Transformation of sulfur dioxide to sulfate at a palladium centre

Article abstract of DOI:10.1039/b205548a

Reaction of 4,4′,4″-tri(tert-butyl)-2,2′:6′,2″-terpyridine (terpy*) with gaseous SO₂-O₂ mixtures and [PdCl₂(MeCN)₂] followed by aqueous workup affords crystallographically characterised [PdCl(terpy*)]Cl and both O- and S-bound [Pd(SO₃)(terpy*)]. The former is also available from [PdCl₂(MeCN)_n] (n = 0, 2) and terpy*; the latter from aerobic oxidation of [Pd(dba)₂]-terpy*-SO₂ mixtures (dba = dibenzylideneacetone). Oxidation of [Pd(SO₃)(terpy*)] with O₂ (at 210°C in PhNO₂) yields crystallographically characterised O-bound [Pd(SO₄)(terpy*)]. The crystal structures of two related compounds are also reported: [Pd(OAc)(terpy*)]Cl and [PdCl₂(bipy*)].

Full text of DOI:10.1039/b205548a

View Article Online / Journal Homepage / Table of Contents for this issue
Transformation of sulfur dioxide to sulfate at a palladium centre
Elizabeth J. MacLean,a Richard I. Robinson,b Simon J. Teat,a Claire Wilsonb and
Simon Woodward*b
a Daresbury Laboratory, Daresbury, Warrington, Cheshire, UK WA4 4AD
b School of Chemistry, The University of Nottingham, University Park, Nottingham,
UK NG7 2RD. E-mail: simon.woodward@nottingham.ac.uk
Received 7th June 2002, Accepted 23rd July 2002
First published as an Advance Article on the web 14th August 2002
Reaction of 4,4Ј,4Љ-tri(tert-butyl)-2,2Ј:6Ј,2Љ-terpyridine (terpy*) with gaseous SO2–O2 mixtures and [PdCl2(MeCN)2]
followed by aqueous workup affords crystallographically characterised [PdCl(terpy*)]Cl and both O- and S-bound
[Pd(SO3)(terpy*)]. The former is also available from [PdCl2(MeCN)n] (n = 0, 2) and terpy*; the latter from aerobic
oxidation of [Pd(dba)2]–terpy*–SO2 mixtures (dba = dibenzylideneacetone). Oxidation of [Pd(SO3)(terpy*)] with O2
(at 210 ЊC in PhNO2) yields crystallographically characterised O-bound [Pd(SO4)(terpy*)]. The crystal structures of
two related compounds are also reported: [Pd(OAc)(terpy*)]Cl and [PdCl2(bipy*)].
the structures was initially complicated by the lack of con-
Introduction
venient spectroscopic handles in the co-ordinated anions and
There is an increasing interest in reactions featuring “soft”
by the tendency of anions to readily dissociate during positive
transition metals in combination with ligands that are normally
ion mass spectrometry.
considered “hard”.1 One under-utilised substrate in this respect
Literature reactions of palladium bipy/terpy complexes with
is sulfur dioxide and in particular its oxidation reactions. Sulfur
SO2 appear to be limited to the studies of Esmardi6 (who pre-
dioxide is employed on industrial scale in aerobic, free radical-
pared non-crystalline [PdCl2(SO2)(terpy)], [Pd(SO4)(bipy)]
based, alkylsulfonic acid production from hydrocarbons.2
and related species by reaction of [Pd(H)Cl2(CO)]Ϫ with SO2
Similarly, SO2 is used as a promoter in heterogeneous platinum-
followed by ligand addition). Proton NMR investigation of
catalysed alkane oxidations.3 While the basic features of sulfur
our reaction mixtures suggested only the presence of [PdCl-
dioxide co-ordination chemistry have been well defined4 there
(terpy*)]ϩ cations after treatment with SO2–O2. This raises the
have been few homogeneous models of such SO2 oxidation
possibility that the sulfite ligand in 1 arises from sulfurous acid
processes.5 We report here our studies of SO2 oxidation and
generated by hydrolysis of SO2 during crystallisation of the
hydrolysis reactions in the presence of palladium complexes
reaction mixture. This idea was tested by exposure of an
of the ligands 4,4Ј-di(tert-butyl)-2,2Ј-bipyridine (bipy*) and
4,4Ј,4Љ-tri(tert-butyl)-2,2Ј:6Ј,2Љ-terpyridine (terpy*).
acetonitrile solution of [PdCl(terpy*)]Cl to an aqueous solution
of SO2. The sulfite complex 1 is not prepared by this chemistry
but a new species is formed; however, its high lability prevented
its isolation. To overcome these problems we undertook a
campaign of synthesis and crystallographic study to see if
the electronic nature of the co-ordinated anion(s) in these
species could be related in any way to the chemical shift of the
diagnostic pyridyl protons in the co-ordinated terpy* and bipy*
ligands.
Preparation of sulfite complexes
Results and discussion
The non-reproducibility of our initial SO2 studies led us to
seek an alternative preparation of [Pd(SO3)(terpy*)] 1(1a).
Saturation of THF solutions [Pd(dba)2] (dba = dibenzylidene-
acetone) containing terpy* led to the formation 1 in low (30%),
but reproducible, yield. It is possible that this reaction proceeds
via a labile palladium SO2 species that undergoes subsequent
terpy* ligation and aerobic oxidation as under these anhydrous
conditions no SO2 hydrolysis is possible. Such behaviour would
account for the variable yield of 1(1a) in the initial studies using
PdII precursors where the multiple Pd0/PdII redox reactions
required would be critically dependent on the local SO2 and
O2 concentrations. Attempts to prepare complexes from SO2
saturated CDCl3 solutions containing [Pd(dba)2] and terpy* led
only to very labile species. Complex 1(1a) could be crystallised
as a trihydrate from aqueous acetonitrile or as a solvate from
CDCl3. Both morphologies were subjected to X-ray analysis.
The molecular structure of the chloroform-derived crystal
is shown in Fig. 1, while selected bond distance and angle data
SO2 Activation studies
Bubbling mixtures of gaseous SO2–O2 through an acetonitrile
solution of [PdCl2(MeCN)2] in the presence of terpy* leads to
the formation of a lemon yellow solution. Subsequent addition
of aqueous methanol effected the production of up to three
compounds. Under favourable conditions precipitation of
single crystals occurred which were subsequently identified as
the sulfite complex [Pd(SO3)(terpy*)] 1/1a while on other occa-
sions only [PdCl(terpy*)]Cl 2 was formed. The ratio of these
species was found to be dependent of the reaction micro-
conditions; minor variations in the SO2–O2 flow rates and other
conditions had dramatic and variable effects on the 1(1a) : 2
ratio. If the reaction was allowed to stand for extended periods
then samples of MeC(᎐O)NH were obtained through the
᎐
2
hydrolysis of the acetonitrile solvent. This latter reaction
appears to be catalysed by the 1–2 mixtures. Confirmation of
3518
J. Chem. Soc., Dalton Trans., 2002, 3518–3524
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b205548a

View Article Online
Fig. 2 Molecular structure of the O-bound component [Pd(SO3)-
(terpy*)] 1a (acetonitrile–water morphology). Hydrogen atoms are
omitted for clarity as is lattice water.
the sulfite ligand may also be a contributing factor in the
isolation of 1(1a) in our SO2 activation studies in aqueous
acetonitrile. The [Pd(SO3)(terpy*)] 1/1a complex (either mor-
phology) gave strong IR bands at 1260 cmϪ1 [ν(S᎐O)asym],
᎐
1133 cmϪ1 [δ(S᎐O)sym], 1100 cmϪ1 [δ(S᎐O)sym] and 983 cmϪ1
᎐
᎐
[ν(S–O), where the assignments have been made based on litera-
ture precedent.6 Isolated 1a also exhibited absorption bands
due to lattice water [3488 cmϪ1, ν(O–H); 1612 cmϪ1, ν(H–O–H)].
At the onset of our studies few palladium complexes of bipy*
and terpy* had been prepared. To allow NMR and structural
comparisons with 1 two chloro and one acetate complexes
were prepared. Direct reaction of PdCl2 and the relevant ligand
fashioned in moderate yields [PdCl(terpy*)]Cl 2 but was an
acceptable route for the preparation [PdCl2(bipy*)] 3. Near
quantitative yields of 2 are realised using [PdCl2(MeCN)2] as
a soluble palladium source and during the course of our
studies Osborn and co-workers reported an essentially identical
procedure using [PdCl2(PhCN)2].9 Both complexes 2 and 3
could be crystallised and their molecular structures are shown
Fig. 1 Molecular structure of S-bound [Pd(SO3)(terpy*)] 1 (CDCl3
morphology). Hydrogen atoms are omitted for clarity as are hydrogen
bonded and lattice CDCl3. Only the major component of a disordered
But fragment is shown.
is summarised in Table 1. Interestingly, the equivalent aceto-
nitrile–water morphology shows disorder in the sulfite unit and
25% of the SO3 units are O-bound. An ORTEP plot of O-
bound [Pd(SO3)(terpy*)] 1a is shown in Fig. 2 with associated
data presented in Table 1. Such linkage isomerism in palladium
sulfite complexes does not appear to have been reported before;
however this phenomenon has been noted in Pt() complexes.7
Previously crystallographically characterised palladium sulfite
complexes appear to be limited to cis-Na2[Pd(SO3)2(en)] which
also contains an S-bound SO3 (Pd–S, 2.269; S–Oave, 1.484 A;
O–S–Oave, 108.5Њ).8 The similarity of these values to 1 is con-
sistent with its formulation as a sulfite rather than a sulfur tri-
oxide complex. In particular, no evidence for the presence of
any O-bound protons could be found in the residual electron
density surrounding the SO3 ligand. In the CDCl3 morphology
the terpy* ligand of 1 shows a “gull-wing” deformation. This
distortion probably arises out of conflict between the terpy*
bite angle [N(1Ј)–Pd–N(1Љ) = 158.7Њ] and the favoured 180Њ
geometry of d8 square planar complexes. In the CDCl3 mor-
phology a solvent molecule resides near the Pd centre, while
three CDCl3 molecules sit in a pyramidal fashion around the
sulfite ligand and hydrogen bond to the O-atoms of the ligand.
Similar hydrogen bonds (to water) are seen in the trihydrate
morphology. In all these cases the Oؒ ؒ ؒH(D) contacts are in
the range 2.03–2.11 Å. The 1H solution NMR spectra of S- and
O-bound 1 and 1a are identical suggesting equilibration
between the two forms on an NMR timescale. Such lability for
Fig. 3 Molecular structure of [PdCl(terpy*)]Cl 2. Hydrogen atoms
and the chloride counter anion are omitted for clarity.
in Figs. 3 and 4 respectively. The co-ordination geometries dis-
played by 2–3 are largely unaffected by the electron rich tert-
butyl substituted ligands. For example, the Pd–Cl distances
are very similar to those in the unsubstantiated complexes: [Pd-
Cl(terpy)]Cl (2.134 Å)10 and [PdCl2(bipy)] (2.276, 2.319 Å).11
Reaction of Pd(OAc)2 with terpy* led to clean formation of
the acetate; in this case the product readily crystallised as [Pd-
(OAc)(terpy*)]Cl 4 from dichloromethane–hexane. The halo-
genated solvent is apparently the source of the outer-sphere
counter ion. The molecular structure of [Pd(OAc)(terpy*)]Cl 4
J. Chem. Soc., Dalton Trans., 2002, 3518–3524
3519

View Article Online
Table 1 Selected intramolecular distance (Å) and angle (Њ) data for complexes 1–2 and 4–6
[Pd(S-SO3)-
(terpy*)] 1
[Pd(O-SO3)-
(terpy*)] 1a
[PdCl-
(terpy*)]Cl 2
[Pd(OAc)-
(terpy*)]Cl 4
[Pd(SO4)-
(terpy*)] 5
[(NHSO2Ar)-
(terpy*)] 6a
Pd–N(1)
Pd–N(1Ј)
Pd–N(1Љ)
Pd–Cl
Pd–S
Pd–O
2.008(4)
2.069(5)
2.075(5)
—
2.2457(13)
—
1.986(4)
2.083(4)
2.085(4)
—
1.929(2)
2.030(2)
2.020(2)
2.2980(7)
—
—
—
1.928(2)
2.012(2)
2.009(2)
—
1.918(4)
2.207(5)
2.018(5)
—
—
2.001(4)
—
1.425(5)
1.453(5)
1.474(5)
1.530(4)
1.935(6)
2.023(7)
2.022(7)
—
—
—
2.027(7)
1.431(8)
1.445(7)
2.267(2)b
2.03(2)
—
—
2.0448(15)
—
—
Pd–N(4)
S–O
—
1.448(4)
1.456(5)
1.457(4)
1.483(4)b
1.52(2)
1.498(5)b
1.50(2)
1.536(15)b
1.33(5)
—
N(1)–Pd–X
N(1Ј)–Pd–X
N(1Љ)–Pd–X
176.84(11)
(X = S)
170.98(10)
(X = S)
178.02(8)
(X = Cl(1))
175.23(7)
(X = O(1))
173.1(2)
(X = O(1))
176.4(3)
(X = N(4))
161.8(5)
(X = O(1A)
96.74(10)
(X = S)
101.10(13)
(X = S)
100.07(7)
(X = Cl(1))
96.83(6)
(X = O(1))
106.6(2)
(X = O(1))
96.5(3)
(X = N(4))
117.8(5)
(X = O(1A))
104.47(10)
(X = S)
101.34(12)
(X = S)
98.26(7)
(X = Cl(1))
100.99(6)
(X = O(1))
91.9(2)
(X = O(1))
101.9(3)
(X = N(4))
83.3(5)
(X = O(1A))
79.29(14)
79.40(14)
158.68(14)
118.2(9)
(X = O(1A)
111.8(3)b
109.9(7)b
110.4(7)b
111(2)
N(1)–Pd–N(1Ј)
N(1)–Pd–N(1Љ)
N(1Ј)–Pd–N(1Љ)
Pd–X–S
79.0(2)
78.7(2)
157.5(2)
—
80.59(10)
81.01(10)
161.53(10)
—
81.41(7)
80.82(7)
162.18(6)
—
80.2(2)
81.3(2)
81.0(3)
80.5(3)
161.5(3)
117.1(4)
(X = N(4))
119.0(5)
161.5(2)
123.4(2)
(X = O(1))
113.0(3)
110.9(3)
111.6(3)
107.7(3)
108.1(2)
105.2(3)
O–S–O
111.2(3)
109.8(3)
110.4(3)
—
—
99(2)
104.8(9)
a From ref. 9 which uses a different atomic numbering scheme translated here as N(2) = N(1), N(1) = N(1Ј), N(3) = N(1Љ), with N(4) being the
sulfonamido nitrogen. b S-bound component in crystal.
Fig.
4
Molecular structure of [PdCl2(bipy*)] 3. Selected bond
distances and angles Pd1–N1 2.015(3), Pd1–N1Ј 2.022(3), Pd1–Cl1
2.2763(9), Pd1–Cl2 2.2834(9) Å; N1–Pd1–N1Ј 79.59(10), N1–Pd1–Cl1
95.36(8), N1Ј–Pd1–Cl1 174.56(7) N1–Pd1–Cl2 174.02(7) N1Ј–Pd1–Cl2
95.31(7), Cl1–Pd1–Cl2 89.84(3)Њ. Hydrogen atoms are omitted for
clarity.
Fig. 5 Molecular structure of [Pd(OAc)(terpy*)]Cl 4. Hydrogen atoms
and the chloride counter anion are omitted for clarity.
[PdCl(terpy*)]OAc resulting in a 1 : 1 mixture of these two
species in solution.
is shown in Fig. 5 while selected bond distance and angle
data appear in Table 1. When dissolved in dichloromethane
crystallographically pure 4 gives rise to two sets of exchanging
signals in its H NMR spectrum. This feature appears to be
due to facile exchange between [Pd(OAc)(terpy*)]Cl 4 and
Aerobic oxidation of [Pd(SO3)(terpy*)] 1 to [Pd(SO)4(terpy*)] 5
Because of the HSAB (Hard–Soft–Acid–Base) miss-matching
of the sulfate ligand and palladium centre we were keen to
explore the preparation of [Pd(SO4)(terpy*)] 5 via aerobic
1
3520
J. Chem. Soc., Dalton Trans., 2002, 3518–3524

View Article Online
oxidation. When d5-PhNO2 solutions of [Pd(SO3)(terpy*)]
1
were heated to 210 ЊC under an oxygen atmosphere H NMR
spectroscopy indicated smooth formation of a single new
species. This species could be easily isolated and characterised
crystallographically as the sulfate complex 5. The preparation
could also be obtained on a preparative scale, although in this
case it was necessary to bubble oxygen gas through the refluxing
solution. Attempts to carry out the transformation of [Pd(SO3)-
(terpy*)] 1 to [Pd(SO4)(terpy*)] 5 aerobically in chlorobenzene
at 135 ЊC led to only traces of the product being formed. The
isolated [Pd(SO4)(terpy*)] 5 complex shows a number of
characteristic IR bands typical for S–O vibrational frequencies.
this only a handful of η1-SO4 complexes of platinum have been
reported.14
All other attempts to prepare complex 5 by alternative routes
were unsuccessful. For example, no coordination of terpy* with
PdSO4 was observed even under forcing conditions. Ligand
substitution reactions of [PdCl(terpy*)]Cl 2 with Li2SO4 or 2 M
H2SO4 were unsuccessful. Use of more forcing conditions using
concentrated H2SO4 with [PdCl(terpy*)]Cl 2 resulted in the pro-
tonation of terpy* and precipitation of palladium black, while
reactions of [Pd(SO3)(terpy*)] 1 with 30% H2O2 or chloride
abstraction from 2 with AgPF6 followed by addition of various
sulfate sources both led to uncharacterisable materials.
Bands at 1259, 1134, 981 cmϪ1 were assigned to ν(S᎐O)asym,
᎐
δ(S᎐O)sym and ν(S–O) stretches based on the limited literature
᎐
precedent.6 Two very strong peaks are observed at 1522 and
1345cmϪ1 in the sulfate complex 5 which are not present in the
IR spectrum of the sulfite 1. However, we cannot assign these
peaks as there are few appropriate examples with fully assigned
IR data. As the IR spectra did not allow unambiguous assign-
ment of the bonding mode of the sulfate ligand an X-ray study
was undertaken. The structure of 5 is shown in Fig. 6, while a
Spectroscopic comparison of 1–6
The ortho protons of the bipy* and terpy* ligands give charac-
1
teristic doublets in their H NMR spectra in the range 10.01–
8.71 ppm. The values for compounds 1–6 are summarised in
Table 2. The nature of the anion bound to the metal centre
clearly affects the metal’s electronic demand on the terpy* lig-
and. This effect gives a clue to why the chloride species appears
to be predominantly isolated during the course of these
1
experiments. Using H NMR spectroscopy the chloride anion
can be seen to place a much lower electronic demand (effective
inductive electron withdrawal from the terpy*) via the pal-
ladium centre, than the corresponding sulfite and sulfate. This
strongly suggests that in solution the sulfite and sulfate groups
are rather labile. The viability of such ionisation processes is
further demonstrated by the chloride–acetate exchange demon-
strated by [Pd(OAc)(terpy*)]Cl 4 in solution. The lability of the
sulfite and sulfate ligands in 1/1a/3 leads to a higher frequency
shift in the H6Ј,6Љ protons of the terpy* ligand due to the greater
contribution of the [Pd(terpy*)]2ϩ cation. The lability of the
sulfite and sulfate groups also leads to preferred isolation of
[PdCl(terpy*)]Cl 2 under the majority of reaction conditions.
Fig. 6 Molecular structure of [Pd(SO4)(terpy*)] 5. Hydrogen atoms
are omitted for clarity. Only the major component of a disordered But
fragment is shown.
summary of the useful bond distance and angle data is given in
Table 1. Based on inspection of the Cambridge Crystallo-
graphic Database,12 no structures of palladium() sulfate com-
plexes are available, only some solution species are known.13
The sulfate in 5 adopts an η1 bonding mode with an Pd–O
distance of 2.001(4) Å and a Pd–O–S angle of 123.4(2)Њ. The
structure of 5 is clearly closely related to that of Osborn’s
sulfonamido complex 6 which shows a Pd–N distance of
2.027(7) Å and a Pd–N–S angle of 117.1(4)Њ.9 Aside from
Conclusions
We have been able to demonstrate the sequential oxidation of
sulfur dioxide to sulfate at a palladium centre via an intermedi-
ate S- or O-bound sulfite complex [Pd(SO3)(terpy*)] 1/1a. The
oxidation of the latter only proceeds under forcing aerobic con-
ditions and provides a rare example of a crystallographically
characterised palladium η1-sulfate complex, [Pd(SO4)(terpy*)]
J. Chem. Soc., Dalton Trans., 2002, 3518–3524
3521

View Article Online
Table 2 Selected 1H NMR dataa for complexes 1–2 and 4–6
δ (J/Hz)
[Pd(SO3)(terpy*)] 1
[PdCl(terpy*)]Cl 2
[Pd(OAc)(terpy*)]Cl 4b
[Pd(SO4)(terpy*)] 5
[Pd(NHSO2Ar)(terpy*)]BF4 6c
H6Ј/6Љ
H3/5
H3Ј/3Љ
H5Ј/5Љ
10.01 (6.0)
7.97
7.93 (1.0)
7.62 (6.0, 1.0)
8.76 (6.0)
9.04
9.01 (2.0)
7.58 (6.0, 2.0)
8.71 (broad d)
8.68
8.54
9.27 (6.0)
8.14
8.09 (1.0)
7.61 (6.0, 1.0)
8.14 (6.0)
8.28
8.28
7.88 (6.0, 2.0)
7.54 (6.0, 2.0)
a In CDCl3 at ambient temperature. b 1 : 1 mixture with 2. c From ref. 9.
5. The nature of the bound ligand, X, in [PdX(terpy*)] com-
plexes engenders dramatic shifts in the proton chemical shifts
of the H6Ј,6Љ terpyridyl protons when these are labile (SO32– and
SO42–).
followed by the addition of terpy* (155 mg; 0.39 mmol) in one
portion. The reaction mixture was refluxed for (48 h) after
which time TLC indicated only trace amounts of terpy*. The
reaction mixture was allowed to cool to room temperature
and an orange precipitate which formed was filtered off, dried
under high vacuum to yield 197 mg (93%). Alternatively, PdCl2
(220 mg; 1.12 mmol) in water (20 ml) was boiled with terpy*
(500 mg; 1.2 mmol) for 3 h and the resulting brown precipitate
purified by trituration with diethyl ether, then ethyl acetate to
yield orange microcrystals (233 mg; 32%).
Experimental
General
THF was distilled from sodium benzophenone ketyl under an
argon atmosphere, hexane was dried over Na wire, dichloro-
methane was distilled from CaH2, petrol refers to the fraction
boiling at 40–60 ЊC unless otherwise stated. Proton and 13C
NMR spectra were recorded on the Bruker AM400, AV400,
DRX500, and JEOL GX270 machines at ambient as stated. IR
spectra were recorded on a Perkin-Elmer FTIR 1600 spectro-
meter. Mass spectra were recorded using either FAB or ES
ionisation mode on a VG micromass 70E or AIMS902 at the
University of Nottingham, or by ES at the EPSRC service,
University of Swansea. Mixtures of bipy* and terpy* were
prepared by a literature method.15 The mixture was absorbed
onto silica gel and was heated in a Kugelrohr oven at 160 ЊC
(0.3 mmHg) until all the bipy* had sublimed, leaving terpy*
adsorbed onto the silica surface. The terpy* was recovered
by dichloromethane extraction. The following compounds
were prepared by literature methods: [PdCl2(MeCN)2],16 [Pd-
(dba)2].17
[Pd(SO3)(terpy*)] 1/1a. Mp > 240 ЊC (dec., from aq. MeCN).
Anal. Calc. for C27H35N3O3PdSؒ3H2O: C, 50.5; H, 6.4; N, 6.5%.
Found: C, 50.1; H, 6.1; N, 6.5%. δH (400 MHz, CDCl3): 10.01
(d, 2 H, J = 6.0, H6Ј/6Љ); 7.97 (s, 2 H, H3/5); 7.93 (d, 2 H, J = 1.0,
H3Ј/3Љ); 7.62 (dd, 2 H, J = 6.0, 1.0, H5Ј/5Љ); 1.55 (s, 9 H, But);
1.42 (s, 18 H, But). δc (67.80 MHz, CDCl3): 167.0, 165.0, 155.6,
155.1, 152.6, 125.0, 119.5, 118.7, 36.2 (But), 35.7 (But), 30.7
(But), 30.3 (But). νmax (KBr disc)/cmϪ1 3488s (included H2O),
3049m, 2960s (C–H), 2870m, 2360s, 2342, 1612s, x1560w
(C᎐C), 1542w, 1475m (C᎐C), 1426m, 1365w, 1260m ν(S᎐O)
,
᎐
᎐
᎐
asym
1133s δ(S᎐O) , 1100s δ(S᎐O)sym, 983s (S–O), 898.2m ν (S–O),
᎐
᎐
sym
858w, 743w, 668m, 638s, 612m. MS: m/z (ESϩ) 588 (Mϩ, 20%);
610 (MNaϩ, 100%) [Found (HRMS): MNaϩ 610.1344,
C27H35NaSPdN3O3 requires 610.1332].
[PdCl(terpy*)]Cl 2. Mp > 240 ЊC (dec., from CDCl3). Anal.
Calc. for C27H35N3Cl2Pdؒ2.5CDCl3: C, 40.6; H, 4.1; N, 4.9%.
Found: C, 40.9; H, 4.4; N, 4.6%. δH (400 MHz, CDCl3); 9.04
(s, 2 H, H3/5); 9.01 (d, 2 H, J = 2.0, H3Ј/3Љ); 8.76 (d, 2 H, J = 6.0,
H6Ј/6Љ); 7.58 (dd, 2 H, J = 6.0, 2.0, H5Ј/5Љ); 1.63 (s, 9 H; But); 1.53
(s, 18 H; But). δc (67.8 MHz, CDCl3) 170.0, 168.3, 158.0, 154.5,
152.0, 124.5, 123.9, 123.0, 37.7 (But), 36.5 (But), 30.8 (But), 30.4
(But). νmax (KBr disc)/cmϪ1 3422m (included water), 2965s
Preparations
[Pd(SO3)(terpy*)] 1/1a or [PdCl(terpy*)]Cl 2 initial method.
Crystalline [PdCl2(MeCN)2] (250 mg; 0.96 mmol) was dissolved
up into acetonitrile (15 ml) at room temperature and stirred
vigorously for 10 min to give a bright orange solution. The
reaction mixture was flushed with O2 (50 ml minϪ1)–SO2
(100 ml minϪ1) bubbling directly into the solution. Solid terpy*
(100 mg; 0.25 mmol) was added in one portion to give a suspen-
sion, after 10 min the ligand had fully dissolved. The remaining
terpy* (233 mg; 0.58 mmol) was added and the reaction mixture
was left stirring under O2–SO2 gas flow at room temperature.
The gas flow was stopped after 2 h and the reaction mixture
flushed with O2 to remove SO2 gas. Water (20 ml) and methanol
(2 ml) were added to the reaction mixture and the yellow
solution was left to stand (48 h). Five nominally identical repe-
titions led to the isolation of [Pd(SO3)(terpy*)] 1/1a (53–68%, as
a trihydrate 1/1a that could be recrystallised from CDCl3 as 1,
runs 1–2); [PdCl(terpy*)]Cl 2 (95%, run 3–4); MeC(=O)NH2
(formed continuously, >2 g isolated typically, run 5).
(C–H), 1614s, 1561m (C᎐C), 1469m (C᎐C), 1413m, 1368w,
᎐
᎐
1261m, 1178w, 1022w, 920w, 872m, 735w, 608w. MS: m/z (ESϩ)
542 (Mϩ, 100%).
Acetamide, MeC(؍O)NH2. Acetamide was identified by
comparison of its 1H NMR spectrum with that of an authentic
sample. The crystals that separated from the aqueous
acetonitrile–methanol mixture showed the following data:
rhombohedral form space group R3c; a 11.526, c 13.589 Å
(CCDC code: ACEMIDO1).
[PdCl2(bipy*)] 3. Palladium dichloride (220 mg, 1.12 mmol)
in water (20 ml) was boiled with bipy* (300 mg, 1.12 mmol).
After heating for 30 min further portions of water (5 ml) and
ethanol (2.5 ml) were added to improve solubility. The reaction
was left to reflux for 50 h, yielding the crude product as a black
crude suspension. The crude solid was stirred vigorously in
dichloromethane for 1 h resulting in a yellow solution and fine
black precipitate (Pd0). The precipitated metal was filtered off
through Celite, and the filtrate was removed under reduced
pressure to yield a yellow solid. The solid was taken up in the
minimum volume of dichloromethane and recrystallisation was
achieved by addition of hexane and cooling in the freezer to
yield fine orange needles (332 mg; 67%). Mp >193 ЊC (dec.).
Anal. Calc. for C27H35N3Cl2Pd: C, 48.5; H, 5.3; N, 6.3%.
Found: C, 48.00; H, 5.3; N, 6.3%. δH (400 MHz, CDCl3): 9.30
(d, 2 H, J = 6.0, H6/6Ј); 7.89 (d, 2 H, J = 2.0, H3/3Ј); 7.52 (dd, 2 H,
[Pd(SO3)(terpy*)] 1 alternative method. Solid [Pd(dba)2]
(143 mg; 0.25 mmol) (dba = dibenzylideneacetone) and terpy*
(100 mg; 0.25 mmol) were dissolved in dry THF (5 ml) under an
argon atmosphere. The resulting dark red solution was stirred
at room temperature under a constant stream of SO2 gas (80 ml
minϪ1) bubbling through the solution (2 h) after which the
solution had turned yellow. A very pale green precipitate was
filtered off yielding [Pd(SO3)(terpy*)] 1 40 mg (30%) with
identical properties to that prepared above.
[PdCl(terpy*)]Cl 2 alternative methods. Solid [PdCl2(Me-
CN)2] (100 mg; 0.39 mmol) was suspended in dry THF (15 ml),
3522
J. Chem. Soc., Dalton Trans., 2002, 3518–3524

View Article Online
J = 6.0, 2.0, H5/5Ј); 1.46 (s, 18 H; But). δc (67.8 MHz, CDCl3)
165.1, 156.1, 150.4, 123.9, 119.3, 35.8 (But), 30.3 (But). νmax
(KBr disc)/cmϪ1 3434s, 2973s (C–H), 1614s, 1543w, 1472m
(1, Nottingham); 2 and 3 were grown from CDCl3 solutions,
4 and 5 were grown by CH2Cl2–hexane solutions.
All crystals were mounted in a perfluoropolyether oil
film mounted on a dual stage fibre and flash frozen to 150 K
using an Oxford Cryosystem open-flow nitrogen cryostat.18
The trihydrate morphology of compound 1a was collected
on Station 9.819 at the Daresbury Synchrotron Radiation
Source, using a Bruker SMART CCD area detector diffrac-
tometer and silicon monochromated radiation (λ = 0.6890 Å).
Data for all other compounds were collected on a SMART1000
CCD area detector diffractometer using graphite mono-
chromated MoKα radiation (λ = 0.71073 Å). SAINT v6.01
(v6.02a for compound 4)20 was used to integrate the data
and apply the Lorentz and polarisation corrections. Crystal
data and details of the data collection and refinement are given
in Table 3. The data were corrected for absorption using
semi-empirical methods21 for all structures (including that of
compound 1a where SADABS was also used to apply a beam
decay correction) except for 4 where numerical methods were
used.22
The structures were solved by direct methods using
SHELXS-9723 for all structures except 1 where SIR9224 was
used. The structures were refined on F2 using full-matrix
least squares (SHELXL-97). Unless otherwise stated, all fully
occupied non-hydrogen atoms were refined with anisotropic
atomic displacement parameters (adps). Hydrogen atoms were
placed in geometrically calculated positions except those of
the water molecules in compounds 1a and 2 and those of the
MeCN in compound 4 which were located by difference Fourier
synthesis. Geometrically placed hydrogen atoms were refined as
part of a riding model with the hydrogen atoms assigned iso-
tropic adps 1.2 times the parent atom Ueq, except for the methyl
hydrogen atoms where it was 1.5 times, the water hydrogen
atoms were refined with restraints and the MeCN hydrogen
atoms were refined as a rigid rotating group. Suitable geometric
restraints were applied to all disorder models. Neutral atom
scattering factors and anomalous dispersion corrections were
taken from ref. 25.
(C᎐C), 1414m, 1251m, 1092w, 902w, 882w, 853s. MS: m/z
᎐
(ESϩ) 432 ( Mϩ Ϫ Cl ϩ MeCN, 100%).
[Pd(OAc)(terpy*)]Cl 4. Palladium acetate (100 mg; 0.45
mmol) was added in one portion to acetonitrile (3 ml) under
nitrogen supply at 20 ЊC and was stirred for 5 min, followed by
addition of terpy* (180 mg; 0.45 mmol). The reaction mixture
was heated to reflux for 10 min, after which time not all the
terpy* had dissolved, the reaction mixture was allowed to cool
and dichloromethane (3 ml) was added. The terpy* dissolved
immediately and the reaction was allowed to stir at 20 ЊC for
15 h. The crude reaction mixture was filtered through a small
plug of Celite, washing with dichloromethane (2 x 10 ml). The
organic washings were reduced under vacuum to give an orange
coloured solid. The solid was taken up in dichloromethane
(1 ml) and was layered with hexane (3 ml). Upon standing pale
yellow crystals formed which were filtered and dried under high
vacuum to yield 4 (75 mg; 30%). Mp >230 ЊC (dec., from
CH2Cl2). δH (400 MHz, DMSO, 298 K) 8.78 (br s, 4H; OAc ϩ
Cl); 8.72 (br s, 2H; Cl); 8.69 (br s, 2H; Cl); 8.56 (d, 2H, JHH
=
6.0 Hz; OAc); 8.13 (d, 2H, JHH = 6.0 Hz, OAc); 7.85 (d, 2H,
JHH = 6.0 Hz; OAc) overlapped by 7.85 (br s, 2H; Cl); 2.03
(s, 3H; CH3CO2); 1.54 (s, 9H; But); 1.45 (s, 18H; But); δH (400
MHz, DMSO, 373 K) 8.71 (apparent d, 2H, J = 2.0); 8.68 (br s,
2H); 8.54 (br s, 2H); 7.88 (dd, 2H, J = 6.0, 2.0, H5Ј/5Љ); 1.84 (br s,
3H; CH3CO2); 1.57 (s, 9H; But); 1.47 (s, 18H; But). δc (100 MHz,
DMSO, 373 K) 167.5, 166.7, 157.7, 154.1, 151.1, 124.7,
122.6, 121.4, 36.7 (tBu), 35.7 (tBu), 29.9 (tBu), 29.5 (tBu). ν(solid
state)/ cmϪ1 2958w; 1613vs (C–O), 1556w; 1480s, 1425m,
1401m, 1367s, 1326s, 1300w, 1264s, 1023m, 916s, 856m. MS:
m/z (ESϩ) 544 (terpy*PdCl37, 100%); 542 (terpy*PdCl35,
100%); 507 (terpy*Pd, 70%) [Found (HRMS): (Mϩ Ϫ Cl)
566.2005, C29H38N3O2Pd requires 566.1999].
[Pd(SO4)(terpy*)] 5. A d5-nitrobenzene solution (2.5 ml) of
[Pd(SO3)(terpy*)] 1 (50 mg; 0.085 mmol) was heated to 100 ЊC
in for 3 h under a constant stream of oxygen. Proton NMR
spectroscopy of the reaction mixture showed the formation of
new signals atributed to 5 as well as those of 1. The reaction
mixture was left heating for a further 10 h under O2 at 210 ЊC
In compound 1 4.5 molecules of CHCl3 per asymmetric unit
were incorporated in the structure, one of which was modelled
with the three Cl atoms each over three sites with occupancies
0.5:0.3:0.2. One But group also showed disorder and was
modelled over two sites with occupancies 0.6:0.4. In compound
1a the SO3 is disordered with 75% S-bound to the palladium
and 25% O-bound to the palladium.
1
after which time H NMR spectroscopy indicated 5 was the
major product but traces of 1 remained. The new species 5
was isolated by removing all the d5-nitrobenzene solvent and
liquid/liquid diffusion (dichloromethane–hexane) to yield large
orange shaped wedges and small yellow crystals (1). X-Ray
characterisation of the orange crystals showed formation of the
desired sulfate species. Repetition of the experiment but with O2
bubbling directly through the PhNO2 solution lead to clean
formation of 5 (93% isolated yield) after 12 h. Lower boiling
solvents were ineffective. Mp >240 ЊC (dec., from CH2Cl2–
hexane). δH (400 MHz, CDCl3): 9.27 (d, 2 H, J = 6.0, H6Ј/6Љ); 8.14
(s, 2 H, H3/5); 8.09 (d, 2 H, J = 1.0, H3Ј/3Љ); 7.61 (dd, 2 H, J = 6.0,
1.0, H5Ј/5Љ); 1.60 (s, 9 H; But); 1.44 (s, 18 H; But). δc (67.8 MHz,
CDCl3); 169.2, 167.9, 158.3, 154.1, 134.6, 129.3, 123.5, 122.6,
37.7 (But), 36.4 (But), 30.6 (But), 30.3 (But). νmax (KBr disc)/
Dichloromethane is included at three sites in compound 2,
one site fully occupied, one site modelled as half occupied and a
third as a quarter occupied, with one Cl atom modelled over
two sites with occupancies 0.15 and 0.10, the latter refined with
an isotropic adp.
A molecule of CH2Cl2 incorporated in the structure of com-
pound 5 showed disorder and was modelled with the two Cl
atoms over two sites with occupancies 0.60 and 0.40. The three
methyl C atoms of one tBu group were also modelled over two
sites with occupancies 0.55 and 0.45. The Flack parameter
(0.00(4)) for complex 5 (space group P212121) establishes that
the correct axial directions have been chosen.
CCDC reference numbers 186002 (1), 186163 (1/1a), 186001
(2), 186003 (3), 186000 (4) and 186004 (5).
See http://www.rsc.org/suppdata/dt/b2/b205548a/ for crystal-
lographic data in CIF or other electronic format.
cmϪ1 3422s, 2961s (C–H), 2360s, 2342s, 1612s, 1560w (C᎐C),
᎐
1522vs, 1474m (C᎐C), 1425w, 1367w, 1345vs, 1259m ν(S᎐O)
,
᎐
᎐
asym
1134 δ(S᎐O) sym, 1020w, 981vs (S–O), 850w, 797w, 712m, 680w,
᎐
668w, 638m, 609w. MS: m/z (ESϩ) 628 (MNaϩ, 2%); 542
(Pd(MeCN)(terpy*), 40%); 524 (NaPd(terpy*), 100%) [Found
Acknowledgements
(HRMS): (Mϩ
507.1866].
Ϫ SO4) 507.1864, C27H35N3Pd requires
We thank EPSRC for a project studentship to R.I.R. (through
grant GR/N09824) and Johnson-Matthey for a generous loan
of palladium salts. We are grateful to Professor Laurent Barloy
for helpful advice and to a referee for pointing out to us the
work of L. I. Elding. We thank the EPSRC mass spectrometry
service for sample analyses.
X-Ray crystallography
Crystals of 1/1a were grown from aqueous MeCN solutions
saturated with SO2–O2 (Daresbury) or from CDCl3 solutions
J. Chem. Soc., Dalton Trans., 2002, 3518–3524
3523

View Article Online
References
1 S. Woodward, Tetrahedron, 2002, 58, 1017 and references therein.
2 (a) See for example: Y. Ishii, K. Matsunaka and S. Sakaguchi, J. Am.
Chem. Soc., 2001, 123, 1545; (b) Y. Ishii, K. Matsunaka and
S. Sakaguchi, J. Am. Chem. Soc., 2000, 123, 7390; (c) E. D. Weil and
S. R. Sandler, in Kirk-Othmer Encyclopedia of Chemical Technology,
John Wiley, New York, 1990, vol. 23, p. 302 and references therein.
3 A. F. Lee, K. Wilson, R. M. Lambert, C. P. Hubbard, R. G. Hurley,
R. W. McCabe and H. S. Gandhi, J. Catal., 1999, 184, 491.
4 (a) R. Mews, E. Lork, P. G. Watson and B. Gortler, Coord. Chem.
Rev., 2000, 197, 277; (b) C. Brandt and R. van Eldik, Chem. Rev.,
1995, 95, 119; (c) G. J. Kubas, Acc. Chem. Res., 1994, 27, 183.
5 (a) J. Valentine, D. Valentine and J. P. Collman, Inorg. Chem., 1971,
10, 219; (b) S. P. Mehandru and A. B. Anderson, Inorg. Chem.,
1985, 24, 2571; (c) Th. Hoffmann, H. Hoffmann, M. Karabit,
P. Zdunneck, E. Wenschuh and J. R. Schüler, Z. Anorg. Allg. Chem.,
1988, 565, 91; (d ) D. H. Farrar and R. R. Gukathasan, J. Chem.
Soc., Dalton Trans., 1989, 557; (e) O. Heyke, A. Neher and
I.-P. Lorenz, Z. Anorg. Allg. Chem., 1992, 608, 23; ( f ) V. C. Ginn,
P. F. Kelly, C. Papadimitriou, A. M. Z. Slawin, D. J. Williams and
J. D. Woollins, J. Chem. Soc., Dalton Trans., 1993, 1805.
6 F. T. Esmardi and M. K. Dohhaidel, Polyhedron, 1990, 9, 1633.
7 J. Berglund, R. Voigt, S. Fronaeus and L. I. Elding, Inorg. Chem.,
1994, 33, 3346.
8 D. K. Breitinger, G. Schottner, M. Raidel and H. P. Beck, Z. Anorg.
Allg. Chem., 1986, 539, 18.
9 L. Barloy, R. M. Gauvin, J. A. Osborn, C. Sizun, R. Graff and
N. Kyritsakas, Eur. J. Inorg. Chem., 2001, 1699.
10 G. M. Intille, C. E. Pfluger and W. A. Baker, J. Cryst. Mol. Struct.,
1973, 3, 47.
11 A. J. Canty, B. W. Skelton, P. R. Traill and A. H. White, Aust. J.
Chem., 1992, 45, 417.
12 Searches of the CSD database were carried out using ConQuest 1.3
(http://cds3.dl.ac.uk/cds/cds.html); D. A. Fletcher, R. F. McMeekin
and D. Parkin, J. Chem. Inf. Comput. Sci., 1996, 36, 746.
13 T. Shi and L. I. Elding, Acta Chem. Scand., 1998, 52, 897.
14 (a) Structure FAGSEL, see: G. Dongyao, W. Fengshan, Q. Yun and
T. Wenxia, Gaodeng Xuexiao Huaxue Xuebao (Chem. J. Chin.
Univ.), 1986, 7, 261; (b) Structure FAMMIP, see: H. A. Meinema,
F. Verbeek, J. W. Marsman, E. J. Bulten, J. C. Dabrowiak,
B. S. Krishnan and A. L. Spek, Inorg. Chim. Acta, 1986, 114, 127;
(c) Structure FEJBED, see: F. D. Rochon and R. Melanson, Inorg.
Chem., 1987, 26, 989; (d ) Structure NOYXUU, see: J. Bruce,
D. Johnson, W. Cordes and R. Sadoski, J. Chem. Crystallogr., 1997,
27, 695; (e) Structure NURTID, see: U. Klement, K.-J. Range and
R. Gust, Z. Kristallogr.–New Cryst. Struct., 1998, 213, 33.
15 T. B. Hadda and H. L. Bozec, Polyhedron, 1988, 7, 575.
16 R. A. Michelin, G. Facchin and P. Uguagliati, Inorg. Chem., 1984,
23, 961.
17 M. F. Rettig and P. M. Maitlis, Inorg. Synth., 1990, 28, 110.
18 J. Cosier and A. M. Glazer, J. Appl. Crystallogr., 1986, 19, 105.
19 (a) R. J. Cernik, W. Clegg, C. R. A. Catlow, G. Bushnell-Wye,
J. V. Flaherty, G. N. Greaves, I. Burrows, D. J. Taylor, S. J. Teat and
M. Hamichi, J. Synchrotron Radiat., 1997, 4, 279; (b) W. Clegg,
M. R. J. Elsegood, S. J. Teat, C. Redshaw and V. C Gibson, J. Chem.
Soc., Dalton Trans., 1998, 3037.
20 SAINT, v6.01 (1997) and v6.02a (2000), Bruker AXS, Inc.,
Madison, WI.
21 G. M. Sheldrick, SADABS: Program for Scaling and Correction of
Area Detector Data, University of Göttingen, Göttingen, 1997.
22 SHELXTL, Bruker AXS, Inc., Madison, WI, 1997.
23 G. M. Sheldrick, SHELXL-97: Program for the Refinement of
Crystal Structures, University of Göttingen, Göttingen, 1997.
24 A. Altomare, G. Cascarano, C. Giacovazzo and A. Guagliardi,
J. Appl. Crystallogr., 1993, 26, 343.
25 A. J. C. Wilson (ed.), International Tables for Crystallography,
Kluwer Academic, Dordrecht, 1992, vol. C.
3524
J. Chem. Soc., Dalton Trans., 2002, 3518–3524