A novel heteroleptic iridium complex supported on graphitic carbon nitride was synthesized and used for photoreduction of carbon dioxide under visible light irradiation. The methanol yield obtained after 24 h irradiation was 9934 mmol g
1cat (TON 1241 with respect to Ir) by using triethylamine (TEA) as a sacrificial donor, which was significantly higher as compared to the semiconductor carbon nitride 145 mmol g
1cat under identical conditions. The presence of triethylamine was found to be vital for the higher methanol yield. After the reaction, the photocatalyst could easily be recovered and reused for subsequent six runs without significant loss in photo activity.

1. Metal-organic hybrid: Photoreduction of CO2 using graphitic carbon
nitride supported heteroleptic iridium complex under visible light
irradiation
Anurag Kumar a, b
, Pawan Kumar a, b
, Rajnikant Borkar c
, Amit Bansiwal c
,
Nitin Labhsetwar c
, Suman L. Jain a, *
a
Chemical Sciences Division, CSIR-Indian Institute of Petroleum, Mohkampur, Dehradun 248005, India
b
Academy of Scientific and Industrial Research (AcSIR), New Delhi 110001, India
c
CSIR-National Environmental Engineering Research Institute (CSIReNEERI), Nehru Marg, Nagpur 440020, India
a r t i c l e i n f o
Article history:
Received 14 May 2017
Received in revised form
11 July 2017
Accepted 23 July 2017
Available online 25 July 2017
Keywords:
Visible light
Photocatalysis
Hybrid material
Graphitic carbon nitride
CO2 reduction
a b s t r a c t
A novel heteroleptic iridium complex supported on graphitic carbon nitride was synthesized and used
for photoreduction of carbon dioxide under visible light irradiation. The methanol yield obtained after
24 h irradiation was 9934 mmol gÀ1
cat (TON 1241 with respect to Ir) by using triethylamine (TEA) as a
sacrificial donor, which was significantly higher as compared to the semiconductor carbon nitride
145 mmol gÀ1
cat under identical conditions. The presence of triethylamine was found to be vital for the
higher methanol yield. After the reaction, the photocatalyst could easily be recovered and reused for
subsequent six runs without significant loss in photo activity.
© 2017 Elsevier Ltd. All rights reserved.
1. Introduction
Photoreduction of carbon dioxide to energy rich chemicals using
solar energy is an area of tremendous importance in view of
providing alternative energy source as well as to mitigate concen-
tration of CO2 in atmosphere [1,2]. Photocatalytic CO2 reduction
using a semiconductor based photocatalyst under mild reaction
conditions has been appeared to be a potential approach for CO2
conversion due to its simplicity and potential scalability for prac-
tical applications [3]. However, most of the semiconductors re-
quires ultra-violet light due to large band gap and provide poor
efficiency and conversion to the desired product. For the efficient
utilization of solar light which comprises 5% UV and 45% visible
light, a photocatalyst that is able to harvest visible light is desired
[4,5]. Due to large band gap, most of the semiconductor photo-
catalysis suffers from the drawback of poor visible light absorption
and faster electron-hole recombination which resulted to the poor
conversion efficiency and selectivity [6]. Unlike to these
semiconductor photocatalysts, molecular complexes of transition
metals such as Ir, Ru(II) or Re(II) bipyridyl complexes possess good
visible light absorbance and long lived excited states; hence they
provide better electron transfer which resulted to the more effi-
cient reduction of CO2 [7e9]. In addition, in metal complexes, metal
centre can bind with CO2, which in turn provide higher CO2 con-
centration and higher conversion efficiency [10,11].
Despite of several advantages, molecular complexes suffer from
the inherent disadvantages of homogeneous catalysis of difficult
recovery and non-recyclability. Thus, combining a visible light
responsive semiconductor with molecular photocatalyst to make
the system heterogeneous seems to be a straight forward approach
for carrying out the CO2 reduction. In this regard, recently our
group has developed a number of CO2 reduction systems by using
organic semiconductors such as graphene oxide/reduced graphene
oxide in combination of transition metal complexes for carrying out
CO2 reduction under visible light irradiation [12e15]. A significant
improvement due to the better charge separation efficiency was
observed in hybrid photocatalysts and they provided higher con-
version/yield of methanol than the individual components with the
added benefits of facile recovery and recyclability for several runs* Corresponding author.
E-mail address: suman@iip.res.in (S.L. Jain).
Contents lists available at ScienceDirect
Carbon
journal homepage: www.elsevier.com/locate/carbon
http://dx.doi.org/10.1016/j.carbon.2017.07.080
0008-6223/© 2017 Elsevier Ltd. All rights reserved.
Carbon 123 (2017) 371e379

2. with nearly consistent activity.
Recently, nanoporous graphitic carbon nitride (npg-C3N4) an
organic semiconductor owing to its easy accessibility from low cost
raw materials, higher stability and visible light responsiveness, has
been extensively used for various applications including as elec-
trochemical electrode, photocatalyst and gas sensors [16,17]. It
contains sp2
hybridized carbon and nitrogen atoms-forming con-
jugated graphitic planes and exhibit higher photoconductivity,
better separation and transport of photo-induced carriers [18]. In
the recent years a number of photocatalysts has been developed by
combining the semiconductors with npg-C3N4 for various appli-
cations such as dye degradation, hydrogen evolution and waste
water treatment [19]. However, the use of npg-C3N4 as metal free
semiconductor [20,21] for photocatalytic reduction of CO2 is rela-
tively less explored. In this regard, recently, Kuriki et al. reported a
Z-scheme photocatalytic system comprising of npg-C3N4 and
ruthenium (II) binuclear complex for the conversion of CO2 to for-
mic acid with higher TON (33,000) and selectivity (87e99% for
formic acid) under visible light [3].
In the present work, we describe an efficient metal-organic
hybrid photocatalyst i.e. nanoporous graphitic carbon nitride
(npg-C3N4) grafted iridium heteroleptic complex (npg-C3N4/Ir-T)
[22,23] for photoreduction of CO2 to give methanol selectively in
significantly higher yield (9.93 mmol gÀ1
cat) by using triethyl-
amine (TEA) as a sacrificial donor under visible light irradiation
(Scheme 1). This value is much greater than those reported for
molecular complexes, semiconductors and semiconductor-metal
complex hybrid photocatalysts for CO2 reduction. Furthermore,
formation of methanol as selective product is beneficial as it is
useful liquid fuel which contains higher energy density and can be
transport easily [24].
2. Experimental
2.1. Materials
2,20-Bipyridine (99%), dicyandiamide (99%), thiourea (!99%),
sodium azide (!99.99%), iridium chloride hydrate (99.8%), copper
(II) sulfate pentahydrate (98%) and 2-cyanopyridine purchased
from Sigma-Aldrich were of analytical grade and used without
further purification. Solvents like dimethylsulfoxide (DMSO),
ethanol, chloroform, hydrogen peroxide (30%) and water were of
HPLC grade and procured from Alfa Aesar. All other chemicals were
of analytical grade and used without further purification.
2.2. Techniques used
The surface morphology of the synthesized material was
determined with the help of field emission scanning electron
microscopy (FE-SEM) by using Jeol Model JSM-6340F. Ultrafine
structure of material was determined by high resolution trans-
mission electron microscopy (HR-TEM) using FEI-TecnaiG2
Twin
TEM operating at an acceleration voltage of 200 kV. Dilute
aqueous suspension of material was deposited on carbon coated
copper grid for the preparation of sample. FTIR spectra of the
compounds were collected on PerkineElmer spectrum RX-1 IR
spectrophotometer containing potassium bromide window. X-
ray diffraction pattern for determining the phase purity and
crystallinity of the material was carried out at Bruker D8 Advance
diffractometer at 40 kV and 40 mA with Cu Ka radiation
(l ¼ 0.15418 nm). Sample for XRD was prepared on glass slide by
adding well dispersed catalyst in slot and drying properly. Ab-
sorption spectra of iridium complex in acetonitrile and solid UV
of oxidized carbon nitride supported iridium complex was
collected on Perkin Elmer lambda-19 UV-VIS-NIR spectropho-
tometer using a 10 mm quartz cell, using BaSO4 as reference. The
Surface properties like BET surface area (SBET), BJH porosity, mean
pore diameter of samples were examined by N2 adsorption-
desorption isotherm at 77 K by using VP; Micromeritics
ASAP2010. Thermogravimetric analyses (TGA) of Ir complex and
C3N4 grafted Ir complex for comparing the thermal stability was
carried out using a thermal analyzer TA-SDT Q-600. Analysis was
carried out in the temperature range of 40e800 C under nitro-
gen flow with heating rate 10 C minÀ1
. Iridium content of
catalyst was determined by ICP-AES analysis which was carried
out through inductively coupled plasma atomic emission spec-
trometer (ICP-AES, DRE, PS-3000UV, Leeman Labs Inc, USA).
Sample for ICP-AES were made by digesting a particular amount
of sample with nitric acid followed by filtration and dilution it up
to 10 ml volume by adding distilled water. ESI-High Resolution
Mass spectra of iridium complex was performed on (Thermo
Exactive Orbitrap system in HESI mode). Electrochemical studies
were performed by using an Electrochemical Workstation Bio-
analytical Systems (BASi) Epsilon potentiostat and three-
electrode, single chamber cell. Solutions were made in 0.1 M
tetra-n-butylammonium hexafluorophosphate (Bu4NPF6) in
spectrophotometric grade acetonitrile (Himedia). The working
electrode was a Pt disk, the auxiliary electrode was a glassy
carbon Pt wire and potentials were measured against either a Ag/
AgCl (3 M NaCl) reference electrode or Ag wire quasi-reference
electrode (QRE), each calibrated against a FeCp2
0/þ
internal stan-
dard (0.46 V vs Ag/AgCl). The scan rate was 100 Mv/s and the
initial concentration of analyte was changed 50 mg to 200 mg for
getting better response.
2.3. Synthesis of 2-(1H-tetrazol-5-yl)pyridine ligand [22]
In a typical synthesis, sodium azide (0.0650 g, 1 mmol) and 2-
cyanopyridine (0.104 g, 1 mmol) were refluxed in DMSO (2 ml) at
140 C for 3 h in the presence of copper (II) sulfate pentahydrate
(0.0050 g, 2 mol%) as catalyst. The progress of the reaction was
checked with TLC. After completion of reaction, the reaction
mixture was cooled at room temperature and 10 mL HCl (4 mol lÀ1
)
was added to it and then with 10 mL ethyl acetate. The organic layer
was separated, washed with 20 mL distilled water, dried over
anhydrous sodium sulfate, and concentrated to give the crude solid
Scheme 1. Synthesis of npg-C3N4/Ir-T photocatalyst.
A. Kumar et al. / Carbon 123 (2017) 371e379372

3. 2-(1H-Tetrazol-5-yl)pyridine. The crude product was purified by
recrystallization using n-hexane: ethyl acetate (1:1) obtaining
0.1430 g of colourless crystals (98% yield). The identity of the syn-
thesized ligand was confirmed by 1
H and 13
C NMR supporting in-
formation (Fig. S4-S5).
2.4. Synthesis of [(bpy)2Ir(m-Cl)]2 complex [15]
The desired [(bpy)2Ir (m-Cl)]2 complex was synthesized by
heating the mixture of bipyridine (0.82 g, 5.25 mmol) and iridium
chloride hydrated (0.35 g, 1.18 mmol) in 40 ml of 2-ethoxyethanol:
water (3:1) at 140 C for 24 h. The mixture was allowed to cool at
0 C and the separated solid was isolated by filtration, washed with
80 ml of acetone: ethanol (1:1) and then dried under vacuum. The
crude product was further purified by recrystallization using hex-
ane: toluene (2:5) to give pure product.
2.5. Synthesis of heteroleptic iridium complex (Ir-T) [23]
Heteroleptic iridium complex was synthesized by following
the literature procedure [23]. In brief, [(bpy)2Ir (m-Cl)]2 (0.500 g,
0.411 mmol), 2- (1H-tetrazol-5-yl)pyridine (0.250 g, 1.7 mmol)
and sodium carbonate (0.479 g, 4.52 mmol) were refluxed in 30 ml
of 2-ethoxyethanol for 12 h under nitrogen atmosphere. After
cooling down to room temperature, the mixture was diluted with
ethyl acetate (120 mL) and washed thoroughly with distilled
water. The combined organic layer was dried over anhydrous
Na2SO4 and concentrated under reduced pressure to give yellow
solid of heteroleptic Ir-T complex. The crude product was purified
column chromatography over activated neutral alumina using
ethyl acetate as eluting agent. Further purification was done by
recrystallization by using 50 ml of ether: hexane (1:4) to give dark
yellow orange crystalline solid as the desired product (64%). The
successful synthesis of heteroleptic Ir-T complex was confirmed
1
H, 13
C NMR and ESI-MS spectra as given in supporting informa-
tion (Fig. S6-S9).
2.6. Synthesis of npg-C3N4 [25]
The synthesis of npg-C3N4 was done by mixing the dicyandia-
mide (1 g) and thiourea (4 g) in a ball mill for 50 min at
300 rpm minÀ1
. The acquired material was heated in muffle furnace
by programmed heating rate 8 C minÀ1
from room temperature to
300 C, 2 C minÀ1
from 300 C to 500 C, 1 C minÀ1
from 500 to
550 C and finally heated at 550 C for 4 h.
2.7. Synthesis of eOH functionalized npg-C3N4 [26]
The oxidation of npg-C3N4 to get eOH functionalized npg-C3N4
(O-npg-C3N4) was carried out by solvothermal treatment of as-
prepared npg-C3N4 with H2O2. In brief, 1 g of npg-C3N4 was taken
in a 100 mL Teflon-lined stainless-steel autoclave and hydrogen
peroxide (30%) was filled up to 80% of vessel total volume. The
autoclave was sealed tightly and heated at 100 C for 6 h. After
cooling the vessel at room temperature, the product was isolated by
centrifugation and washed with water to remove the trace of
hydrogen peroxide. The pale-yellow brown solid was dried at 50 C.
2.8. Synthesis of npg-C3N4/Ir-T [27]
Graphitic carbon nitride supported iridium complex was syn-
thesized by the mixing of Ir complex (0.1 g, 0.15 mmol) and eOH
functionalized npg-carbon nitride (2 g) in ethanol and stirred the
resulting mixture in dark for 48 h. After that the solid was isolated
by filtration and washed with ethanol and dried under reduced
pressure over night at 80 C to give heterogeneous npg-C3N4/Ir-T.
The loading of iridium content in the heterogeneous material was
determined by ICP-AES and was found to be 2.04 wt% (0.132 mmol
iridium gÀ1
catalyst).
3. Results and discussion
3.1. Synthesis and characterization of photocatalyst
At first 2-(1H-tetrazol-5-yl)pyridine ligand was synthesized
from the reaction of 2-cyanopyridine and sodium azide using
copper (II) sulfate as catalyst by following the literature pro-
cedure [22]. As prepared 2-(1H-tetrazol-5-yl)pyridine ligand was
used for the synthesis of heteroleptic iridium complex [23] (Ir-T)
as shown in Scheme 2. Nanoporous graphitic carbon nitride
(ngp-C3N4) was synthesized by the calcination of dicyandiamide
and thiourea at programmed temperature [25]. The surface area
(SBET), total pore volume (Vp) and mean pore diameter (rp) of
npg-C3N4 was found to be 34.67 m2
gÀ1
, 0.21 cm3
gÀ1
, and
8.21 nm, respectively (Fig. S3a). The synthesized npg-C3N4 was
oxidized to get eOH functionalized npg-C3N4 [26]. The hetero-
leptic iridium complex was immobilized to chemically func-
tionalized npg-C3N4 by mixing and stirring the suspension in
ethanol in dark for 48 h at room temperature (Scheme 1). After
immobilization of Ir complex, the SBET, Vp and rp of npg-C3N4/Ir-T
were found to be 54.28 m2
gÀ1
, 0.34 cm3
gÀ1
and 2.69 nm
Scheme 2. Synthesis of heteroleptic iridium complex (Ir-T).
A. Kumar et al. / Carbon 123 (2017) 371e379 373

4. respectively (Fig. S3b). The increased surface area after immo-
bilization of the complex indicated the intercalation of Ir com-
plex units in between the nanosheets of graphitic carbon nitride.
The basic nitrogen and hydroxyl groups on the npg-C3N4 acted as
ligands to provide stable chemical bonding with iridium
complex.
The physico-chemical characterization of the synthesized pho-
tocatalysts (npg-C3N4 and npg-C3N4/Ir-T) was done using various
techniques. The results of FTIR and Raman analysis of photo-
catalysts are given in the supporting information (Fig. S1eS2). The
crystal structure of the synthesized materials was determined with
the help of X-ray diffraction patterns (Fig. 1). In the diffractrogram
of graphitic carbon nitride a characteristic peak at the 2q value of
27.4 (002) with 0.32 nm interlayer distance is observed which
matches well with JCPDS card No. 87e1526 (Fig. 1a) [28e30]. After
the grafting of Ir complex on the surface of -OH functionalized npg-
C3N4 the XRD pattern (Fig. 1b) [31] remains almost similar which
indicated that grafting of complex does not influence the phase
structure of npg-C3N4 or it is due to the lower loading of the
complex on the surface.
The surface morphology and the micro-structure of the syn-
thesized materials were determined by HR-TEM analysis (Fig. 2).
HR-TEM image of the npg-C3N4 showed the layered graphitic and
crumpled nanosheets like structure [32] at 50 nm scale (Fig. 2a).
After the immobilization of Ir-complex the sheet structure
remained almost similar which indicates that immobilization did
not change the morphology of the surface (Fig. 2b). Furthermore,
the existence of large dark spots indicates the successful immobi-
lization of iridium complex between the scaffolds of graphitic
carbon nitride. At 20 nm resolution, some crystallite fringes were
designated to the 0.32 nm interplaner d-spacing of 002 plane of
npg-C3N4 (Fig. 2c). Presence of crystallite fringes with similar
interplaner d-spacing as obtained from XRD confirm that sheets
remains intact during immobilization step (Fig. 2d).
Fig. 1. XRD Pattern of a) npg-C3N4, b) npg-C3N4/Ir-T. (A colour version of this figure can
be viewed online.)
Fig. 2. HR-TEM images of npg-C3N4, a) at 50 nm scale, and npg-C3N4/Ir-T, b) at 100 nm, c) at 20 nm showing lattice fringes, d) at 10 nm scale bar showing interplaner d-spacing. (A
colour version of this figure can be viewed online.)
A. Kumar et al. / Carbon 123 (2017) 371e379374

5. The elemental mapping indicates the homogeneous distribu-
tion of the C, N and Ir elements in the hybrid photocatalyst (Fig. 3).
Furthermore, the presence of all the desired elements in the EDX
pattern suggested the successful formation of npg-C3N4/Ir-T
photocatalyst (Fig. 3). EDX spectra of npg-C3N4/Ir-T showed the
presence of iridium with 1.92 wt% that was in good agreement
with the value obtained by ICP-AES analysis (2.04 wt%). Further-
more, SEM elemental mapping confirmed the existence and uni-
form distribution of the Ir-T complex on the surface of npg-C3N4
(Fig. S10).
The surface chemical properties to confirm the immobilization
of iridium complex on the graphitic carbon nitride surface was
identified by XPS analysis (Fig. 4). Survey scan of npg- C3N4/Ir-T
catalyst showed the presence of peaks due to C1s, N1s, O1s, Ir4f at
binding energy 288, 401, 535, 65 eV respectively which show
iridium complex was present in npg-C3N4/Ir-T catalyst (Fig. S9). The
high resolution XPS spectra in C1s region revealed two character-
istic peak components at binding energy 289.7 and 285.9 eV due to
N]CeN2 and CeC carbons, respectively which are corresponding
to the graphitic carbon nitride (Fig. 4a) [32,33]. In N1s region two
peak components at 402 and 400 eV originated due to the sec-
ondary nitrogen C]NeC and tertiary nitrogen (Ne(C)3), respec-
tively (Fig. 4b) [31e35]. The O1s region showed two peak
components at 533 and 535 eV due to CeOH and C3eNþ
eO-
types
of oxygen which confirmed the successful functionalization of
graphitic C3N4 by eOH groups (Fig. 4c) [26]. In the Ir 4f region
appearance of two peaks at 64.44 eV (Ir 4f7/2) and 67.41 eV (Ir 4f5/2)
suggested that iridium is presented in Irþ3
form in the synthesized
hybrid photocatalyst (Fig. 4d) [36].
The electrochemical properties of the synthesized npg-C3N4 and
npg-C3N4/IrT were explained by cyclic voltammetry at 298 K with
0.1 M tetra-n-butylammonium hexafluorophosphate (nBu4NPF6) as
supporting electrolyte (Fig. 5). There was not much difference
observed in the CV of neat npg-C3N4 and hybrid npg-C3N4/Ir-T
photocatalyst, which was assumed due to the lower loading of the
iridium complex on the support. The photocatalyst displayed two
pairs of peaks on the forward and backward sweeps (Fig. 5b). From
the CV curve, we estimated the band gap (HOMO-LUMO) to be
2.1 eV which was nearly equal to the band gap calculated by Tauc
plot (Fig. 7). The onset of reduction appeared at 0.9 V, leading to a
LUMO level at À3.9 eV, and the oxidation onset at À1.2 V, leading to
a HOMO level at À6.0 eV by using the following equations:
ELUMO ¼ Eonset(red) þ 4.8 eV and EHOMO ¼ Eonset(ox) þ 4.8 eV, where
Eonset(ox) and Eonset(red) are the onset potentials for the oxidation
and reduction processes [37].
UVevisible spectroscopy is also a powerful non-destructive tool
to analyze the optical properties of material. Fig. 6 displays
UVevisible spectra of synthesized materials. The UV adsorption
spectrum of npg-C3N4 indicates the characteristic absorption bands
near to 260 and 377 nm due to pep* and nep* transitions,
respectively (Fig. 6a). In addition, a steep absorption edge near
450 nm, due to the band to band electron transition which is in well
accordance to the literature value [38]. After the immobilization of
the iridium complex units, a new absorption band near to 470 nm
due to the metal to ligand charge transfer (MLCT) corresponding to
the iridium complex is appeared. Furthermore, a slight decrease in
the absorbance in case of npg-C3N4/IrT is due to the shortage in
nep* transitions (Fig. 6b).
The optical band gap of the synthesized photocatalyst was
determined by Tauc plot with linear extrapolation as shown in
Fig. 7. The optical band gap value of npg-C3N4 was found to be
2.4 eV and after the immobilization of complex, the optical band
gap value was found 2.24 eV which confirmed the visible light
responsiveness of the hybrid photocatalyst.
Thermal stability and decomposition behavior of the synthe-
sized npg-C3N4 and npg-C3N4/Ir-T with temperature was deter-
mined by TGA analysis as shown in Fig. 8. Thermogram of npg-C3N4
showed a constant weight loss between 560 and 710 C due to the
degeneration of carbon nitride nanosheets at elevated temperature
(Fig. 8a) [39]. For npg-C3N4/Ir-T catalyst mainly two types of weight
loss were observed (Fig. 8b). In the first type, a continuous slow
weight loss near to 200e400 C was due to the degradation of
iridium complex; whereas another major weight loss between 600
and 720 C was due to the degradation of graphitic carbon nitride
nanosheets.
3.2. Photocatalytic activation of CO2
The photocatalytic activity of synthesized npg-C3N4 and npg-
C3N4/Ir-T was tested for the reduction of CO2 using mixture of
water/DMF/triethylamine as reaction media and sacrificial donor,
respectively under visible light irradiation. At first, the vessel was
charged with DMF (30 ml), triethylamine (10 ml) and deionized
water (10 ml) and then the solution was degassed by purging ni-
trogen for 30 min under vigorous stirring. After that the resulting
mixture was added with photocatalyst (100 mg) and the resulting
suspension was saturated with CO2. Reaction vessel (dia. 5 cm) was
kept about 3 cm away from the light source (20 W, LED light) and
the intensity of the light at the vessel was 85 W mÀ2
as measured by
Fig. 3. HR-TEM elemental mapping of npg-C3N4/Ir-T, a) electron image, b) carbon, c)
nitrogen, d) iridium, and e) EDX pattern. (A colour version of this figure can be viewed
online.)
A. Kumar et al. / Carbon 123 (2017) 371e379 375

6. Fig. 4. High resolution XPS of npg-C3N4/Ir-T, a) C 1s, b) N 1s, c) O 1s and d) Ir 4f region. (A colour version of this figure can be viewed online.)
Fig. 5. Cyclic voltammogram of a) npg-C3N4 and b) npg-C3N4/Ir-T using Pt wire and
glassy carbon disk electrode at a scan rate of 100 mV sÀ1
. (A colour version of this
figure can be viewed online.)
Fig. 6. UVeVis absorption spectra of, a) npg-C3N4/Ir-T, b) npg-C3N4. (A colour version
of this figure can be viewed online.)
A. Kumar et al. / Carbon 123 (2017) 371e379376

7. an intensity meter. The progress of the reaction was measured by
withdrawing the sample after a certain interval of time and
analyzed by GC/FID equipped with a 30 m long Stabilwax®
w/
Integra- Guard®
column. The peak area was correlated to a standard
calibration curve for quantitative determination of methanol. As
the reduction of CO2 provided selective formation of methanol, the
methanol yield was used to evaluate the performance of the pho-
tocatalysts. For the analysis of gaseous products, 20 mL of sample
was injected in the GC-RGA (TCD-FID). The gas phase analysis did
not show the presence of any possible CO2 reduction by-product,
such as CO, CH4 etc. However, a small amount of H2 (64 mmol gÀ1
cat) was observed. The absence of peaks in the GC-FID and the GC-
MS for any other possible liquid product, and in the GC-RGA (TCD-
FID) for any possible gaseous product inferred the higher selectivity
of the npg-C3N4/Ir-T catalyst for methanol formation. As methanol
was selectivity obtained as photoreduction product, its formation
rate RMeOH (mmol gÀ1
cat) as a function of reaction time was
calculated and plotted in Fig. 9. It can be clearly seen that after 24 h
of illumination the methanol yield for npg-C3N4/Ir-T photocatalyst
was found to be 9934 mmol gÀ1
cat (TON 1241 with respect to Ir) by
using TEA as a sacrificial agent. However, the use of npg-C3N4
provided only 145 mmol gÀ1
cat yield of methanol under identical
experimental conditions (Fig. 9) (Table 1). The quantum yield (f)
for methanol formation was estimated to be 0.946 for npg-C3N4/Ir-T
catalyst and 0.013 for npg-C3N4.
These results indicated a significant enhancement in the
methanol yield after immobilization of the iridium complex to
npg-C3N4 support, which might be due to the better charge sep-
aration and mobility of the electrons on the npg-C3N4 support.
Control experiments showed that methanol formation using npg-
C3N4/IrT photocatalyst was negligible when either CO2 or visible
light was absent. Similarly, no reaction was occurred in the
absence of photocatalyst under otherwise identical reaction
conditions (Table 1).
After the photocatalytic reaction, the catalyst could easily be
recovered by centrifugation, washed with water, dried and used for
subsequent runs under described experimental conditions. It can
be seen from Fig. 10 that the recovered photocatalyst exhibited
almost similar activity for six runs for the photoreduction of CO2 as
the fresh one. There was not any leaching of metal/ligand observed,
which was further ascertained by ICP-AES analysis of recovered
catalyst. The amount of iridium content after fifth reuse was found
to be nearly similar (1.98 wt%) to the fresh catalyst (2.04 wt%).
Although the exact mechanism of the reaction is not clear at this
stage; based on the existing reports and band edge position a
possible reaction mechanism is shown in Scheme 3 [40e43]. Owing
to the higher band gap i.e 2.4 eV of npg-C3N4, it can absorb only in
the blue region of the visible light and therefore afforded low yield
of CO2 photoreduction product. However, in the heterogeneous Ir/
npg-C3N4 photocatalyst, visible light absorbance increases due to
the better absorption profile of iridium complex [44]. Iridium
complex after absorption of visible light transformed from singlet
(S0
) to triplet (S1
) state with the transfer of the electrons from
HOMO to LUMO. The electrons from LUMO of iridium complex can
readily be transferred to conduction band of carbon nitride due to
the small difference in the potential [45,46]. Hence, the systems
acted like a p-n heterojunction, where Ir-T complex works like p-
type semiconductor while npg-C3N4 works as n-type semi-
conductor [47,48]. The combined band gap for the npg-C3N4/Ir-T
system was found to be 2.10 eV as determined from cyclic
Fig. 7. Tauc plot for band gap determination of a) npg-C3N4/Ir-T, b) npg-C3N4. (A colour
version of this figure can be viewed online.)
Fig. 8. TGA diagram of, a) npg-C3N4; b) npg-C3N4/Ir-T. (A colour version of this figure
can be viewed online.)
Fig. 9. Conversion of CO2 to methanol with time using, a) blank reaction; b) npg-C3N4;
c) npg-C3N4/Ir-T. (A colour version of this figure can be viewed online.)
A. Kumar et al. / Carbon 123 (2017) 371e379 377

8. voltammetry suggested the visible light absorbance of the photo-
catalyst. In the next step, the adsorbed CO2 on the surface of carbon
nitride accepted electrons from the conduction band of npg-C3N4
for reduction. High surface area of photocatalyst and presence of
nitrogen rich basic sites on carbon nitride, provided more surface
bounded CO2 concentration, which further enhanced the efficient
electron transfer [49].
The selective formation of methanol can be explained on the
basis of band structure of npg-C3N4/Ir-T photocatalyst. From cyclic
voltammetry, the value of LUMO position (conduction band for
heterogeneous semiconductor) for npg-C3N4/Ir-T was found to
be À3.9 V which represent reduction potential at NHE scale equal
to À0.54 V, from equation ENHE ¼ À4.44eEvacuum [50]. Thus, the
conduction band position of photocatalyst was more negative than
reduction potential of HCOOH (CO2/HCOOH ¼ À0.61 V), so ther-
modynamically photoreduction of CO2 to HCOOH was not possible
[51e53]. Further, the position of conduction band for npg-C3N4/Ir-T
was nearly similar to the reduction potential of CO (CO2/
CO ¼ À0.53 V) and HCHO (CO2/HCHO ¼ À0.48 V) respectively,
which was also unfavorable as for the efficient electron transfer the
conduction band potential should be more positive than the
reduction potential of products. However, the reduction potential
for the methanol formation from CO2 (CO2/CH3OH) was
only À0.38 V which was more positive and favorable for the effi-
cient electron transfer from the conduction band of photocatalyst.
Similarly, the more positive value of reduction potential of methane
(CO2/CH4 ¼ À0.24 V) was also favorable for the efficient electron
transfer and its formation as a by-product. However, during the
product analysis we did not observe the formation of methane as
by-product which might be due to the fact that reduction of CO2 to
methane requires 8 electrons and 8 protons which was thermo-
dynamically less favorable than methanol (6 electrons and 6 pro-
tons) formation [54].
Triethylamine works as sacrificial donor and provides required
electrons for the reduction of CO2 while itself get oxidize to
degradation products like diethylamine, acetaldehyde etc. [55].
However, some electrons may be derive from water oxidation as
the valance band position of npg-C3N4/Ir-T photocatalyst
was þ1.56 V vs NHE, which was positive enough than water
oxidation potential (H2O/O2 ¼ þ1.23 V) [56].
4. Conclusions
The present work demonstrated a highly efficient and selective
system developed by using a metal free semiconductor npg-C3N4
coupled with iridium complex for the photocatalytic reduction of
CO2 to methanol under visible light irradiation. The results
showed that npg-C3N4 coupled with iridium complex provided
selectively methanol in 9934.47 mmol gÀ1
cat yield which is
significantly higher as compared to the organic semiconductor
(145 mmol gÀ1
cat). The analysis results confirmed the selective
formation of methanol without any evidence for the formation of
any other gaseous as well as liquid product from the photore-
duction of CO2. The recovered photocatalyst showed almost
similar activity for six runs and provided almost similar yield of
methanol selectively in all cases, which is a significant finding
considering that immobilization of metal complex to photoactive
semiconductor supports will open up new possibilities for
developing efficient photocatalysts for the production of high
value products from CO2 in a sustainable way.
Table 1
Photocatalytic reduction of CO2 with different photocatalysts.[a]
.
Entry Photocatalyst Reactant Visible light source Methanol Yield (mmol gÀ1
cat) TONb
1 Blank CO2 Yes e e
2 npg-C3N4 CO2 Yes 145 e
3 npg-C3N4/Ir-T CO2 Yes 9934 1241
4 npg-C3N4/Ir-T CO2 Noc
e e
5 npg-C3N4/Ir-T N2 Yes e e
a
Reaction conditions: photocatalyst, 100 mg, solvent: DMF/water/triethylamine (30/10/10), visible light irradiation using 20 Watt LED (l 400 nm), time: 24 h.
b
TON is defined as the ratio of the amount of CH3OH produced to that of the active site Ir in the catalyst.
c
Without light (dark condition).
Fig. 10. Results of recycling experiments. (A colour version of this figure can be viewed
online.)
Scheme 3. Possible mechanism of CO2 photoreduction using npg-C3N4/Ir-T.
A. Kumar et al. / Carbon 123 (2017) 371e379378

9. Acknowledgments
Authors are thankful to Director IIP for granting permission to
publish these results. AK and PK are thankful to Council of Scientific
and Industrial Research (CSIR) New Delhi for providing research
fellowships. Analytical Science Division of the Institute is kindly
acknowledged for providing support in analysis. Chetan Joshi, USF,
Tampa, is acknowledged for technical help in synthesis of complex
and providing ESI-Mass analysis from INMAS, DRDO, New Delhi.
We are thankful to Abhishek Pathak, Division of Advanced Mate-
rials and Devices, CSIR-NPL, Delhi, India for XPS analysis of samples.
DST is acknowledged for financial help in project GAP-3125.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.carbon.2017.07.080.
References
[1] C. Song, Catal. Today 115 (2006) 2e32.
[2] X. Xiaoding, J.A. Moulijn, Energy Fuels 10 (1996) 305e325.
[3] R. Kuriki, H. Matsunaga, T. Nakashima, K. Wada, A. Yamakata, O. Ishitani,
K. Maeda, J. Am. Chem. Soc. 138 (2016) 5159e5170.
[4] H. Zhang, X. Fan, X. Quan, S. Chen, H. Yu, Environ. Sci. Technol. 45 (2011)
5731e5736.
[5] J. Yu, G. Dai, B. Huang, J. Phys. Chem. C 113 (2009) 16394e16401.
[6] Y. Izumi, Coord. Chem. Rev. 257 (2013) 171e186.
[7] A. Genoni, D.N. Chirdon, M. Boniolo, A. Sartorel, S. Bernhard, M. Bonchio, ACS
Catal. 7 (2017) 154e160.
[8] P. Voyame, K.E. Toghill, M.A. Mendez, H.H. Girault, Inorg. Chem. 52 (2013)
10949e10957.
[9] Y. Kou, Y. Nabetani, D. Masui, T. Shimada, S. Takagi, H. Tachibana, H. Inoue,
J. Am. Chem. Soc. 136 (2014) 6021e6030.
[10] J.M. Smieja, C.P. Kubiak, Inorg. Chem. 49 (2010) 9283e9289.
[11] B. Gholamkhass, H. Mametsuka, K. Koike, T. Tanabe, M. Furue, O. Ishitani,
Inorg. Chem. 44 (2005) 2326e2336.
[12] P. Kumar, A. Kumar, B. Sreedhar, B. Sain, S.S. Ray, S.L. Jain, Chem. Eur. J. 20
(2014) 1e9.
[13] P. Kumar, B. Sain, S.L. Jain, J. Mater. Chem. A 2 (2014) 11246e11253.
[14] R. Gusain, P. Kumar, O.P. Sharma, S.L. Jain, O.P. Khatri, Appl. Catal. B 181 (2016)
352e362.
[15] P. Kumar, A. Bansiwal, N. Labhsetwar, S.L. Jain, Green Chem. 17 (2015)
1605e1609.
[16] J. Zhu, P. Xiao, H. Li, S.A.C. Carabineiro, ACS Appl. Mater. Interfaces 6 (2014)
16449e16465.
[17] S. Verma, R.B.N. Baig, M.N. Nadagouda, R.S. Varma, ACS Sustain. Chem. Eng. 4
(2016) 1094e1098.
[18] Y.-S. Yun, J. Park, S.U. Lee, A. Thomas, W.H. Hong, G.D. Stucky, Angew. Chem.
125 (2013) 1e6.
[19] X. Wang, S. Blechert, M. Antonietti, Acs. Catal. 2 (2012) 1596e1606.
[20] G. Zhang, M. Zhang, X. Ye, X. Qiu, S. Lin, X. Wang, Adv. Mater 26 (2014)
805e809.
[21] S. Verma, R.B.N. Baig, C. Han, M.N. Nadagouda, R.S. Varma, Green Chem. 18
(2016) 251e254.
[22] B. Akhlaghinia, S. Rezazadeh, J. Braz. Chem. Soc. 12 (2012) 2197e2203.
[23] S. Kumar, P. Kumar, A. Deb, D. Maithi, S.L. Jain, Green Chem. 18 (2016)
2514e2521.
[24] E.V. Kondratenko, G. Mul, J. Baltrusaitis, G.O. Larrazabal, J.P. Ramirez, Energy
Environ. Sci. 6 (2013) 3112e3135.
[25] J. Xu, Y. Wang, Y. Zhu, Langmuir 29 (2013) 10566e10572.
[26] G. Dong, Z. Ai, L. Zhang, Rsc. Adv. 4 (2014) 5553e5560.
[27] K. Takanabe, K. Kamata, X. Wang, M. Antonietti, J. Kubota, K. Domen, Phys.
Chem. Chem. 12 (2010) 13020e13025.
[28] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen,
M. Antonietti, Nat. Mater 8 (2009) 76e80.
[29] L. Ge, Mater. Lett. 65 (2011) 2652e2654.
[30] M. Groenewolt, M. Antonietti, Adv. Mater. 17 (2005) 1789e1792.
[31] J. Yu, S. Wang, B. Cheng, Z. Lin, F. Huang, Catal. Sci. Technol. 3 (2013)
1782e1789.
[32] M.J. Bojdys, J.O. Muller, M. Antonietti, A. Thomas, Chem. Eur. J. 14 (2008)
8177e8182.
[33] H.C. Kolb, M.G. Finn, K.S. Sharpless, Angew. Chem. 40 (2001) 2004e2021.
[34] Q. Xiang, J. Yu, M. Jaroniec, J. Phys. Chem. C 115 (2011) 7355e7363.
[35] L. Ge, C. Han, J. Liu, J. Mater. Chem. 22 (2012) 11843e11850.
[36] U. Hintermair, S.W. Sheehan, A.R. Parent, D.H. Ess, D.T. Richens, P.H. Vaccaro,
G.W. Brudvig, R.H. Crabtree, J. Am. Chem. Soc. 135 (2013) 10837e10851.
[37] Y. Zhu, X. Li, Q. Cai, Z. Sun, G. Casillas, M.J. Yacaman, R. Verduzco, J.M. Tour,
J. Am. Chem. Soc. 134 (2012) 11774e11780.
[38] D.W. Thompson, A. Ito, T.J. Meyer, Pure. Appl. Chem. 85 (2013) 1257e1305.
[39] M. Sadhukhan, S. Barman, J. Mater. Chem. A 1 (2013) 2752e2756.
[40] H. Cheng, K. Fuku, Y. Kuwahara, K. Mori, H. Yamashita, J. Mater. Chem. A 3
(2015) 5244e5258.
[41] S. Sato, T. Morikawa, T. Kajino, O. Ishitani, Angew. Chem. 52 (2013) 988e992.
[42] H. Yang, C. Guo, G.H. Guai, Q. Song, S.P. Jiang, C.M. Li, ACS Appl. Mater. In-
terfaces 3 (2011) 1940e1945.
[43] J. Grodkowski, P. Neta, J. Phys. Chem. A 104 (2000) 1848e1853.
[44] R. Kuriki, M. Yamamoto, K. Higuchi, Y. Yamamoto, M. Akatsuka, D. Lu, S. Yagi,
T. Yoshida, O. Ishitani, K. Maeda, Angew. Chem. Int. Ed. 56 (2017) 4867e4871.
[45] J. Willkomm, K.L. Orchard, A. Reynal, E. Pastor, J.R. Durrant, E. Reisner, Chem.
Soc. Rev. 45 (2016) 9e23.
[46] A. Bachmeier, V.C.C. Wang, T.W. Woolerton, S. Bell, J.C. Fontecilla-Camps,
M. Can, S.W. Ragsdale, Y.S. Chaudhary, F.A. Armstrong, J. Am. Chem. Soc. 135
(2013) 15026e15032.
[47] B. Kumar, M. Llorente, J. Froehlich, T. Dang, A. Sathrum, C.P. Kubiak, Annu. Rev.
Phys. Chem. 63 (2012) 541e569.
[48] P. Kumar, C. Joshi, A. Barras, B. Sieber, A. Addad, L. Boussekey, S. Szunerits,
R. Boukherroub, S.L. Jain, Appl. Catal. B 205 (2017) 654e665.
[49] Y.Y. Lee, H.S. Jung, Y.T. Kang, J. CO2 Util. 20 (2017) 163e177.
[50] N. Sato, Electrochemistry at Metal and Semiconductor Electrodes, Elsevier,
1998.
[51] S.N. Habisreutinger, L.S. Mende, J.K. Stolarczyk, Angew. Chem. Int. Ed. 52
(2013) 7372e7408.
[52] N.M. Dimitrijevic, B.K. Vijayan, O.G. Poluektov, T. Rajh, K.A. Gray, H. He,
P. Zapol, J. Am. Chem. Soc. 133 (2011) 3964e3971.
[53] V.P. Indrakanti, J.D. Kubicki, H.H. Schobert, Energy Environ. Sci. 2 (2009)
745e758.
[54] B. Yu, Y. Zhou, P. Li, W. Tu, P. Li, L. Tang, J. Ye, Z. Zou, Nanoscale 8 (2016)
11870e11874.
[55] R. Reithmeier, C. Bruckmeier, B. Rieger, Catalysts 2 (2012) 544e571.
[56] C. Ye, X.Z. Wang, J.X. Li, Z.J. Li, X.B. Li, L.P. Zhang, B. Chen, C.H. Tung, L.Z. Wu,
ACS Catal. 6 (2016) 8336e8341.
A. Kumar et al. / Carbon 123 (2017) 371e379 379