Carboxylate-rich hybrid ligands in Mn(ii) complexes as precursors for water oxidation reactions

Article abstract of DOI:10.1039/c3dt51666k

A new family of Mn^II complexes stabilized by carboxylate-rich hybrid ligands has been synthesized, characterized and applied as catalytic precursors for MnO₂ nanoparticles in water oxidation by (NH ₄)₂Ce(NO₃)₆. The Royal Society of Chemistry.

Full text of DOI:10.1039/c3dt51666k

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Carboxylate-rich hybrid ligands in Mn(II)
complexes as precursors for water oxidation
reactions†
Cite this: Dalton Trans., 2014, 43,
1821
C. R. Raymond Gan,a Zhaolin Liu,b Shi-Qiang Bai,b Kian Soo Ongb and
T. S. Andy Hor*b,c
Received 22nd June 2013,
Accepted 29th October 2013
A new family of MnII complexes stabilized by carboxylate-rich hybrid ligands has been synthesized,
characterized and applied as catalytic precursors for MnO2 nanoparticles in water oxidation by (NH4)2-
Ce(NO3)6.
DOI: 10.1039/c3dt51666k
www.rsc.org/dalton
Water oxidation promoted by homogeneous catalysts remains Nano manganese oxides and supported nano manganese
one of the most pursued frontiers today. Formidable chal- oxide on silica are relatively accessible through surfactant-
lenges lie in the energetic and kinetics of the process as well mediated and solvothermal routes but often involve energy
as ligand design and tuning of its associated stereochemical intensive heat treatment.10,11 Facile decomposition into con-
features.1,2 Many biomimetic molecular systems have fared trollable MnO2 of the desired size and shape aspect ratio often
much poorer compared to the water-oxidizing complex (WOC) involves an understanding of the surfactant, pH and tempera-
of photosystem II (PSII)3,4 in the photosynthetic organism. ture influences on the precursor. As part of our continual
Artificial mimics using nitrogen based ligands perform well interest in precursor based water oxidation and hetero-
for some electron-rich metals such as ruthenium and iridium, metallic manganese systems,12–16 herein we have reported four
but face serious constraints when coordinated to mid-tran- MnII carboxylate complexes and their explicit ligand facilitated
sition metals such as manganese, which is ironic as it is the benign metamorphosis into differentiated MnO2 nanoparticles
key metal for PSII. This is evident in a reported homogeneous in the presence of CAN which are active in water oxidation
MnIII/IV water oxidation catalyst5 with turnover number (TON) reactions.
of 4 as compared to a RuIII catalyst6 with a TON of 13 using
similar nitrogen donor based ligands. Although hybrid carboxy-
Experimental section
Materials and physical measurements
late/pyridyl type ligands, which can stabilize Mn in high oxi-
dation states, demonstrate catalytic activity towards water
oxidation (TON: ∼13–20), half the oxygen source is derived
All the chemicals including ethylenediaminetetraacetic acid
from the tert-butyl hydroperoxide oxidant.7,8 Fostering water
(H4L1) were used as received. Diethyl 2,2′-azanediyldiacetate
oxidation in which both oxygen sources derive from H2O by
(L2a), diethyl 2,2′-(2-hydroxy-5-nitrobenzylazanediyl)diacetate
homogeneous Mn complexes has, to date, largely proven
(L2b), and 2,6-bis(bromomethyl)-4-nitrophenol were syn-
difficult. On the other hand, MnO2 has been shown to be an
thesized based on literature reports.17,18 Elemental analyses
excellent water oxidation catalyst due to its robustness,
were performed on a VarioMICRO Elementar system. 1H and
environmental compatibility and the ability to extract both
13C NMR spectra were recorded on a Bruker AMX 500 MHz
oxygen from H2O in the presence of (NH4)2Ce(NO3)6 (CAN).9
NMR spectrometer. Electrochemical analysis was performed
on an Autolab assembly with glassy carbon as the working
electrode, 3 M KCl, Ag/AgCl as the reference electrode and
aNUS Graduate School for Integrative Sciences and Engineering, Center for Life
platinum as the counter electrode. Oxygen was detected on a
Sciences, #05-01, 28 Medical Drive Singapore 117456, Singapore
bInstitute of Materials Research and Engineering, Agency for Science, Technology and
Research, 3 Research Link, S117602, Singapore
Rank Brothers Clark electrode and the isotope O2 was analyzed
by GC-MS. IR spectra were recorded using a Varian 3100 FT-IR
spectrometer using KBr pellets. UV-Vis absorption spectra
were recorded on a UV2550 Shimadzu spectrometer. The fol-
lowing buffer solutions were prepared for UV-Vis studies: pH 6:
0.1 M NaH2PO4 (87.7 mL) + 0.1 M Na2HPO4 (12.3 mL), pH 5:
0.1 M acetic acid (357 mL) + 0.1 M sodium acetate (643 mL),
cDepartment of Chemistry, National University of Singapore, 3 Science Drive 3,
S117543, Singapore. E-mail: andyhor@nus.edu.sg
†Electronic supplementary information (ESI) available: The CIF files of the com-
plexes can be obtained from Cambridge Crystallographic Data Centre. CCDC
885184 (1), 885242 (2), 885186 (3) and 885187 (4). For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c3dt51666k
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1821–1828 | 1821

View Article Online
Paper
Dalton Transactions
pH 4: 0.1 M acetic acid (847 mL) + 0.1 M sodium acetate
Synthesis of tetraethyl 2,2′,2″,2′′′-(2-hydroxy-5-nitro-1,3-phe-
(153 mL), pH 3: 0.1 M acetic acid (982.3 mL) + 0.1 M sodium nylene)bis(methylene)bis(azanetriyl)tetraacetic acid sodium
acetate (17.7 mL), pH 2: 0.2 M KCl (50 mL) + 0.2 M HCl salt (Na5L3). Aqueous solution (50 ml) of NaOH (1.75 g,
(13 mL) + H2O (137 mL), pH 1: 0.2 M KCl (50 mL) + 0.2 M HCl 43.75 mmol) was added into a THF solution (150 mL) of L3a
(134 mL) + H2O (16 mL). The pH of the solution was measured (4.5 g, 8.33 mmol), and the mixture was refluxed for 24 h. The
using a Hanna Instruments pH meter. The pH was monitored solution was vacuumed to yield a (very hydroscopic) yellow
1
of the solution containing the Mn catalyst (2.36 μmol in crude product. Yield: 4.51 g, 100%. H NMR (D2O, 500 MHz,
2.5 mL of H2O) and the addition of CAN (182 μmol) into the δ): 8.04 (2H, s, phenyl), 3.50 (s, 4H, CH2), 3.10 (s, 8H, CH2). 13
C
solution. ESI-MS was performed on a Finnigan LCQ mass NMR (D2O, 500 MHz, δ): 179.9, 179.8, 130.0, 129.1, 127.5, 59.1,
spectrometer. (Conditions: capillary temperature was main- 55.4. ESI-MS (m/z, %): [H5L3 + H]+ (428.8, 100), [Na5L3 + Na]+
tained at 50 °C, and the capillary voltage was 17 V for the low (563.9, 76). Anal. Calcd for C16H14N3Na5O11·H2O (557.26): C,
flow conditions. The concentrations are 0.59 mM for com- 34.49; H, 2.89; N, 7.54. Found: C, 34.58; H, 3.45; N, 7.20%.
plexes 1–4 and 1.18 mM for (NH4)2Ce(NO3)6 (CAN), respect-
Synthesis of complex 1. Ligand H4L1 (1 g, 3.42 mmol) and
ively. The solution upon mixing was immediately subjected to NaOH (0.55 g, 13.75 mmol) were dissolved in water (10 mL),
mass spectrometric analysis. The samples were injected via a Mn(ClO4)2·6H2O (1.86 g, 5.16 mmol) was added and the reac-
syringe fitted with a 5 mL sample loop.) Transmission Electron tion was refluxed for 30 min. Slow evaporation of the resulting
Microscopy (TEM) characterization was performed using a solution at r.t. afforded crystals of 1 after ∼4 months. Yield:
Philips CM300-FEG instrument with an operating voltage of 1.45 g, 100% (based on Mn). ESI-MS (m/z, %): [1 − H]− (850.2,
300 kV. The amount of Mn and O elements was determined 100). Anal. Calcd for C20H38N4O22Mn3·4H2O (923.40): C, 26.01;
using an Oxford INCA Energy Dispersive X-ray (EDX) spectro- H, 5.02; N, 6.07. Found: C, 25.72; H, 4.74; N, 5.94%. Main
meter. Sample preparation involves separating the resulting IR (cm−1): 3489 (s), 3267 (s), 2960 (m), 2917 (m), 1704 (m),
product upon reaction of the complex and CAN by centrifu- 1609 (s), 1409 (s), 1345 (m), 1292 (m), 1253 (m), 1111 (m),
gation, washed with deionized water, suspended in EtOH and 1030 (m), 983 (m), 851 (m), 786 (m), 696 (m) and 646 (m).
ultrasonicated before loading on a carbon coated copper grid
Synthesis of complex 2. H3L2 (0.1 g, 0.35 mmol), NaOH
for imaging. X-ray photoelectron spectroscopy (XPS) data were (0.04 g, 1.05 mmol) and Mn(ClO4)2·6H2O (0.13 g, 0.35 mmol)
recorded with a Thermo Scientific Theta Probe and binding were mixed and stirred in H2O (5 mL) at r.t. for 10 min. The
energies were calibrated with respect to the residual C 1s peak mixture was filtered and the filtrate was layered with EtOH
at 285.0 eV.
(5 mL) to afford crystals of 2 after ∼3 months. Yield: 0.04 g,
Synthesis of 2,2′-(2-hydroxy-5-nitrobenzylazanediyl)diacetic 32% (based on Mn). ESI-MS (m/z, %): [2 − Na + H2O + OH]−
acid (H3L2). L2b (1 g, 2.94 mmol) was mixed with NaOH (389.8, 100). Anal. Calcd for C11H11MnN2NaO8·4H2O (449.20):
(0.24 g, 6 mmol) in H2O (20 mL) and refluxed for 30 min. HCl C, 29.41; H, 4.26; N, 6.24. Found: C, 29.77; H, 4.11; N, 6.29%.
(20%, 10 mL) was subsequently added to precipitate out the Main IR (cm−1): 3380 (s), 3257 (s), 159I8 (vs), 1479 (s), 1400 (s),
2,2′-(2-hydroxy-5-nitrobenzylazanediyl)diacetic acid. Yield: 1385 (s), 1292 (s), 1262 (s), 1187 (m), 1100 (s), 1010 (m),
0.4 g, 48%. 1H NMR (DMSO, 500 MHz, δ): 8.17 (1H, s, phenyl), 936 (m), 891 (m), 724 (w), 673 (w), 644 (w) and 518 (w).
8.07 and 8.06 (1H, d, J = 8.8 Hz, phenyl), 6.97 and 6.95 (1H, d,
Synthesis of complex 3. Na5L3 (0.05 g, 0.09 mmol) was dis-
J = 8.8 Hz, phenyl), 3.93 (2H, s, CH2), 3.48 (4H, s, CH2). solved in H2O (2 mL) and Mn(ClO4)2·6H2O (0.067 g,
13C NMR (DMSO, 500 MHz, δ): 172.7, 163.0, 139.9, 0.185 mmol) was added at r.t. to give a red solution. A few
126.1, 125.2, 125.1, 116.2, 54.2 and 53.0. ESI-MS (m/z, %): drops of formic acid were added to the mixture. The resulting
[H3L2 + H]+ (284.92, 100). Anal. Calcd for C11H12N2O7·1.25 red solution was slowly evaporated to afford crystals of 3 in
(HCl) (329.80): C, 40.06; H, 4.05; N, 8.49. Found: C, 39.97; H, ∼6 months. Yield: 0.019 g, 35% based on Mn. ESI-MS (m/z, %):
3.76; N, 8.19%.
Synthesis of tetraethyl 2,2′,2″,2′′′-(2-hydroxy-5-nitro-1,3-phe- 2(CHO2)
nylene)bis(methylene)bis(azanetriyl)tetraacetate
[3 − 3(Na) − 2(CHO2) − 2(H2O)]− (588.02, 50), [3 − 3(Na) −
5(H2O)]− (534.11, 100). Anal. Calcd for
(L3a). THF C18H26Mn2N3Na3O20·H2O: C, 26.98; H, 3.52; N, 5.24. Found: C,
−
solutions (20 mL) of 2,6-bis(bromomethyl)-4-nitrophenol 26.74; H, 3.30; N, 5.48%. Main IR (cm−1): 3424 (s), 2927 (m),
(0.72 g, 3.05 mmol) and L2a (1.16 g, 6.14 mmol) were mixed 1594 (vs), 1471 (m), 1406 (s), 1300 (s), 1098 (s), 977 (w),
and stirred at r.t. for 30 min, and then Et3N was added and the 921 (m), 751 (m) and 629 (m).
solution was stirred for 24 h. The resulting mixture was filtered
Synthesis of complex 4. Citric acid (0.2 g, 0.95 mmol) and
and the filtrate was collected and concentrated to yield a MnCO3 (0.11 g, 0.95 mmol) were mixed and refluxed in H2O
yellow oil. The product was purified by column chromato- (5 mL) for 30 min under stirring. The mixture was then fil-
graphy on silica gel. Yield: 1.37 g, 95%. 1H NMR (CDCl3, tered, and the filtrate was layered with acetone (5 ml) to afford
500 MHz, δ): 8.12 (2H, s, phenyl), 4.22 (8H, q, J = 7 Hz, CH2), crystals of 4 after ∼3 months. Yield: 0.15 g, 59.5% (based on
4.02 (4H, s, CH2), 3.57 (8H, s, CH2), 1.29 (12H, t, J = 7.55 Hz, Mn). ESI-MS (m/z, %): [4 − 2(H2O) − 2(H)]− (487.53, 40), [4 − 2
Me). 13C NMR (CDCl3, 500 MHz, δ): 171.0, 161.6, 140.2, 125.3, (H2O) − H − Mn]− (434.71, 100). Anal. Calcd for C12H16O16Mn2
124.3, 61.0, 54.6, 53.6 and 14.2. ESI-MS (m/z, %): [L3a + Na]+ (526.12): C, 27.39; H, 3.07. Found: C, 27.02; H, 3.49%. Main
(564.0, 100). Anal. Calcd for C24H35N3O11(541.55): C, 53.23; IR (cm−1): 3481 (s), 3397 (s), 2974 (m), 2922 (m), 2810 (m),
H, 6.51; N, 7.76. Found: C, 53.75; H, 6.14; N, 7.57%.
2531 (m), 1733 (s), 1594 (vs), 1476 (m), 1418 (s), 1322 (m),
1822 | Dalton Trans., 2014, 43, 1821–1828
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 1 Crystallographic data of complexes 1–4
Complex
1
2
3
4
Formula
Mw
C
20H46Mn3N4O26
C11H11MnN2NaO8
377.15
C18H26Mn2N3Na3O20
783.27
C6H8MnO8
263.06
923.43
T/K
100(2)
Monoclinic
P21/n
9.158(1)
16.064(2)
11.822(2)
90.743(3)
1739.0(4)
2
223(2)
Monoclinic
P21/c
13.141(2)
9.325(1)
11.093(2)
101.347(4)
1332.8(3)
4
100(2)
Monoclinic
C2/c
30.224(3)
9.6362(9)
19.887(2)
99.506(2)
5712.4(9)
8
100(2)
Crystal system
Space group
a/Å
b/Å
Orthorhombic
P212121
6.0017(7)
10.443(1)
13.510(2)
90
846.8(2)
4
2.064
c/Å
β/°
V/Å3
Z
Dcalc/Mg m−3
1.764
1.178
1.880
1.071
1.822
1.025
μ/mm−1
1.584
θ range/°
2.14–27.49
12 174
3978 [0.0229]
285
2.70–27.50
9301
3056 [0.0647]
214
2.22–25.00
16 293
5031 [0.0579]
451
2.47–27.49
6007
1938 [0.0275]
146
Reflections collected
Independent reflections [Rint
Parameters
GOF
]
1.110
1.122
1.042
1.057
R1 [I > 2σ(I)]
wR2 (all data)
0.0288
0.0721
0.0621
0.1321
0.0444
0.1229
0.0232
0.0575
1284 (m), 1257 (m), 1192 (m), 1153 (m), 1072 (m), 889 (m), 849 (m),
798 (m), 667 (m), 609 (m), 581 (m), 538 (m) and 490 (m).
Clark cell measurements
The chosen Mn catalyst (2.36 μmol, 1 eq. in 2.5 mL of H2O) was
bubbled with Ar until O2 saturation reaches below 5% threshold.
182 μmol of CAN (80 eq.) was then added into the solution care-
fully. The parameters were recorded on a data logger pico
ADC-20. The gas output from the catalytic reaction was con-
ducted using 16OH2 and 18OH2 to eliminate the possibility of O2
contamination from the environment during the measurement
process by GC-MS. The atmospheric air background was used as
a reference against the gas output from the activity of 1–4.
Fig. 1 Single-crystal molecular structures of complexes 1–4.
X-ray crystallography
from commercially available Mn(ClO4)2 and ethylenediamine-
tetraacetic acid (H4L1), which contains four carboxylic acid
groups, in water. [Mn2(L2)2Na2(OH2)2] (2) was obtained from
the reaction of Mn(ClO4)2 with H3L2, which contains two
carboxylic acid groups. A similar procedure from Mn(ClO4)2,
formic acid and Na5L3 (L3 = tetraethyl 2,2′,2″,2′′′-(2-hydroxy-5-
nitro-1,3-henylene)bis(methylene)bis(azanetriyl)tetraacetic acid),
which has four carboxylates, gives [Mn2(L3)(HCO2)2Na3(OH2)]
(3). [Mn2(H2L4)2(OH2)2] (4) was synthesized from MnCO3 and
tribasic citric acid (H4L4).
Single-crystal X-ray diffraction analysis (Fig. 1) of these com-
plexes indicates that they are stabilised by multidentate coordi-
nation groups of carboxylate, phenolic oxygen and amine
nitrogen donors (Scheme 1). A common feature of these com-
plexes is that the Mn centers are invariably stabilised by
primary hydration, or have coordination sites for water
affiliation. Complex 1 shows a linear trinuclear MnII structure
in which a central [Mn(H2O)4] moiety is flanked by two
[Mn(H2O)(HL1)] units with HL1 as bridges. Complex 2 pre-
sents a Mn–Na heterometallic 3D coordination polymer, in
which all the carboxylates, phenolate, and para-nitro groups
All measurements were conducted on a Bruker AXS SMART
APEX diffractometer equipped with a CCD area-detector using
Mo-Kα radiation (λ = 0.71073 Å) (Table 1). The collecting
frames of data, indexing reflection and determination of
lattice parameters and polarization effects were performed
using the SMART suite programs.19 The integration of the
intensity of reflections and scaling was carried out using
SAINT.20 The empirical absorption correction was performed
using SADABS.21 The space group determination, structure
solution and least-squares refinements on |F|2 were carried
out using SHELXS-97 and SHELXL-97. Anisotropic displace-
ment parameters were refined for the rest of the non-hydrogen
atoms. CCDC reference numbers: 885184 (1) 885242 (2),
885186 (3) and 885187 (4).
Results and discussion
The ligands used are [N,O] hybrid ligands heavily fortified
with carboxylate groups for enhanced donicity and water com-
patibility. The complex [Mn3(HL1)26H2O] (1) was prepared
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1821–1828 | 1823

View Article Online
Paper
Dalton Transactions
Scheme 1 Structural representation of complexes 1–4.
Fig. 2 Oxygen evolution output of 1–4 with CAN.
are fully involved in coordination. Complex 3 contains a di-
nuclear Mn2 unit in which three of the carboxylates and one
para-nitro group further link to six Na centers and extend into
a 3D framework. Complex 4 crystallizes in an orthorhombic
chiral space group. It shows a 1D polymeric structure in which
the MnII centers are bridged by citric carboxylates. Stabilisation
from multidentate citrate and hydrate is evident, as was
reported in other systems such as PdII citrate.22 In order to
gain an understanding of the possible structures of the com-
plexes in solution phase, ESI-MS studies were conducted on
the complexes (Fig. S1–4†). For 1, the spectrum gives a parent
peak [1 − H]− at m/z = 850, which suggests that the integrity
of 1 is preserved in solution. For 2, the peak at m/z = 389.8
[2 − Na + H2O + OH]− points to sodium dissociation giving
a mononuclear Mn species in solution. For 3, m/z = 588.02
represents the dissociation of Na+ and formic acid, generating
a hydrated dinuclear Mn species depicted as [3 − 3(Na) −
2(CHO2) − 2(H2O)]− as well as the dehydrated form at m/z =
534.11, [3 − 3(Na) − 2(CHO2) − 5(H2O)]−. For 4, the peak at
m/z = 487.53 suggests a dimeric Mn form that exists as [4 −
2(H2O) − 2(H)]−.
Table 2 Water oxidation activities of 1–4 using CAN as a sacrificial
oxidant
TOF/mmolO
per molMn
2
Mn per
32O2 to
34O2 to
36O2 to
Complex complexa s−1 b
air ratioc air ratiod air ratioe
1
2
3
4
3
0.012
0.006
0.006
0.011
0.24
0.04
0.09
0.15
0.015
0.005
0.011
0.013
0.101
0.017
0.033
0.067
1f
2
2
a No. of Mn atoms per complex in solution. b Based on the expected
amount of MnO2 formed. c The gas ratio is calculated by (intensity of
32O2 evolved subtract the background 32O2 intensity)/background 32O2
intensity. d 34O2 evolved/32O2 in argon purged background atmospheric
air in 18OH2 (95%). e 36O2 evolved/32O2 in argon purged background
atmospheric air in 18OH2 (95%). f Complex 2 is a polymeric complex in
solid but appears to exist as a mononuclear species in solution.
from oxygen exchange of the initial coordinated water–water of
crystallization in the complex with 18OH2 (95%).
While the presence of 34O2 is common in water oxidation
driven by homogeneous Mn complexes known to date, the
observation of 36O2 came as a surprise as it is known to arise
only as a result of catalytic activity from the reaction of solid
manganese oxides with CAN.23,24 In a bid to determine the
actual catalyst that is active in this system, the precipitates
formed were subjected to transmission electron microscopy
(TEM) and energy-dispersive X-ray spectroscopy (EDX) ana-
lyses, both of which point to MnOx nanoparticles (Fig. 3 and
S15–18†). Those formed in the reaction for 1 and 4 are spheri-
cal but irregular, with a size distribution of 20 to 50 nm in dia-
meter. Those formed from 2 and 3 appear as irregular prolate
Oxygen evolution, pH, TEM and XPS studies
The oxygen evolution ability of complexes 1–4 was tested using
(NH4)2Ce(NO3)6 (CAN) and measured using a Clark electrode
(Fig. 2 and Table 2). Control experiments were done using
CAN, and CAN with ligands; Mn(ClO4)2 did not result in cata-
lytic O2 evolution whereby treatment of 1–4 with CAN
(182 μmol, 80 eq.) led to the evolution of 0.15, 0.025, 0.05 and
0.09 μmol of O2 at the end of the reaction respectively (Fig. S5†).
These data revealed that 1 is most active, followed by 4, 3 and 2.
The oxygen output was further analysed by GC-MS. Evolu-
tion of excess 32O2 was observed when using H2O, while more
36O2 than 34O2 with reference to the background was observed
when 18OH2 (95%) was used during the water oxidation for all
the complexes (Table 2 and Fig. S6–14†). These isotopic experi-
ments also suggested that both oxygen atoms of O2 are from
water, and that the presence of 34O2 most certainly resulted
spheroid particles with
a
larger size distribution of
100–200 nm in length and 50–90 nm in width. Also evident is
the aggregation of small into larger particles in these MnO2
samples. EDX analysis of the decomposed product from 1–4
confirms the bulk material to be made up of Mn and O, with
traces of residual Ce at 4.9 keV. X-ray photoelectron spec-
troscopy (XPS) analysis of the residues from 1–4 and CAN
1824 | Dalton Trans., 2014, 43, 1821–1828
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 5 Relative amounts of residual Mn of 1–4 from ICP-MS.
MnO2 produced. As 1 is a trinuclear complex, it can theoreti-
cally give rise to three molar equivalents of MnO2. Complex 2
is a mononuclear manganese species in solution and is
expected to decompose into one equivalent of MnO2 while 3
and 4 are dimanganese complexes in solution, thus leading to
two equivalents of MnO2 (Table 2). The manganese content of
1, (3, 4) and 2 is hence in the ratio of 3 : 2 : 1 which also corre-
sponds well to the relative quantities of the Mn observed from
the residue of 1–4 by inductively coupled plasma mass spec-
trometry (ICP-MS) (Fig. 5). This also provides a rough guide to
the relative quantities of the O2 generated. The O2 evolution
data also suggest that smaller size MnO2 particles are more
active in water oxidation activity. This is clear in the case of 3
and 4 where equivalent amounts of MnO2 are generated but 4
is more active than 3 by a factor of ∼1.8 due to its smaller size
and shape aspect ratio. The TOF calculated based on the
expected amount of MnO2 formed in the reaction therefore
Fig. 3 TEM images of MnO2 formed from reactions of (A) 1, (B) 2,
(C) 3 and (D) 4 with (NH4)2Ce(NO3).
correspond to 0.012 and 0.011 mmolO /molMn s−1 for complex
2
1 and 4 respectively and 0.006 mmolO /molMn s−1 for com-
2
plexes 2 and 3 (Table 2). It appears that irregular spherical
MnO2 of size 20–50 nm in diameter are more catalytically
efficacious than irregular prolonged spherical MnO2 of size
100–200 nm in length and 50–90 nm in width.
Fig. 4 XPS spectra of MnO2 residues from the reactions between 1–4
and (NH4)2Ce(NO3)6.
(Fig. 4) revealed the peaks of Mn 2p3/2 and 2p1/2 at 642 and
653.8 eV, 654 and 642.2 eV, 653.7 and 641.9 eV, 642 and
653.8 eV respectively. The difference in binding energy between
the two peaks for all samples corresponds to a spin-energy sep-
The rapid conversion of the complexes under oxidative con-
ditions to MnO2 nanoparticles can be followed by pH changes.
A rapid drop from an initial pH of ∼7 for the complex solution
to pH ∼ 1.4 within 60 s is attributed to the introduction of
80 eq. of CAN solution (Fig. 6). Progressive reduction of the pH
to ∼1.04 after 900 s could be due to the release and accumu-
lation of H+ during O2 evolution in water. Unlike photoelectro-
chemical cells, the H+ ions cannot be converted to H2.26
We further examined the reaction mixtures (containing 1–4
with two equivalents of CAN) by ESI-MS to identify possible
intermediates en route to degradation to MnO2 (Fig. S19–27†).
The dominant species in the mixture containing 1 is 428.0
aration of 11.8 eV, which is in agreement with the reported
25
values of Mn 2p3/2 and Mn 2p1/2
,
thus lending more support
for the generation of MnO2 through oxidation of 1–4 with CAN.
The use of 1 and 4 as catalytic precursors thus led to MnO2
of similar size and shape while 2 and 3 gave rise to particles of
bigger size and wider shape distribution. This disparity can be
rationalized as 1 and 4 have a similar carboxylate coordination
sphere whilst 2 and 3 are unique in having phenoxide coordi-
nation. The O2 evolution data suggest that the water oxidation
activity is dependent on the amount and physical state of
m/z, which corresponds to [Mn2(L1)(H2O)2(μ-O)4Ce2(NO3)]2−
.
This suggests that oxidation of 1 occurs with cleavage of the
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1821–1828 | 1825

View Article Online
Paper
Dalton Transactions
effect.27 (NH4)2Ce(NO3)6, similarly to 1 and 4, also exhibits
deep UV absorption at <220 nm. Complex 2 exhibits an absorp-
tion peak at 387 nm, which is significantly red-shifted from
that of 1 and 4. Complex 3 has an absorption band at 369 nm.
By changing the solution pH to 6 and 5, the absorption peak
of 2 is blue shifted to 378 and 366 nm respectively. Further
reduction to pH 4 resulted in an absorption shoulder at
375 nm and a peak at 316 nm. Lowering the pH to below 4
causes a red shift of the absorption peak of 2 to 312 nm. The
absorption peak of 3 red shifts to 384 and 378 nm when the
solution pH changes from 7 to 6 and 5 respectively. Two absorp-
tion peaks for 3 occur at 311 and 384 nm under pH 4 while only
the 311 nm peak was seen in the pH range 3–1. As the peaks at
∼310 nm are representative of the free ligand, these studies
suggested that carboxylate dissociation starts below pH 4.
The course of the water oxidation reaction was also followed
spectrophotometrically for 1–4 upon the addition of one equi-
valent of CAN (Fig. S32–35†). Complex 1 revealed the formation
of a peak at 487 nm which is consistent with high valence
MnIV species.28 Complex 2 with one equivalent of CAN also
shows the formation of a new intermediate species at 320 nm
which is blue shifted from 387 nm. For 3, there are two new
peaks at 305 and 379 nm which are also suggestive of water
oxidation intermediates. For 4, it appears that the intermedi-
ates formed could not be readily detected by UV-Vis. The peak
at 487 nm, which was present for 1, suggests the formation of
a soluble high-valent Mn species that is sufficiently stable to
be measured before decomposition into MnO2 nanoparticles.
On the other hand, addition of CAN to 2–4 gave no evidence of
any stable intermediate but a more rapid decomposition to
MnO2 nanoparticles.
Fig. 6 pH versus time plots of complexes 1–4 with CAN.
[MnL1] moiety and entry of two Ce ions. The mixture of 2 and
CAN shows the formation of 387.77 m/z [Mn(L2)(OH)2(OH2)]−,
which is the product of deprotonation of a hydrate to a
hydroxide complex. For 3, addition of CAN to the solution also
points to the entry of Ce, as evidenced in a new set of peaks
at 798.3 [Mn2(L3)(OH)2(OH2)5Ce]+ and 797.4 [Mn(L3)-
(OH)3(OH2)4Ce]+. The mixture of 4 and CAN shows the peak at
m/z 485.6 [Mn2(L4)2]−. These data collectively point to proton
release and hydroxide generation from a coordinated hydrate
prior to MnO2 formation, which is consistent with the
observation of a sharp drop in pH.
Although water oxidation described in this work is brought
about by the MnO2 nanoparticles, a cyclic voltammetry analy-
sis of the complexes conducted in aqueous medium of pH 6
(Fig. 8, Table 3) reveals that water oxidation by complexes 1–4
could be feasible under electrochemical conditions or chemi-
cal oxidants that can maintain the oxidation function at high
UV absorption and cyclic voltammetry
The UV-Vis absorption of 1–4 was studied as a function of pH
in order to investigate the stability of the complexes under
various pH conditions (Fig. 7 and Fig. S28–31†). In deionised
H2O (pH 7), the absorption of 1 and 4 both lies in the deep UV
region of less than 220 nm, generally interpreted as the LMCT
Fig. 7 Absorption spectra in deionised H2O. (Complex: 0.02 mM; CAN:
Fig. 8 Cyclic voltammetry of 1–4 in 0.1 M Na2HPO4 buffer at pH 6.0.
0.02 mM.)
(Background voltammogram has been subtracted.)
1826 | Dalton Trans., 2014, 43, 1821–1828
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 3 Cyclic voltammetry data of complexes 1–4
Notes and references
Complex
[Epa, Epc]/V
ΔEp/V
E1/2/V
1 V. Balzani, A. Credi and M. Venturi, ChemSusChem, 2008, 1,
26–58.
2 N. S. Lewis and D. G. Nocera, Proc. Natl. Acad. Sci. U. S. A.,
2006, 103, 15729–15735.
3 J. Dasgupta, R. T. van Willigen and G. C. Dismukes, Phys.
Chem. Chem. Phys., 2004, 6, 4793–4802.
1
[1.16, not observed], [1.0, 0.58],
[0.72, 0.14]
[0.83, 0.69], [0.23, 0.09]
[0.82, 0.67], [0.25, 0.14]
[0.94, 0.57], [0.71, 0.09]
0.42, 0.58
0.79, 0.43
2
3
4
0.14, 0.14
0.15, 0.09
0.38, 0.62
0.76, 0.16
0.75, 0.20
0.76, 0.40
4 J. Messinger, Phys. Chem. Chem. Phys., 2004, 6, 4764–4771.
5 J. Limburg, J. S. Vrettos, L. M. Liable-Sands,
A. L. Rheingold, R. H. Crabtree and G. W. Budvig, Science,
1999, 283, 1524–1527.
6 S. W. Gersten, G. J. Samuels and T. J. Meyer, J. Am. Chem.
Soc., 1982, 104, 4029–4030.
7 A. K. Poulsen, A. Rompel and C. J. McKenzie, Angew. Chem.,
Int. Ed., 2005, 44, 6916–6920.
8 R. K. Seidler-Egdal, A. Nielsen, A. D. Bond, M. J. Bjerrum
and C. J. McKenzie, Dalton Trans., 2011, 40, 3849–3858.
9 M. M. Najafpour and D. J. Sedigh, Dalton Trans., 2013, 42,
12173–12178.
10 F. Jiao and H. Frei, Chem. Commun., 2010, 46, 2920–2922.
11 M. M. Najafpour, F. Rahimi, E.-M. Aro, C.-H. Lee and
S. I. Allakhverdiev, J. R. Soc. Interface, 2012, 9, 2383–2395.
12 G. C. Dismukes, R. Brimblecombe, G. A. N. Felton,
R. S. Pryadun, J. E. Sheats, L. Spiccia and G. F. Swiegers,
Acc. Chem. Res., 2009, 42, 1935–1943.
pH. An irreversible oxidation wave at 1.16 V and two redox
waves at E1/2 (MnII/III) = 0.43 and E1/2 (MnIII/IV) = 0.79 V for 1,
E1/2 (MnII/III) = 0.16 and E1/2 (MnIII/IV) = 0.76 V are attributed to
2. Two redox waves at E1/2 (MnII/III) = 0.20 and E1/2 (MnIII/IV) =
0.76 V are attributed to 3. For 4, a redox wave at E1/2 (MnII/III) =
0.40 and E1/2 (MnIII/IV) = 0.76 V is assigned. The redox potential
of these complexes meets the thermodynamic requirement of
water oxidation, i.e. E0 (complex) > E0 = 0.61 V vs. Ag/AgCl at
pH = 6. Significantly enhanced catalytic water oxidation waves
were observed at potentials from 0.5 to 1.4 V. A similar study
on an Ir complex confirmed that the increased oxidation wave
with respect to the electrolyte was a clear indication of catalytic
water oxidation.29
Conclusion
13 H. Likura and T. Nagata, Inorg. Chem., 1998, 37, 4702–4711.
14 T. Nagata, S. K. Zharmukhameodov, A. A. Khorobyrkh,
V. V. Klimov and S. I. Allakhverdiev, Photosynth. Res., 2008,
98, 277–284.
15 Y. R. Hong, Z. Liu, S. F. Al-Bukhari, C. J. J. Lee, D. L. Yung,
D. Chi and T. S. A. Hor, Chem. Commun., 2011, 47, 10653–
10655.
16 Y. Liu, K. H. Lee, J. J. Vittal and T. S. A. Hor, J. Chem. Soc.,
Dalton Trans., 2002, 2747–2751.
17 A. M. D’Hardemare, O. Jarjayes and F. Mortini, Synth.
Commun., 2004, 34, 3975–3988.
18 J. Mendoza, P. M. Nieto, P. Prados and C. Sanchez, Tetrahe-
dron, 1990, 46, 671–682.
19 SMART, version 5.631, Software Reference Manuals; SAINT,
version 6.63, Software Reference Manuals, Bruker AXS
GmbH, Karlsruhe, Germany, 2000.
20 G. M. Sheldrick, Software for Empirical Absorption Correc-
tion: SADABS, University of Göttingen, Göttingen, Germany,
2001.
21 (a) G. M. Sheldrick, Program for Crystal Structure Solution:
SHELXS-97, University of Göttingen, Göttingen, Germany,
1997; (b) G. M. Sheldrick, Program for Crystal Structures
Refinement: SHELXL-97, University of Göttingen, Göttingen,
Germany, 1997.
We have demonstrated the use of carboxylate-rich Mn com-
plexes for catalytic water oxidation. This mimics the natural
PSII system which also shows a similar carboxylate coordi-
nation to the [Mn4Ca] cluster.30 However, under oxidative con-
ditions, our complexes serve as catalytic precursors of MnO2
nanoparticles, whose water oxidation activities depend on
their size and shape which are dependent on the nature and
coordination mode of the multidentate hybrid ligands. The
manganese carboxylate/hydroxyl complexes in the presence
of a large excess of CAN led to irregular spherical MnO2
(20–50 nm) which are more catalytically active than prolate
spherical MnO2 nanoparticles (100–200 nm in length and
50–90 nm in width) formed from the reaction between manga-
nese phenoxide/carboxylate complexes. The ligands used in
the catalytic precursor complexes have a notable influence on
the physical form, and thus the catalytic efficacy, of the nano-
particles formed. We are currently investigating the mechanis-
tic role of these ligands in the complex-to-nanoparticle
transformation process.
Acknowledgements
This research was supported by NUS Graduate School for 22 P. Gunari, S. Krishnasamy, S-Q. Bai and T. S. A. Hor, Dalton
Science and Engineering, the Institute of Materials Research Trans., 2010, 39, 9462–9464.
and Engineering and NUS Department of Chemistry. We also 23 D. Shevela, S. Koroidov, M. M. Najafpour, J. Messinger and
acknowledge Dr Z. Zheng for XPS measurements, H. R. Tan for P. Kurz, Chem.–Eur. J., 2011, 17, 5415–5423.
TEM imaging, Dr S. W. Tay for discussions and G. K. Tan for 24 M. M. Najafpour, F. Rahimi, M. Amini, S. Nayeria and
single crystal X-ray diffraction and refinements.
M. Bagherzadehd, Dalton Trans., 2012, 41, 11026–11031.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1821–1828 | 1827

View Article Online
Paper
Dalton Transactions
25 G. Du, X. Liu, Y. Zong, T. S. Andy Hor, A. Yu and Z. Liu, 28 B. Bahramian, V. Mirkhani, M. Moghadam and A. H. Amin,
Nanoscale, 2013, 5, 4657–4661. Appl. Catal., A, 2006, 315, 52–57.
26 R. Brimblecombe, A. Koo, G. C. Dismukes, G. F. Swiegers 29 D. G. H. Hetterscheid and J. N. H. Reek, Chem. Commun.,
and L. Spiccia, J. Am. Chem. Soc., 2010, 132, 2892–
2011, 47, 2712–2714.
2894.
30 M. M. Najafpour, A. N. Moghaddam, Y. N. Yang, E. Aro,
R. Carpentier, J. J. Eaton-Rye, C. Lee and S. I. Allakhverdiev,
Photosynth. Res., 2012, 114, 1–13.
27 O. M. Suleimenov and T. M. Seward, Chem. Geol., 2000,
167, 177–192.
1828 | Dalton Trans., 2014, 43, 1821–1828
This journal is © The Royal Society of Chemistry 2014