Coordination chemistry of Co complexes containing tridentate SNS ligands and their application as catalysts for the oxidation of n-octane

Article abstract of DOI:10.1039/c4dt01750a

The selective oxidation of saturated hydrocarbons to terminal oxygenates under mild catalytic conditions has remained a centuries long challenge in chemical catalysis. In an attempt to address this challenge, two series of tridentate donor ligands {2,6-bis(RSCH₂)pyridine and bis(RSCH₂CH₂)amine [R = alkyl, aryl]} and their respective cobalt complexes {Co[2,6-bis(RSCH₂)pyridine]Cl₂ and Co[bis(RSCH₂CH₂)amine]Cl₂} were synthesized and characterized. Crystal structures of Co[2,6-bis(RSCH₂)pyridine]Cl₂ [R = -CH₃ (1), -CH₂CH₃ (2), -CH₂CH₂CH₂CH₃ (3) and -C₆H₅ (4)] are reported in which 1 crystallized as a homo-bimetallic dimer that incorporated two bridging chloride atoms in an octahedral geometry around each cobalt center, while 2, 3 and 4 crystallized as mono-metallic species characterized by trigonal bipyramidal arrangement of ligands around each cobalt center. As catalysts for the homogeneous selective oxidation of n-octane, the catalysts yielded ketones as the dominant products with a selectivity of ca. 90% for the most active catalyst Co[bis(CH₂CH₂SCH₂CH₂)amine]Cl₂ (6) at a total n-octane conversion of 23%. Using tert-butyl hydroperoxide (TBHP) as an oxidant, optimization of reaction conditions is also reported.

Full text of DOI:10.1039/c4dt01750a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Coordination chemistry of Co complexes
containing tridentate SNS ligands and their
application as catalysts for the oxidation of
n-octane†
Cite this: Dalton Trans., 2014, 43,
15968
Lynette Soobramoney, Muhammad D. Bala* and Holger B. Friedrich*
The selective oxidation of saturated hydrocarbons to terminal oxygenates under mild catalytic conditions
has remained a centuries long challenge in chemical catalysis. In an attempt to address this challenge,
two series of tridentate donor ligands {2,6-bis(RSCH2)pyridine and bis(RSCH2CH2)amine [R = alkyl, aryl]}
and their respective cobalt complexes {Co[2,6-bis(RSCH2)pyridine]Cl2 and Co[bis(RSCH2CH2)amine]Cl2}
were synthesized and characterized. Crystal structures of Co[2,6-bis(RSCH2)pyridine]Cl2 [R = –CH3 (1),
–CH2CH3 (2), –CH2CH2CH2CH3 (3) and –C6H5 (4)] are reported in which 1 crystallized as a homo-bi-
metallic dimer that incorporated two bridging chloride atoms in an octahedral geometry around each
cobalt center, while 2, 3 and 4 crystallized as mono-metallic species characterized by trigonal bipyramidal
arrangement of ligands around each cobalt center. As catalysts for the homogeneous selective oxidation
of n-octane, the catalysts yielded ketones as the dominant products with a selectivity of ca. 90% for the
most active catalyst Co[bis(CH2CH2SCH2CH2)amine]Cl2 (6) at a total n-octane conversion of 23%. Using
tert-butyl hydroperoxide (TBHP) as an oxidant, optimization of reaction conditions is also reported.
Received 12th June 2014,
Accepted 26th August 2014
DOI: 10.1039/c4dt01750a
www.rsc.org/dalton
oxygen atom from an appropriate donor to the activated
species.4
Introduction
Selective activation of paraffins (alkanes) has remained an area
Although a number of researchers have reported on the use
of active research interest due to the relative inertness of the of transition metal complexes in the catalytic activation of
saturated hydrocarbon bonds. The inertness of paraffins is due paraffins,4–11 the employment of pincer complexes in this
to strong, localized C–C and C–H bonds, which makes it regard has remained challenging to scientists worldwide.
difficult for them to participate in chemical reactions. This is Hence, there are only a handful of reported examples on the
because they lack empty orbitals of low energy or filled orbitals utilization of pincer ligand-based metal complexes for the cata-
of high energy necessary for most chemical transformations.1,2 lytic activation of mainly aryl and vinyl C–H bonds.2,12–23
Most attempts at paraffin activation aim to convert the inert
Since their discovery in the mid-1970s, pincer-type ligands
alkane to relatively more reactive or value-added products have become increasingly important in chemistry,24 and owing
which amongst other uses serve as feedstock, additives or sol- to their ability to tailor the reactivity, selectivity and stability of
vents for pharmaceutical and other chemical industries.3 Oxy- the ligands towards a specific reaction pathway or product, the
genates (alcohols, ketones and carboxylic acids), which result application of pincer-based metal complexes has dramatically
from oxidative functionalization of alkanes, are the most increased.18,25–27 These advantages are due to the terdentate
desired products of paraffin activation in chemistry. These are binding mode of the ligands which implies that they are able
generally believed to be obtained by activation of the C–H to stabilize metal centers more effectively than related mono
bond via bond cleavage, followed by insertion of a single or bidentate variants. In addition to imparting stability, simul-
taneous occupation of multiple coordination sites around a
metal center allows the ligands to control access to the center
which in turn affects binding of reactive species, thus provid-
School of Chemistry and Physics, University of KwaZulu-Natal, Private Bag X54001,
ing enhanced selectivity towards a desired substrate during
Durban 4000, South Africa. E-mail: bala@ukzn.ac.za, friedric@ukzn.ac.za;
catalysis. The ligands are also easily tuned, both electronically
Fax: +27 31 260 3091
†Electronic supplementary information (ESI) available. CCDC 984366–984369.
and sterically, to ensure catalyst versatility.25,27 It is worth
noting that in spite of the increased interest in pincer com-
For ESI and crystallographic data in CIF or other electronic format see DOI:
plexes, those complexes stabilized by ligands typified by SNS
10.1039/c4dt01750a
15968 | Dalton Trans., 2014, 43, 15968–15978
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
donor atom combination have in general received relatively employed which involves the use of cheaper reagents to
less attention. prepare, and the method followed was adapted from a study by
However, since the discovery of SNS complexes of Cr28 as Reger et al.41 In a 500 ml round bottom flask (RBF), a solution
excellent ethylene trimerization catalysts, interest in SNS containing 8.0 g (0.20 mol) of NaOH and 2.78 g (0.020 mol) of
ligands has also increased amongst researchers. In addition to 2,6-pyridinedimethanol was prepared in 150 ml THF–water
the seminal report by McGuinness, Wasserscheid and co- (1 : 1). The solution was then cooled to 0 °C to which a mixture
workers,28 SNS complexes of Mo,29 Pt,30 Pd,31 Ru,32 Cu,33 Ni,34 of p-toluenesulfonyl chloride in 75 ml THF was added. After
and Zn35 have to date been synthesized and characterized. One stirring at room temperature for four hours, the mixture was
of the key features of the SNS pincer ligand relevant to its poured into 200 ml water and extracted with 75 ml dichloro-
application in catalysis is that it may be designed to adopt a methane (DCM) and this was repeated three times. The
hemilabile binding mode.36 This is usually achieved by exploit- organic phase was washed with a saturated NaCl solution and
ing the relative difference in binding abilities between the distilled water, and dried with Na2SO4. The solvent was
hard nitrogen donor and the soft sulfur donor atoms to metal removed under vacuum yielding a crystalline white product
centers.37 Being hemilabile also affords the metal complex (6.57 g, 73%). 1H NMR (400 MHz, CDCl3): δ 2.4 (s, 6H), 5.1
with greater flexibility and balance between stability and (s, 4H), 7.3 (d, 6H), 7.7 (t, 1H), 7.8 (d, 4H).
reactivity.38,39
Synthesis of 2,6-bis(CH3SCH2)pyridine (L1). To prepare this
To the best of our knowledge, there has been no work ligand a modified protocol adapted from a study by Canovese
reported on Co-based SNS complexes or the application of SNS et al. was followed.31 In a 50 ml two neck RBF fitted with a
pincer complexes for C–H or paraffin activation. Hence this nitrogen tap, 1.79 g of PMT was dissolved in ∼40 ml of THF
study reports on the first examples of a new range of Co-based and this solution was cooled in an ice bath. A mass of 0.62 g
SNS pincer complexes and their successful application as cata- of sodium methanethiolate was added to the ice cooled solu-
lysts in the activation and functionalization of n-octane.
tion after which the mixture was stirred for 2 hours at room
temperature. At the end of the reaction the solvent was evacu-
ated under reduced pressure and the resulting cream residue
was partitioned between DCM (200 ml) and water (100 ml).
After washing several times with water, the organic phase was
dried with MgSO4 and the solvent was removed to yield the
Experimental section
General
All manipulations were carried out using standard Schlenk product as impure pale yellow oil. The product was purified by
techniques under a nitrogen atmosphere. Solvents were dried elution with DCM through a silica packed column (0.53 g,
1
according to established methods40 and purged with high 66%). H NMR (400 MHz, CDCl3): δ 2.0 (s, 6H), 3.8 (s, 4H), 7.2
purity nitrogen gas prior to use. Diethyl ether (Et2O) and tetra- (d, 2H), 7.6 (t, 1H). 13C NMR (400 MHz, CDCl3): δ 15.2 (CH3–S),
hydrofuran (THF) were dried over sodium wire and benzophe- 40.0 (py–CH2–S), 121.1 (CH–py), 137.3 (CH–py), 158.2 (C–py).
none, absolute ethanol (EtOH) was dried over magnesium 13C DEPT 135 (400 MHz, CDCl3): δ 15.2 (CH3–S) neg, 40.0
turnings/iodine, and dichloromethane (DCM) was dried over (CH2–S) pos, 121.1 (CH–py) neg, 137.3 (CH–py) neg. IR νmax
phosphorous pentoxide. All other reagents were purchased (cm−1): 3058 (w), 2966 (m), 2914 (m), 2856 (m), 1589 (s), 1572
commercially and used as received. All NMR spectra were (s), 1451 (s), 747 (s), 813 (m).
recorded using a Bruker Avance III 400 MHz spectrometer at
Synthesis of 2,6-bis(CH2CH3SCH2)pyridine (L2). The pro-
ambient temperature. The 1H NMR data are reported as chemi- cedure followed for the synthesis of L2 was adapted from a
cal shifts (δ, ppm) and referenced to the solvent peak CDCl3. study by Teixidor et al.35 with a few modifications. Sodium
The proton decoupled 13C NMR data are presented as chemi- metal (0.23 g, 10 mmol) and ethanethiol (0.5 ml, 10 mmol)
cal shifts (δ, ppm) and referenced to the solvent peak CDCl3 were stirred together in ethanol (10 ml) for 20 min in a 20 ml
with the specific carbon indicated in parentheses. The 13C Schlenk tube. This solution was added to another ethanol
DEPT 135 NMR data, which distinguish between CH, CH2 and solution (50 ml) of PMT (2.24 g, 5 mmol) in a 100 ml two neck
CH3, are listed as chemical shifts (δ, ppm) and positive (pos) RBF fitted with a nitrogen inlet, and the reaction mixture was
or negative (neg) with the corresponding carbons in paren- left to reflux overnight. The solvent was removed in vacuo and
theses. The IR spectra were recorded on a Perkin Elmer Attenu- the residue was extracted twice with diethyl ether (150 ml) and
ated Total Reflectance (ATR) spectrophotometer and elemental the remaining insoluble residue was discarded. The extract was
analyses were performed on a LECO CHNS elemental analyzer, washed with aqueous Na2CO3 and twice with water (150 ml) and
while the melting points were determined using Stuart Scienti- dried with MgSO4. The solvent was removed in vacuo to yield the
fic melting point apparatus.
impure product which was purified analogously to L1 to give a
1
Synthesis of the precursor 2,6-pyridine-dimethylene-dito- pale yellow oil (0.62 g, 54%). H NMR (400 MHz, CDCl3): δ 1.2
sylate (PMT). The reported protocols for the preparation of the (t, 6H), 2.5 (q, 4H), 3.8 (s, 4H), 7.2 (d, 2H), 7.6 (t, 1H). 13C NMR
pyridine-based ligands involve expensive starting materials like (400 MHz, CDCl3): δ 14.5 (CH3CH2–S), 25.6 (CH3CH2–S), 37.8
2,6-bis(chloromethyl)pyridine, therefore an alternate synthetic (py–CH2–S), 121.0 (CH–py), 137.2 (CH–py), 158.5 (C–py). 13C
route was followed to reduce the costs. A more suitable starting DEPT 135 (400 MHz, CDCl3): δ 14.5 (CH3–S) neg, 25.6 (CH2–S)
material, 2,6-pyridine-dimethylene-ditosylate (PMT), was pos, 37.8 (py–CH2–S) pos, 121.0 (CH–py) neg, 137.2 (CH–py)
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15968–15978 | 15969

View Article Online
Paper
Dalton Transactions
neg. IR νmax (cm−1): 3056 (w), 2971 (m), 2926 (m), 2869 (m), NMR (400 MHz, CDCl3): δ 25.9 (CH2–cy), 26.1 (CH2–cy), 33.5
1590 (s), 1573 (s), 1451 (s), 747 (s), 788 (m). (CH2–cy), 36.6 (py–CH2–S), 43.4 (CH–cy), 121.0 (CH–py), 137.3
Synthesis of 2,6-bis(CH2CH2CH2CH3SCH2)pyridine (L3). A (CH–py), 158.8 (C–py). 13C DEPT 135 NMR (400 MHz. CDCl3):
similar procedure to that for L2 was followed for L3 with the δ 25.9 (CH2–cy) pos, 26.1 (CH2–cy) pos, 33.5 (CH2–cy) pos, 36.6
following masses and volumes: sodium metal (0.14 g, (py–CH2–S) pos, 43.4 (CH–cy) neg, 121.0 (CH–py) neg, 137.3
6.18 mmol), butanethiol (0.7 ml, 6.18 mmol), PMT (1.38 g, (CH–py) neg. IR νmax (cm−1): 3064 (w), 2962 (m), 2923 (m),
3.09 mmol). No further purification was required as the TLC 1590 (s), 1571 (s), 1480 (s), 1449 (s), 737 (s), 809 (m).
showed only one spot; however the oil was dried under
Synthesis of bis(CH2CH3SCH2CH2)amine (L6). The method
reduced pressure for several hours to remove any solvent, thus for the preparation of L6 was adapted from a study by Konrad
yielding the product as a yellow oil (0.55 g, 63%). 1H NMR et al.34 In a 250 ml Schlenk flask, 8.79 g (50 mmol) of bis(2-
(400 MHz, CDCl3): δ 0.9 (t, 6H), 1.4 (m, 4H), 1.6 (m, 4H), 2.5 chloroethyl)amine hydrochloride was dissolved in 100 ml
(t, 4H), 3.8 (s, 4H), 7.3 (d, 2H), 7.6 (t, 1H). 13C NMR (400 MHz, ethanol. The solution was added to a 500 ml Schlenk flask
CDCl3): δ 13.7 (CH3CH2CH2CH2–S), 22.0 (CH3CH2CH2CH2–S), which contained a mixture of NaOH (5.99 g, 150 mmol) and
31.3 (CH3CH2CH2CH2–S), 31.4 (CH3CH2CH2CH2–S), 38.1 ethanethiol (11.0 ml, 150 mmol) in ethanol (150 ml) at 0 °C.
(py–CH2–S), 121.0 (CH–py), 137.2 (CH–py), 158.5 (C–py). 13C The resulting mixture was stirred for two hours, filtered via a
DEPT 135 (400 MHz, CDCl3): δ 13.7 (CH3–S) neg, 22.0 (CH2–S) cannula and the filtrate was dried in vacuo. Diethyl ether was
pos, 31.4 (CH2–S) pos, 38.1 (CH2–S) pos, 121.0 (CH–py) neg, added to the residue and once again the mixture was filtered
137.2 (CH–py) neg. IR νmax (cm−1): 3056 (w), 2956 (m), 2926 and the filtrate was dried in vacuo to yield the product as a
1
(m), 2871 (m), 1589 (s), 1573 (s), 1451 (s), 747 (s), 812 (m).
pale yellow oil (6.66 g, 69%). H NMR (400 MHz, CDCl3): δ 1.3
Synthesis of 2,6-bis(C6H5SCH2)pyridine (L4). This method (t, 6H), 1.8 (s, 1H, N–H), 2.6 (q, 4H), 2.7 (t, 4H), 2.8 (q, 4H). 13
C
was adapted from a study by Teixidor et al.37 Sodium metal NMR (400 MHz, CDCl3): δ 14.9 (CH3CH2–S), 25.9 (CH3CH2–S),
(0.14 g, 6 mmol) was stirred with thiophenol (0.61 ml, 31.9 (CH2CH2–N), 48.3 (CH2CH2–N). 13C DEPT 135 NMR
6 mmol) in ethanol (10 ml) in a 20 ml Schlenk tube for (400 MHz. CDCl3): δ 14.9 (CH3–S) neg, 25.9 (CH2–S) pos, 31.9
10 min. The solution was then added dropwise to an ice (CH2–N) pos, 48.3 (CH2–N) pos. IR νmax (cm−1): 3288 (w), 2962
cooled solution of PMT (1.52 g, 3 mmol) in THF (20 ml) and (s), 2921 (s), 2864 (s), 2823 (s), 1451 (s), 739 (m).
the resulting solution was stirred for a further three hours at
Synthesis of bis(CH2CH2CH2CH3SCH2CH2)amine (L7). L7
0 °C. The solvent was then removed in vacuo and the resulting was prepared analogously to L8 with the following masses
residue was extracted with 70 ml diethyl ether and washed and volumes: bis(2-chloroethyl)amine hydrochloride (8.81 g,
with 1 M NaOH solution (25 ml × 2) after which the organic 50 mmol), NaOH (6.13 g, 150 mmol) and butanethiol (16.0 ml,
layer was dried with MgSO4. The solvent was removed in vacuo 150 mmol). The product was obtained as yellow oil (10.71 g,
to yield the product as pale yellow oil and no further purifi- 86%). 1H NMR (400 MHz, CDCl3): δ 0.9 (t, 6H), 1.3 (m, 4H), 1.5
1
cation was required (0.90 g, 82%). H NMR (400 MHz, CDCl3): (m, 4H), 1.8 (s, 1H, N–H), 2.5 (t, 4H), 2.6 (t, 4H), 2.8 (t, 4H).
δ 4.2 (s, 4H), 7.1 (d, 4H), 7.2 (t, 4H), 7.3 (d, 4H), 7.5 (t, 1H). 13
C
13C NMR (400 MHz, CDCl3): δ 13.7 (CH3CH2CH2CH2–S),
NMR (400 MHz, CDCl3): δ 40.3 (py–CH2–S), 121.3 (CH–py), 22.0 (CH3CH2CH2CH2–S), 31.7 (CH3CH2CH2CH2–S), 31.8
126.2 (CH–ph), 128.9 (CH–ph), 129.6 (CH–ph), 135.7 (C–ph), (CH3CH2CH2CH2–S), 32.4 (CH2CH2–N), 48.4 (CH2CH2–N). 13C
137.2 (CH–py), 157.3 (C–py). 13C DEPT 135 NMR (400 MHz. DEPT 135 NMR (400 MHz, CDCl3): δ 13.7 (CH3–S) neg, 22.0
CDCl3): δ 40.3 (py–CH2–S) pos, 121.3 (CH–py) neg, 126.2 (CH– (CH2–S) pos, 31.7 (CH2–S) pos, 31.8 (CH2–S) pos, 32.4 (CH2–N)
ph) neg, 128.9 (CH–ph) neg, 129.6 (CH–ph) neg, 137.2 (CH–py) pos, 48.4 (CH2–N) pos. IR νmax (cm−1): 3294 (w), 2955 (vs), 2925
neg. IR νmax (cm−1): 3064 (w), 2962 (m), 2923 (m), 1590 (s), (vs), 2872 (vs), 1458 (s), 741 (m).
1571 (s), 1450 (s), 1436 (s), 737 (s), 809 (m).
Synthesis of Co[2,6-bis(CH3SCH2)pyridine]Cl2 (1). A mixture
Synthesis of 2,6-bis(C6H11SCH2)pyridine (L5). This method of L1 (0.4917 g, 2.5 mmol) in ethanol (5 ml) was added drop-
was adapted from a study by Temple et al.42 In a 20 ml wise to a solution of CoCl2·6H2O (0.5717 g, 2.4 mmol) in
Schlenk tube, NaOH (0.32 g, 8.2 mmol) and cyclohexylmercap- ethanol (10 ml) in an almost 1 : 1 mole ratio with the ligand in
tan (1.0 ml, 8.2 mmol) were stirred together in ethanol (10 ml) slight excess. After an overnight reflux, the resulting indigo
for 30 min. This mixture was added dropwise to a THF solu- solution was concentrated by reducing the volume of the
tion (20 ml) of PMT (1.79 g, 4 mmol) in a 100 ml two neck RBF solvent to ∼5 ml. At this point the product precipitated out
fitted with a nitrogen inlet. After stirring at room temperature was separated via cannula filtration and washed with a small
overnight, the solvent was removed in vacuo and the residue amount of ethanol. Drying in vacuo for several hours afforded
was partitioned between DCM and deionized water (50 ml the product as a blue microcrystalline solid (0.28 g, 36%). Crys-
each). The organic layer was collected, while the aqueous layer tals suitable for X-ray diffraction were grown from a concen-
was washed with DCM (20 ml × 3), and then the organics were trated acetonitrile solution layered with Et2O. Elemental
combined and dried with MgSO4. After the solvent was analysis for C9H13Cl2CoNS2: calcd C, 32.8; H, 4.0; N, 4.3; found
removed in vacuo the product was obtained as yellow oil and C, 32.4; H, 4.4; N, 4.0. Melting point: 176–177 (dimer),
no further purification was required (1.21 g, 91%). 1H NMR 196–197 °C.
(400 MHz, CDCl3): δ 1.3 (m, 10H), 1.6 (m, 2H), 1.7 (m, 4H), 1.9
Synthesis of Co[2,6-bis(CH2CH3SCH2)pyridine]Cl2 (2).
(m, 4H), 2.6 (m, 2H), 3.8 (s, 4H), 7.2 (d, 2H), 7.6 (t, 1H). 13C Complex 2 was prepared analogously to 1 using 0.1235 g
15970 | Dalton Trans., 2014, 43, 15968–15978
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
(0.52 mmol) CoCl2·6H2O and 0.1193 g (0.53 mmol) L2. A brilli- SAINT+ while face indexed and multi-scan absorption correc-
ant blue crystalline solid was obtained (0.086 g, 47%). Crystals tions were made using SADABS. The structures were solved by
suitable for X-ray diffraction were deposited from
a
direct methods using SHELXS.43 Non-hydrogen atoms were
concentrated ethanol solution. Elemental analysis for first refined isotropically followed by anisotropic refinement by
C11H17Cl2CoNS2: calcd C, 37.0; H, 4.8; N, 3.9; found C, 37.3; H, full matrix least-squares calculations based on F2 using
5.2; N, 3.9. Melting point: 164–165 °C.
SHELXS. Hydrogen atoms were first located in the difference
Synthesis of Co[2,6-bis(CH2CH2CH2CH3SCH2)pyridine]Cl2 map then positioned geometrically and allowed to ride on
(3). Complex 3 was prepared analogously to 1 using 0.0843 g their respective parent atoms. Diagrams were generated using
(0.35 mmol) CoCl2·6H2O and 0.1021 g (0.36 mmol) L3. An SHELXTL, PLATON44 and ORTEP-3.45 In preliminary refine-
indigo blue crystalline solid was obtained (0.069 g, 48%). Crys- ments for complex 3 the butyl groups in the structure were
tals suitable for X-ray diffraction were grown from a concen- found to be disordered and were as a consequence refined
trated
C13H21Cl2CoNS2: calcd C, 44.5; H, 6.8; N, 3.1; found C, 45.0; H, atoms in the butyl groups were refined isotropically with a
6.6; N, 3.4. Melting point: 107–108 °C. common Uiso parameter. Crystallographic data and additional
ethanol
solution.
Elemental
analysis
for over two positions with isotropic thermal parameters. Carbon
Synthesis of Co[2,6-bis(C6H5SCH2)pyridine]Cl2 (4). Complex information regarding the crystal structure determination are
4 was prepared analogously to 1 using 0.1190 g (0.50 mmol) available as ESI.† Furthermore, CCDC 984366–984369 respect-
CoCl2·6H2O and 0.1635 g (0.51 mmol) L4. An aqua blue ively for compounds 1–4 obtainable from the Cambridge
powder was obtained (0.17 g, 77%). Crystals suitable for X-ray Crystallographic Data Centre contain supplementary crystallo-
diffraction were grown by vapor diffusion of Et2O into a con- graphic data for this contribution.
centrated
DCM
solution.
Elemental
analysis
for
Oxidation of n-octane
C19H17Cl2CoNS2: calcd C, 50.3; H, 3.8; N, 3.1; found C, 50.1; H,
4.3; N, 3.0. Melting point: 211–213 °C.
The paraffin oxidation studies were carried out using n-octane
as the substrate with pentanoic acid as the internal standard.
The catalytic reactions were performed under a nitrogen
atmosphere in a 50 ml two-neck pear shaped flask equipped
with a condenser. The reaction mixture consisted of 5 ml
degassed acetonitrile (solvent), TBHP as the oxidant, n-octane
and the respective catalyst (3 mg). The reaction mixture was
stirred in an oil bath at the optimum temperature (80 °C) and
after the time period (24 and 48 hours for the amine and pyri-
dine-based catalysts respectively), an aliquot of the sample was
removed with a Pasteur pipette and filtered through a cotton
wool plug, after which 0.5 μL of the aliquot was injected into a
GC for analysis and quantification. Masses and volumes of
components of the reaction mixture were dependent on the
specific catalyst; however catalyst loading was kept constant at
1 mol%, i.e. a catalyst to substrate mole ratio of 1 : 100. Fur-
thermore, the n-octane to oxidant ratio was varied from 1 : 3,
1 : 6, 1 : 9, 1 : 12, 1 : 20, 1 : 30 to 1 : 40 from which the optimum
ratio of 1 : 20 was established. At higher oxidant concen-
trations, substrate reactivity decreased drastically indicating
that the catalyst was possibly deactivated by the excess oxidant
in the reaction vessel. All reactions (including blanks) were
carried out in duplicates. The average conversion was
expressed as a percentage of total moles of products/initial
moles of substrate, while selectivity was expressed as a percen-
tage of moles of each product/sum total moles of all products.
Synthesis of Co[2,6-bis(C6H11SCH2)pyridine]Cl2 (5). Complex
5 was prepared analogously to 1 using 0.1188 g (0.50 mmol)
CoCl2·6H2O and 0.1717 g (0.51 mmol) L5. A fine purple
powder was obtained (0.10 g, 43%). Elemental analysis for
C19H29Cl2CoNS2: calcd C, 49.0; H, 6.3; N, 3.0; found C, 48.6; H,
6.8; N, 2.8. Melting point: 142–143 °C.
Synthesis of Co[bis(CH2CH3SCH2CH2)amine]Cl2 (6). A
mixture of L6 (0.2448 g, 1.26 mmol) in ethanol (5 ml) was
added dropwise to a solution of CoCl2·6H2O (0.2615 g,
1.09 mmol) in ethanol (10 ml) in an almost 1 : 1 mole ratio,
with the ligand in slight excess, and a color change from dark
blue to indigo-purple was observed. The reaction mixture was
allowed to stir for two hours at room temperature after which
the solution was concentrated to ∼5 ml and the product preci-
pitated out as a lavender colored solid which was separated via
cannula filtration (0.23 g, 66%). Elemental analysis for
C8H19Cl2CoNS2: calcd C, 29.7; H, 5.9; N, 4.3; found C, 29.6; H,
6.3; N, 4.2. Melting point: 171–172 °C.
Synthesis of Co[bis(CH2CH2CH2CH3SCH2CH2)amine]Cl2
(7). Complex 7 was prepared similarly to 6 using 0.3061 g
(1.29 mmol) CoCl2·6H2O and 0.3606 g (1.45 mmol) L7. A laven-
der powder was obtained (0.31 g, 64%). Elemental analysis for
C12H27Cl2CoNS2: calcd C, 38.0; H, 7.2; N, 3.7; found C, 37.9; H,
7.6; N, 3.7. Melting point: 144–145 °C.
X-ray structure determination
Single crystals were selected and glued onto the tip of a glass
fiber, mounted in a stream of cold nitrogen at 173 K and cen-
tered in the X-ray beam using a video camera. Intensity data
were collected on a Bruker APEX II CCD area detector diffract-
Results and discussion
Syntheses and characterization of new cobalt complexes
ometer with graphite monochromated Mo Kα radiation (50 kV, In this study, the synthesis and application of seven new Co
30 mA) using the APEX 2 data collection software. The collec- complexes containing SNS ligands are reported in which two
tion method involved ω-scans of width 0.5° and 512 × 512 bit sets of ligands were synthesized based on variation in the
data frames. Data reduction was carried out using the program architecture of the backbone N-donor atom; these are (i) a con-
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15968–15978 | 15971

View Article Online
Paper
Dalton Transactions
Table 1 Selected IR data for ligands and the corresponding complexes
6 and 7
IR νmax/cm−1
Ligand/
complex
N–H
C–H
C–H bend
stretcha
(alkyl)b
(alkyl)b
C–Sa
L6
6
L7
7
3228
3228
3294
3219
2864
2870
2872
2867
1458
1465
1458
1464
N/A
694
N/A
744
a Weak (for the coordinated S donor). b Medium.
as weakening of the N–H stretching frequency of 7 from 3294
to 3219 cm−1, are interpreted as indicators of successful com-
plexation of the SNS ligands to the metal center.
Scheme 1 Synthetic route to SNS-Co(II) complexes 1–7.
Crystal structures of 1–4
strained six-membered pyridine ring and (ii) a linear straight
chained amine. Synthesis of all ligands was adapted from the
literature, however novel methodologies were developed for
preparation of the SNS-Co(II) complexes as summarized in
Scheme 1. Successful formation of products was monitored
and confirmed by a variety of techniques, discussed later.
However, it is noted that due to the paramagnetic high spin
nature of the cobalt(II) complexes, their unresolved NMR data
are not reported in this study.
With the exception of the methyl substituted complex 1, which
was isolated as a chloro-bridged bimetallic species containing
octahedrally bonded cobalt centers (Fig. 1), the longer chain
(S-bonded R substituents) analogues all crystallized with a tri-
gonal bipyramidal geometry around each cobalt(II) center
(Fig. 2, 4 and 5) with selected bond lengths and bond angles
listed in Table 2. Consequently, 2, 3 and 4 all crystallized in
the monoclinic crystal system, while 1 crystallized in the lower
ˉ
Refluxing the metal salt (CoCl2·6H2O) with isolated SNS
ligands {2,6-bis(RSCH2)pyridine [R = methyl, L1; ethyl, L2;
butyl, L3; phenyl, L4 and cyclohexyl, L5] and bis(RSCH2CH2)-
amine [R = ethyl, L6 and butyl, L7]} afforded the corres-
ponding pyridyl-centered SNS complexes Co[2,6-bis(RSCH2)-
pyridine]Cl2 (R = methyl, 1; ethyl, 2; butyl, 3; phenyl, 4 and
cyclohexyl, 5) and amine-based SNS complexes Co[bis-
(RSCH2CH2)amine]Cl2 (R = ethyl, 6 and butyl, 7). The initial
indicators of complex formation were the absence of resolvable
NMR data combined with sharp melting points and distinct
color changes during the reaction. For the pyridyl complexes
(1–5), accurate elemental analyses were combined with data
from X-ray diffraction (with the exception of 5) for confir-
mation of the molecular composition, structure and geometry.
The amine-based complexes (6–7) were very air sensitive,
comparatively less stable than the corresponding pyridyl com-
plexes (2–3) and decomposed on exposure to the atmosphere.
Hence, all attempts at growing crystals suitable for X-ray crys-
tallography failed. However, data from other techniques
including IR spectroscopy, elemental analysis and melting
point determination, conclusively confirmed isolation of the
intended complexes in high purity. Table 1 presents a
summary of the most important bands in the IR spectra which
correspond to the N–H stretching vibration (3500–3300 cm−1),
the C–H alkyl stretch and bend as well as C–H rocking
vibrations. However, IR spectroscopy was unsuitable for detect-
ing the C–S stretch for all three ligands due to overlap by
strong and broad C–H rocking vibrations, but it is worth
noting that the signals did intensify upon ligand coordination
to the Co center. Relevant shifts in vibration frequencies, such
symmetry triclinic P1 space group. The molecular structure of
1, as presented in Fig. 1, shows tridentate SNS ligands each
chelated in a meridional fashion to cobalt centers via two
sulfur atoms and a nitrogen atom from the pyridine ring. Rela-
tively weak metal–metal interactions were observed between
the two octahedral centers of 1 separated through space by a
Co⋯Co distance of 3.684 Å, which in perspective is weaker
than similar interactions in reported square-planar SNS metal
complexes bearing t-butyl groups on the S-donor atoms,
having dimeric centers (respectively, 3.223 and 3.255 Å for
Ir⋯Ir and Rh⋯Rh).36
Complex 2 crystallizes in the monoclinic Cc space group
with four twins (two independent molecules, Fig. 2) in the
asymmetric unit cell. The Co(II) ion in each molecule exists as
a five coordinate center stabilized by the neutral SNS ligand
and two terminal chloride ions. Due to the increased size of
the S-atom substituent (R group), the geometries of the latter
members of this series of complexes differ significantly in
solid state structural conformations from 1. This is due to the
non-constrained free rotation around C–C single bonds that
result in increased steric bulk and less compact geometries.
For instance, the bulkier and electronically richer ethyl R
group in 2 may be partly responsible for its propensity to exist
as a five coordinate monomer, while 1 easily dimerizes in an
octahedral fashion. A comparison of selected bond lengths
and angles shows that the N(1)–Co(1) bond of 2 is shorter than
that of 1 [2.048(7) Å vs. 2.1131(17) Å], while the S(1)–Co(1) and
S(2)–Co(1) bond lengths are longer, indicative of a stronger
C–N bond and comparatively weaker C–S bonds in 2. This
observation may be ascribed to increased back-donation and
15972 | Dalton Trans., 2014, 43, 15968–15978
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 1 Molecular structure and atomic numbering scheme of 1 with thermal ellipsoids at the 50% probability level.
Fig. 2 Molecular structure and atomic numbering scheme of 2 showing two crystallographically independent molecules in the asymmetric unit
cell. Thermal ellipsoids are drawn at the 50% probability level.
chlorides and the N-donor atom, while the two axial positions
are occupied by the relatively trans S-donor atoms. A related Zn
(SNS) complex reported by Teixidor and co-workers35 revealed
a conformation similar to 2, but, due to the presence of
bulkier bromide ions, exhibited an acute S(1)–Zn–S(2) bite
angle of 157.0°, compared to 161.90° for 2.
Observation of the crystal packing between molecules of 2
revealed an ordered arrangement of crystal units containing
the two molecules in alternating planes (Fig. 3), which resulted
in a higher melting point of 164–165° compared to 107–108°
for 3. Primarily, the more ordered crystal in the solid state
required a greater input of energy to disrupt intermolecular
cohesion in the complex.46
Table 2 Selected bond lengths (Å) and bond angles (°) for 1–4
1
2
3
4
N(1)–Co(1)
S(1)–Co(1)
S(2)–Co(1)
S(2)–Co(1)–S(1)
N(1)–Co(1)–S(1)
N(1)–Co(1)–S(2)
2.1131(17)
2.4980(5)
2.4889(6)
162.85(2)
81.84(5)
2.048(7)
2.521(2)
2.518(3)
161.90(8)
82.3(2)
2.133(3)
2.057(5)
2.4496(15)
2.4604(14)
152.55(7)
82.07(11)
80.80(10)
2.5460(17)
2.5208(17)
161.84(6)
81.94(15)
81.13(14)
81.33(5)
81.5(2)
overlap of d-orbitals of the Co(II) center and the pi-antibonding
orbitals of the pyridyl N-donor as a result of the ethyl group in 2.
A distorted trigonal bipyramidal geometry is observed for 2
due to a slightly narrower ‘bite angle’ defined as the S(2)–Co
(1)–S(1) angle. The geometry around the metal center is
defined by three equatorial positions occupied by the two
Geometrically, complex 3 (Fig. 4) is very similar to 2 with
distorted trigonal bipyramidal geometry in which the pyridyl
moiety and the two chloride groups occupy equatorial posi-
tions, while the two S-donor atoms occupy axial positions. The
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15968–15978 | 15973

View Article Online
Paper
Dalton Transactions
(107–108 °C) for 3 compared to the other complexes reported
herein.46
Complex 4, represented in Fig. 5, crystallizes in the P21/c
space group with the phenyl substituents positioned on the
opposite sides of the S–N–S–Co plane, which correlates with a
related Cr complex reported by Temple et al.42 In comparison
with 3, the methylene linkers C(7) and C(13) are situated on
either side of S–N–S–Co plane deviating from the plane by
+0.950 and −0.825 Å respectively.
Application of complexes 1–7 in the oxidation of n-octane
As noted earlier, the main goal of this study was to prepare cat-
alysts for the selective activation of paraffins represented by
n-octane. To test the effectiveness of complexes 1–7, tert-butyl
hydroperoxide (TBHP) was used as a source of oxygen for
n-octane. The choice of oxidant was based on the knowledge
that it is a more effective and milder source of reactive oxygen
than the more common hydrogen peroxide (H2O2).47,48
Exploratory work with the current set of catalysts has also
showed that TBHP is the more active and stable source of
oxygen, hence all data presented here are based on its use as
the oxidant.
Fig. 3 Wireframe representation of the crystal packing of 2 with hydro-
gen atoms omitted for clarity.
A series of exploratory experiments were conducted in order
to establish optimum reaction conditions for optimal substrate
conversion and product selectivity. The optimum octane to
oxidant ratio was found to be 1 : 20 while 24 and 48 hours were
the most favorable reaction times for the amine- and pyridine-
based complexes respectively. Results showed that beyond
these time frames no significant change in substrate conver-
sion was obtained. Furthermore, too low conversions were
observed at shorter reaction times. Blank reactions with only
the oxidant TBHP in the absence of any catalyst gave a total
n-octane conversion of 1% and as expected, no conversion was
observed for blank runs that contained the catalyst in the
absence of any oxidant. A blank run with the metal precursor
(CoCl2·6H2O) as the catalyst was also performed and the
results revealed 5 and 7% n-octane conversions after 24 and
48 hours respectively. Furthermore, only 2-,3- and 4-octanone
were formed and no 1-octanol or other products of primary
carbon activation were detected, indicating poor selectivity to
primary products for the blank runs.
The oxidation of n-octane with catalysts 1–7 under the
optimum conditions produced a mixture of isomeric octa-
nones and octanols that were oxygenated at carbon positions
1, 2, 3 and 4 (Scheme 2), as well as other terminal products,
mainly octanal and octanoic acid. Furthermore, no cracking of
the substrate was observed and only C8 products were
obtained. Table 3 presents data on total octane conversion and
the regioselectivity parameter (C1 : C2 : C3 : C4). The regioselec-
tivity parameter compares the relative reactivity of hydrogen
atoms at carbon positions 1, 2, 3 and 4 along the n-octane
backbone leading to various oxygenates (Scheme 2).
Fig. 4 Molecular structure and atomic numbering scheme of
showing one conformation of the disordered butyl group with thermal
ellipsoids at the 50% probability level.
3
N(1)–Co(1), S(1)–Co(1) and S(2)–Co(1) bond lengths for 3 are
very similar to 1 (Table 2), but a notably more acute bite angle
of 152.55(7)° is observed in comparison with either 1 or 2. The
pyridyl ring is twisted such that C(2) and C(6) reside on the
S–N–S–Co plane, whereas C(3), C(4) and C(5) atoms deviate
from the [N(1), C(2), C(3), C(4), C(5)] plane by 0.383, 0.485 and
0.313 Å respectively, leading to a dihedral angle of 11.81°.
Unlike in complexes 1 and 2, the substituent ‘arms’ C(1)
and C(7) are positioned on the same side of the S–N–S–Co
plane, deviating by just 0.155 and 0.287 Å respectively. The
steric bulk and disorder brought about by the free rotation of
the butyl groups accounted for the relatively low melting point
Results of the catalytic study suggest that amongst the pyri-
dine-based catalysts, 1 was the most active, with the highest
(12%) total conversion of the substrate n-octane. Such an
outcome was not unexpected considering that 1 has the least
15974 | Dalton Trans., 2014, 43, 15968–15978
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 5 Molecular structure and atomic numbering scheme of 4 with thermal ellipsoids at the 50% probability level.
Scheme 2 Product distribution in the oxidation of n-octane.
Table 3 Total conversion and regioselectivity parameter C1 : C2 : C3 : C4 in the oxidation of n-octane using TBHP as the oxidant
Total
conversion/%
Octanone
selectivity/%
Octanol
C1 : C2 : C3 : C4
Octanone
C2 : C3 : C4
Totala
C1 : C2 : C3 : C4
Catalyst
1
2
3
4
5
6
7
12
11
9
9
7
23
17
83
79
83
82
77
83
90
1 : 3.5 : 3.1 : 0
1 : 1.7 : 0 : 1.9
2 : 3.6 : 1 : 3.6
1 : 1.5 : 0 : 2.5
1 : 2.5 : 0 : 5
1 : 1.9 : 2.1 : 1.4
1 : 1 : 0 : 3.3
1.3 : 1 : 1
1.5 : 1.1 : 1
1.4 : 1 : 1
1.4 : 1 : 1
1.4 : 1 : 1
1.4 : 1.1 : 1
1.4 : 1.1 : 1
1 : 5.8 : 4.6 : 3.7
1 : 3.9 : 2.5 : 2.8
1 : 5.6 : 3.9 : 4.2
1 : 3.8 : 2.6 : 3.2
1 : 4.7 : 3.2 : 4.1
1 : 4.1 : 3.3 : 3.0
1 : 8.3 : 6.4 : 6.4
a The total regioselectivity parameter takes into account all products (octanones, octanols, octanal and octanoic acid). Reactions were carried out
at 80 °C with a catalyst loading of 1 mol%, an octane to oxidant ratio of 1 : 20, pentanoic acid as the internal standard and reaction times of 24
and 48 hours for the amine- and pyridine-based catalysts respectively.
sterically hindered active site due to the relatively small of the R group on the S-donor atoms, which implies that the
methylene groups bonded to the ligand S-donor atoms. This, catalytic activity is dominated by steric factors.
in theory, should render greater access to the metal center for
However, in the instances where the side groups are straight
the substrate and lead to the enhanced catalytic activity chain alkyl substituents, electronic and steric contributions to
observed for 1. In general terms, there is an inverse relation- reactivity are difficult to separate because the longer chain
ship between the reactivity of the metal complexes and the size members show enhanced electronic contributions to the che-
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15968–15978 | 15975

View Article Online
Paper
Dalton Transactions
lating S-donors, but they are also more sterically hindered. The hols to ketones, an experiment was set up under similar con-
question that arose in these instances is whether sterics or ditions as used for the octane oxidation. Complex 6 was used
electronics dominate the activity of the catalysts. Complexes 4 as a representative catalyst, into which a fresh solution of
and 5 contain phenyl and cyclohexyl side groups respectively 2-octanol was oxidized with TBHP over a 5 h period. The result
and the difference in reactivity between these two catalysts led revealed a time dependent production of over 86% 2-octanone
to the belief that steric access to the metal center is the prime within the 5 h period. This observation suggests that the over-
determinant of catalytic activity. The planar phenyl side group oxidation of octanols (including 1-octanol to octanal) pro-
in 4 is less sterically demanding than the electronically richer ceeded quite early and relatively quickly in the current systems
but sterically bulkier cyclohexyl in 5. The result is consistent such that any alcohol that was formed almost immediately got
and showed better total conversion and selectivity for 4.
In addition to R-group steric variation as a determinant of context, closely related reports are summarized below:
catalytic activity in complexes 1–5, the amine N-donor contain- • Kirillova et al. have described the use of a tetracopper(II)
further oxidized. To put the results of this catalytic study into
ing complexes (6 and 7) displayed significantly higher catalytic complex for the oxidation of alkanes under mild conditions
activities with conversions of 23 and 17% respectively which is using TBHP as the oxidant.49 The results obtained for the oxi-
likely due to the additional flexibility of the amine backbone. dation of n-octane in particular showed that the reaction did
This observation has implications for the design of catalysts not involve free hydroxyl radicals due to the high regioselecti-
stabilized by carbon chain chelated donor atoms.
vity profile (1 : 65 : 32 : 30) after 30 min reaction time. Follow-
The degree of flexibility of the carbon chain linker between ing an increase in the reaction time to 180 min, the parameter
the chelated donor atoms is important for the set of SNS dropped to 1 : 10 : 6 : 6. In addition, over-oxidation of alcohols
ligands reported herein. A flexible two carbon spacer between to the corresponding ketones was observed.
a simple amine N-donor and the S-donor atoms yielded more
• Lau and co-workers have reported on the oxidation of
effective catalysts than a rigid pyridine N-donor linked to the alkanes with an [OsVI(N)Cl4]−/Lewis acid system using various
S-donors via a methylene spacer. Direct comparison of the two peroxides including TBHP.50 Yields of up to 93% based on
structurally related amine backbone complexes
6 (SNS- [TBHP] consumption were obtained with cyclohexane as the
etamine) and 7 (SNS-butamine) further confirmed the impor- substrate. A yield of up to 69% alcohol production was
tance of the side chain steric influence as a key determinant of achieved and further oxidation of the alcohol to cyclohexanone
catalytic activity. Hence, in summary, complex 6 functionalized was also observed. Studies using open-chain alkanes (n-hexane
with a relatively smaller ethyl R group showed the best poten- and n-heptane) as substrates were also carried out but no oxi-
tial for n-octane oxidation amongst the whole range of seven dation of the primary C–H bond was achieved although total
catalysts studied.
yields of up to 83% were obtained.
Consistent with reported observations, the total regioselec-
• Shul’pin et al. have demonstrated the oxidation of a
tivity for each catalyst (Table 3) showed that internal hydrogens variety of substrates using a manganese(IV) complex salt.48 In
at the carbon position C2 (Scheme 2) were the most reactive, particular, the results obtained for the oxidation of n-octane
while terminal hydrogens at position C1 were the least reac- with TBHP showed the production of a mixture of isomeric
tive.1,47 Hence, the total selectivity up to 90% recorded for alcohols and ketones, with negligible activation of the primary
octanone isomers via activation of internal carbon positions C–H bond.
(mainly C2 but also including C3 and C4) as the dominant
• Pombeiro, Shul’pin and co-workers have reported a tetra-
products of the oxidation reaction is an improvement over copper(II) triethanolaminate complex for the oxidation of
the selectivity of ca. 80% previously reported by Pombeiro alkanes using H2O2 and acid co-catalysts.51 For n-octane,
et al.47 Specifically on octanols, it was observed that the regioselectivity
parameters
highest selectivity to 1- and 2-octanols was displayed by 3, to 1 : 5.1 : 5.2 : 4.3 were reported.
3-octanol by 1 and to 4-octanol by 5, implying that the more • Pombeiro et al. have also described the oxidation of
of
1 : 4.3 : 3.7 : 3.4
and
rigid pyridine N-donor based complexes are more selective alkanes using a homogeneous and immobilized Mn(salen)
against over-oxidation of octanols to octanones than com- complex.47 The results obtained for the oxidation of n-octane
plexes 6 and 7. However, it is evident that overall, 7 displayed with TBHP are as follows: no activation of the primary C–H
the highest selectivity at positions C2, C3 and C4 which bond was observed, isomeric alcohols and ketones were formed
accounted for the higher overall octanone production at these which were oxygenated at positions 2, 3 and 4 of the hydro-
positions. The trends of these parameters are comparable to carbon chain and yields of ca. 1% were reported with TONs of
those reported in the literature47,49 for catalytic systems that up to 110. Furthermore, low regioselectivity profiles [C(2) :
utilized TBHP as the oxidant. Furthermore, these results imply C(3) : C(4)] of 1 : 1 : 1.3 and 1.7 : 1 : 1.1 for the homogeneous and
that a radical initiated mechanism is followed as proposed by immobilized catalysts, respectively, were obtained.
Pombeiro and co-workers.47
• Finally, a report from Pombeiro, Shul’pin and co-workers
Thus far it is safe to summarize that in these catalytic described the oxidation of alkanes using a H2O2–NaVO3–
systems, over-oxidation of internal carbons has led to the pro- H2SO4 catalyst system.52 Regioselectivity parameters of
duction of ketones as the main dominant products. In an 1 : 10.1 : 10.7 : 8.4 and 1 : 7 : 6 : 5 were obtained for this system
effort to further understand the rate of this oxidation of alco- in MeCN and H2O respectively.
15976 | Dalton Trans., 2014, 43, 15968–15978
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
13 M. Albrecht, P. Dani, M. Lutz, A. L. Spek and G. van Koten,
Conclusion
J. Am. Chem. Soc., 2000, 122, 11822–11833.
The preparation of a series of cobalt complexes containing 14 M. Kanzelberger, B. Singh, M. Czerw, K. Krogh-Jespersen
pincer-type SNS ligands with two different backbones, a con-
strained six-membered pyridine ring and a linear straight
and A. S. Goldman, J. Am. Chem. Soc., 2000, 122,
11017–11018.
chain amine, was achieved. The crystal structures of complexes 15 R. A. Baber, R. B. Bedford, M. Betham, M. E. Blake,
1–4 confirmed that the SNS ligands were terdentate and co-
ordinated to the metal centre via the N- and S-donor atoms.
All the Co complexes (1–7) were tested for the oxidative
S. J. Coles, M. F. Haddow, M. B. Hursthouse, A. G. Orpen,
L. T. Pilarski, P. G. Pringle and R. L. Wingad, Chem.
Commun., 2006, 3880–3882.
functionalization of n-octane with TBHP as the oxidant. Only 16 D. F. MacLean, R. McDonald, M. J. Ferguson, A. J. Caddell
linear C8 oxygenated products were obtained. Complex 1 was and L. Turculet, Chem. Commun., 2008, 5146–5148.
the most active catalyst in the pyridine-based series with a 17 W. H. Bernskoetter, S. K. Hanson, S. K. Buzak, Z. Davis,
total conversion of 12%, while 6 was the most active amongst
P. S. White, R. Swartz, K. I. Goldberg and M. Brookhart,
the two amine-based catalysts with a recorded total conversion
J. Am. Chem. Soc., 2009, 131, 8603–8613.
of 23%. Overall, the amine-based complexes proved to be the 18 J. Meiners, A. Friedrich, E. Herdtweck and S. Schneider,
most efficient catalysts in this study, as significantly higher Organometallics, 2009, 28, 6331–6338.
conversions were exhibited compared to the pyridine-based 19 B.-S. Zhang, W. Wang, D.-D. Shao, X.-Q. Hao, J.-F. Gong
catalysts. But, the more rigid pyridine-based complexes
showed better alcohol selectivity.
and M.-P. Song, Organometallics, 2010, 29, 2579–2587.
20 D. Milstein, Top. Catal., 2010, 53, 915–923.
According to the selectivity profile of each catalyst, octa- 21 J.-L. Niu, X.-Q. Hao, J.-F. Gong and M.-P. Song, Dalton
nones were more abundant, with 2-octanone being the domi- Trans., 2011, 40, 5135–5150.
nant product formed, which is in line with the regioselectivity 22 P. Majumder, S. Baksi, S. Halder, H. Tadesse, A. J. Blake,
parameter that revealed C2 as the prominent position of attack
in the hydrocarbon chain for all the catalysts studied.
M. G. B. Drew and S. Bhattacharya, Dalton Trans., 2011, 40,
5423–5425.
23 N. Nebra, J. Lisena, N. Saffon, L. Maron, B. Martin-Vaca
and D. Bourissou, Dalton Trans., 2011, 40, 8912–8921.
24 C. J. Moulton and B. L. Shaw, J. Chem. Soc., Dalton Trans.,
1976, 1020–1024.
25 M. E. van der Boom and D. Milstein, Chem. Rev., 2003, 103,
1759–1792.
26 J. T. Singleton, Tetrahedron, 2003, 59, 1837–1857.
27 J. Choi, A. H. R. MacArthur, M. Brookhart and
A. S. Goldman, Chem. Rev., 2011, 111, 1761–1779.
28 D. S. McGuinness, P. Wasserscheid, W. Keim, D. Morgan,
J. T. Dixon, A. Bollmann, H. Maumela, F. Hess and
U. Englert, J. Am. Chem. Soc., 2003, 125, 5272–5273.
29 S. P. Downing, M. J. Hanton, A. M. Z. Slawin and
R. P. Tooze, Organometallics, 2009, 28, 2417–2422.
Acknowledgements
We are very grateful to the University of KwaZulu-Natal, c*
change and the NRF for generous financial support. We thank
Dr Manuel Fernandes and Dr Bernard Omondi for X-ray crys-
tallographic data collection and refinement.
References
1 J. A. Labinger and J. E. Bercaw, Nature, 2002, 417, 507–514.
2 B. C. Bailey, H. Fan, J. C. Huffman, M.-H. Baik and 30 B. Pitteri and M. Bortoluzzi, Polyhedron, 2008, 27,
D. J. Mindiola, J. Am. Chem. Soc., 2007, 129, 8781–8793.
3 J. Brazdil, Top. Catal., 2006, 38, 289–294.
4 A. E. Shilov and G. B. Shul’pin, Chem. Rev., 1997, 97,
2879–2932.
2259–2264.
31 L. Canovese, G. Chessa, G. Marangoni, B. Pitteri,
P. Uguagliati and F. Visentin, Inorg. Chim. Acta, 1991, 186,
79–86.
5 R. H. Crabtree, J. Chem. Soc., Dalton Trans., 2001, 32 C. Viñas, P. Anglès, G. Sánchez, N. Lucena, F. Teixidor,
2437–2450.
L. Escriche, J. Casabó, J. F. Piniella, A. Alvarez-Larena,
R. Kivekäs and R. Sillanpää, Inorg. Chem., 1998, 37,
701–707.
6 R. H. Crabtree, J. Organomet. Chem., 2004, 689, 4083–4091.
7 G. W. Parshall, Acc. Chem. Res., 1975, 8, 113–117.
8 Y. N. Kozlov, G. V. Nizova and G. B. Shul’pin, J. Phys. Org. 33 S.-Q. Bai, L. L. Koh and T. S. A. Hor, Inorg. Chem., 2009, 48,
Chem., 2008, 21, 119–126.
1207–1213.
9 J. A. Labinger, Fuel Process. Technol., 1995, 42, 325–338.
10 T. A. van den Berg, J. W. de Boer, W. R. Browne, G. Roelfes
and B. L. Feringa, Chem. Commun., 2004, 2550–2551.
11 G. B. Shul’pin, C. R. Chim., 2003, 6, 163–178.
34 M. Konrad, F. Meyer, K. Heinze and L. Zsolnai, J. Chem.
Soc., Dalton Trans., 1998, 199–206.
35 F. Teixidor, L. Escriche, J. Casabo, E. Molins and
C. Miravitlles, Inorg. Chem., 1986, 25, 4060–4062.
12 A. Vigalok, O. Uzan, L. J. W. Shimon, Y. Ben-David, 36 Y. Klerman, E. Ben-Ari, Y. Diskin-Posner, G. Leitus,
J. M. L. Martin and D. Milstein, J. Am. Chem. Soc., 1998,
120, 12539–12544.
L. J. W. Shimon, Y. Ben-David and D. Milstein, Dalton
Trans., 2008, 3226–3234.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 15968–15978 | 15977

View Article Online
Paper
Dalton Transactions
37 F. Teixidor, G. Sanchez-Castello, N. Lucena, L. Escriche, 45 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
R. Kivekas, M. Sundberg and J. Casabo, Inorg. Chem., 1991, 46 R. J. C. Brown and R. F. C. Brown, J. Chem. Educ., 2000, 77,
30, 4931–4935.
724–731.
38 P. Braunstein and F. Naud, Angew. Chem., Int. Ed., 2001, 40, 47 T. C. O. Mac Leod, M. V. Kirillova, A. J. L. Pombeiro,
680–699.
M. A. Schiavon and M. D. Assis, Appl. Catal., A, 2010, 372,
191–198.
48 G. B. Shul’pin, G. Suss-Fink and L. S. Shul’pina, J. Mol.
Catal. A: Chem., 2001, 170, 17–34.
39 M. Bassetti, A. Capone and M. Salamone, Organometallics,
2003, 23, 247–252.
40 B. S. Furniss, A. J. Hannaford, P. W. G. Smith and
A. R. Tatchell, Vogel’s textbook of practical organic chemistry, 49 M. V. Kirillova, A. M. Kirillov, D. Mandelli, W. A. Carvalho,
J. Wiley and sons, New York, 5th edn, 1989. A. J. L. Pombeiro and G. B. Shul’pin, J. Catal., 2010, 272, 9–17.
41 D. L. Reger, R. F. Semeniuc and M. D. Smith, Cryst. Growth 50 S. Yiu, W. Man and T. Lau, J. Am. Chem. Soc., 2008, 130,
Des., 2005, 5, 1181–1190. 10821–10827.
42 C. N. Temple, S. Gambarotta, I. Korobkov and 51 M. V. Kirillova, Y. N. Kozlov, L. S. Shul’pina, O. Y. Lyakin,
R. Duchateau, Organometallics, 2007, 26, 4598–4603.
43 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr.,
2008, 64, 112–122.
A. M. Kirillov, E. P. Talsi, A. J. L. Pombeiro and
G. B. Shul’pin, J. Catal., 2009, 268, 26–38.
52 L. S. Shul’pina, M. V. Kirillova, A. J. L. Pombeiro and
G. B. Shul’pin, Tetrahedron, 2009, 65, 2424–2429.
44 A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7–13.
15978 | Dalton Trans., 2014, 43, 15968–15978
This journal is © The Royal Society of Chemistry 2014