe þ k ½Clꢁꢀeg½NiðS2CPhÞðtriphosÞ ꢀ ð11Þ
ꢁ2
85.25
179.93
If KP1h[HCl]e < 1, we can estimate that KP1h ≤ 6.7 dm3 mol−1
.
94.69
Comparison of the data in Table 4 shows that proton transfer
to [Ni(S2CPh)(triphos)]+ (K1PhkP2h) is slower than to the alkyl
carboxydithioate complexes and this appears to be due, princi-
pally, to the lower value of KP1h (k2Ph ≥ 0.33 s−1). This obser-
vation indicates that the basicities of the carboxydithioate
ligands are significantly different when R = aryl compared to
when R = alkyl, and that this affects the association between
HCl and the complex {i.e. formation of the hydrogen-bonded
precursor (C), KR1}, but has less effect on the rates of the intra-
molecular proton transfer step (kR2).
102.33
mized at the B3LYP/Lanl2dz levels of theory and including
solvent, see ESI†) we have explored the protonation of each
sulfur in [Ni(S2CR)(triphos)]+ (R = Me or Ph). The calculations
indicate, for both R = Me and Ph, that protonation of the equa-
torial sulfur (S2) results in chelate ring-opening to form
species ER, with a pendant S2H group (Fig. 9). Selected calcu-
lated bond lengths and bond angles for R = Ph are shown in
Table 5 and the analogous data for R = Me, together with pic-
tures of ER, are shown in ESI.† The calculated bond angles for
ER indicate that the Ni has a square-planar geometry (Table 5
and ESI†). This result is in line with the interpretation of the
kinetics {eqn (8) and (9)}. Interestingly, calculations indicate
that protonation of the axial sulfur (S1) in [Ni(S2CR)(triphos)]+
can also occur but is associated with only relatively minor
changes in the structure and no chelate ring-opening. Com-
parison of the selected calculated bond lengths and bond
angles for EPh and FPh (Table 5) and the analogous dimensions
for [Ni(S2CPh)(triphos)]+ from the X-ray crystallographic data
(Table 3) demonstrate the major dimension changes involved
The site of protonation: computational studies
One complication in presenting a rigorous analysis of the kine-
tics for protonation of [Ni(S2CR)(triphos)]+ is that there are two
distinct sulfur sites in the complex. X-ray crystallography
reveals a trigonal bipyramidal structure for all [Ni(S2CR)-
(triphos)]+, in which one sulfur of the carboxydithioate ligand
sits in the equatorial plane while the other occupies an axial
position (Fig. 2 and 3). The kinetics described above cannot
distinguish between a mechanism involving protonation of a
single sulfur site or two parallel pathways, each associated
with protonation of a different sulfur. Using molecular mech-
anics calculations (GAUSSIAN03 package,16 geometries opti-
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
rate description of the kinetics. For simplicity, in the remain-
der of the discussion, the results will be interpreted in terms
of a single-site protonation {Fig. 7 and eqn (8)} but where the
interpretation would be compromised if a two-site protonation
model is more applicable {Fig. 10 and eqn (12)} the issues will
be discussed.
Isotope effects for reactions of [Ni(S2CR)(triphos)]+ (R = Me,
Et or Bun) with HCl
Because proton transfer from HCl to [Ni(S2CR)(triphos)]+ in
MeCN is slow (and rate-limiting for R = Et, Bun or Ph) these
reactions would be expected to be associated with a kinetic
isotope effect when HCl is replaced by DCl.16 Furthermore,
proton transfer to [Ni(S2CR)(triphos)]+ (R = Me, Et or Bun)
involves initial formation of the hydrogen-bonded species (KR1)
followed by intramolecular proton transfer (kR2) and these
steps are distinguishable in the analysis of the kinetics. Thus,
measuring the kinetics of the reactions of DCl with [Ni(S2CR)-
(triphos)]+ allow us to determine how the isotopic change in
the acid affects both KR1 and k2R.
Fig. 10 Mechanism for the reactions of [Ni(S2CR)(triphos)]+ with HCl in
the presence of Cl−, accomodating the results of the DFT calculations
which indicate that protonation can occur at both sulfurs but with
different structural consequences.
when protonation occurs at S2, but the relatively minor
changes associated with protonation at S1.
It seems likely that chelate ring-opening occurs after proto-
nation of S2 because reorganization to the square-planar geo-
metry is energetically rather undemanding, requiring only a
slight opening up of the P1–Ni–P3 angle. In contrast, it seems
likely that the reorganization to a square-planar geometry after
protonation at S1 and cleavage of the Ni–S1 bond is energeti-
cally more demanding since it requires opening up of the
P1–Ni–P3 angle and movement of the Ni–S2 bond.
In the light of the results from these calculations we need
to consider a modification to the mechanism in Fig. 7 to
accommodate the protonation of both sulfur sites and the
different geometrical changes which accompany the protona-
tions (Fig. 10). Assuming that the rates of proton transfer to
both S1 and S2 are slow and that the mechanism for protona-
tion of both sites is the same, the modified mechanism is
associated with the rate law shown in eqn (12). Comparison of
eqn (8) (single site protonation) and eqn (12) yields the
relationships shown in eqn (13)–(15).
In principle, we could also measure the isotope effect for
the reaction between DCl and [Ni(S2CPh)(triphos)]+. However,
there is a problem with analysing the kinetic data in this case.
As outlined earlier, mixtures of HCl and Cl− are involved in the
homoconjugation equilibrium reaction shown in eqn (3) and
the concentrations of [HCl]e and [Cl−]e in such mixtures can
be calculated using the literature homoconjugation equili-
brium constant (KH). Unfortunately, the homoconjugation
equilibrium constant for mixtures of DCl and Cl− is unknown,
and would be expected to be different from KH. Thus, in mix-
tures containing DCl and Cl−, the concentrations of [DCl]e and
[Cl−]e cannot be calculated. The same problem is encountered
in attempting to measure the isotope effect for the reaction
between DCl and [Ni(S2CR)(triphos)]+ (R = Me, Et or Bun) in
the presence of Cl−. Consequently, we have only measured the
kinetics of the reactions of [Ni(S2CR)(triphos)]+ (R = Me, Et or
Bun) with DCl in the absence of Cl−.
(
)
ꢄ
ꢅ
RS1
RS2
2
K1RS k2 þ K1RS k2 ½HClꢀe
1
ꢄ
ꢅ
RS1
RS2
ꢁ2
Rate ¼
þ k þ k
½Clꢁꢀe
The kinetics of the reactions between [Ni(S2CR)(triphos)]+
(R = Me, Et or Bun) and DCl in MeCN follow the same general
behaviour as the analogous reactions with HCl (Fig. 5); the
ꢄ
ꢅ
ꢁ2
1 þ K1RS þ K1RS ½HClꢀe
1
2
ꢀ
ꢁ
ꢂ NiðS2CRÞðtriphosÞþ
ð12Þ reactions exhibit a first-order dependence on the concentration
of complex and a non-linear dependence on the concentration
RS1
K1R ¼ K1 þ K1RS
ð13Þ
2
of DCl. Analysis of the dependence of the reaction rate on the
concentration of DCl (using plots of 1/kobs versus 1/[DCl]),
RS1
RS2
K1RS k2 þ K1RS k2
1
2
yields the values of (K1R)D and (kR2)D shown in Table 4. The data
k2R
¼
ð14Þ
K1RS þ K1RS
1
2
show that the overall isotope effect for transfer of the proton
from acid to [Ni(S2CR)(triphos)]+ {(K1Rk2R)H/(KR1kR2)D} is small for
all complexes and varies from an apparent inverse to a normal
kꢁR2 ¼ kꢁRS21 þ kꢁRS22
ð15Þ
Although the results from the DFT calculations indicate isotope effect (0.79–1.14). Inspection of the individual isotope
both S1 and S2 can be protonated (but chelate ring-opening is effects for KR1 and k2R reveals that the small isotope effects for
only associated with protonation of S2), the calculations give (KR1k2R)H/(K1Rk2R)D are a consequence of larger, but opposing,
no insight into the relative rates of each pathway. For example, isotope effects for the individual (KR1)H/(KR1)D and (kR2)H/(k2R)D.
protonation at S2 could be much faster than at S1, in which Thus, for all complexes, the formation of the hydrogen-bonded
case the simpler rate law {eqn (8)} associated with a single-site species {[Ni(S2CR)(triphos)]⋯HCl}+ (C) is associated with an
protonation mechanism is applicable. However if the rates of inverse equilibrium isotope effect {(KR1)H/(K1R)D = 0.44–0.85}
protonation of S1 and S2 are similar then eqn (12) is the accu- and the intramolecular proton transfer step is associated with
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
a normal primary isotope effect {(kR2)H/(k2R)D = 1.33–1.81}. It is derivative is slightly more acidic than the R = alkyl derivatives.
to be anticipated that DCl will be a slightly stronger acid than The different kinetics for the reaction of [Ni(S2CMe)(triphos)]+
HCl in MeCN and this is reflected in the values of (KR1)H/(KR1)D with mixtures of HCl and Cl− means that K1MeKM2 e, and hence
(<1.0). The normal kinetic isotope effects for (kR2)H/(k2R)D (>1.0) pKaMe, cannot be calculated.
are consistent with this step involving the intramolecular
transfer of a proton.17
The pKRa values indicate that these carboxydithioic acid
ligands (when bound to the {Ni(triphos)}2+ site) are signifi-
cantly more acidic than the free acids (in MeCN: EtCS2H, pKa =
13.4; PhCS2H, pKa = 13.2).19,22 Thus, coordination of the car-
The rates of protonation of [Ni(S2CR)(triphos)]+
The studies presented herein show that the rates of proton boxydithioic acid results in an increase in the acidity {Δ(pKa −
transfer to coordinated carboxydithioate in [Ni(S2CR)- pKEat) = 5.5; Δ(pKa − pKaPh) = 6.3}. Interestingly, a similar effect
(triphos)]+ (R = Me, Et, Bun or Ph), and deprotonation of the has been observed with PhSH coordinated to the
corresponding coordinated conjugate acid, are significantly {Ni(triphos)}2+ site {Δ(pKa − pKaPhSH) = 5.4}.5
slower than the diffusion-controlled limit. At least in part this
These calculations need to be treated with considerable
is because the reactions are thermodynamically unfavourable caution for two reasons. First, the pKas have been calculated
18
{e.g. (pKaHCl ꢁ pKaPhCS H) = 2.0 (vide infra)}. The kinetic ana- from kinetic data where proton transfer is slow. Previous
lyses presented above allow us to calculate the rate constants studies have indicated that proton transfer reactions with
for the intramolecular proton transfer step (kR2) within the lutH+ to [Ni(XPh)(triphos)]+ are slow, at least in part, because
hydrogen-bonded intermediate (C; R = Et, Bun or Ph). The steric interactions of the bulky phenyl substituents on triphos
results in Table 4 show that the values of kR2 are essentially interfere with the approach of the sterically demanding lutH+
independent of R. It is also evident from the data in Table 4 towards the basic ligand.7 We have suggested that calculating
that the protonation of [Ni(S2CMe)(triphos)]+ is appreciably the pKa of the coordinated PhXH, in these cases, does not cor-
faster than that of [Ni(S2CEt)(triphos)]+ or [Ni(S2CBun)- respond to the true basicity of the ligand, but rather reflects
(triphos)]+ (K1Mek2Me : K1Etk2Et : KB1ukB2u = 236 : 2 : 1). The reason for the rates of proton transfer to and from the PhX ligand.
the increased reactivity of [Ni(S2CMe)(triphos)]+ is not clear. Clearly, HCl is much less sterically demanding than lutH+ and
The difference in rates seems too large to be attributed to an this, together with the less sterically demanding 5-coordinate
electronic effect of the R group modulating the basicity of [Ni(S2CR)(triphos)]+, suggests that the values of pKaR deter-
sulfur; after all, the pKas of MeCO2H, EtCO2H and BunCO2H mined from the kinetics described in this paper are more
are all the same {pKa (in water) = 4.79 0.01}.19 It seems more likely to reflect the true acidity of the coordinated carboxy-
likely that the difference in rates is a consequence of the size dithioic acid. Secondly, if we consider the entirely general
of the R = alkyl substituents. Inspection of the values of K1R mechanism involving a two-site protonation model (Fig. 10),
and k2R shows that K1R = 76 9 dm3 mol−1, essentially indepen- then the associated rate law is eqn (12). In this case the pKas
dent of the R substituent. The difference in the rates is calculated from the kinetic data would be meaningful only
2
RS1
RS2
because the rate of intramolecular proton transfer for the R = if one pathway dominates (e.g. K1RS k2 > K1RS k2
and
1
2
Me derivative is at least 150 times faster than for the other R = kRS > kꢁRS2 ), or if both pathways occur at the same rate (i.e.
2
ꢁ21
RS1
RS2
2
RS1
RS2
alkyl derivatives. This suggests that the larger R substituents K1RS k2 ꢃ K1RS k2 and kꢁ2 ꢃ kꢁ2 ).
1
interfere with HCl hydrogen bonding to [Ni(S2CR)(triphos)]+
(in C), and attaining the optimal geometry for proton transfer.
An alternative possible explanation is that the significant
Conclusions
difference in the rates of protonation of [Ni(S2CMe)(triphos)]+
and the other derivatives is a consequence of a change in the The studies reported herein show that the rates of protonation
dominant site of protonation (S1 or S2).
of coordinated carboxydithioates are significantly slower than
the diffusion-controlled limit (kdiff ∼ 3.7 × 1010 dm3 mol−1 s−1
in MeCN).23 The reason for this slowness is, in part, because
The basicities of [Ni(S2CR)(triphos)]+
In principle, the elementary rate and equilibrium constants the reactions are thermodynamically unfavourable. However,
determined from analysis of the kinetics of the reactions of the rates of proton transfer are ca. 107–1010 slower than the
[Ni(S2CR)(triphos)]+ (R = Et, Bun or Ph) with mixtures of HCl diffusion-controlled limit and the thermodynamic argument
and Cl− allow calculation of the pKRa of the coordinated cannot account for this large difference. It seems likely, as
carboxydithioic acids.
suggested earlier for the protonation reactions of [Ni(XPh)-
The equilibrium constant for the protonation of [Ni(S2CR)- (triphos)]+ (X = O, S or Se), that steric factors contribute to the
(triphos)]+ in the single site model is KR1KR2 = K1RkR2/kR−2. These slowness of the reactions.7
values are presented in Table 4. Using the literature pKa of HCl
The kinetics for the protonation of [Ni(S2CR)(triphos)]+ (R =
in MeCN (8.9),20 and the value of KR1KR2, the pKaR of [Ni(S2CR)- alkyl) with HCl indicate a two-step equilibrium mechanism.
(triphos)]+ in MeCN can be calculated: pKaEt = 7.9; pKBau = 8.3; The initial step involves the rapid formation of a hydrogen-
pKPah = 6.9. The notable feature is that pKaR is rather insensitive bonded precursor (KR1) and this is followed by intramolecular
to the R-substituent. Similar levelling effects for acidities of proton transfer (kR2). It seems likely that for all [Ni(S2CR)-
ligands have been noted before.7,21 As expected, the R = Ph (triphos)]+ (R = alkyl or aryl) the same mechanism operates but
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
only when R = alkyl does the precursor accumulate to concen-
trations sufficient to perturb the kinetics. The measured
kinetic isotope effects for the complete transfer of a proton
from HCl to [Ni(S2CR)(triphos)]+ (R = Me, Et or Bun) are small
{(KR1k2R)H/(K1RkR2)D = 0.79–1.14}. This is the result of a small
inverse equilibrium isotope effect for the formation of the pre-
cursor hydrogen-bonded adduct {(KR1)H/(K1R)D = 0.44–0.85} and
a significant kinetic primary isotope effect for the intramolecu-
6 C. Cauquis, A. Deronzier, D. Serve and E. Vieil, J. Electro-
anal. Chem. Interfacial Electrochem., 1975, 60, 205.
7 A. Alwaaly and R. A. Henderson, Chem. Commun., 2014, 50,
9669.
8 W. Clegg and R. A. Henderson, Inorg. Chem., 2002, 41,
1128.
9 C. Bonnans-Plaisance and J.-C. Gressiw, J. Chem. Educ.,
1988, 65, 93, and refs therein.
lar proton transfer step {(kR2)H/(k2R)D = 1.33–1.81}. Finally, DFT 10 (a) CrysAlisPro software, Agilent Technologies Ltd., Oxford,
calculations indicate that protonation can occur at either the
equatorial or the axial sulfur sites, but that chelate ring-
opening of the carboxydithionic acid occurs only after protona-
tion of the equatorial sulfur. The ways in which parallel proto-
nation of the two sulfur sites would complicate the analyses
and interpretation of the kinetics have been discussed.
UK, 2013; (b) APEX2, SADABS and SHELXTL, Bruker AXS
Inc., Madison, Wisconsin, USA, 2014; (c) CrystalClear,
Rigaku Corp., Tokyo, Japan, 2013; (d) G. M. Sheldrick,
SHELXL-2014, University of Göttingen, Germany, 2014;
(e) OLEX2, O. V. Dolomanov, L. J. Bourhis, R. J. Gildea,
J. A. K. Howard and H. Puschmann, J. Appl. Crystallogr.,
2009, 42, 339.
11 W. Addison, T. Nageswara Rao, J. Reedijk, J. van Rijn and
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349.
This paper defines the angular structural parameter, τ =
(β − α)/60 (where α = smaller of basal angles and β = larger
of basal angles). The parameter τ is an index of trigonality
in 5-coordinate complexes. For a perfect square-based
pyramid, τ = 0 and for a perfect trigonal bipyramid τ = 1.0.
For the complexes described herein β = P2NiS1 and α =
P1NiS2. Hence, R = Et, τ = 0.84; R = Ph, τ = 1.15; R =
C6H4OMe-4, τ = 1.01; R = C6H4Cl-4, τ = 0.96.
Acknowledgements
The authors acknowledge the use of the EPSRC UK National
Service for Computational Chemistry Software (NSCCS) at
Imperial College London in carrying out this work. A.A. thanks
the Iraqi Ministry of Higher Education and Scientific Research
for a studentship. Crystallographic studies were part-funded by
an EPSRC equipment grant, facilitated by access to equipment
at Durham University Chemistry Department for one of the
structures during refurbishment at Newcastle, and made poss-
ible for two of the compounds by access granted to beamline
I19 at Diamond Light Source, for which we are very grateful.
12 A. Petrou, A. D. Koutselos, H. S. Wahab, W. Clegg,
R. W. Harrington and R. A. Henderson, Inorg. Chem., 2011,
50, 847.
13 R. G. Wilkins, Kinetics and Mechanisms of Reactions of Tran-
sition Metal Complexes, VCH, Weinheim, 2nd edn, 1991,
p. 24.
Notes and references
1 (a) K. W. Kramarz and J. R. Norton, Prog. Inorg. Chem., 14 J. F. Coetzee, Prog. Phys. Org. Chem., 1967, 4, 45.
1994, 42, 1, and refs therein; (b) R. A. Henderson, Angew. 15 V. Autissier, W. Clegg, R. W. Harrington and
Chem., Int. Ed. Engl., 1996, 35, 946, and refs therein.
R. A. Henderson, Inorg. Chem., 2004, 43, 3098.
2 (a) M. Eigen, Angew. Chem., Int. Ed. Engl., 1964, 3, 1; 16 M. J. Frisch, et al., Gaussian 03, revision D.01, Gaussian,
(b) M. M. Kreevoy, D. S. Sappenfield and W. Schwabacher,
J. Phys. Chem., 1965, 69, 2287.
3 (a) C. F. Bernasconi and R. L. Montanez, J. Org. Chem.,
1997, 62, 8162; (b) C. F. Bernasconi, D. M. Weirsema and
W. Stronach, J. Org. Chem., 1993, 58, 217, and refs therein.
4 (a) R. H. Holm, P. Kennepohl and E. I. Solomon, Chem.
Inc., Wallingford, CT, 2004.
17 (a) S. E. Scheppele, Chem. Rev., 1972, 72, 511, and refer-
ences therein; (b) G. Parkin and J. E. Bercaw, Organometal-
lics, 1989, 8, 1172.
18 R. P. Bell, The Proton in Chemistry, Chapman and Hall,
London, 2nd edn, 1973, ch. 11.
Rev., 1996, 96, 2239; (b) S. C. Lee and R. H. Holm, Chem. 19 Advanced Chemistry Development (ACD/Labs) Software
Rev., 2004, 104, 1135; (c) D. W. Mulder, M. W. Ratzloff, V11.02 (©1994–2014 ACD/Labs).
E. M. Shepard, A. S. Byer, S. M. Noone, J. W. Peters, 20 K. Izutsu, Acid–Base Dissociation Constants in Dipolar
J. B. Broderick and P. W. King, J. Am. Chem. Soc., 2013, 135, Aprotic Solvents, Blackwell Scientific, Oxford, UK, 1990.
6921; (d) B. M. Hoffman, D. Lukoyanov, Z.-Y. Yang, 21 B. L. Westcott, N. E. Gruhn and J. H. Enemark, J. Am.
D. R. Dean and L. C. Seefeldt, Chem. Rev., 2014, 114, 4041; Chem. Soc., 1998, 120, 3382.
(e) A. Jablonskyte, L. R. Webster, T. R. Simmons, 22 Calculated pKa (in water) from ref. 19 then correcting to the
J. A. Wright and C. J. Pickett, J. Am. Chem. Soc., 2014, 136,
13038.
5 V. Autissier, P. M. Zarza, A. Petrou, R. A. Henderson,
pKa in MeCN using equations in, B. G. Cox, Acids and Bases
Solvent Effects on Acid–Base Strength, Oxford University
Press, UK, 2013.
R. W. Harrington and W. Clegg, Inorg. Chem., 2004, 43, 23 F. Wilkinson, A. F. Olea, D. J. McGarvey and D. R. Worrall,
3106.
J. Braz. Chem. Soc., 1995, 6, 211–220.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.