Chapter 8

Cubic CdO2: Synthesis, Photo

Catalytic Activities, Catalytic

Oxidation Activities and Optical

Studies

Chapter 8

289

CHAPTER 8

Cubic CdO2: Synthesis, Photo Catalytic Activities, Catalytic

Oxidation Activities and Optical Studies

Binary oxides containing metal ions with the d10

electronic configuration are

of interest due to various interesting properties exhibited by them which in turn result

in many applications. Zinc oxide (ZnO) and cadmium oxide (CdO) are typical

examples that are extensively used as transparent electrode, phototransistor,

photodiode, catalysts and gas sensors [8.1-8.4]. Although the peroxides of these metal

ions are known earlier [8.5, 8.6]; however their properties are being explored in the

last two decades [8.7-8.23]. Binary peroxides can be used as catalysts, sensors,

oxygen source in organic synthesis, and precursors for the synthesis of metal oxides

[8.21-8.24]. These peroxides are also used in surgical treatments, vulcanization of

rubbers and concrete mixers [8.25-8.30]. In the literature, synthesis of cadmium

peroxide (CdO2) has been performed employing hydrothermal methods, direct

reaction of oxygen with cadmium metal in either in presence of oxygen gas or in

liquid ammonia or thermal decomposition of the cadmium superoxide [8.5, 8.6, 8.18-

8.23]. Of these, the preparations involving the oxidation of cadmium salts with

hydrogen peroxide (H2O2) appear to be the nonhazardous and viable procedure to

obtain CdO2 [8.6]. Its distinct advantage is to induce the formation of high-

crystallized powders with narrow particle size distribution as well as high purity

without the requirement of post-annealing at high temperatures [8.16].

CdO2 has been reported to possess pyrite type structure exhibiting semi

conducting properties [8.6, 8.18]. Our studies on the use of pyrite type structured zinc

Chapter 8

290

peroxide (ZnO2) as photocatalyst for the degradation of aqueous dye solutions have

been discussed in chapter 7. Like ZnO2, CdO2 has been evaluated as a catalyst for the

photo-degradation of aqueous dye solution and described in this chapter. It is

noteworthy that the sulfur deficient pyrite FeS2 has recently been shown to be a

photocatalyst for the degradation of aqueous dye solution [8.31]. Additionally fluoride

doping in cubic CdO2 nanocrystals has been attempted. Use of CdO2 for the organic

transformation reactions are also described in this chapter.

Synthesis, characterization and properties of CdO2

Cd(NO3)2·4H2O (99%, CDH, India), CdSO4·8/3H2O (99%, CDH, India) and

Cd(CH3COO)2·2H2O (99%, SRL, India) were used as the cadmium sources for the

preparation of CdO2 following the procedure reported in the literature [8.6].

Typically, 1 g of cadmium salt was dissolved in excess of NH4OH to ensure the

formation of ammoniacal cadmium salt solutions i.e., [Cd(NH3)4]2+

. Rapid addition of

10 ml of 30% H2O2 solution to the ammoniacal cadmium salt solutions produced

white colored powders. The powders were separated by filtration and dried at room

temperature.

PXRD patterns of the products obtained with H2O2 solution from

Cd(CH3COO)2·2H2O, Cd(NO3)2·4H2O and CdSO4·8/3H2O are shown in Figure 8.1

(a), (b) and (c), respectively. Reflections in the PXRD patterns were sharp and

intense, implying good crystallinity. The positions and intensities of the peaks

matched very well with the CdO2 JCPDS File No: 78-1125. No reflections due to

either CdO or Cd(OH)2 phase were observed in the PXRD patterns. A Diamond

drawing of the cubic CdO2 lattice using the DIAMOND 3 program is shown in Figure

8.2. To extract the structural parameters further, Rietveld refinement of the

Chapter 8

291

Figure 8.1: PXRD patterns of CdO2 prepared from (a) Cd(CH3COO)2·2H2O (b)

Cd(NO3)2·4H2O and (c) CdSO4·8/3H2O.

experimental profile was carried out using the pyrite structural model with T6

h-Pa-3 (#

205) space group as reported in the JCPDS File No: 78-1125. Rietveld fit of the

PXRD pattern of the CdO2 prepared from the Cd(CH3COO)2·2H2O is shown in Figure

8.3 confirming the cubic symmetry and ruling out the presence of impurities. The

cubic lattice constant of CdO2 from the fit is a = 5.3580 (3) Å. The broadness of the

Bragg reflections clearly indicated the low crystallite size of CdO2 and the average

crystallite size as estimated by the Scherrer analysis was 7 nm. The crystal data and

structure refinement parameters of the cubic CdO2 are summarized in Table 8.1. The

refined unit cell and position parameters, after the final cycle of refinement, are

provided in Table 8.2. The TEM image of CdO2 nanocrystals from

Cd(CH3COO)2·2H2O is presented in Figure 8.4 (A).

JCPDS File no: 78-1125

Chapter 8

292

Figure 8.2: Structure of cubic CdO2.

Figure 8.3: Final Rietveld fit (red line) of the observed PXRD pattern of cubic CdO2

nanocrystals (blue line) and residuum.

Chapter 8

293

Table 8.1: Crystallographic data of cubic CdO2.

Table 8.2: Refined Atomic parameters after the Final cycle of Refinement.

Formula CdO2

Crystal system Cubic

Space Group Pa-3 (# 205)

a [Å] 5.3580 (3)

V [Å3] 153.8290 (2)

Z 2

ρ calc [g/cm3] 6.2355 (1)

Rexp (%) 4.91

Rp (%) 6.06

Rwp (%) 4.72

GOF (S) 1.23

Step size/ Step time 0.01°/1.5sec per step

Number of data points 4000 (2θ = 20-60º)

Number of parameters 21

Number of restraints 2

Temperature 298 K

Atom Site S.O.F x y z Beq

Cd 4a 1.0 0.0 0.0 0.0 1.0

O 8c 1.0 0.38504 0.38504 0.38504 1.0

Chapter 8

294

Aggregation of the nanosized crystals is observed in the TEM image [8.18].

The absence of use of surfactants as morphology controlling agents and the high

surface energy of the nanocrystals may be promoting agglomeration. In the SAED

pattern of the crystallites (Figure 8.4 (B)), well defined spots were present suggesting

the higher crystallinity of the CdO2. It was indexed corresponding to the (200) and

(400) planes of the cubic CdO2 structure. A low-magnification SEM image of the

CdO2 prepared from the Cd(CH3COO)2·2H2O is provided in Figure 8.5, where

uniform spherical morphology of the crystals has been observed. The magnified SEM

image is shown as an inset in Figure 8.5.

Figure 8.4: (A) TEM image and (B) SAED pattern of the CdO2 nanocrystal prepared using

Cd(CH3COO)2·2H2O.

The FT-IR spectrum of CdO2 (obtained using Cd(NO3)2·4H2O) is presented in

Figure 8.6. The signature band of the O-O stretching is observed at 1030 cm-1

while

Cd-O stretching band is observed at 660 cm-1

. The presence of bands at 3417 cm-1

and

1566 cm-1

are attributed to the OH stretching and bending modes of surface adsorbed

water [8.14]. A sharp band at 1407 cm−1

indicated the presence of of N-O stretching

(A) (B)

Chapter 8

295

Figure 8.5: FE-SEM image of CdO2 obtained using Cd(CH3COO)2·2H2O. Inset shows the

same at lower magnification.

vibration. This may arise from the traces of nitrate in the product. Raman spectrum of

the CdO2 is shown in the Figure 8.7 in which scattering resonances are observed at

345 and 830 cm-1

. CdO2 is proposed to possess cubic structure with Pa-3 (Th6) space

group which gives rise to the following irreducible representation:

Г = Ag + Eg + Tg + Au + Eu + Tu

Among these five irreducible representations, only two of them are Raman active (Ag

+ Eg), one is only infra red active (Tu) while the other is both infra red and Raman

active (Tg). The third band may be quite weak at room temperature and probably is

not observed. The absence of CdO even in very minute quantities is evident from the

Raman spectrum of CdO2 [8.18, 8.32]. Base line shift observed in the Raman

spectrum hinted the luminescent nature of the sample.

Chapter 8

296

Figure 8.6: FTIR spectrum of CdO2 obtained using Cd(NO3)2·4H2O.

Figure 8.7: Raman spectrum of CdO2 obtained using Cd(CH3COO)2·2H2O.

Chapter 8

297

The peroxide of cadmium, in fact, releases oxygen exothermically resulting in

explosion. The TGA trace of CdO2 in Figure 8.8 showed that the weight loss of about

10% in the range of 180-280°C. Based on the reaction of CdO2 → CdO + 1/2O2, the

calculated weight loss is 11%. The observed and the theoretically estimated weight

loss matched within the limits of error of measurement. An exothermic peak at around

213°C has been observed in its DSC trace (Figure 8.8).

Figure 8.8: TGA and DSC traces of CdO2 obtained using Cd(CH3COO)2·2H2O.

UV-Visible diffuse reflectance spectrum of CdO2 nanocrystals (from

Cd(NO3)2·4H2O) showed absorption maximum at around 325 nm, which is slightly

blue-shifted in comparison with the excitonic absorption reported for the CdO2 (345

nm) thin film (Figure 8.9 (A)) [8.21]. Using the Kubelka-Munk function, the

estimated band gap of CdO2 sample was 3.10 eV (Figure 8.9 (B)). The band gap of

Chapter 8

298

CdO2 from the other reactant sources {Cd(CH3COO)2·2H2O and 3CdSO4·8H2O},

have been estimated to be still lower viz., 3.04 eV and 2.97 eV respectively from their

UV-Visible diffuse reflectance spectra data (Figure 8.9 (B)). All these values are

lower as compared to the reported ones for the bulk (~ 3.6 eV) [8.18, 8.21].

Figure 8.9: (A) UV-Visible diffuse reflectance spectrum of CdO2 nanocrystals prepared from

Cd(NO3)2·4H2O. (B) Estimation of band gaps of CdO2 prepared from (a) 3CdSO4·8H2O, (b)

Cd(NO3)2·4H2O and (c) Cd(CH3COO)2·2H2O.

Room temperature PL spectrum of CdO2 (from Cd(NO3)2·4H2O) shows a

strong emission at 438 nm (2.83 eV) and a medium intense emission at 418 nm (2.97

eV) (Figure 8.10 (A)). Of these two, the one occurring at the lower wave length

corresponded to the band edge and the other one was the anti-Stokes emission. If

CdO2 has been assumed to be stoichiometric, then one would not observe PL at room

temperature ignoring the bonding aspects (covalency). The fact that a visible emission

is observed clearly demonstrated the presence of intrinsic defects. A similar spectrum

has been observed for the CdO2 from CdSO4·8/3H2O and Cd(CH3COO)2·2H2O as well

(A) (B)

Chapter 8

299

(Figure 8.10 (B)). The higher emission intensities of CdO2 (from Cd(NO3)2·4H2O)

may possibly due to higher defect density in those samples. A visible emission has

been observed in nanocrystalline CdO2. The lifetime of emission transition (438 nm)

with the excitation wavelength (370 nm) is presented in Figure 8.11. The decay time

of the emission transition is 11ns. Our report is the first observation of a visible

emission in CdO2.

Figure 8.10: (A) RTPL spectrum of CdO2 obtained using Cd(NO3)2·4H2O. (B) RTPL spectra

of CdO2 obtained using Cd(CH3COO)2·2H2O, Cd(NO3)2·4H2O and CdSO4·8/3H2O.

Magnetization versus magnetic field plot of CdO2 nanocrystals (from

Cd(CH3COO)2·2H2O) at room temperature (300 K) is reproduced in Figure 8.12. The

sample showed diamagnetic behaviour negating the presence of oxygen vacancies.

Moreover, Vannerberg [8.33-8.35] has reported weak paramagnetic behavior for the

metal peroxide samples that are dried in air at 110-120ºC. The formation of

superoxide, Cd(O2)2 would indeed be consistent with the paramagnetic behavior and

the observed diamagnetic behaviour for our samples suggested that the synthesized

CdO2 is devoid of superoxide.

(A) (B)

Chapter 8

300

Figure 8.11: RTPL decay curve of CdO2 sample obtained using Cd(CH3COO)2·2H2O

recorded at an emission of 438 and excitation of 370 nm.

Figure 8.12: Room temperature M-H curve of CdO2 sample prepared from

Cd(CH3COO)2·2H2O.

Chapter 8

301

CdO + CdTiO3 thin films prepared on glass substrates by the sol-gel method

have been reported to photocatalytically decompose aqueous solutions of methylene

blue (MB) under UV light irradiation [8.36]. Hydrothermally prepared Cd(OH)2

nanowires and CdO nanobelts, CdS nanowires and CdSe nanoparticles, have been

demonstrated to be good photocatalysts for the degradation of organic dyes such as

Safranine T and Pyronine B, under UV light irradiation [8.37]. The order of catalytic

activity is Cd(OH)2 < CdO < CdS < CdSe. It has been reasoned out that higher

photocatalytic activities of these systems are related to the smaller crystallite size of

these cadmium compounds. Earlier, CdO2 nanocrystals have been reported to degrade

methyl orange (MO) dye solution in 5 hours under UV light irradiation [8.19]. The

photocatalytic degradation of MB dye over CdO2 nanocrystals prepared from

Cd(CH3COO)2·2H2O under visible light irradiation has been carried out and the

results are presented in Figure 8.13 (A) and (B). CdO2 nanocrystals decomposed MB

dye solution under visible light in nearly 2 h. For comparison, the adsorption of the

catalyst in the dark and the photolysis of the dye are also shown in Figure 8.13 (A)

and (B). From these data, it is quite clear that the decolorization of the dye has been

accelerated by the CdO2. The pseudo first order rate constant for the photo

degradation reaction is estimated to be 0.59 h- 1. Figure 8.14 shows the N2 adsorption-

desorption isotherms of the CdO2 sample. The surface area of CdO2 prepared from the

Cd(CH3COO)2·2H2O has been estimated to be 52.8 m2/g, which is high, thus favoring

the rapid degradation of MB under visible light irradiation. PXRD pattern of the

catalyst after the photocatalytic reactions showed that the structure of the catalyst is

intact suggesting its reusability (Figure 8.15).

Chapter 8

302

Figure 8.13: (A) Photocatalytic decomposition of methylene blue over CdO2 under visible

light irradiation (a) adsorption experiment (performed in the absence of light) (b) MB

photolysis (c) decrease in concentration of MB with time. (B) Temporal changes in the

maximum absorbance of MB by CdO2 under visible light irradiation.

Figure 8.14: Charts of absorption-desorption of N2 isotherms on the CdO2 sample.

(A) (B)

Chapter 8

303

Figure 8.15: PXRD pattern of CdO2 after photo catalytic decomposition of MB.

The oxidation of benzylic alcohols and benzyl amines using CdO2

nanocrystals as the oxidation catalyst to their corresponding en-one and imines has

been studied. All reactions have been carried out in a round bottom flask equipped

with a magnetic stirrer, schlenk line fitted with water condenser, and a temperature

controller. In a typical reaction, a mixture of CdO2 and the substrate in toluene under

air/O2 atmosphere was stirred at boiling conditions under ambient pressure for the

desired duration. After cooling to room temperature, the products were separated by

extracting with ethyl acetate. Further purification has been carried out by column

chromatography. Under air atmosphere, benzyl amine required higher amount of

catalyst with longer reaction time as compared to the reaction carried out under O2

atmosphere (Table 8.3, entry 1-3). The change of atmosphere from air to O2 did not

Chapter 8

304

Table 8.3: Summary of the substrates used, experimental conditions and moles of catalysts

used for the organic transformation reaction.

Oxidative dimerization of benzyl amines

S. No. Atm. CdO2

(mmol) Substrate Product Time (h) Isolated

Yield (%)

1 Air 15

26 80

2 Air 6

26 37

3 O2 6

11 84

Oxidative dehydrogenation of benzyl alcohols to corresponding carbonyl compounds

4 Air 2.5

12 77

6 O2 2.5

12 78

6 Air 5

19 75

*substrate (5 mmol), toluene as solvent (2.5 ml), temperature (120˚C)

Chapter 8

305

have any remarkable improvement in yields or the reaction time required for the

oxidation of benzylic alcohol (Table 8.3, entry 4-6). Among solvents used (toluene,

water, DMF and acetonitrile), yields in toluene is quite high. Both electron-rich and

electron-deficient substrates with different substituents have been observed to get

easily easily oxidized with yields varying from moderate to excellent. The catalyst

after three reaction cycles for oxidation reactions of benzyl amine has been found to

show an activity similar to that of the fresh ones (Table 8.4).

Table 8.4: Summary of yields obtained after three cycles of oxidation of benzylamine.

Synthesis, characterization and properties of CdO2:F

The following trails were performed to incorporate fluoride in CdO2

nanocrystals. In most of the reactions, KCdF3 is used as the Cd source. Experimental

details such as reagents, molar ratios and reaction conditions are listed below:

Trails using hydrogen peroxide:

HP-1: 0.5 g (2.5 mmol) of KCdF3 was mixed with 50 ml of 30% H2O2

followed by refluxing for 6 h at 120°C. The colour of the suspension turned

off-white after refluxing. After the reaction, the flask was cooled to room

No. of Cycle Isolated Yield

1st Cycle 84

2nd

Cycle 80

3rd

Cycle 77

*substrate (5 mmol), toluene as solvent (2.5 ml), temperature (120˚C)

Chapter 8

306

temperature and the sample was filtered. The powders were dried at room

temperature in air.

HP-2: 5 mmol (1.0 g) of KCdF3 was mixed with 50 ml of 30% H2O2 followed

by refluxing for 6 h at 120°C. A similar observation as in the trail HP-1 was

noticed and the powders were dried after filtration.

HP-3: 1.0 g (5 mmol) of KCdF3 was mixed with 70 ml of 30% H2O2 followed

by refluxing for 6 h at 120°C. A similar observation as in HP-1 was noticed

and the powders were dried after filtration.

HP-4: 0.8 g (4 mmol) of KCdF3 was mixed with 150 ml of 30% H2O2

followed by refluxing for 5 h at 120°C. A similar observation as in HP-1 was

noticed and the powders were dried after filtration.

HP-5: 1.0 g (5 mmol) of KCdF3 was mixed with 50 ml of 30% H2O2 followed

by overnight stirring. The white coloured suspension was dried after filtration.

HP-6: 0.8 g (4 mmol) of KCdF3 was mixed with 75 ml of 30% H2O2 followed

by heating at 120°C in a hydrothermal vessel for 12 h. No solid product was

obtained in this trail.

HP-7: 0.4 g (2 mmol) of KCdF3 was mixed with a mixture of 80 ml of 30%

H2O2 and 20 ml of CH3OH followed by refluxing for 15 h at 120°C. The

powders were dried after filtration.

HP-8: 0.5 g (2.5 mmol) of KCdF3 was mixed with a mixture of 50 ml of 30%

H2O2 and 50 ml of CH3OH followed by refluxing for 15 h at 120°C. The

powders were dried after filtration.

Chapter 8

307

HP-9: 0.2 g (1 mmol) of KCdF3 was mixed with a mixture of 20 ml of 30%

H2O2 and 80 ml of CH3OH followed by refluxing for 15 h at 120°C. The

powders were dried after filtration.

Trails using sodium hydroxide:

SH-1: 1.5 g (7.5 mmol) of KCdF3 was dissolved in 50 ml of d.d. water

followed by the addition of NaOH solution till the completion of precipitation.

The resultant suspension was refluxed for 3 h at 120°C. The powders were

dried after filtration.

SH-2: 1.0 g (5 mmol) of KCdF3 was dissolved in 50 ml of d.d. water and then

dilute NaOH solution was added to increase the pH=7. The resultant

suspension was heated at 120°C in a hydrothermal vessel for 20 h. The

powders were dried after filtration.

Reaction using sodium hypochlorite:

HS-1: 0.8 g (4 mmol) of KCdF3 was mixed with 100 ml of 40 % NaOCl

followed by refluxing for 4 h at 120°C. The powders were dried after

filtration.

Reaction using hydrogen peroxide and ammonium hydroxide:

HA-1: 1.0 g (5 mmol) of KCdF3 was dissolved in 20 ml of NH4OH and to it

20 ml of 30% H2O2 was added. The solution was refluxed for 5 h at 120°C.

The powders obtained after refluxing were filtered and dried.

PXRD patterns of the products obtained from all the above trails are shown in Figure

8.16-8.19. Product from the HP-1 showed CdO2 as the major phase with additional

reflections due to unidentified impurities in its PXRD pattern (Figure 8.16). In the

Chapter 8

308

HP-2 and HP-3 trails, the amount of KCdF3 has been increased as compared to the

HP-1 trail keeping the volume of the H2O2 same. The intensity of the additional

unidentified reflections increased with increase in the amount of KCdF3 in the PXRD

patterns of the products from these trails (Figure 8.17 (a) and (b)). An increase in the

H2O2 amount did not yield CdO2, instead resulted in a product whose PXRD pattern

could not be assigned to any known combination of K, Cd, F, and O (Figure 8.17 (c)).

An X-ray amorphous product resulted when the reaction between KCdF3 and H2O2

was conducted at room temperature (Sample HP-5) (Figure 8.17 (d)). Reactions of

KCdF3 with H2O2 under hydrothermal conditions did not yield any solid product

(Sample code HP-6). A mixture of H2O2 and CH3OH solution in 4:1 ratio when

reacted with KCdF3 yielded CdO2 with unidentified impurities (Sample HP-7; Figure

8.18 (a)). Further reduction in the ratio of H2O2 to CH3OH (1:1; Sample HP-8) yielded

Figure 8.16: PXRD pattern of CdO2 obtained from the trial HP-1. # impurity reflections.

JCPDS File no: 78-1125

Chapter 8

309

Figure 8.17: PXRD patterns of (a) Sample HP-2, (b) Sample HP-3, (c) Sample HP-4 and (d)

Sample HP-5.

Figure 8.18: PXRD patterns of (a) Sample HP-7, (b) Sample HP-8, and (c) Sample HP-9.

Chapter 8

310

Figure 8.19: PXRD patterns of (a) Sample SH-1, (b) Sample SH-2, (c) Sample HS-1 and (d)

Sample HA-1.

a product whose PXRD assignment from the ICSD was not possible (Figure 8.18 (b)).

The starting precursor KCdF3 did not undergo any reaction by taking the H2O2 to

CH3OH in the ratio of 1:4 (Sample HP-9) as revealed by its PXRD pattern (Figure

8.18 (c)). Upon refluxing KCdF3 with NaOH, crystalline Cd(OH)2 resulted (Sample

SH-1; Figure 8.19 (a)). PXRD pattern of the product obtained by subjecting the

mixture of KCdF3 and NaOH under hydrothermal condition at 120ºC for 20 h has

been found unassignable to any known from the ICSD database (Sample SH-2;

Figure 8.19 (b)). A similar conclusion has been arrived for the products of KCdF3

with NaOCl as well as with a mixture of H2O2 and NH4OH (Sample HS-1 and HA-1;

Figure 8.19 (c) and (d)). The summary from all these trails is presented in Table 8.5.

Chapter 8

311

Table 8.5: Summary of reactions of KCdF3 with some oxidizing agents.

Code Experimental Condition Product

HP-1 0.5 g KCdF3 + 50 ml 30% H2O2, refluxed for 6 h CdO2, Minor unidentified

impurity

HP-2 1.0 g KCdF3 + 50 ml 30% H2O2, refluxed for 6 h CdO2, Major unidentified

impurity

HP-3 1.0 g KCdF3 + 70 ml 30% H2O2, refluxed for 6 h CdO2, Major unidentified

impurity

HP-6 0.8 g KCdF3 + 75 ml 30% H2O2, hydrothermally

heated at 120°C for 12 h

No solid product

HP-7 0.4 g KCdF3 + 80 ml 30% H2O2 + 20 ml CH3OH,

refluxed for 15 h

CdO2, Minor unidentified

impurity

HP-9 0.2 g KCdF3 + 20 ml 30% H2O2 + 80 ml CH3OH,

refluxed for 15 h

KCdF3

SH-1 1.5 g of KCdF3 solution + NaOH solution, refluxed

for 3 h

Cd(OH) 2

As we have been successful to dope F- in crystalline CeO2 using

(CH3CH2CH2CH2)4N+F

- (tert-butyl ammonium fluoride) solution in THF (as

described in chapter 6), the same approach has been extended to dope F- in CdO2 as

well. For this, 2.30 g of (10 mmol) of Cd(CH3COO)2·2H2O was dissolved in 50 ml of

diluted NH4OH. To the obtained solution, 10 ml of 30% H2O2 solution was added.

White colored product initially formed was aged at room temperature for 12 h in a

closed flask. The product was separated by filtration and dried at room temperature.

Details of the other trails for the synthesis of CdO2:F using (CH3CH2CH2CH2)4N+F

-

are presented below:

Chapter 8

312

TA-5: To synthesize of 5 mol% fluoride doped in CdO2: 2.18 g (9.5 mmol) of

Cd(CH3COO)2·2H2O was dissolved in 50 ml of diluted NH4OH. To this, 0.5

ml of 1M (CH3CH2CH2CH2)4N+F

- solution in THF was added followed by the

addition of 10 ml of 30% H2O2 solution.

TA-10: 10 mol% fluoride doped CdO2 has been attempted by dissolving 2.07

g (9 mmol) of Cd(CH3COO)2·2H2O in 50 ml of diluted NH4OH followed by

addition of 1 ml of 1M (CH3CH2CH2CH2)4N+F

- solution in THF and 10 ml of

30% H2O2 solution.

TA-20: 20 mol% fluoride doped CdO2 has been attempted by dissolving 1.84

g (8 mmol) of Cd(CH3COO)2·2H2O in 50 ml of diluted NH4OH followed by

addition of 2 ml of 1M (CH3CH2CH2CH2)4N+F

- solution in THF and 10 ml of

30% H2O2 solution.

PXRD patterns of products from the above experiments are shown in Figure 8.20. The

reflections due to cubic CdO2 are observed in the PXRD patterns. Additional

reflections due to CdF2 are not observed in the patterns. The refined lattice parameter

for the samples are a = 5.3637 (5) Å, 5.3645 (3) Å and 5.3648 (6) Å for the

experiments in which 5 mol%, 10 mol% and 20 mol% fluoride doping has been

attempted. The TEM image of cubic CdO2:F (20 mol%) shows aggregation of

nanosized crystals (Figure 8.21 (A)). EDX analysis of the TEM images revealed the

presence of 9.7% of fluoride in addition to cadmium and oxygen (Figure 8.21 (B)).

UV-Visible diffuse reflectance spectrum of CdO2 and CdO2:F nanocrystals are shown

in Figure 8.22 (A). Pure CdO2 showed absorption maxima at around 325 nm in the

UV-Visible diffuse reflectance spectrum and a band gap value of 3.28 eV has been

Chapter 8

313

Figure 8.20: PXRD patterns of the F- (5, 10 and 20 mol%) doped cubic CdO2 nanocrystals.

Figure 8.21: (A) TEM image and (B) TEM-EDX of the cubic CdO2:F (20 mol%).

estimated using the Kubelka-Munk function. On F- doping, the absorption maxima

shifted to 329 nm, 341 nm and 347 nm for the trails with the 5, 10 and 20 mol%

(A) (B)

Chapter 8

314

Figure 8.22: (A) UV-Visible diffuse reflectance spectra and (B) Estimation of band gaps of

CdO2 and CdO2:F nanocrystals.

fluoride doped CdO2 samples. The calculated band gaps were 3.25 eV, 3.16 eV and

3.10 eV for the 5, 10 and 20 mol% fluoride doped CdO2 respectively (Figure 8.22

(B)). No substantial change in the band gap value has been observed in F- doping. The

FT-IR spectra of CdO2 and fluoride doped CdO2 are presented in Figure 8.23. Cd-O

and O-O stretching band observed at 660 cm-1

and 1025 cm-1

remained unchanged in

both the pure and F- doped samples. Additionally, bands at 3430 cm

-1 and 1526 cm

-1

arising from the stretching and bending modes of surface adsorbed water are observed

[8.13]. Presence of trace amount of acetate groups has been suggested by the sharp

band at 1380 cm−1

[8.38]. Raman spectra of CdO2 and CdO2:F nanocrystals are shown

in Figure 8.24. Apart from peaks due to CdO2 (345 and 830 cm-1

), few unidentified

peaks are observed in all of the F- doped samples. These peaks did not match with the

finger print Raman bands of CdF2 [8.39, 8.40].

(A) (B)

Chapter 8

315

Figure 8.23: FTIR spectra of CdO2 and CdO2:F nanocrystals.

Figure 8.24: Raman spectra of the pure and F- (5, 10 and 20 %) doped cubic CdO2

nanocrystals.

Chapter 8

316

Conclusion

CdO2 nanocrystals has been synthesized from Cd(NO3)2·4H2O, CdSO4·8/3H2O and

Cd(CH3COO)2·2H2O by precipitation reaction with H2O2 at room temperature.

Reduction in the band gap from 3.6 eV (bulk) to 2.9-3.1 eV (nanocrystals) has been

observed for the CdO2 both from the UV-Visible diffuse reflectance spectrum as well

as from the photoluminescence measurements. For the first time, phosphorescence in

the visible range has been observed. Also, the effective use of CdO2 as a photocatalyst

for the degradation of the hazardous dye methylene blue has been demonstrated.

Surface area of CdO2 from this procedure is 52.8 m2/g. The CdO2 nanocrystals

showed excellent catalytic activity toward conversion of alcohols to corresponding

en-one and amine to corresponding imine. The catalyst oxidation using CdO2 is a

simple, highly efficient, environmentally benign method and reusable.

(CH3CH2CH2CH2)4N+F

- solution in THF has been employed as a fluoride source for

doping of fluoride in CdO2. PXRD patterns suggested the presence of CdO2 only,

however their Raman showed unidentified bands in addition to the ones due to CdO2.

Chapter 8

317

References

[8.1] Minami, T. Semicond. Sci. Technol. 2005, 20, S35.

[8.2] Su, L. M.; Grote, N.; Schmitt, F. Electron. Lett. 1984, 20, 716.

[8.3] Akhavan, O. ACS Nano 2010, 4, 4174.

[8.4] Roy, S.; Basu, S. Bull. Mater. Sci. 2002, 25, 513.

[8.5] Schechter, D. L.; Kleinberg, J. J. Am. Chem. Soc. 1954, 76, 3297.

[8.6] Hoffman, C. W. W.; Ropp, R. C.; Mooney, R. W. J. Am. Chem. Soc. 1959, 81,

3830.

[8.7] Chen, W.; Lu, Y. H.; Wang, M.; Kroner, L.; Paul, H.; Fecht, H. J.; Bednarcik,

J.; Stahl, K.; Zhang, Z. L.; Wiedwald, U.; Kaiser, U.; Ziemann, P.; Kikegawa, T.; Wu,

C. D.; Jiang, J. Z. J. Phys. Chem. C 2009, 113, 1320.

[8.8] Cheng, S.; Yan, D.; Chen, J. T.; Zhuo, R. F.; Feng, J. J.; Li, H. J.; Feng, H. T.;

Yan, P. X. J. Phys. Chem. C 2009, 113, 13630.

[8.9] Sun, M.; Hao, W.; Wang, C.; Wang, T. Chem. Phys. Lett. 2007, 443, 342.

[8.10] Hsu, C. C.; Wu, N. L. J. Photochem. Photobiol. A 2005, 172, 269.

[8.11] Rosenthal-Toib, L.; Zohar, K.; Alagem, M.; Tsur, Y. Chem. Eng. J. 2008, 136,

425.

[8.12] Yang, L. Y.; Feng, G. P.; Wang, T. X. Mater. Lett. 2010, 64, 1647.

[8.13] Bai, H.; Liu, X. Mater. Lett. 2010, 64, 341.

[8.14] Gondal M. A.; Drmosh Q. A.; Yamani Z. H. Appl. Surf. Sci. 2009, 256, 298.

[8.15] Sebok, D.; Szabo, T.; Dekany, I. Appl. Surf. Sci. 2009, 255, 6953.

[8.16] Zhang, Y. C.; Wu, X.; Hu, X. Y.; Guo, R. J. Cryst. Growth, 2005, 280, 250.

[8.17] Han, X.; Liu, R.; Chen, W.; Xu, Z. Thin Solid Films, 2008, 516, 4025.

Chapter 8

318

[8.18] Liu, Y.; Zhang, Y. C.; Zhang, M. Mater. Lett. 2010, 64, 1779.

[8.19] Liu, Y.; Zhang, Y. C.; Xu, X. F. J. Hazard. Mater. 2009, 163, 1310.

[8.20] Zhang, Y. C.; Wang, G. L.; Hu, X. Y.; Zhou, W. D. J. Cryst. Growth 2005,

285, 600.

[8.21] de León-Gutiérreza, L. R.; Cayente-Romeroa, J. J.; Peza-Tapiaa, J. M.;

Barrera-Calvab, E.; Martínez-Floresb, J. C.; Ortega-Lópeza, M. Mater. Lett. 2006,

60, 3866.

[8.22] Han, X. Liu, R.; Xu, Z.; Chen, W.; Zheng, Y. Electrochem. Commun. 2005, 7,

1195.

[8.23] Zhang, Y. C.; Wang, G. L. Mater. Lett. 2008, 62, 673.

[8.24] Lin, J. M.; Goto, H.; Yamada, M. J. Chromatogr. A 1999, 844, 341.

[8.25] Balbina A.; Johnson, B. A.; Meleney, F. L. Annals of Surgery 1939, 109, 881.

[8.26] Meleney, F. L. Surg. Clin. N. Amer. 1936, 16, 691.

[8.27] Meleney, F. L.; Johnson, B. A. Surgery 1937, 1, 169.

[8.28] Ibarra, L; Alzorriz, M.; Polym. Int. 2000, 49, 115.

[8.29] Ibarra L.; Alzorriz, M. Polym Int. 1999, 48, 580.

[8.30] Nazari A.; Riahi, S. J. Exp. Nanosci., 2012, 7, 491.

[8.31] Kirkeminde, A.; Ren, S. J. Mater. Chem. A, 2013, 1, 49.

[8.32] Popovic, Z. C.; Stanisic, G.; Stojanovic, D.; Kostic, R. Phys. Stat. Sol. b 1991,

165, K109.

[8.33] Vannerberg, N.G. Ark. Kemi. 1959, 49, 57.

[8.34] Vannerberg, N.G. Ark. Kemi. 1959, 14, 119.

[8.35] Vannerberg, N.G. Ark. Kemi. 1956, 10, 455.

Chapter 8

319

[8.36] Mayén-Hernández, S.; Andrea, T. -D.; Gerardo, C. -P.; Rebeca, M. -A.;

Gregorio, Z. -A.; Orlando, J. Adv. Oxidation Technol. 2007, 10, 85.

[8.37] Li, J.; Ni, Y.; Liu, J.; Hong, J. J. Phys. Chem. Solids, 2009, 70, 1285.

[8.38] Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coordination

Compounds, John Wiley & Sons, Inc. (2009).

[8.39] O'Horo, M. P.; White, W. B. Phy. Rev. B 1973, 7, 3748.

[8.40] Sato, T.; Tateya, S. Phys. Stat. Sol. (b) 1982, 113, 291.

PUBLICATIONS

Published: May 03, 2011

r 2011 American Chemical Society 10131 dx.doi.org/10.1021/jp201292d | J. Phys. Chem. C 2011, 115, 10131–10139

ARTICLE

pubs.acs.org/JPCC

Application of KZnF3 as a Single Source Precursor for the Synthesisof Nanocrystals of ZnO2:F and ZnO:F; Synthesis, Characterization,Optical, and Photocatalytic PropertiesShahzad Ahmad,† Mamta Kharkwal,† Govind,‡ and R. Nagarajan†,*†Materials Chemistry Group, Department of Chemistry, University of Delhi, Delhi 110007 India‡Physics of Energy Harvesting, National Physical Laboratory, Dr. K. S. Krishnan Marg, New Delhi 110 012 India

bS Supporting Information

1. INTRODUCTION

In the recent past, anion substitutions have attracted theattention of researchers for engineering the band gap of thematerials suitable for desired applications. Anions such as C, N, S,and other halides have been successfully incorporated in TiO2

and have been demonstrated to produce H2 from water and alsoto decompose harmful organics under visible light irradiation.1

As compared to cation substitutions, influence of anionic sub-stitutions on the structure and properties has not been investigatedextensively due to the difficulty associated with their synthesis.However, it is gaining momentum as shown by the recentdiscovery of novel superconducting materials, La[O1�xFx]FeAs(x = 0.05�0.12) with a maximum Tc of 26 K.

2

Among the attempts to improve the conductivity and opticaltransparency in ZnO, dopants producing highest electronmobilitywould be most preferred. It has been found that anion doping of

F� in ZnO would be more appropriate as compared to group IIIelements at the cation site as they presumably become electricallyactive n-type dopants. Conduction band in ZnO derives mainlyfrom themetal orbital, and therefore dopants which substitute foroxygen would cause perturbation of the valence band leading toless scattering of electrons in the conduction band resulting inhigh electronmobility. Earlier, F� doping in ZnO films have beencarried out by spray pyrolysis using various fluorinating sourcessuch as NH4F, HF, and F2 gas.

3�7

As compared to ZnO, studies on the various properties ofZnO2 are limited partly due to the incomplete structural informa-tion from single crystal X-ray diffraction and the uncertainty in its

Received: February 9, 2011Revised: April 21, 2011

ABSTRACT:Mixed metal fluoride, KZnF3, possessing a cubicperovskite structure has successfully been employed as a singlesource precursor for the synthesis of fluoride-doped ZnO2

nanocrystals by a simple low-temperature oxidation procedure.Utilizing the fact that ZnO2 is a precursor for ZnO, F

�-dopedwurtzite ZnO was readily obtained by a straightforward decom-position procedure. The structure, optical, and photocatalyticproperties of doped ZnO2:F and ZnO:F were studied andcompared with the undoped ones. The preservation of the cubic pyrite structure of ZnO2 by the inclusion of F

�-ions was revealedby the powder X-ray diffraction pattern. Uniform cubemorphology of the nanocrystals of ZnO2:F was observed in both the scanningelectron microscopy and transmission electron microscopy images with the crystallite size of 20 nm. The IR and Ramanspectroscopy analysis implied the absence of any Zn�F direct bonding in ZnO2:F. The high-resolution core level X-rayphotoelectron spectroscopy (XPS) spectrum of F 1s observed at 687.9 eV confirmed the presence of fluoride ions in the ZnO2

lattice. By fitting the core level F 1s spectrum, the concentration of the F� ion was found to be 8.6%. A red shift in the excitonicabsorption was observed on F� doping in ZnO2. A similar trend was also observed in the band edge emission from thephotoluminescence spectrum recorded at 300 K. The intensity of the violet emission in the pure ZnO2 (with a decay time of 18 ns)decreased on F� doping (with a decay time of 13 ns). While ZnO2 nanocrystals efficiently degraded methylene blue (MB) solutionunder UV radiation andmoderately under visible radiation, F�-doped samples showed lesser efficiency for the photo degradation ofthe MB solution. F�-doped ZnO was obtained by decomposing the ZnO2:F in air at 450 �C for 3 h. The symmetry remainedhexagonal on F�-doping as revealed by the powder X-ray diffraction pattern. The intensity of the Raman bands of fluoride-dopedZnO nano powders were in general less as compared to the undoped ZnO, except the one observed at 582 cm�1, which indicated thepresence of higher oxygen vacancies in ZnO:F. Core level XPS measurements provided conclusive evidence for the doping offluorine (6.1%) in ZnO. The band gap value of ZnO:F, estimated from the diffuse reflectance spectrum, was 3.0 eV, and it showedbroad visible emission. As a consequence of higher oxygen vacancies, ZnO:F exhibited efficient photocatalytic activity under visibleirradiation for the degradation of aqueous MB dye solution.

10132 dx.doi.org/10.1021/jp201292d |J. Phys. Chem. C 2011, 115, 10131–10139

The Journal of Physical Chemistry C ARTICLE

exact composition.8�18 However, it is extensively used as theprecursor formaking hexagonal ZnO.9�11,15�18 Therefore, if onecould achieve F�-doped ZnO2, it would pave way for thesynthesis of ZnO:F by a simple decomposition procedure. It isvery important to recognize that anionic substitution of ZnO2 athigh temperatures cannot be performed easily as it is known toliberate oxygen violently on heating at around 230 �C.9 Also, thehazardous nature of fluorine gas makes it extremely difficult toexecute. An ideal approach toward the doping of F� in ZnO2

would essentially require low-temperature oxidation of a singlesource precursor that is air stable. Toward this objective,successful anionic doping by F� in ZnO2 has been performedby the controlled oxidation of the cubic perovskite KZnF3 basedon the following facts; mixed metal complex fluorides arehydroxyphilic due to same size of the fluoride ion and hydroxideion.19Also, the preparation of metal peroxides proceed throughthe formation of metal hydroxides. It is known earlier thatcomplex oxyfluoride nanocrystals, such as YOF, can be preparedfrom single source fluoro carbonate precursors through con-trolled decomposition reactions.20 F�-doped ZnO was obtainedby the controlled decomposition of the F�-doped ZnO2 whichdid not require intricate and sophisticated setup that is essentialfor fluorination reactions and eliminated the use of hazardousand corrosive fluorinating agents.

In this paper, the synthesis and characterization of ZnO2:F andZnO:F from KZnF3 is described. Additionally, pyrite structuredZnO2 nanocrystals and ZnO2:F and ZnO:F systems have alsobeen evaluated as photocatalysts for the degradation of harmfuldye methylene blue (MB) under both UV and visible radiations.

2. EXPERIMENTAL SECTION

2.1. Synthesis. White-colored ZnO2 was prepared fromZnSO4 3 7H2O and Zn(CH3COO)2 3 2H2O.

10,11 To the aqueoussolution of 1 g of ZnSO4.7H2O, dilute NH4OHwas added to geta white-colored colloidal suspension. It was further oxidized with50mLof 30%H2O2 at 70 �C for 1 h.11 Zn(CH3COO)2.2H2O(1 g)was added to a mixture of 50 mL of deionized water and 5 mL of30%H2O2, which resulted in a suspension. It was stirred overnightand then dried at 60 �C.10 KZnF3 was synthesized hydrothermallyfrom Zn(CH3COO)2 3 2H2O and KF.21 F�-ion-doped ZnO2 wasobtained by treating 0.5 mols of KZnF3 in a round-bottom flaskwith 150 mL of 30% H2O2, and the resultant suspension wasrefluxed for 6 h at 120 �C. The suspension turned yellow afterrefluxing, which was cooled to room temperature and filtered. Thepowders were dried at room temperature over the desiccator.2.2. Characterization. The powder X-ray diffraction patterns

of the samples were recorded using Rigaku Miniflex-II X-raydiffractometer employing Cu KR1

radiation over the range of2θ = 5�60�. Thermogravimetric analysis (Perkin-Elmer Dia-mond TG/DTA) was performed from ambient temperature to500 �C at a heating rate of 5 �C min�1 with a sample mass of∼6 mg in an aluminum pan. The transmission electron micro-scopy (TEM) images and energy dispersive X-ray analysis(EDX) measurements were performed on the Philips TecnaiG2 30 transmission electron microscope operating at an accel-erating voltage of 300 kV. The scanning electron microscopy(SEM) micrographs of the samples were recorded on a HitachiS-3700 M microscope. FT-IR spectra of the samples weregathered using Perkin-Elmer FT-IR spectrometer model 2000employing KBr as dispersal medium. Raman spectra of thesamples (in compact form) were obtained using a Renishaw

spectrophotometer equipped with microscope and Arþ laser(λ = 514.5 nm). Diffuse reflectance spectra of the samples werecollected on Perkin-Elmer UV�vis spectrophotometer Lambda-35 attached with an integrating sphere and using BaSO4 as thereference. The photoluminescence (PL) measurements wereexecuted using Horiba Jobin Yuvon Fluoro log modular spectro-fluorometer at room temperature employing CW Xenon lampsource. Photoluminescence decay curves at room temperaturewere collected using Horiba Jobin Yvon fluorohub v (2.0). XPSmeasurements were accomplished with a Perkin-Elmer seriesXPS using an Al KR X-ray line (1486.6 eV) for photoelectronexcitation at a base pressure of 2 � 10�9 Torr.2.3. Photocatalytic Experiments. Photocatalytic studies

were carried out using a 450 W xenon arc lamp (Oriel, Newport,USA) along with a water filter to cut down IR radiation and glasscut off filters, Melles Griot- 03SWP602 to permit only UV light(λ < 400 nm) radiation and Melles Griot-03FCG057 to permitonly visible light (400 nm e λ e 800 nm) radiation as desired.Irradiation was carried out over an external pyrex container with avolume of 250 mL (9.5 cm height and 6 cm diameter), and watercirculation was carried out to avoid any thermal effects. Theappropriate dye solution to be decomposed was taken along withthe required amount of the catalyst in the pyrex container andwas constantly stirred to maintain a homogeneous suspension.The dyes and the catalyst were dissolved in double distilled water.All the experiments were carried out at room temperature.A typical experiment of degradation was carried out as follows:

The catalyst (0.5 g) was added to 150 mL aqueous solution of MBwith an initial concentration of 15� 10�6mol/L forUV and visibleirradiation experiments. Prior to irradiation, the suspension of thecatalyst and dye solutionwas stirred in the dark for 30�60min so asto reach adsorption equilibrium. Five-milliliter aliquots were takenout periodically from the reaction mixture. The solutions werecentrifuged, and the concentration of the solutions was determinedby measuring the maximum absorbance (λmax = 665 nm).

3. RESULTS AND DISCUSSION

3.1. Structural and Morphology Characterization. Thepowder X-ray diffraction pattern of pure ZnO2 prepared fromZn(CH3COO)2 3 2H2O and ZnSO4 3 7H2O are shown in parts aand b of Figure 1, respectively. The positions and intensitiesof the peaks were matching very well with the JCPDS File No.78-1124 reported for ZnO2. The broadness of the peaks sug-gested the low crystallite size of the product with the averagecrystallite size of 20 and 5 nm (as estimated by the Scherreranalysis) from the acetate and sulfate salts of zinc, respectively. Inthe inset of Figure 1, Le-Bail fit of the powder X-ray diffractionpattern of the pure ZnO2 is shown.

22 The lattice constants ofZnO2 derived from the fits were a = 4.870(1) Å and 4.902(1) Åfor the preparations from Zn(CH3COO)2 3 2H2O and ZnSO4 37H2O, respectively. The PXRD pattern of the cubic perovskiteKZnF3 is shown in Figure 2b, confirming its monophasic naturewith lattice constant a = 4.0545(1) Å. The PXRD pattern ofthe oxidized product from KZnF3 is presented in Figure 2c. Theproduct showed peaks whose positions and intensities matchedvery well with the reported ZnO2 confirming the completeoxidation of KZnF3. The average crystallite size as estimatedby the Scherrer analysis was 20 nm with the refined cubic latticeconstant of a = 4.902(1) Å.HRTEM image of pure ZnO2 from Zn(CH3COO)2 3 2H2O is

shown in Figure 3A, and it showed agglomeration due to high

10133 dx.doi.org/10.1021/jp201292d |J. Phys. Chem. C 2011, 115, 10131–10139

The Journal of Physical Chemistry C ARTICLE

surface energy resulting from nanosized crystallites. Homo-geneously dispersed crystals of ZnO2, from the oxidation ofKZnF3, were observed in the SEM image (Figure 3B). Squaremorphology of the cubic nanocrystals were viewed in the TEMimage (Figure 4A). The presence of fluoride in ZnO2 wasconfirmed by SEM-EDX and TEM-EDX analysis on variouslocations of the sample during analysis indicating that uniformdoping across all ZnO2 nanocrystals occurred. To quantify thefluoride ion content, X-ray photoelectron spectrum of ZnO2:F(from the oxidation of KZnF3) was collected. The surveyspectrum is presented in Figure 4B. Peaks at energies of1021.5, 687.9, and 530.2 eV were attributed to Zn 2p3/2, F 1s,and O 1s levels, respectively. Peaks due to potassium were absent

indicating its absence in the final product. A very less intense peakdue to carbon was detected in the XPS measurements, whichmight be due to the contamination of the sample from reactingwith the atmospheric CO2 due to high surface energy of thenanocrystals. Carbon 1s peak was taken as the internal standardfor the samples and the peak positions were normalized withrespect to it. The high resolution core level spectrum of F 1s hasbeen acquired, where the observed peak of F 1s at 687.9 eVsuggested that fluoride ion was present in the ZnO2 lattice. Theabsence of a peak at around 684 eV in the XPS spectrum negatedthe presence of fluoride ion adsorbed on the surface of ZnO2.

23

The shape of the F 1s core peak (inset of Figure 4B) could befitted as the Gaussian distribution. The percentage compositionof each element was calculated using the following relationship

Cx ¼AxSx

∑AxSx

� 100

Figure 3. (A) HRTEM image of ZnO2. (B) SEM image of ZnO2:F.

Figure 1. Powder X-ray diffraction pattern of (a) ZnO2 from Zn-(CH3COO)2 3 2H2O and (b) ZnO2 from ZnSO4 3 7H2O. Inset showsthe Le-Bail fit of ZnO2 from Zn(CH3COO)2 3 2H2O.

Figure 2. Powder X-ray diffraction pattern of (a) ZnO2 from Zn-(CH3COO)2 3 2H2O, (b) KZnF3, and (c) oxidized product fromKZnF3.

10134 dx.doi.org/10.1021/jp201292d |J. Phys. Chem. C 2011, 115, 10131–10139

The Journal of Physical Chemistry C ARTICLE

where Ax is the area under the curve for element x and Sx is thecorresponding sensitivity factor. On the basis of this equation,the concentration of the zinc, oxygen, and fluorine were 36.6,53.8, and 8.6%, respectively. From the XPS analysis, F� dopingfor the peroxide ion in ZnO2 has been established conclusively.The following mechanism involving the sequences of chemical

reactions could be conceived for the formation of inherentlyF�-doped ZnO2

KZnF3 þH2O2 f KZnF3�xðOHÞxKZnF3�xðOHÞx f ZnðOHÞ2 þHFþ KOH

ZnðOHÞ2 þHFþ KOH f ZnO2:F

H2O2, being a powerful oxidizing agent, dissociated KZnF3resulting in Zn(OH)2 and KOH. Zinc hydroxide (Zn(OH)2) onrefluxing at 120 �C, under the highly alkaline conditions providedby the generated KOH, could have resulted in ZnO2. Highlyreactive HF, presumed to be generated in situ, might beresponsible for the doping of some of the hydroxyl groups withF�. The possible evidence for such a mechanism was to analyze

the products after 3 h of the oxidation reaction with H2O2, i.e.,halfway through the reaction. In the powder X-ray diffraction ofsample after 3 h of oxidation, diffraction peaks pertaining toKZnF3, Zn(OH)2, andZnO2 (Figure S1 in Supporting Information)were observed which were quite supportive of the proposedmechanism of formation of F�-doped ZnO2.The FT-IR spectra of ZnO2 and F�-doped ZnO2 are pre-

sented in Figure 5A. The Zn�O stretching band observed at432 cm�1 remained unchanged in both the pure and F�-dopedsamples of ZnO2. The signature band of the O�O stretchingobserved at 1048 cm�1 in the pure ZnO2 has been shifted to1011 cm�1 on F� doping. This could very well be understoodbased on the electronegativity difference between oxygen andfluorine. Also, this result ruled out the direct Zn�F bonding inthe doped sample. The presence of bands in the range3300�3450 cm�1 and 1630�1640 cm�1 were attributed tothe OH stretching and bending modes of water, respectively, and

Figure 4. (A) HRTEM image of ZnO2:F. (B) Survey XPS spectrumof fluoride-doped ZnO2. Inset shows core level spectrum of F 1s.

Figure 5. (A) FTIR spectrum of ZnO2 (dotted) and ZnO2:F(solid).(B) Raman spectrum of ZnO2 (dotted) and ZnO2:F (solid).

10135 dx.doi.org/10.1021/jp201292d |J. Phys. Chem. C 2011, 115, 10131–10139

The Journal of Physical Chemistry C ARTICLE

their presence could very well be rationalized due to the aqueousmedium employed for the synthesis.15

Raman spectra of pure and F�-doped ZnO2 are compared inFigure 5B. Both the samples showed strong band at around830�840 cm�1, which was assigned to the stretching vibration ofO�O bond of the peroxy group.14 F�-ion-doped ZnO2 showedshifting of bands toward lower wave numbers arising due to theelectronegativity difference between oxygen and fluorine. Thecationic field provided by Zn2þ ion in fluoride-ion-doped ZnO2

was comparatively less than in pure ZnO2 because of the samereasons of difference in the electro negativity. Consequently, theO�O vibration shifted to lower frequency as lesser numbers ofelectrons are polarized away from the peroxide ion in F�-dopedZnO2.

24 In other words, it could be recognized that higherthe electropositive character of the central metal ion, thengreater would be the shifting of the position of the peroxideband. The broadening of the Raman band at around 468 cm�1

clearly indicated the presence of oxygen vacancies onF�-doping. Additional weak band of the O�O vibration ofthe peroxide ion was observed at 934 cm�1 in pure ZnO2 and at1557 cm�1 in F�-doped ZnO2 which could be the overtonebands .24,25

The thermogravimetric trace of pure ZnO2 was comparedwith the F�-doped ZnO2 in Figure 6A. Minor weight loss below200 �C, in both the samples, can very well be attributed to theevaporation of water as aqueous medium was employed for thepreparation.8 The mass loss from 40 to 210 �C (corresponded to0.2 mol of water) followed by a single step decompositioncompleting at 260 �C accounted for the liberation of 0.5 molof oxygen. Doped ZnO2 began to decompose at 260 �C con-firming the change in the anion composition, i.e., the substitutionof F� ions for some of O2

2� ions. The initial mass loss until260 �C corresponded to 0.3 mol of water. F�-doped samplesshowed greater stability in terms of sensitivity to air andmoisture as compared to undoped ZnO2. The fluoride contentremained more or less the same even after few weeks. However,care must be exercised to preserve it under inert atmosphere.3.2. Optical Properties. ZnO2 is supposed to be possessing

pyrite-type structures; band structure calculations from the firstprinciples have shown it to be an indirect band gap semiconductorwith a band gap of 2.3 eV.1 However, the calculations based onthe density functional theory have been assessed to under-estimate the band gap value by about 50�100%.8 The real bandgap based on the optical spectroscopy measurements was in therange of 3.3�4.6 eV.9 Excitonic absorption band at 265 nm wasobserved for ZnO2 and using the Kubelka�Munk function, theband gap value was calculated to be 3.7 eV (Figure 6B). On F�

doping, the excitonic absorption band shifted to 275 nm yieldinga band gap of 3.3 eV (Inset of Figure 6B). The reduction in theband gap of F�-ion-doped ZnO2 might be due to the change inelectronic structure caused by the doping. Apart from thefundamental absorption band, doped sample gave an additionalabsorption band in the visible range which could be ascribed tothe F centers.The widespread application of PL spectra to investigate the

efficiency of charge carrier trapping, surface oxygen vacancies,and defects present in the compounds motivated us to study thephoto luminescent behavior of ZnO2 and ZnO2:F. PL spectra ofpure ZnO2 (prepared from Zn(CH3COO)2 3 2H2O and ZnSO4 37H2O) are reproduced in Figure 7A. Emission peaks centered at404, 430, and 454 nm were observed for pure ZnO2 matchingwith the earlier reports.14 ZnO2 prepared from ZnSO4 3 7H2O

showed higher intensity of the bands as compared from to thoseprepared from Zn(CH3COO)2 3 2H2O, which could possibly bedue to reduction in the size of the crystals.26 PL spectrum of pureZnO2 prepared from Zn(CH3COO)2 3 2H2O was comparedwith F�-doped ZnO2 in the inset of Figure 7A since the averagecrystallite size of both being equal. For the F�-doped ZnO2, theemission peaks were Stokes shifted and observed at 412, 434, and460 nm, respectively. The difference in band gap calculated usingthe Kubelka�Munk function and from the position of theemission peak is mainly caused by the Stokes shift due toFranck�Condon effect.27 A red shift of bands similar to thetrend observed in the diffuse reflectance spectra was observed inF�-doped ZnO2 as compared to undoped ZnO2. The intensity ofemission at 404 and 430 nm from F�-doped ZnO2 was higherthan the emissions from the pure ZnO2 occurring at 412 and

Figure 6. (A) TGA trace of pure ZnO2 (dotted) and ZnO2:F (solid).(B) UV�visible diffuse reflectance spectrum of pure ZnO2 (dotted) andZnO2:F (solid). Estimation of band gap of ZnO2 and ZnO2:F samplesare shown in inset.

10136 dx.doi.org/10.1021/jp201292d |J. Phys. Chem. C 2011, 115, 10131–10139

The Journal of Physical Chemistry C ARTICLE

434 nm. The first emission band observed in both the pure andF�-doped ZnO2 corresponded to the band edge. It was con-clusive that the reduction in band gap of ZnO2 occurred uponfluoride doping from the room temperature PL studies as well.The third band observed at around 460 nm was attributed topresence of oxygen vacancies.28 It was obvious from thedecrease in the intensity of the band at around 460 nm in thedoped ZnO2 (as compared to pure ZnO2 prepared from Zn-(CH3COO)2 3 2H2O) that fluoride doping decreased the oxygenvacancies. The phosphorescence in the violet region correspond-ing to the emission at around 430 nm in both the pure andF�-doped ZnO2 was observed with decay time (fitted ex-ponentially) of 18 and 13 ns, respectively.

3.3. Photocatalytic Properties. Semiconductor-based photo-catalysts have a number of applications including generation ofhydrogen from water, environmental remediation and watertreatment, self-cleaning windows and odor control.29 A goodphotocatalytic material must meet the criteria of having theabsorption spectrum overlapping with the solar spectrum, goodredox capability, stability in water, and excellent capacitance of theelectron�hole pair. ZnO2 met the majority of these criteria andtherefore further investigated for the heterogeneous photocatalyticdecomposition of aqueous methylene blue (MB) solution undervisible light (400 nm e λ e 800 nm) as well as the ultravioletregion (λ < 400 nm). Also, the prototype compound FeS2(pyrite) is known to exhibit many interesting electronic proper-ties including efficient solar energy converter in a pyrite p�njunction cell material.30 The complete decolorization of the MBdye occurred in 2 h duration with the ZnO2 nanocrystals undervisible light irradiation as shown in Figure 7B. The absorbancevariations of MB solutions during the photo degradation areprovided in the inset of Figure 7B. For comparison, the adsorp-tion of the catalyst in dark and the photolysis of the dyeare also provided. To distinguish between the photocatalysisand the photolysis of MB, experiments were carried out in thepresence and in the absence of catalyst under visible lightirradiation as the aqueous solution of MB is known to undergoself-photolysis under visible light radiation. The rate of photo-lysis of theMB solution was much lower than the photocatalyticdecomposition of the MB solution by ZnO2 nanocrystals whichclearly confirmed that the decomposition is indeed due to thepresence of ZnO2 as the photocatalyst. The pseudo-first-orderrate constant, using the equation, ln(C0/C) = kt, for the photodegradation reaction was 1.8 � 10�2 min�1. Similar set ofexperiments for the F�-ion-doped ZnO2 nanocrystals showedthat the initial rate of decomposition of MB in the first 30 minwas comparable to that of pure ZnO2 nanocrystals. However,the rate of the degradation turned sluggish after 30 min(Figure 8A). This observation pointed out that the criticalrole of oxygen vacancies for the increased photocata-lytic activity and possible blocking of the active sites of thephotocatalyst by the fluoride ions leading to lower rate ofmineralization of MB.In the photo decomposition of MB over ZnO2 under UV (λ <

400 nm) light radiation, and in the presence of ZnO2 as catalyst,the decolorization was observed within 90 min of irradiation(Figure 8B). The pseudo-first-order rate constant for the photodegradation reaction was 1.9 � 10�2 min�1. Temporal changesin the maximum absorbance of MB by ZnO2:F under UV lightirradiation are also illustrated in the inset of Figure 8B. There wasno considerable decomposition ofMB by doped ZnO2 under UVlight irradiation.3.4. Formation of ZnO:F from ZnO2:F. Heating ZnO2:F at

450 �C for 3 h in air resulted in a yellow-colored solid whosepowder X-ray diffraction pattern confirmed hexagonal ZnO asshown in Figure 9A. Presence of yellow coloration indicated theexistence of nonstoichiometry in ZnO and a possible doping withF� ions. The broader reflections illustrated low crystallite size ofZnO (size of 4 nm from Scherrer analysis). XPS of ZnO:F wasalso collected to quantify the dopant (fluoride ion) concentra-tion. The Gaussian distribution of F 1s core peak is presented ininset of Figure 9A. The F 1s peak was observed at 686 eV for theZnO:F nanopowders. The concentration of the zinc, oxygen, andfluorine were 36.4, 56.5, and 6.1%, respectively. Consequently,this methodology offered a safe and a reliable way of producing

Figure 7. (A) RTPL spectrum of ZnO2 prepared from Zn-(CH3COO)2 3 2H2O (dotted) and ZnSO4 3 7H2O (solid). RTPL spectraof ZnO2 prepared from Zn(CH3COO)2.2H2O (dotted) and ZnO2:F(solid) also shown in inset. (B) Photocatalytic decomposition of MBdye over ZnO2 under visible light irradiation (a) adsorption experi-ment (performed in the absence of light), (b) MB photolysis, (c)decrease in concentration of MB with time. Temporal changes in themaximum absorbance ofMB by ZnO2 under visible light irradiation arealso shown in the inset.

10137 dx.doi.org/10.1021/jp201292d |J. Phys. Chem. C 2011, 115, 10131–10139

The Journal of Physical Chemistry C ARTICLE

F�-doped ZnO nano powders. TEM image of ZnO:F showedagglomeration of the nanocrystals (Figure 9B).UV�visible diffuse reflectance spectrum of F�-doped ZnO is

shown in Figure 10A. The ZnO:F showed an excitonic ab-sorption at 360 nm, which was in good agreement with thepreviously reported work on ZnO:F nanoparticles.4 The absor-bance beyond 400 nm was attributed to the presence of largenumber of defect levels just below the conduction band. ZnO:Falso showed a small and wide absorption band around 650 nm,which has earlier been observed in F�-doped ZnO films3 arisingfrom the transition of oxygen vacancies (inset of Figure 10A).The estimated band gap using the Kubelka�Munk function was3.00 eV (inset of Figure 10B). The reduction in band gap could

arise from the formation of large number of defect levels justbelow the conduction band upon fluoride doping.3

Raman spectrum of F�-doped ZnO is presented in Figure 10B.For the purpose of comparison, Raman spectrum of commercialZnO is also provided in the Figure 10B. The Raman active modesof hexagonal wurtzite structure of ZnO, based on the normal-mode analysis, are E2 (low), E2 (high), A1 (longitudinal optical),A1 (transverse optical), and E1 (transverse optical). The intensityof Raman bands of pure ZnO were higher than that of fluoride-doped ZnO nano powders except one observed at 582 cm�1

indicating higher oxygen vacancies in ZnO:F. Peaks observedat 98 and 332 cm�1 corresponded to E2 (low) mode and second-order Ramanmodes, respectively. The peak centered at 430 cm�1

was the characteristic E2 (high) peak of wurtzite lattice whilethe peak at 582 cm�1 E1 (longitudinal optical) signified thepresence of oxygen vacancies. The broad band that appearedbetween 1000 and 1200 cm�1 were assigned to two phononmodes (two longitudinal optical).31

Figure 8. (A) Photocatalytic decomposition of MB over ZnO2:F undervisible light irradiation (decrease in concentrationofMBwith time).Temporalchanges in the maximum absorbance of MB by ZnO2 under visible lightirradiation shown in the inset. (B) Photocatalytic decomposition of MB overundoped ZnO2 under UV light irradiation in the (a) adsorption experiment(performed in the absence of light) and (b) decrease in concentration of MBwith time. Temporal changes in the maximum absorbance of MB by ZnO2:Funder UV light irradiation are also shown in the inset.

Figure 9. (A) PXRD pattern of hexagonal ZnO:F obtained by heatingZnO2:F at 450 �C in air. The XPS core level spectrum of F 1s of ZnO:F isshown in the inset. (B) HRTEM image of ZnO:F.

10138 dx.doi.org/10.1021/jp201292d |J. Phys. Chem. C 2011, 115, 10131–10139

The Journal of Physical Chemistry C ARTICLE

RTPL spectrum of ZnO:F (shown in Figure 11A) exhibitedbands at 445 and 600 nm when excited with λex 340 nm aswell as 370 nm. The first emission band at 445 nm (2.8 eV)corresponded to the band edge emission. It is noteworthy thatundoped ZnO shows band edge emission in the UV range only.32

The shifting of band edge emission to the visible region clearlyindicated the doping effect of fluoride ions in ZnO lattice. Thestrong emission band at around 600 nm could be related to thepresence of lot of oxygen vacancies created on fluoride ion doping.This strong emission band could arise from the radiative recombi-nation of a delocalized electron close to conduction band with adeeply trapped hole in the Oi center.The heterogeneous photocatalytic decomposition of aqueous

methylene blue (MB) solution under visible light (400 nme λe800 nm) as well as the ultraviolet region (λ < 400 nm) over ZnO:F

was studied. The corresponding photodecomposition plots ofthe MB dye over ZnO:F is shown in Figure 11B. It was clearlyobserved that the ZnO:F was photocatalytically active underboth UV and visible light irradiation for the photodecompositionof MB dye. The dye was decolorized in approximately 2 h underUV and visible light irradiation. The pseudo-first-order rateconstants for the photo degradation reactions were 5.3 � 10�3

and 5.4 � 10�3 min�1 under UV and visible light irradiation,respectively. Although, visible photocatalytic activity in ZnOnano rods has been reported earlier,33 the present study is thefirst of its kind in which the visible light photocatalytic activityinduced upon doping ZnO with F.

4. CONCLUSIONS

From the single source precursor, KZnF3, fluoride-ion-dopedZnO2 nanocrystals possessing pyrite structure were obtained by a

Figure 10. (A) UV�visible diffuse reflectance spectrum of ZnO:F. Anexpanded portion of UV�visible diffuse reflectance spectrum of ZnO:Fin the visible range is shown in the inset. (B) Raman spectrum of pureZnO (dotted) and ZnO:F (solid). Estimation of band gap of ZnO:F isalso shown in the inset.

Figure 11. (A) RTPL spectrum of ZnO:F at λex = 340 nm (dotted) and370 nm (solid). (B) Photocatalytic decomposition MB over ZnO:Funder UV light irradiation (decrease in concentration of MB with time).Temporal changes in the maximum absorbance of MB by ZnO:F undervisible light irradiation is shown in the inset.

10139 dx.doi.org/10.1021/jp201292d |J. Phys. Chem. C 2011, 115, 10131–10139

The Journal of Physical Chemistry C ARTICLE

simple and safe oxidation procedure. Optical band gap value ofZnO2 decreased from 3.7 to 3.3 eV on F� doping. Analysis of IRand Raman spectra confirmed the introduction of F� ions in theperoxide without the presence of any direct Zn�F bonding.F�-ion concentration was estimated to be 8.6% from the XPSmeasurements. In the PL spectrum of ZnO2 at room tempera-ture, violet emission with a decay time of 18 ns was observed inaddition to the bands due to the band-edge and the oxygenvacancies. The decay time of the violet emission was reduced to13 ns on F�-ion doping and the intensity of the emissiondecreased substantially due to the replacement of some of theoxygen with fluorine in the peroxide. ZnO2 degraded theaqueous MB under UV irradiation and a moderate degradationwas observed under visible irradiation. F�-doped hexagonalstructured ZnO was obtained by the simple decomposition ofZnO2:F in air. XPS measurement yielded fluoride concentrationof 6.1% in ZnO. The band gap of the F-doped ZnO was 3.0 eVmuch lower than the undoped ones. ZnO:F showed a strongemission in the visible range on photo excitation. F�-doped ZnOdegraded aqueous MB dye solution efficiently under visible andUV irradiations.

’ASSOCIATED CONTENT

bS Supporting Information. PowderX-ray diffractionpatternof the product after oxidation of KZnF3 for 3 h. This information isavailable free of charge via the Internet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*E-mail: rnagarajan@chemistry.du.ac.in.

’ACKNOWLEDGMENT

The authors sincerely thank and acknowledge DST, Govt ofIndia (Nanomission), for the financial support to carry out thiswork. The authors thank Dr. S. Uma for many useful discus-sions. One of the authors, Shahzad Ahmad, wishes to recordhis sincere thanks to CSIR, India for a JRF fellowship. Theauthors also thank Miss Sudha from DCE (DU), Delhi, forproviding SEM-EDX viewgraphs.

’REFERENCES

(1) Carp, O.; Huisman, C. L.; Reller, A. Prog. Solid. State. Chem. 200432, 33–177.(2) Kamihara, Y.; Watanabe, T.; Hirano, M.; Hosono, H. J. Am.

Chem. Soc. 2008, 130, 3296–3297.(3) Kumar, P. M. R.; Kartha, C. S.; Kumar, K. P. V.; Singh, F.;

Avasthi, D. K. Mater. Sci. Eng., B 2005, 117, 307–312.(4) Gonzalez-Hernandez, R.; Martinez, A. I.; Falcony, C.; Lopez,

A. A.; Pech-Canul, M. I.; Hdz-Garcia, H. M. Mater. Lett. 2010, 64,1493–1495.(5) Sanchez-Juarez, A.; Tiburcio-Silver, A.; Ortiz, A. Sol. Energy

Mater. Sol. Cells 1998, 52, 301–311.(6) Guillen-Santiago, A.; Olvera, M.; de la, L.; Maldonado, A.;

Asomoza, R. Phys. Stat. Sol. (a) 2004, 201, 952–959.(7) Hichou, A. E.; Bougrine, A.; Bubendoroff, J.; Ebothe, J.; Addou,

M.; Troyon, M. Semicond. Sci. Technol. 2002, 17, 607–613.(8) Chen, W.; Lu, Y. H.; Wang, M.; Kroner, L.; Paul, H.; Fecht, H. J.;

Bednarcik, J.; Stahl, K.; Zhang, Z. L.; Wiedwald, U.; Kaiser, U.; Ziemann,P.; Kikegawa, T.; Wu, C. D.; Jiang, J. Z. J. Phys. Chem. C 2009, 113,1320–1324.

(9) Cheng, S.; Yan, D.; Chen, J. T.; Zhuo, R. F.; Feng, J. J.; Li, H. J.;Feng, H. T.; Yan, P. X. J. Phys. Chem. C 2009, 113, 13630–13635.

(10) Sun, M.; Hao, W.; Wang, C.; Wang, T. Chem. Phys. Lett. 2007,443, 342–346.

(11) Hsu, C. C.; Wu, N. L. J. Photochem. Photobiol. A 2005, 172,269–274.

(12) Rosenthal-Toib, L.; Zohar, K.; Alagem,M.; Tsur, Y.Chem. Eng. J.2008, 136, 425–429.

(13) Yang, L. Y.; Feng, G. P.; Wang, T. X. Mater. Lett. 2010, 64,1647–1649.

(14) Bai, H.; Liu, X. Mater. Lett. 2010, 64, 341–343.(15) Gondal, M. A.; Drmosh, Q. A.; Yamani, Z. H. Appl. Surf. Sci.

2009, 256, 298–304.(16) Sebok, D.; Szabo, T.; Dekany, I. Appl. Surf. Sci. 2009, 255,

6953–6962.(17) Zhang, Y. C.; Wu, X.; Hu, X. Y.; Guo, R. J. Cryst. Growth 2005,

280, 250–254.(18) Han, X.; Liu, R.; Chen, W.; Xu, Z. Thin Solid Films 2008,

516, 4025–4029.(19) Machin, D. J.; Martin, R. L.; Nyholm, R. S. J. Chem. Soc. 1963,

148, 1490–1500.(20) Zeng, J. H.; Lou, T. J.; Wang, Y. F.; Guo, J. C.; Di, D.; Ma, R. L.

J. Phys. Chem. C 2008, 113, 597–602.(21) Tyagi, N.; Kumar, P. S.; Nagarajan, R. Chem. Phys. Lett. 2010,

494, 284–286.(22) Le Bail, A.; Duroy, H.; Fourquet, J. L.Mater. Res. Bull. 1988, 23,

447–452.(23) Yu, J. C.; Yu, J.; Ho, W.; Jiang, Z.; Zhang, L. Chem. Mater. 2002,

14, 3808–3816.(24) Eysel, H. H.; Thym, S. Z. Anorg. Allg. Chem. 1975, 411, 97–192.(25) Bates, J. B.; Stidham, H. D. Chem. Phys. Lett. 1976, 37, 20–24.(26) Liqianga, J.; Yichuna, Q.; Baiqia, W.; Shudana, L.; Baojianga, J.;

Libina, Y.; Weia, F.; Hongganga, F.; Jiazhong, S. Sol. Energy Mater. Sol.Cells 2006, 90, 1773–1787.

(27) Rahman, M. M.; Krishna, K. M.; Soga, T.; Jimbo, T.; Umeno,M. J. Phys. Chem. Solids 1999, 60, 201–210.

(28) Li, D.; Ohashi, N.; Hishita, S.; Kolodiazhnyi, T.; Haneda, H.J. Solid State Chem. 2005, 178, 3293–3302.

(29) Fox, M. A.; Dulay, M. Chem. Rev. 1995, 83, 341–357.(30) Birkholz, M.; Fiechter, S.; Hartmann, A.; Tributsch, H. Phys.

Rev. B 1991, 43, 11926–11936.(31) Lo, S. S.; Huang, D. Langmuir 2010, 26, 6762–6766.(32) Wenckstern, H. V.; Schmidt, H.; Brandt,M.; Lajn, A.; Pickenhain,

R.; Lorenz,M.; Grundmann,M.; Hofmann, D.M.; Polity, A.; Meyer, B. K.;Saal, H.; Binnewies, M.; Borger, A.; Becker, K. D.; Tikhomirov, V. A.; Jug,K. Prog. Solid State Chem. 2009, 37, 153–172.

(33) Chu, D.; Masuda, Y.; Ohji, T.; Kato, K. Langmuir 2010, 26,2811–2815.

Hexagonally Ordered KLaF4 Host: Phase-Controlled Synthesis andLuminescence StudiesShahzad Ahmad,† G. Vijaya Prakash,‡ and R. Nagarajan*,†

†Materials Chemistry Group, Department of Chemistry, University of Delhi, Delhi 110007, India‡Nanophotonics Laboratory, Department of Physics, Indian Institute of Technology, Delhi, New Delhi 110016, India

*S Supporting Information

ABSTRACT: Experiments resulting in the successful synthesis ofhexagonally ordered KLaF4 have been described for the first time.Syntheses from three different lanthanum precursors and KF undernonaqueous conditions and at atmospheric pressure are presented. Thetemperature, time of the fluorination reactions, and lanthanum precursorinfluenced the formation of hexagonal KLaF4. While La(OiPr)3 andLa(acac)3 yielded hexagonal KLaF4 by their reaction with KF in methanolat 65 °C, LaCl3 favored only the formation of cubic KLaF4 at 25 °C (roomtemperature). Size-induced phase transformation from cubic KLaF4 to itshexagonal polymorph has been proposed for the reactions involvingLa(acac)3 and La(OiPr)3 and KF. Rietveld refinement of the powder X-raydiffraction pattern of the hexagonally ordered KLaF4 was successfullycarried out in space group P6 2m (No. 189) with lattice constants a =6.5842(3) Å and c = 3.8165(3) Å. A relatively lower effective phonon energy of 262 cm−1 observed for the hexagonally orderedKLaF4 (determined from its Raman spectrum) suggests its potential as a host for optically active elements with the possibility ofminimized nonradiative processes. The hexagonal KLaF4 sample was doped with Er3+ ion (3 mol %) and systematicallyinvestigated by diffuse reflectance, normal emission, and upconversion studies. Strong green emission (4S3/2,

2H11/2 →4I15/2) has

been observed upon 980 and 460 nm excitation. A highly transparent light-emitting polymer [poly(methyl methacrylate)]composite containing hexagonal KLaF4:Er

3+ phosphor has also been effectively demonstrated for many potential applications.

1. INTRODUCTION

Research on the lanthanide-doped upconversion (UC) energysystems continues to be a vibrant and growing interdisciplinaryfield, essentially in two directions.1 One of them is tuning of theoptical properties such as the high UC efficiency and emissionprofile in well-established energy UC systems, adoptingdifferent synthetic strategies, surface modification, and multi-color emission optimization. The other direction has beencontrolling the size, shape, and phase purity of the crystals ofthese systems from the applications point of view. In addition,the search for newer host fluoride lattices that are moreeffective for the UC process is also being pursued with theconcomitant aim of understanding its role.2

The NaLnF4 (Ln3+ = rare earth) system has attracted theattention of researchers as an efficient host matrix for green,red, and blue UC emission through appropriate doping.3 Thechoice of fluorides as the host matrix for the UC studies hasbeen justified because of its low phonon energy (preventingnonradiative relaxation) and multisite character of the hostcrystal lattice, i.e., occupation of the rare-earth active center intwo or more nonequivalent crystallographic sites. Among theavailable synthetic strategies for solids, simple, robust, effectiveand convenient, energy-efficient solution-based techniques arebeing investigated for the synthesis of technologically importantmaterials including nanophosphors.4

The critical role played by the size of the alkali-metal ionsand the rare-earth ions in determining the structure type, latticesymmetry, and stoichiometry was exemplified by the research ofTyagi and co-workers,5 in which Li3ScF6 (with Na2GeF6-typestructure) was readily obtained at atmospheric pressure, whilethe scheelite structure type LiScF4 was realized only at highpressures. Generally, KREF4 has two polymorphs at roomtemperature and ambient pressure: orthorhombic andhexagonal, except in the case of KLaF4 and KCeF4, whichpossess a cubic phase.6 Earlier, our group has reported efficientgreen UC emission by Er3+ doping7 and strong red-emittingEu3+-doped KLaF4,

8 a heavier analogue of the NaYF4 system(crystallizing in the Fm3 m cubic space group) by a green wet-chemical synthetic procedure. Following our studies, Liu et al.9

have recently reported the controlled synthesis and opticalproperties of monodispersed nanoparticles of rare-earth-ion-doped cubic KLaF4. Except for the identification of hexagonalKLaF4 from the phase diagram of the KF−LaF3 system byZachariasen,10 there are no further literature reports on thepossible synthesis of this ordered lattice and it remains achallenge to date.

Received: July 18, 2012Published: November 20, 2012

Article

pubs.acs.org/IC

© 2012 American Chemical Society 12748 dx.doi.org/10.1021/ic301566e | Inorg. Chem. 2012, 51, 12748−12754

In the case of NaYF4, it is observed that hexagonal ordering isthermodynamically more stable than its cubic form. Generally,the hexagonal phase has been stabilized by wet-chemicalmethods, with the hydrothermal method being the mostinvestigated. Stabilization of the hexagonal phase of NaYF4 hasbeen achieved by the addition of excessive amounts of afluoride source, conducting the reactions at high temperaturesand by an aging process.11 Usually, a chelating ligand [e.g.,ethylenediaminetetraacetic acid (EDTA), cetyl trimethylammo-nium bromide, citrate, etc.] is reported to act as a phasecontroller, facilitating the slow and directed growth of thehexagonal phase. Optimization of various reaction parameters,such as the pH of the medium of growth and annealingtemperature, was found to be a reason for the preferentialsynthesis of the hexagonal form without the presence of a cubicphase.12 Recently, Chen et al.13 developed a new strategy ofsynthesizing hexagonal NaYF4 by doping-induced phasetransition via a solvothermal method using Ti4+, Zr4+, andCa2+ ions as dopants. Grzechnik et al.14 studied the effect ofhigh pressures (up to 11 GPa) at ambient temperature as well

as at higher temperatures (up to 1100 °C) on the cubic-to-hexagonal phase transformation of NaYF4.In this paper, detailed investigations into the synthesis of

hexagonally ordered KLaF4 in pure form by the fluorinationreaction of three different precursors of lanthanum and KF in anonaqueous medium and at atmospheric pressures aredescribed. It has been observed that the preferred phaseformation, purity, and size of KLaF4 crystals are sensitive to thestarting lanthanum precursor, reaction time, and temperature.The phonon energy of this host matrix has also beendetermined from the Lorentzian fitting of its Raman spectrumfor the first time. Doping of this host lattice with the opticallyactive Er3+ ion was carried out to functionalize it to be a robustfluorescence host system.

2. EXPERIMENTAL SECTIONA series of experiments were conducted by reacting LaCl3 (SigmaAldrich 99.9%, 0.2452 g, 1 mol), La (acac)3·xH2O (Sigma Aldrich,0.4362 g, 1 mol), and La(OiPr)3 (Sigma Aldrich, 0.3162 g, 1 mol) withKF (Merck, GR 0.2324 g, 4 mol). Methanol was employed as the

Figure 1. (a) Structure of hexagonal KLaF4 using the DIAMOND 3 program. (b) PXRD pattern of the hexagonal KLaF4 sample [observed,calculated (profile matching), and difference profiles given respectively as blue, red, and olive-green lines and Bragg positions as blue vertical lines].

Inorganic Chemistry Article

dx.doi.org/10.1021/ic301566e | Inorg. Chem. 2012, 51, 12748−1275412749

solvent medium for the reactions involving LaCl3 and La(acac)3. Amixed-solvent system of isopropyl alcohol and methanol was used forthe reaction of La(OiPr)3 and KF. Typically, the reactants weredissolved in 20 mL of the solvent. The solution containing thelanthanum ion was added dropwise under constant stirring to a KFsolution. For doping studies, Er(acac)3, prepared from ErCl3·6H2O(Sigma-Aldrich, 99.9%), was used.15

The powder X-ray diffraction (PXRD) patterns were collected usinga Bruker Discover D8 high-resolution diffractometer employing Cu Kαradiation (λ = 1.5418 Å) over the range of 2θ = 10−70°. The structurerefinement of the PXRD data was carried out by the Rietveld methodusing TOPAS3 software.16 The thermogravimetric analysis of thesamples was carried out using Perkin Elmer Diamond TG/DTAsystem from room temperature to 500 °C at a heating rate of 10 °Cmin−1. The high-resolution transmission electron microscopy(HRTEM) images of the samples were collected using a PhilipsTecnai G2 30 transmission electron microscope operating at anaccelerating voltage of 300 kV. A Raman spectrum of the sample, incompact form, was collected using a Renishaw spectrophotometerequipped with a microscope and an Ar+ laser (λ = 785 nm, 10 mW).Diffuse-reflectance spectra of the samples were obtained using aPerkin-Elmer Lambda-35 UV−vis spectrophotometer attached to anintegrating sphere. BaSO4 was used as the reference for thesemeasurements. The conventional excitation and emission spectralmeasurements of the sample [dispersed in water (1 wt %)] was carriedout using Horiba Jobin Yvon Fluorolog modular spectrofluorimeter atroom temperature employing a continuous-wave xenon lamp source.The steady-state and time-resolved UC emission measurements insolid form were performed employing a 980 nm diode laser as theexcitation source using the setup as shown in ref 7. The emission lightwas dispersed into a monochromator (Acton SP 2300) coupled to aphotomultiplier tube (calibrated with mercury emission) through anappropriate lens system. For time-resolved emission, a mechanicalchopper (12 Hz), a lock-in amplifier, and a digital storage oscilloscopewere used to record the transient response.

3. RESULTS AND DISCUSSION3.1. Synthesis, Structure, and Morphology. In order to

establish the optimal conditions for the synthesis of pure

hexagonal KLaF4, a series of reactions in a nonaqueous mediumwere carried out by systematically varying the lanthanumprecursor, utilizing KF as the fluorinating and structure-directing agent. The addition of a lanthanum precursor solution(in methanol or isopropyl alcohol) to KF (in methanol) atroom temperature resulted in a white suspension. Two differentapproaches were followed after this step. In one, the reaction

was continued at room temperature under constant stirring for∼12 h, and in the other, the suspension was aged at 65 °C for∼12 h.The reaction between LaCl3 and KF, driven by salt

elimination, was very fast at room temperature, and thePXRD of the product contained reflections pertaining to KCl,cubic KLaF4, and LaF3. After repeated washing with methanol,the complete removal of KCl occurred, resulting in cubic KLaF4(as the major phase) and LaF3 (as the minor phase). The agingreaction performed at 65 °C for 12 h yielded only LaF3.Relevant data have been provided in Figure S1 in the

Table 1. Crystallographic Data of Hexagonal KLaF4

formula (KLaF4)1.5cryst syst hexagonalspace group P6 2m (No. 189)a (Å) 6.5842(3)c (Å) 3.8165(3)V (Å3) 143.289(2)Z 1ρcalc (g/cm

3) 4.415(5)Rexp (%) 4.38Rp (%) 5.38Rwp (%) 6.67GOF (S) 1.5step size (deg)/step time (s/step) 0.01/1.5no. of data points (2θ = 20−70°) 5000no. of param 21no. of restraints 2temperature (K) 298

Table 2. Refined Positional Parameters after the Final Cycleof Refinementa

atom site SOF x y z Beq

La1 1a 1.0 0.0 0.0 0.0 1.0K1/La2 2d 0.75/0.25 0.3333 0.6667 0.5 1.0F1 3g 1.0 0.6288(2) 0.0 0.0 1.0F2 3f 1.0 0.2152(3) 0.5 0.5 1.0

aRefined unit cell parameters a = 6.5842(3) Å and c = 3.1865(3) Å.

Figure 2. (a) TEM image of the hexagonal KLaF4 sample. (b) HR-TEM image of the hexagonal KLaF4 sample. A SAED pattern of thecrystal is shown in the inset.

Inorganic Chemistry Article

dx.doi.org/10.1021/ic301566e | Inorg. Chem. 2012, 51, 12748−1275412750

Supporting Information (SI). Taking the lead from the fact thatthose ligands capable of expanding the coordination number ofthe metal ions (such as EDTA) favor the formation ofhexagonally ordered NaYF4, two different lanthanum pre-cursors, viz., La(acac)3 and La(OiPr)3, were reacted with KF atroom temperature as well as at 65 °C. The reactions were notas rapid as in the case of LaCl3. The high chelate constants andthe polymeric nature of these precursors might have influencedthe slow formation kinetics of the final products. While theproducts from the room temperature reactions of La(acac)3 andLa(OiPr)3 were identified to be the cubic KLaF4 phase fromtheir PXRD patterns, the aging reactions performed at 65 °Cusing these two lanthanum precursors yielded products whosePXRD patterns matched well with the hexagonal form of KLaF4(JCPDS file no. 75-1927 and Figure S2 in the SI). Thereflections in the PXRD pattern of hexagonal KLaF4 were sharpand intense, implying good crystallinity. No other crystallinebyproducts were observed in these sets of reactions. Thereproducibility as well as the yield of hexagonal KLaF4 from thereaction involving La(OiPr)3 was quite high compared to thatof La(acac)3, indicating the subtle difference in their behavior asprecursors in the fluorination reaction. This might again berelated to the different chelate constants of the ligands as wellas the difference in the vapor pressures of La(acac)3 andLa(OiPr)3 when subjected to aging at 65 °C.It is interesting to analyze these results to find the key factors

responsible for this phase transformation from cubic tohexagonal in KLaF4. First, the observation of LaF3 being theonly crystalline product for the reaction between LaCl3 and KFconducted at 65 °C suggested that the cubic KLaF4 phase was akinetically stable one, undergoing dissociation to thethermodynamically more stable LaF3. Additionally, theformation of cubic KLaF4 at room temperature and hexagonalKLaF4 at 65 °C from the other precursors indicated that thetemperature of the reaction might be one of the factorsinducing the transformation. To verify this, solid cubic KLaF4obtained using La(acac)3 was subjected to thermogravimetric/differential thermal analysis (TG-DTA) in which no exothermicsignal (usually observed for the cubic−hexagonal phasetransition in NaYF4) was observed until 500 °C, signifyingthe absence of a solid−solid phase transition (Figure S3 in the

SI). To ascertain whether the size of the crystallites of theproducts induced the phase transition on aging, the averagecrystallite sizes (estimated by using Scherrer analysis of theirPXRD patterns) of cubic KLaF4 from the reactions of LaCl3,La(acac)3, and La(O

iPr)3 with KF at room temperature (FigureS4 in the SI) and hexagonal KLaF4 from the reactions ofLa(acac)3 and La(O

iPr)3 with KF at 65 °C were compared. Theaverage crystallite sizes of cubic KLaF4 using LaCl3, La(acac)3,and La(OiPr)3 were 16.7, 6.3, and 2.1 nm, respectively. Theaverage crystallite sizes of the hexagonal KLaF4 samples fromthe reactions of La(OiPr)3 and La(acac)3 with KF were 9 and21 nm, respectively. Thus, it is evident that the crystallite sizesof cubic KLaF4 using the lanthanum precursors with chelatinggroups are lower than the one using the lanthanum salt.Because the higher crystallite size of cubic KLaF4 obtained fromLaCl3 did not yield hexagonal KLaF4 and the smaller crystallitesof cubic KLaF4 [obtained from La(acac)3 and La(OiPr)3 atroom temperature] transformed to hexagonal KLaF4 (uponaging at 65 °C), it is proposed that the average crystallite size ofcubic KLaF4 from these reactions is the major factor for

Figure 3. Room temperature Raman spectrum of the hexagonal KLaF4sample.

Figure 4. (a) UC spectra of the hexagonal KLaF4:Er3+ (3 mol %)

sample excited with a 980 nm laser at 150 mW power. UV−visiblediffuse-reflectance spectra of undoped and 3 mol % Er3+-dopedhexagonal KLaF4 and the digital photograph of the green glow UC areshown in the inset. (b) Log−log power dependence of the UCemissions of the hexagonal KLaF4:Er

3+ (3 mol %) sample excited at980 nm.

Inorganic Chemistry Article

dx.doi.org/10.1021/ic301566e | Inorg. Chem. 2012, 51, 12748−1275412751

effecting its phase transformation to hexagonal. Also, the higheraverage crystallite size of hexagonal KLaF4 [9 and 21 nm fromLa(OiPr)3 and La(acac)3, respectively] compared to the loweraverage crystallite size of cubic KLaF4 [2.1 and 6.3 nm fromLa(OiPr)3 and La(acac)3, respectively] clearly indicated that thelarger crystallites have been obtained at the expense of smallerones together with the phase transformation. Such a processhas been explained through the interface nucleation mechanismfor the phase transformation of colloidal In2O3, ZrO2, TiO2,and ZnS nanocrystals.17 Also, the densely aggregated nano-crystals showed higher transformation and growth ratescompared to the loosely aggregated crystals in those systems.17

A comparison of the transmission electron microscopy (TEM)images of cubic and hexagonal KLaF4 obtained from La(acac)3revealed the presence of agglomeration in both samples,suggesting that a similar mechanism might be effecting thephase transformation of the cubic KLaF4 nanocrystals to thehexagonal form (Figure S5 in the SI). It can be further arguedthat the smaller nanocrystallites have higher packing densityrelative to the larger ones, might result in a higher probability ofparticle−particle contact formation, and thereby accelerated thephase transformation rate in KLaF4. The temperature of the

reaction (65 °C) would have contributed further to theincreased probability of nanocrystallites having interaction andrapid hexagonal phase nucleation. It is possible that the largercrystallites (of cubic KLaF4 obtained from LaCl3) might have alow particle−particle contact density, hindering interfacenucleation to transform to a hexagonal polymorph and insteaddissociating to LaF3. Thus, the temperature, fluorinationreaction time, nature of the lanthanum precursor, and size ofthe crystallites (formed initially) influenced the formation ofhexagonal KLaF4.A Diamond18 drawing of the hexagonally ordered KLaF4

lattice is shown in Figure 1a, and the structure consists of anordered array of F− ions with two types of relatively low-symmetry cation sites that are selectively occupied by K+ andLa3+ ions. This arrangement results in significant electron clouddistortion of the cations differing from its cubic polymorph inwhich the high-symmetry cationic sites are randomly occupiedby K+ and La3+ ions.19 It has been well documented that thetransformation from cubic to hexagonal form in these fluoridesystems is of a disorder-to-order character with respect tocations. Hexagonal KLaF4 has been considered to beisomorphous to the β1-K2UF6 structure, with the formulagiven as (KLaF4)1.5.

20 In the structural model employed for theRietveld refinement, the site 2d was considered to be shared by1.5 K+ and 0.5 La3+ ions, while the site 1a was exclusivelyoccupied by La3+ ions. Six F− ions were equally distributed insites 3f and 3g. Assuming spherical particles without strain, thediffractogram was well fit in the hexagonal space group P6 2m(No. 189), with lattice constants of a = 6.5842(3) Å and c =3.8165(3) Å (Figure 1b). The crystal data and structurerefinement parameters of hexagonal KLaF4 are summarized inTable 1. The refined unit cell and position parameters, after thefinal cycle of refinement, are provided in Table 2.Hexagonal KLaF4 obtained from the reaction of La(OiPr)3

and KF also showed agglomeration of crystallites in its TEMimage (Figure 2a). Lattice fringes of the individual crystalliteswere seen in the HRTEM image, indicative of their goodcrystallinity (Figure 2b). The distances between the lattice

Figure 5. (a) Normal excitation (black line) and emission spectra (redline) of the 3 mol % Er3+-doped hexagonal KLaF4 sample. (b) UCluminescence decay curves of the KLaF4:Er

3+ (3 mol %) samplerecorded with an emission of 545 nm. The inset shows a CIE (X, Y)coordinate diagram, with the chromaticity points calculated from theUC and normal emission spectra of the Er3+-doped KLaF4 sample.

Figure 6. Transparent PMMA film containing a 1 wt % hexagonalKLaF4:Er

3+ (3 mol %) film. An intense green emission upon shiningwith white light is shown in the inset.

Inorganic Chemistry Article

dx.doi.org/10.1021/ic301566e | Inorg. Chem. 2012, 51, 12748−1275412752

fringes were measured to be 0.33 nm, corresponding to the dspacing for the (110) lattice plane of the hexagonal KLaF4structure. The selected-area electron diffraction (SAED)pattern of the crystallites (inset of Figure 2b) could be indexedto the (100), (101), (200), (302), and (311) planes of thehexagonal KLaF4 structure.3.2. Optical Properties and UC Luminescence. A room

temperature Raman spectrum of hexagonal KLaF4 [preparedfrom La(OiPr)3 as the precursor] showed four phonon bandscentered at 161, 231, 292, and 383 cm−1 (Figure 3). Theeffective phonon energy was calculated to be 262 cm−1 by theLorentzian fitting procedure,21 which is lower than the onereported for the widely studied NaYF4 system (360 cm−1).22

Other Lorentzian fitted parameters are provided in Table S1 inthe SI. Because lower phonon energies are an important aspectfor minimizing nonradiative (multiphonon) losses as well asincreasing the overall metastable energy lifetime of the UCprocesses, our estimation reinforced the superiority of KLaF4over NaYF4 as the host lattice.Doping of hexagonal KLaF4 with the optically active Er3+

ions was examined to study its solubility and the associatedoptical properties including the UC. The Er3+-doped (3 mol %)sample was prepared under experimental conditions identicalwith those employed for the undoped samples using La(OiPr)3and KF. The close size match between the La3+ (LaVI = 1.17 Å;La VIII = 1.30 Å) and Er3+ (ErVI = 1.03 Å; ErVIII = 1.14 Å) ions23

certainly facilitated its accommodation in a hexagonally orderedfashion. Rietveld refinement of the PXRD pattern of the 3 mol% Er-doped sample is provided in Figure S6 in the SI.The measured UV−visible diffuse-reflectance spectra in the

visible and near-IR regions of the 3 mol % Er3+-doped KLaF4sample are reproduced in Figure 4a. The sharp bands wereassigned to the intraconfigurational f−f transitions from the4I15/2 ground state to the 4I11/2,

4I9/2,4F9/2,

2H11/2,4F7/2, and

4F5/2 excited states.24 Upon excitation of the Er3+-doped (3 mol

%) hexagonal KLaF4 sample with a 980 nm laser operating atthe 150 mW power level, intense green upconverted (UC)emission with predominant emission at 545 nm along with lowintense peaks at 522 and 662 nm was observed (Figure 4a).They were attributed to the transition from the 4S3/2,

2H11/2,and 4F9/2 states to the 4I15/2 ground state, respectively. Thedigital photograph of the intense green UC emission observedfrom the sample is also shown in Figure 4a in the inset. Uponcomparison of the UC emission spectra of cubic KLaF4:Er

3+

with the present data, it is clear that the positions of the greenand red emissions did not change with a change in thesymmetry of the host lattice.7 However, the integralluminescence intensity of the Er3+-doped hexagonal KLaF4enhanced by manifolds is that of the cubic KLaF4:Er

3+ sample.The increase in the intensity could be the result of a differentcrystal-field environment around the Er3+ ions in these two hostlattices. In order to determine the number of photons as well asto understand the mechanism involved in the UC process inthis system, a pumping power (P) dependence of the intensitiesof the UC emission was carried out (Figure S7 in the SI). Thelogarithmic plots of the emission intensity as a function of theexcitation power of a 980 nm laser for the emissions 2H11/2 →4I15/2,

4S3/2 →4I15/2, and

4F9/2 →4I15/2 are shown in Figure 4b.

It is very well-known that, for the UC process, the emissionintensity (I) is proportional to Pn (where n > 1) for one emittedphoton.25 The plot of log I versus log P yields a straight line forthe 2H11/2 → 4I15/2 and 4S3/2 → 4I15/2 transitions with slope

values of n of 1.85 and 1.52, respectively. This implies a two-photon green (545 and 522 nm) emission process initiated atrelatively low excitation powers. The slope for the red (662nm) emission, 4F9/2 → 4I15/2, is 1.01, indicating that energytransfer is the dominant mechanism for the UC process.26

Power dependencies of the UC emissions became nonlinearabove 300 mW, suggesting saturation of the UC processes.25

The normal emission and corresponding excitation spectrafor the 3 mol % Er3+-doped KLaF4 sample dispersed in water(∼1 wt %) are shown in Figure 5a. The excitation spectrummonitored at an emission wavelength of 545 nm consisted oftwo broad excitation lines centered at 366 and 460 nm, whichare attributed to the 4I15/2 → 4G9/2 and 4I15/2 → 4F5/2transitions, respectively.24 The normal emission spectrum ofhexagonal KLaF4:Er

3+ excited at 460 nm followed an intensegreen emission similar to that of its UC emission. Twopredominant peaks centered at 522 and 545 nm were observedto arise out of the transition from the 2H11/2 and

4S3/2 states tothe 4I15/2 ground state, respectively. The CIE1931 XY diagramof the hexagonal KLaF4:Er

3+ (3 mol %) sample is presented inthe inset of Figure 5b. The calculated CIE color coordinates inUC (X = 0.355; Y = 0.623) as well in normal emission (X =0.384; Y = 0.564) shift only slightly and fall well within thegreen region. The UC decay time of the 4S3/2 state of the Er

3+-doped hexagonal KLaF4 sample (Figure 5b) was calculatedfrom a single-exponential fitting as 2.55 ms, suggesting minimalclustering effects.Owing to the smaller crystallite size of the hexagonal

KLaF4:Er3+ sample, it was realized that if they are dispersed in a

polymer matrix, preferably, a transparent one, it will findpotential applications in three-dimensional displays andoptoelectronic and radiation detectors.27 For this purpose, 1wt % of KLaF4:Er (3 mol %) was dispersed in acetone, added tothe monomer solution of methyl methacrylate (MMA) inacetone, and subjected to polymerization using a benzoylperoxide initiator.28 The resultant composite poly(methylmethacrylate) (PMMA) films show intense greenish colorupon shining with white-light radiation (digital picture shownin Figure 6). These preliminary results are to be furtherexplored and are quite promising toward their potential use inaforementioned objectives.

4. CONCLUSIONSInvestigations on the bulk synthesis of the hexagonally orderedKLaF4 by a solution-based method revealed that both thetemperature and ligands attached to the lanthanum precursorinfluenced its formation. While the simple lanthanum salt(LaCl3) did not yield hexagonal KLaF4, La(acac)3 andLa(OiPr)3 yielded hexagonally ordered KLaF4 after aging at65 °C for 12 h. It is reasoned out that the cubic-to-hexagonalphase transformation in KLaF4, from reactions involvingLa(acac)3 and La(OiPr)3 as the reactants, could possibly beinduced by the nanosized crystallites produced in these sets ofreactions via an interface nucleation mechanism. The first timesynthesis of this phase facilitated its structural refinement by theRietveld refinement procedure of its PXRD pattern. The totalconservation of atoms between the reactants and the productsand the nongeneration of hazardous HF during the synthesisare some of the aspects that need to be highlighted from thegreen chemistry point of view. Also, this method does notrequire a calcination step or a postannealing process. Thephonon energy of this hexagonal KLaF4 lattice has beendetermined for the first time and is lower than that of the

Inorganic Chemistry Article

dx.doi.org/10.1021/ic301566e | Inorg. Chem. 2012, 51, 12748−1275412753

widely investigated NaYF4 system. Doping of the hexagonalpolymorph with optically active Er3+ ions illustrated itssuitability as a host lattice for the energy UC process. A two-photon emission process has been demonstrated to result in thestrong green UC in this system, opening a window of additionalhost matrices for various applications. An optically transparentthick film of PMMA containing the upconverting hexagonalKLaF4:Er

3+ phosphor was prepared, suggesting its additionalfuturistic applications.

■ ASSOCIATED CONTENT*S Supporting InformationFigures of PXRD patterns, a TG-DTA trace, TEM images, andUC emission spectra and a table of the Lorentzian peaks fittedparameters of the Raman spectrum of hexagonal KLaF4. Thismaterial is available free of charge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: rnagarajan@chemistry.du.ac.in.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThe authors sincerely thank and acknowledge DU-DSTPURSE Grant and University of Delhi for financial supportto carry out this work. The authors thank Dr. S. Uma for manyuseful discussions and for permitting us to use her DST-fundedfacilities. S.A. expresses his sincere thanks to CSIR, New Delhi,India, for the SRF fellowship.

■ REFERENCES(1) Downing, E.; Hesselink, L.; Ralston, J.; Macfarlane, R. Science1996, 273, 1185−1189. Hinklin, T. R.; Rand, S. C.; Laine, R. M. Adv.Mater. 2008, 20, 1270−1273. de Wild, J.; Meijerink, A.; Rath, J. K.; vanSark, W. G. J. H. M.; Schropp, R. E. I. Energy Environ. Sci. 2011, 4,4835−4848. Trupke, T.; Shalav, A.; Richards, B. S.; Wurfel, P.; Green,M. A. Sol. Energy Mater. Sol. Cells 2006, 90, 3327−3338.(2) Rollet, A.-L.; Allix, M.; Veron, E.; Deschamps, M.; Montouillout,V.; Suchomel, M. R.; Suard, E.; Barre, M.; Ocana, M.; Sadoc, A.;Boucher, F.; Bessada, C.; Massiot, D.; Fayon, F. Inorg. Chem. 2012, 51,2272−2282.(3) Chatterjee, D. K.; Rufaihah, A. J.; Zhang, Y. Biomaterials 2008, 29,937−943. Park, Y.; Kim, J. H.; Lee, K. T.; Jeon, K.-S.; Na, H. B.; Yu, J.H.; Kim, H. M.; Lee, N.; Choi, S. H.; Baik, S.; Kim, H.; Park, S. P.;Park, B.-J.; Kim, Y. W.; Lee, S. H.; Yoon, S.-Y.; Song, I. C.; Moon, W.K.; Suh, Y. D.; Hyeon, T. Adv. Mater. 2009, 21, 4467−4471. Wang,M.; Mi, C.-C.; Wang, W.-X.; Liu, C.-H.; Wu, Y.-F.; Xu, Z.-R.; Mao, C.-B.; Xu, S.-K. ACS Nano 2009, 3, 1580−1586. Boyer, J.-C.; Cuccia, L.A.; Capobianco, J. A. Nano Lett. 2007, 7, 847−852.(4) Wang, D.; Xie, T.; Li, Y. Nano Res. 2009, 2, 30−46. Yin, Y.;Alivisatos, A. P. Nature 2005, 437, 664−670. Murphy, C. J.; Sau, T. K.;Gole, A. M.; Orendorff, C. J.; Gao, J.; Gou, L.; Hunyadi, S. E.; Li, T. J.Phys. Chem. B 2005, 109, 13857−13870.(5) Tyagi, A. K.; Kohler, J.; Balog, P.; Weber, J. J. Solid State Chem.2005, 178, 2620−2625.(6) Du, Y.-P.; Zhang, Y.-W.; Sun, L.-D.; Yan, C.-H. Dalton Trans.2009, 8574−8581.(7) Tyagi, N.; Reddy, A. A.; Nagarajan, R. Opt. Mater. 2010, 33, 42−47.(8) Das, S.; Reddy, A. A.; Ahmad, S.; Nagarajan, R.; Prakash, G. V.Chem. Phys. Lett. 2011, 508, 117−120.(9) Liu, R.; Tu, D.; Liu, Y.; Zhu, H.; Li, R.; Zheng, W.; Ma, E.; Chen,X. Nanoscale 2012, 4, 4485−4491.

(10) Zachariasen, W. H. J. Am. Chem. Soc. 1948, 70, 2147−2151.(11) Sun, Y.; Chen, Y.; Tian, L.; Yu, Y.; Kong, X.; Zhao, J.; Zhang, H.Nanotechnology 2007, 18, 275609 (1−9). Zeng., J.-H.; Su, J.; Li, Z.-H.;Yan, R.-X.; Li, Y.-D. Adv. Mater. 2005, 17, 2119−2123. Liang, X.;Wang, X.; Zhuang, J.; Peng, Q.; Li, Y. Adv. Funct. Mater. 2007, 17,2757−2765. Wang, Z.; Tao, F.; Yao, L.; Cai, W.; Li, X. J. Cryst. Growth2006, 290, 296−300.(12) Ghosh, P.; Patra, A. J. Phys. Chem. C 2008, 112, 3223−3231. Yi,G.; Lu, H.; Zhao, S.; Ge, Y.; Yang, W.; Chen, D.; Guo, L.-H. Nano Lett.2004, 4, 2191−2196.(13) Chen, D.; Huang, P.; Yu, Y.; Huang, F.; Yang, A.; Wang, Y.Chem. Commun. 2011, 47, 5801−5803.(14) Grzechnik, A.; Bouvier, P.; Crichton, W. A.; Farina, L.; Kohler, J.Solid State Sci. 2002, 4, 895−899.(15) Pope, G. W.; Steinbach, J. F.; Wagner, W. F. J. Inorg. Nucl. Chem.1961, 20, 304−313.(16) Rietveld, H. M. J. Appl. Crystallogr. 1969, 2, 65−71. Coelho, A.A. TOPAS, version 3.1; Bruker AXS GmbH: Karlsruhe, Germany,2003.(17) Farvid, S. S.; Radovanovic, P. V. J. Am. Chem. Soc. 2012, 134,7015−7024. Sabyrov, K.; Burrows, N. D.; Penn, R. L. Chem. Mater.2012, DOI: dx.doi.org/10.1021/cm302129a. Zhang, Y. L.; Jin, X. J.;Rong, Y. H.; Hsu, T. Y.; Jiang, D. Y.; Shi, J. L. Mater. Sci. Eng., A 2006,438−440, 399−402. Feigl, C. A.; Barnard, A. S.; Russob, S. P. Phys.Chem. Chem. Phys. 2012, 14, 9871−9879. Huang, F.; Banfield, J. F. J.Am. Chem. Soc. 2005, 127, 4523−4529. Chraska, T.; King, A. H.;Berndt, C. C. Mater. Sci. Eng., A 2000, 286, 169−178.(18) Brandenburg, K.; Putz, H. Diamond, version 3.0; Crystal ImpactGbR: Bonn, Germany 2005.(19) Wang, F.; Han, Y.; Lim, C. S.; Lu, Y.; Wang, J.; Xu, J.; Chen, H.;Zhang, C.; Hong, M.; Liu, X. Nat. Lett. 2010, 463, 1061−1065.(20) Zachariasen, W. H. Acta Crystallogr. 1948, 1, 265−268.(21) Sarakovskis, A.; Grube, J.; Mishnev, A.; Springis, M. Opt. Mater.2009, 31, 1517−1524.(22) Suyver, J. F.; Grimm, J.; van Veen, M. K.; Biner, D.; Kramer, K.W.; Gudel, H. U. J. Lumin. 2006, 117, 1−12. Shan, J.; Uddi, M.; Yao,N.; Ju, Y. Adv. Funct. Mater. 2010, 20, 3530−3537. Renero-Lecuna, C.;Martín-Rodríguez, R.; Valiente, R. Chem. Mater. 2011, 23, 3442−3448.(23) Shannon, R. D. Acta Crystallogr. 1976, 32, 751−767.(24) Reisfeld, R.; Jorgensen, C. K.; Laser and Excited States of RareEarths; Springer-Verlag: Berlin: 1977.(25) Pollnau, M.; Gamelin, D. R.; Luthi, S. R.; Gudel, H. U.; Hehlen,M. P. Phys. Rev. B: Condens. Matter 2000, 61, 3337−3346.(26) Kort, K. R.; Banerjee, S. Inorg. Chem. 2011, 50, 5539−5544.Bednarkiewicz, A.; Wawrzynczyk, D.; Gagor, A.; Kepinski, L.;Kurnatowska, M.; Krajczyk, L.; Nyk, M.; Samoc, M.; Strek, W.Nanotechnology 2012, 23, 145705. Li, Y.; Wang, G.; Pan, K.; Jiang, B.;Tian, C.; Zhou, W.; Fu, H. J. Mater. Chem. 2012, 22, 20381−20386.(27) Li, S.; Toprak, M. S.; Jo, Y. S.; Dobson, J.; Kim, D. K.;Muhammed, M. Adv. Mater. 2007, 19, 4347−4352. Hinklin, T. R.;Rand, S. C.; Laine, R. M. Adv. Mater. 2008, 20, 1270−1273. Downing,E.; Hesselink, L.; Ralston, J.; Macfarlane, R. Science 1996, 273, 1185−1189. Godovsky, D. Y. Adv. Polym. Sci. 2000, 153, 163−205.(28) Brydson, J. A. Plastics Materials; Butterworth-Heinemann:Waltham, MA, 1999.

Inorganic Chemistry Article

dx.doi.org/10.1021/ic301566e | Inorg. Chem. 2012, 51, 12748−1275412754

Synthesis and optical characterization of strong red light emitting KLaF4:Eu3+

nanophosphors

Subrata Das a, A. Amarnath Reddy a, Shahzad Ahmad b, R. Nagarajan b, G. Vijaya Prakash a,⇑a Nanophotonics Laboratory, Department of Physics, Indian Institute of Technology Delhi, New Delhi 110016, Indiab Materials Chemistry Group, Department of Chemistry, University of Delhi, New Delhi 110007, India

a r t i c l e i n f o

Article history:Received 20 February 2011In final form 6 April 2011Available online 9 April 2011

a b s t r a c t

Monophasic KLaF4 possessing cubic symmetry with varied Eu3+ concentrations were synthesized by wet-chemical reaction. The obtained nanophosphor exhibits nanocrystals of 5 nm size and the dopant Eu3+

ions were successfully incorporated into the sites of La3+ ions of the host lattice. The dominant red coloremission at 612 nm from the hypersensitive (5D0 ?

7F2) transition of Eu3+ indicates the inversion anti-symmetry crystal field around Eu3+ ion, which is favorable to improve the red color purity. Furthermore,the emission life times are high enough and our results broadly suggest the potential application forwhite LEDs, mercury-free lamps and display panels.

� 2011 Elsevier B.V. All rights reserved.

1. Introduction

In recent years, a great deal of research effort has been devotedto the synthesis of rare-earth (Re3+) doped nanocrystalline phos-phors due to their novel capabilities resulted from the quantumconfinement effects and a high surface-to-volume ratio comparedto their bulk counterparts [1–5]. Of these, fluoride nanophosphorssuch as YF3 [3], NaYF4 [4,6], GdF3 [7], LiF [8], and LaF3 [9] are of spe-cial interest due to their many interlinked facts that could influ-ence the emissive nature of Re3+ ions. The special class offluorides, Re3+ doped ALnF4 (A = Alkali ion, Ln = rare earth ion)phosphors (for ex: Re3+-doped NaYF4) are attractive, since the hostprovides very low phonon frequencies, optical transparency over awide wavelength range and site-selective doping capability [4–6].Therefore, wide range of applications such as in vivo imaging of tis-sues and cells, solid state light emitting applications, scintillatorsand Thermally Stimulated Luminescence (TSL) dosimetry, havebeen intensely been investigated [6–9]. In general, the emissionmechanism of rare-earth ion is critically dependent on the relativeenergy of the 4f emitting level, site occupation and guest–hostinteractions. So far in ALaF4 (A = Na, K) doped phosphor systems,NaLaF4 nanocrystallines, has been widely studies but relativelyfew studies were reported in KLaF4 nanophosphors [5,10]. Unlikethe other fluorides, KLaF4 usually exists in two phases: metastablecubic phase (a-KLaF4) and more stable hexagonal phase (b-KLaF4),depending on the synthetic conditions [5].

Trivalent europium (Eu3+) ion is widely recognized as an activa-tor for red emission (around 612 nm), which has been used in mostcommercial red phosphors. The intra-4f-shell down-conversion

transitions (5D0 ?7Fj (j = 1, 2, 3, 4)) of Eu3+ ions are strongly

dependent on crystal structure of the host and sensitive to the localenvironment where the rare-earth has been situated [11,12].Among all, the red color transition (5D0 ?

7F2) of Eu3+ is the mostintense and hypersensitive transition therefore the transition prob-abilities are strongly influenced by host lattice, specially the cova-lent nature of host and site symmetry of the occupation [11,12]. Inthis Letter, we report monophasic red emitting Eu3+ doped KLaF4

nanophosphor, which can be effectively excited by UV/blue/greenlights and suitable for use in white LEDs.

2. Experimental

A standard one-step procedure discussed elsewhere was fol-lowed for the preparation of KLaF4 nanophosphor [5,13]. Stoichi-ometric amounts of Potassium Fluoride (KF) and Lanthanum (III)Acetylacetonate (LaC15H21O6�xH2O) and appropriate amount ofEuropium (III) Chloride (EuCl3�6H2O) were individually dissolvedin appropriate amounts of anhydrous methanol and added drop-wise (ca. 15–20 min) with constant stirring. Later, the suspensionwas digested for 1 h after which the product was separated byusing ordinary filtration and dried at room-temperature. X-raydiffraction (XRD) data for all these samples was collected on PANanalytical XPERT-PRO diffractometer with CuKa1 source(k = 1.5405 Å). The steady-state and time-resolved emission mea-surements were carried using home built setups using 532 nmdiode laser as excitation source, wherein the emission light wasdispersed into a monochromator (Acton SP2300) coupled to aphoto multiplier tube (PMT) through appropriate lens system.For time-resolved measurements, mechanical chopper (12 Hz),lock-in amplifier, and digital storage oscilloscope were employed.

0009-2614/$ - see front matter � 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.cplett.2011.04.029

⇑ Corresponding author. Fax: +91(0) 11 2658 1114.E-mail address: prakash@physics.iitd.ac.in (G. Vijaya Prakash).

Chemical Physics Letters 508 (2011) 117–120

Contents lists available at ScienceDirect

Chemical Physics Letters

journal homepage: www.elsevier .com/ locate /cplet t

3. Results and discussions

3.1. XRD and TEM studies

The X-ray diffraction patterns of x mol% (x = 0, 1, 3, 5) Eu3+-doped KLaF4 phosphors are presented in Figure 1A and the allthe obtained patterns are identified as cubic KLaF4 having spacegroup Fm3m (JCPDS File No. 75-2020) [14]. It is known that KLaF4

exhibit two phases, namely, a-phase (cubic) and b-phase (hexago-nal) depending on the synthesis conditions [5]. The a-phase KLaF4

is a metastable high-temperature phase and is isomorphic withthat of CaF2, wherein the K+ and La3+ ions are randomly coordi-nated and each cation is coordinated by F� ion again [5]. As seenfrom Figure 1A, upon Eu3+ ion doping, the diffraction peaks areslightly shifted to the higher side by about 1–2� angles. These shiftsin the XRD peaks are attributed to the substitution of the larger io-nic radius La3+ ions (117 pm) by comparatively smaller ionic radiusEu3+ ions (108 pm) in host lattice [7,15]. This is further indicatingthat Eu3+ ions have been successfully doped into the crystal latticeof KLaF4 host. In addition, the (2 2 0) and (3 1 1) peaks exhibit sig-nificant broadening with enhanced intensity, which may be attrib-uted to the disorder in the (2 2 0) and (3 1 1) sides of KLaF4:Eu3+

[16]. The estimated cubic lattice parameter for the x mol% (x = 0,1, 3, 5) Eu3+-doping were 6.38 (0), 6.26 (1), 5.95 (3), 5.94 (5) Årespectively. The decrease in lattice parameters with the increaseof Eu3+-doping is reasonable, since Eu3+ ions are selectively re-places the sites of much larger ionic radius sites of La3+ ions. The

broad diffraction peaks indicating the decrease in crystal size andthe average crystallite sizes estimated from the Scherrer analysiswere in the range of 5–15 nm. The TEM image (Figure 1B) of5 mol% Eu3+-doped KLaF4 shows that the sample precipitates intoagglomerated nanosized particles of average grain size 5 nm,which is close to the particle size estimated from XRD data.

3.2. Photoluminescence studies

The room temperature PL spectra of x mol% (x = 1, 3, 5, 10) Eu3+-doped KLaF4 nanophosphors under 532 nm laser, is shown in theFigure 2A. The spectra consists of several emission peaks at 579,594, 612, 650 and 701 nm which are corresponds to Eu3+ ion tran-sitions, 5D0 ?

7FJ (J = 0, 1, 2, 3, 4) respectively. Out of all, the hyper-sensitive electric dipole transition at 612 nm (5D0 ?

7F2 of Eu3+)was found to be intense, which is responsible for the bright or-ange-red luminescence and the corresponding intensities showsa systematic enhancement with the increase of Eu3+-doping con-centration. The red color intensity of the phosphor is even visibleto the naked eye (Inset of Figure 2A). The 5D0 ?

7F0 is basically aforbidden transition due to the same J = 0 value. Another intenseline 5D0 ?

7F1 (orange) is a magnetic dipole transition. The hyper-sensitive 5D0 ?

7F2 (at 612 nm) transition is highly sensitive to thesite occupation of Eu3+ ions of inversion or anti-inversion symme-try. The orange color (5D0 ?

7F1) emission dominates when Eu3+

ions occupy the sites of inversion symmetry. However, the lowercrystal symmetry and subsequently the dominant red color

Figure 1. Room temperature (A) X-ray diffraction pattern of x mol% (x = 0, 1, 3, 5) Eu3+-doped KLaF4 nanophosphor. The XRD patterns were indexed according to the JCPDSdata No. 75–2020. (B) TEM image of 5 mol% Eu3+-doped KLaF4 nanophosphor.

Figure 2. Room temperature emission spectra of (A) x mol% (x = 1, 3, 5, 10) Eu3+ doped in KLaF4 nanophosphor (kexe = 532 nm) and (B) 10 mol% Eu3+-doped KLaF4

nanophosphor with different excitation wavelengths.

118 S. Das et al. / Chemical Physics Letters 508 (2011) 117–120

(5D0 ?7F2) purity can be achieved by introducing high degree of

disorder, either by particle size reduction or introducing metallicand nonmetallic elements, as reported by many authors [30–33].The dominant red emission from 5D0 ?

7F2 transition indicatesthe inversion anti symmetry crystal field around Eu3+ ion in thepresent nanophosphor, which is favorable to improve the colorpurity of the red phosphor [16,17]. Moreover, the XRD patternsshow that Eu3+ ions are successfully incorporated into the sites ofLa3+ in KLaF4 lattice, which is further supporting the emissioncharacteristics.

In general, the transition probability of the magnetic-dipoletransition 5D0 ?

7F1 is nearly independent on the host matrixand the electric-dipole allowed 5D0 ?

7F2 transition is stronglyinfluenced by the local structure and site asymmetry around Eu3+

ion [17,18]. Therefore, the emission intensity ratio between redand orange (R/O) color transitions corresponding to 5D0 ?

7F2

and 5D0 ?7F1 is widely known as the asymmetric ratio. The R/O ra-

tios of the present nanophosphors, for different Eu3+-doping con-centration, are calculated from the emission spectra (Figure 2A)and the values are listed in Table 1. The relative R/O ratios as afunction of the Eu3+ content in KLaF4 nanophsohors are nearlysame (1.29–1.38), indicating that the overall Eu3+ local environ-ments are almost same for different concentrations. Therefore, itcan be speculated that the Eu3+ ions are conveniently occupiedthe sites of La3+ in the crystal lattice of KLaF4 [16,19].

To further illustrate the emission characteristics and the excita-tion wavelength dependence, several excitation wavelengths ofcommercially available excitation sources are used (Figure 2B).The emission results show no spectral shift under different excita-tions and gives stable color purity at red-end wavelengths, which isfavorable for LED applications. However, the excitations less than532 nm, the R/O ratios are more than 2 and such values are compa-rable to those reported for oxide phosphors. The R/O ratios of sev-eral reported Eu3+ doped oxide and fluoride based phosphors arealso given for comparison in Table 1. A simple schematic energy le-vel diagram illustrating the excitation and emission transitions ofEu3+ ions is shown in Figure 3A.

The emission intensity decay profiles for the 612 nm (5D0 ?7F2

of Eu3+) emission of the KLaF4:Eu3+ nanophosphors were recordedand the decay curves fits into a single-exponential function I = I0

exp (�t/s) (I0 is the initial emission intensity at t = 0) (see Figure3B). The emission life time values (s) obtained from single expo-nential fits are given in Table 1. The reason for mono-exponentialnature is due to homogeneous distribution of doping ions insidethe host matrix without any cluster formation [20]. The lifetime(s) values increases from 2.90 to 6.90 ms as the Eu3+-concentra-tions increases from 1.0 to 10 mol%. The observed life time valuesfor the present nanophosphors are much higher than the otheroxide phosphors and close to the fluoride phosphors (Table 1).These life time values are also consistent with the analogous phos-phor NaYF4:Eu3+ [23]. While the relative intensities of 5D0 ?

7F2 ofEu3+ emission transition are strongly influenced by their hypersen-sitivity to local environment, the radiative life times are dependenton various factors such as, covenant nature, polarisability, struc-tural defects and lattice arrangement [20,30–32]. Comparativelylarger lifetimes in the present nanophosphors can possibly beattributed to more radiative relaxation caused by surface defects,which can eventually act as luminescent centers. When the surfacearea increases with decrease in particle size, there are more andmore defects which may act as luminescent centers in the sample.Broadly, the emission red-color richness and the larger lifetimesare suitable for potential applications in displays and lights, wherehigh life times are required [21].

4. Conclusion

A strong red-emitting nanophosphor, KLaF4:Eu3+, was synthe-sized from wet-chemical reactions for the first time. These mono-phasic nanophosphors show strong dominating red color richnessand comparatively longer lifetimes when excited by differentwavelengths less than 532 nm. The emission characteristics are

Table 1The emission life times of 612 nm (5D0 ? 7F2 of Eu3+) and Red-to-Orange(5D0 ? 7F2/5D0 ? 7F1 of Eu3+) (R/O) intensity ratios of various Eu3+-doped KLaF4

nanophosphors (kexe = 532 nm). The R/O ratios and emission life times of severalreported Eu3+ doped phosphors are also given for comparison.

Phosphor Life time ‘s’ (ms) R/O ratio Reference

KLaF4:1 mol%Eu3+ 2.86 1.29 Present workKLaF4:3 mol%Eu3+ 4.30 1.33 Present workKLaF4:5 mol%Eu3+ 5.40 1.35 Present workKLaF4:10 mol%Eu3+ 6.90 1.38 Present workGdF3:5 mol%Eu3+ 4.80 1.31* [7]CaF2:1.5 mol%Eu3+ 2.08 1.43* [22]NaYF4:10 mol%Eu3+ 6.20 1.59* [23]LaVO4:20 mol%Eu3+ 0.12 1.43* [24]CaSc2O4:6 mol%Eu3+ 1.00 1.30* [25]CaSiO3:4 mol%Eu3+ 3.30 1.65* [26]Y2SiO5:1 wt%Eu3+ 2.10 2.71 [27]InVO4:30 mol%Eu3+ 0.83 2.33* [28]YVO4:2 mol%Eu3+ 0.54 1.24* [29]YPO4:2 mol%Eu3+ 2.89 1.99* [29]La2O3: Eu3+ 1.38 3.98 [30]

* Estimated R/O values from respective references.

Figure 3. (A) Schematic excitation and emission transition energy level diagram of Eu3+ ion. (B) Representative emission decay curve for 612 nm (5D0 ?7F2 of Eu3+) emission

of 10 mol% Eu3+-doped KLaF4 nanophosphor (kexe = 532 nm).

S. Das et al. / Chemical Physics Letters 508 (2011) 117–120 119

competitive with that of commercial fluoride phosphors such asGdF3:Eu3+ and CaF2:Eu3+. The ability to excite KLaF4:Eu3+ withmany commercially available UV, violet and green excitationsources to generate an intense red emission (612 nm) makes thesephosphors a very promising material for white light LED and otherdisplay applications.

Acknowledgement

Authors acknowledge the financial support from Department ofInformation Technology (DIT), Govt. of India, under PhotonicsDevelopment Program (ref: 12(1)/2008-PDD).

References

[1] S. Lu, J. Zhang, J. Zhang, H. Zhao, Y. Luo, X. Ren, Nanotechnology 21 (2010)365709.

[2] Y. Li, J. Zhang, X. Zhang, Y. Luo, S. Lu, Z. Hao, X. Wang, J. Phys. Chem. C 113(2009) 17705.

[3] D. Chen, Y. Wang, Y. Yu, P. Huang, Appl. Phys. Lett. 91 (2007) 051920.[4] G. Wang, Q. Peng, J. Solid State. Chem. 184 (2011) 59.[5] N. Tyagi, A. Amarnath Reddy, R. Nagarajan, Opt. Mater. 33 (2010) 42.[6] G.S. Yi, H.C. Lu, S.Y. Zhao, Y. Ge, W.J. Yang, D.P. Chen, L.H. Guo, Nano Lett. 4

(2004) 2191.[7] X. Zhang, T. Hayakawa, M. Nogami, Y. Ishikawa, J. Alloys Compd. 509 (2011)

2076.[8] C. Adachi, M.A. Baldo, M.E. Thompson, S.R. Forrest, J. Appl. Phys. 90 (2001)

5048.

[9] S.N. Achary, A.K. Tyagi, T.K. Seshagiri, V.N. Natarajan, Mater. Sci. Eng. B 129(2006) 256.

[10] Z. Wang, C. Liu, Y. Wang, Z. Li, J. Alloys Compd. 509 (2011) 1964.[11] Y. Huang, L. Shi, E.S. Kim, H.J. Seo, J. Appl. Phys. 105 (2009) 013512.[12] Q. Ma, Y. Zhou, A. Zhang, M. Lu, G. Zhou, C. Li, Solid State Sci. 11 (2009) 1124.[13] G.W. Pope, J.F. Steinbach, W.F. Wagner, J. Inorg. Nucl. Chem. 20 (1961) 304.[14] W.H. Zachariasen, Acta Crystallogr. 2 (1949) 388.[15] A. Sarakovskis, J. Grube, A. Mishnev, M. Springis, Opt. Mater. 31 (2009) 1517.[16] M. Zhong, G. Shan, Y. Li, G. Wang, Y. Liu, Mater. Chem. Phys. 106 (2007) 305.[17] L. Zhou, J. Wei, J. Wu, F. Gong, L. Yi, J. Huang, J. Alloys Compd. 476 (2009) 390.[18] B.V. Rao, K. Jang, H.S. Lee, S.S. Yi, J.H. Jeong, J. Alloys Compd. 496 (2010) 251.[19] G. Vijaya Prakash, R. Jagannathan, Spectrochim. Acta A 55 (1999) 1799.[20] N.S. Singh, R.S. Ningthoujam, M.N. Luwang, S.D. Singh, R.K. Vatsa, Chem. Phys.

Lett. 480 (2009) 237.[21] C. Qin, Y. Huang, G. Chen, L. Shi, X. Qiao, J. Gan, H.J. Seo, Mater. Lett. 63 (2009)

1162.[22] L. Song, J. Gao, R. Song, J. Lumin. 130 (2010) 1179.[23] G. Jia, P.A. Tanner, J. Alloys Compd. 471 (2009) 557.[24] S.W. Park et al., Phys. B 405 (2010) 4040.[25] Z. Hao, J. Zhang, X. Zhang, X. Wang, Opt. Mater. 33 (2011) 355.[26] H. Nagabhushana, B.M. Nagabhushana, M. Madesh Kumar, Chikkahanu

mantharayappa, K.V.R. Murthy, C. Shivakumara, R.P.S. Chakradhar,Spectrochim. Acta A 78 (2011) 64.

[27] N. Rakov, D.F. Amaral, R.B. Guimarães, G.S. Maciel, J. Appl. Phys. 108 (2010)073501.

[28] Y.S. Chang, Z.R. Shi, Y.Y. Tsai, S. Wu, H.L. Chen, Opt. Mater. 33 (2011) 375.[29] G. Pan et al., J. Appl. Phys. 104 (2008) 084910.[30] L. Yu, H. Song, Z. Liu, L. Yang, S. Lu, Phys. Chem. Chem. Phys. 8 (2008) 303.[31] X.C. Jiang, C.H. Yan, L.D. Sun, Z.G. Wei, C.S. Liao, J. Solid State. Chem. 175 (2003)

245.[32] L. Wang, Y. Wang, J. Mater. Sci. 43 (2008) 2908.[33] H.Q. Liu, L.L. Wang, S.Q. Chen, B. Zou, J. Lumin. 126 (2007) 459.

120 S. Das et al. / Chemical Physics Letters 508 (2011) 117–120

Indian Journal of Chemistry Vol. 51A, Jan-Feb 2012, pp. 145-154

Anion doped binary oxides, SnO2, TiO2 and ZnO: Fabrication procedures, fascinating properties and future prospects

R Nagarajan*, Vinod Kumar & Shahzad Ahmad

Materials Chemistry Group, Department of Chemistry, University of Delhi, Delhi 110 007, India Email: rnagarajan@chemistry.du.ac.in

Received 14 November 2011

Doping the oxygen in the three technologically important binary oxides SnO2, TiO2, and ZnO, with other anions such as nitrogen, carbon, fluorine, sulfur and chlorine by various synthetic procedures are described. The crystal structures of these oxides along with their electronic structures are summarized. The evolution of many useful properties, such as efficient photocatalysis, high electrical conductivity with high optical transparency on doping with some of these anions, is discussed. An important milestone achieved by our research group for the synthesis of F-doped SnO2 and ZnO powders is highlighted. Heavily F-doped SnO2, obtained by the safe, simple, reliable and reproducible synthetic approach developed by us is the first example among oxide semiconductors to show Moss-Burstein effect, and consequently the defect states are produced in the system on injection of extra charge carriers. These trapped defect states give rise to glow curves in the thermoluminescence spectrum. This observation suggests the use of SnO2:F as a thermal UV sensors in high radiation environments. In the concluding part, the need for the investigation in to the interesting structural, electronic and optical properties, especially in the heavily doped regime, are portrayed as part of the future directions of research in these oxides.

Keywords: Oxides, Binary oxides, Doping, Anion doping, Tin dioxide, Titania, Zinc oxide, Photocatalysis, Sensors, Electrical conductivity

Intentional doping into a crystal with impurity atoms has revolutionized the semiconductor industry due to their wide spread applications in electronic and electro-optic components1. Though cationic doping into the lattice for tailoring the band gap of materials for the desired applications has been studied extensively and is reasonably understood, the mechanism of the evolution of many interesting properties arising from the anionic doping is debatable and requires more studies to achieve clarity. However, in both the scenario, many interesting electronic and optical properties, under the heavily doped regime, are yet to be fully explored and understood. When employing dopants to change the optical response of a material, it is desirable to maintain the integrity of the crystal structure of the host material while changing its electronic structure. The crystal structure of a material is directly related to the ratio of cation and anion size in the crystal lattice2. It appears to be relatively easier to achieve cationic substitution in a controlled fashion than to substitute one anion for the other due to the difference in the charge states and ionic radii.

One of the main reasons for fewer studies on anion doped semiconductors is due to their synthetic challenge. This review focuses on the status of an important and experimentally challenging area of the

synthesis/fabrication of anion doped binary oxides SnO2, TiO2 and ZnO. The choice of these three oxides is based on the very high consumer demand projected due to their ecofriendly nature with applications in various fields, such as photovoltaics, photocatalysis, self-cleaning coatings and flat panel display devices3-14. There exists an obvious heterogeneity in the method of synthesis and hence, the results from the reports vary. They differ greatly in the methods to introduce the dopant, which range from chemical methods such as sol-gel reactions, electrochemical doping, and oxidation of nitride (in the case of titanium) to physical methods, such as magnetron co-sputtering and ion implantation. Therefore, a review of the current status describing the synthesis and properties and the unsolved issues regarding the anion doping in the oxides, SnO2, TiO2 and ZnO is essential. This review is organized in the following way. Fabricating the anion doped SnO2, TiO2 and ZnO by different synthetic methodologies is given primary emphasis, followed by the description of fascinating properties exhibited by the anion doped systems. Finally, the challenges faced by the scientists, especially synthetic chemists in particular, are highlighted to provide an incentive to other scientists to study the effect of doping on the structure and the properties further.

INDIAN J CHEM, SEC A, JAN-FEB 2012

146

Crystal Structures and Electronic Properties of SnO2, TiO2 and ZnO

SnO2 and TiO2 crystallize in rutile structure, but TiO2 is also found in anatase form2; their structures obtained using Diamond software is shown in Fig. 1(a & b). Rutile structure of SnO2 and TiO2 possess tetragonal symmetry, space group: P42/mnm, with a = 4.755 Å, c = 3.199 Å (SnO2) and a = 4.592 Å, c = 2.957 Å (TiO2)2. The anatase form of TiO2 also possesses tetragonal symmetry, but crystallizes in a different space group of I41/amd with a = 3.782 Å, c = 9.502 Å. In the crystal structure of SnO2 and TiO2, slightly distorted octahedral coordination around the central metal ion by the six oxygen ions is present. In anatase, the corner-sharing octahedra form (001) planes the edges of which are connected with the plane of octahedral units below. In both type of structures, the stacking of the octahedral units, result in threefold coordinated oxygen atoms. The wurtzite structure of ZnO has a hexagonal unit cell with c/a = 1.633 crystallizing in P63mc space group2. (Fig. 1c). The structure is composed of two interpenetrating hexagonal close packed (hcp) sub lattices, each of which consists of one type of atom displaced with respect to each other along the threefold c-axis in fractional coordinates.

As a direct band gap n-type semiconductor (Eg = 3.6 eV), SnO2 is a promising material for a variety of applications. The valence bands of the stoichiometric SnO2 originate mainly from the O 2p state with mixing of Sn 5s and 5p states and these levels are responsible for the cohesion, while the conduction band contains mainly 5s orbitals. The origin of the n-type behavior is the native non-stoichiometry caused by the oxygen vacancies. The electronic structure of TiO2 (Eg = 3.2 eV) is governed by a strong hybridization between the Ti 3d and the O 2p states around the Fermi level. The upper valence band is dominated by O 2p levels and the lower conduction band is formed mainly by Ti 3d states. ZnO is a direct energy band gap (3.37 eV) semiconductor in which the conduction band derives mainly from the zinc orbitals13.

Fabrication Procedures and Properties of Anion Doped SnO2, TiO2, and ZnO

Anion doping in SnO2

Fluoride appears to be the most favored dopant and substituent in SnO2 due to the following: (a) its ionic size (F-: 0.133 nm) very closely matches with that of the oxide ion (O2-: 0.132 nm), (b) the energy of the Sn−F bond (~ 26.75 D°/kJ mol-1) is comparable to

that of the Sn−O bond (~ 31.05 D°/kJ mol-1), and, (c) since the charge on the fluoride ion is only half that of the charge on the oxide ion, the Coulomb forces that bind the lattice together are reduced. Among the transparent conducting oxides (TCO), F-doped SnO2 (FTO) tops the list due to their high chemical stability, mechanical strength, and thermal resistance, as well as their low cost4. Their favorable visible transmittance and infrared reflectance in

Fig. 1 – Structural arrangement of (a) rutile SnO2, (b) anatase TiO2 and (c) wurtzite ZnO.

NAGARAJAN et al.: FABRICATION & PROPERTIES OF ANION DOPED BINARY OXIDES OF Sn, Ti & Zn

147

comparison with tin oxide films doped with other impurities make them ideal candidates for preparing low emissivity coating glass (Low-E glass)4,5. FTO has been proposed to be an alternative to ITO due to its cost effective value in the fabrication of organic photovoltaic devices7-9.

F-doping in SnO2 in the powder and thin film forms has been achieved by many synthetic techniques and are summarized in Table 1, along with the significant outcome of such methods. FTO films have been fabricated by chemical vapor deposition techniques15, dip coating technique16, spray pyrolysis17-22, ultrasonication spray pyrolysis23, reactive RF sputtering24, and pulsed laser deposition methods25. Bae et al.26 grew SnO2:F films using a low-pressure metallorganic chemical vapor deposition (LP-MOCVD) with tetramethyltin. FTO films were also obtained using the organo-tin precursor tert-amyloxyfluorodipentan-2,4-dionatotin (IV) complex27. By the atmospheric pressure chemical vapor deposition (APCVD) procedure, Zhao et al.28

deposited FTO films with monobutyl tin trichloride (C4H9SnCl3) and trifluoroacetic acid (CF3COOH) as the sources of tin and fluorine. Similarly, highly conducting FTO films were prepared by Talaty et al.15 employing the single-source precursor (SSP), [SnCl4{OC(H)OC2H5}2] in a APCVD system. Pyrosol method followed by ultrasonication has been applied

to fabricate FTO films29. In dual-source CVD procedures, a tin precursor (e.g. Me4Sn, SnCl4) along with either F2, NH4F, HF, BrCF3, CF3COOH or other fluorocarbons, as the fluorine sources were used to achieve F-doped SnO2

30. Fabrication of SnO2:F films from a soluble and volatile SSP were less common since compounds/complexes with a direct Sn–F bond were found to be insufficiently volatile, as such species commonly generated bridged oligomers/ polymers through F: →Sn interactions, e.g. Ph3SnF 31. This aspect was overcome using Sn(O2CCF3)2

32 and Bu2Sn(O2CCF3)2

33. Purely inorganic complexes Sn[OCH(CF3)2]4·2(HNMe2) and Sn[OCH(CF3)2]2·L (L=HNMe2, C5H5N) were also employed successfully for making FTO films33. The Sn(IV) containing species, when used for making films, generated FTO with good optical transparency (>85 %) but high resistivity (2.1 × 10−3 ohm cm), while the divalent tin analogue, Sn[OCH(CF3)2]2·HNMe2 afforded non-conductive SnO0.9−1.3F0.1−0.4, suggesting that hydrolysis, rather than oxidation, was driving the film deposition; both processes required LPCVD due to limited precursor volatility. Organotin fluoroalkoxides perform poorly in comparison with organotin fluoroalkyls (R3SnRf) as precursors for SnO2:F synthesis34. Noteworthy is the use of organotin fluorocarboxylates (R3SnO2CRf), as they are air stable, cheap and easy to handle35.

Table 1 – Summary of the synthetic conditions and methodology adopted for the successful F-doping in SnO2

Source employed Methodology Dopant Remarks Ref.

C4H2F6O2/CF3COOH/ SF6/ Bu3SnORf/ Et3SnO2CC2F5

APCVD (thin films)

1.02-4.80 at.% Intricate synthesis set-up, poisonous gaseous and presence of carbon as impurities

15, 24, 28, 35, 34

HF CVD (thin films) 4.5 at.% Low sheet resistance and high optical transparency

16, 26

NH4F SP (thin films) ~30 wt.% Low resistivity 21, 22, 29 SnO2+SnF2 Laser deposition

(thin film) 5–20 % Optical transmittance of 87 % 25

DFM (CH2F2) Vapor phase synthesis (powder)

27.1 at.% Sn(CH3)4 as a precursor and hazardous F source 38

SnF2(R1COCHCOR2)2/ (C4H9-C≡C)3-SnRF/ SnF(OtAm)(acac)2

Sol-gel and calcination (powder)

3.0-8.0 at.% Much lower resistivity than ITO, carbon as impurities

39,40

HF Thermal decomposition (powder)

<10 mol% Use of organic Sn precursor, Sn2(C6O7H4).H2O 41

N(CH3)4F.4H2O Microwave polyol approach (powder)

1-10 mol%. Defect sites studied by solid state NMR 42

KSnF3 Oxidation (powder) 21.23 % (63.81 at.%)

Air stable single source inorganic precursor, heavily doped, high surface area & photocatalytic, thermoluminescence

43

INDIAN J CHEM, SEC A, JAN-FEB 2012

148

For the preparation of the SnO2:F powders, the sol-gel method (employing organometallic precursors) and spray pyrolysis techniques are reported. Ha et al.3 obtained SnO2:F by the slow hydrolysis of the single molecular precursor, fluoro-(2-methylbutan-2-oxy)di(pentan-2, 4-dionato)- tin complex. The sol-gel process was coupled with the hydrothermal treatment to obtain SnO2:F nanocrystals by Wu et al.36 Han and co-workers37 prepared FTO nano-powders by combining the sol-gel and the combustion methods with acetylene black as fuel. Suffner et al.38 doped fluorine in SnO2 by passing difluoromethane (DFM) over SnO2. Toupance and co-workers39,40 applied sol-gel routes derived from organo-tin complexes containing fluoride and by which ~3 mol % fluoride was found to be doped in SnO2. A similar approach was also used by Han and co-workers6. Esteves et al.41 prepared F-doped SnO2 powders by Pechini’s process using tin citrate, Sn2(C6O7H4).H2O and ethylene glycol. SnO2:F nanoparticles were also fabricated by a microwave assisted polyol approach42 from SnCl4 and N(CH3)4F.

Synthetic procedures that are known to yield F-doped SnO2 powders employed a tin source (SnCl2 or SnCl4) and the fluorine source (NH4F, CH2F2, SF6 and HF). Whenever organotin complexes were used as SSP, heat treatment after the reactions to get rid of the volatile organic impurities and to improve crystalline nature of the final product was found to be essential. Total elimination of carbon impurities was difficult, thus making it hard to understand the effect of F-doping on the properties of SnO2. In any case, doping higher concentration of fluoride ions in SnO2 was not easily achievable in the powder form; in many instances, it was either restricted by the fluoride concentration in the precursor or the contamination of the final product due to the incomplete removal of the fluorinating agents.

We have reported the successful synthesis of fluoride-doped SnO2 by a novel oxidation procedure in which the inorganic fluoride complex of Sn2+, KSnF3, was employed as the SSP43. KSnF3 is air stable, easy to prepare and easy to handle. Its controlled oxidation with H2O2 was carried out at 100 ºC to obtain F-doped SnO2. The choice of this precursor was based on the fact that the mixed metal complex fluorides are highly hydroxyphilic due to same size of the fluoride and hydroxide ions. Also, SnO2 could very easily be obtained from the hydroxides of tin. More importantly, the reactions

were conducted at low temperatures which did not require any intricate set up or handling of hazardous fluorinating agents. The use of inorganic precursor prevented inclusion of carbon in the product, which is usually the contaminant starting with an organotin precursor. Core level X-ray photoelectron spectroscopy (XPS) measurements provided conclusive evidence for the heavy doping of fluorine in SnO2 (21.3 %), the highest concentration achieved in powders till now. SnO2:F nanocrystals, obtained by our method, showed greater thermal stability up to 300ºC (without the evolution of the hazardous HF) as revealed by the hyphenated TG-MS techniques. These results will find greater applications for the fabrication of large scale coatings of heavily F-doped SnO2 by spray pyrolysis and dip coating or electrophoretic thin film deposition procedures.

Enormous blue shift in the band gap was observed in our F-doped SnO2 sample which was related to the heavily doped situation due to the Moss-Burstein effect44 with increase in the carrier concentration. This is the first time that this effect has been demonstrated in an oxide semiconductor. Evaluation of SnO2:F as a photo catalyst for the degradation of aqueous Rhodamine B (Rh B) dye solution showed efficiency matching with the commercial Degussa P-2543. Without external irradiation, the heavily F-doped SnO2 nano crystals showed a broad glow curve at 454 K in the thermoluminescence spectrum, signifying the presence of F-centres (Fig. 2). On irradiating the SnO2:F samples with UV light, the glow curve resolved into two, one centred at 440 K and the other at 520 K (Inset of Fig. 2). The results from this study indicated its use as potential thermal and UV sensors.

Fig. 2 – Thermoluminescence glow curve of (a) SnO2:F obtained by the low temperature oxidation of KSnF3 and (b) undoped SnO2. Inset shows TL glow curve of SnO2:F recorded at 3 K/s after irradiating with UV light for 10 minutes.

NAGARAJAN et al.: FABRICATION & PROPERTIES OF ANION DOPED BINARY OXIDES OF Sn, Ti & Zn

149

Though considerable literature on the effects of other dopants such as N and S in SnO2 exist, very few reports are available on the possible N-doping of SnO2. Undoped SnO2 is an n-type semiconductor (due to the existence of intrinsic defects) while p-type SnO2 films have been reported by Al or In doping45. Yan et al.46 suggested that p-type SnO2 can also be achieved by nitrogen doping. Pan et al.47 have grown nitrogen-incorporated SnO2 thin films on Si (100) and quartz substrates by reactive sputtering of a Sn target in gaseous mixture of N2–O2. Anion doping in TiO2

First reports of anion-doped TiO2 began to appear in the early 1990s48, but the study of Asahi and co-workers49 in 2001 on the N-doped TiO2 exhibiting excellent visible light photocatalytic activity turned the researchers’ attention towards anion doping as a prelude to produce second generation materials that would increase the photocatalytic activity of TiO2 over the UV and much of the visible-light region. Subsequently, many studies appeared on the preparation of N-doped, C-doped, S-doped, F-doped TiO2 and N and C co-doped TiO2 materials and their evaluation as visible light catalysts. The use of N3-, C4-, S2-, or halides (F-, Cl-, Br-, I-) as doping agents, in TiO2, have been subjected to intense research and are summarized in several review articles50. The application of this oxide in water electrolysis, heterojunction solar cells and gas sensing in addition to photocatalysis adds to the list of its fascinating properties which are described in detail in many reviews51,52. The various method of obtaining the anion doped TiO2 are presented in Table 2.

Solution phase methods such as the sol-gel and solvothermal process are low-cost, convenient and robust approaches to prepare anion doped TiO2. By a simple wet process in which the hydrolysis product

of Ti(SO4)2 with ammonia was calcined using an ordinary electric furnace in dry air at 400 ºC, tracely nitrogen-doped TiO2 powders (yellow in color) were prepared52. The nitrogen-doped titania were also obtained by hydrolysis of titanium tetrachloride with a nitrogen-containing base such as ammonia, ammonium carbonate, or ammonium bicarbonate, followed by the calcination in air at 400 ºC, yielding a dopant concentration of 0.08 ± 0.13 wt% of nitrogen53. In solution phase synthesis, additives such as urea and alkyl ammonium salts were used as dopant sources to prepare N-doped TiO2. These nitrogen precursors were either added during TiO2 synthesis or used to treat TiO2 powders in order to accomplish anion doping54. Belver and coworkers55 synthesized a series of N-containing titanium precursors by modifying titanium tetraisopropoxide with different amine type ligands. By reacting TiCl4 with 2, 2’-bipyridine, Sano and coworkers56 obtained N-doped TiO2 through Ti4+-bipyridine complex. Anion doped TiO2 was synthesized by a simple solvothermal method57 starting with a molecular titanium precursor containing nitrogen and carbon, viz., titanium tri-thanolaminato isopropoxide. Nitrogen concentration in the TiO2 was enhanced up to 8 % by the direct amination of the 6-10 nm sized titania particles58. The nitrogen doped titanium dioxide (TiO2) nanocrystals were also prepared by heating TiN at 450, 550 and 650 ºC for 2 h in air yielding a mixture of anatase and rutile phases59. The incorporation of nitrogen ions in TiO2 single crystals was also achieved by sputtering with N2

+/Ar+ mixtures and subsequent annealing to 900 K under ultrahigh vacuum conditions60.

The evolution of N-doped TiO2 occurred on thermal treatment of amorphous ammonium titanium oxychloride precursor, synthesized using varying Ti/Cl ratios61. Nitrogen and sulfur co-doped and

Table 2 – Summary of the different type of dopants and the synthetic methodology used to dope in TiO2

Source employed Methodology Remarks Ref.

Ammonia Hydrolysis of Ti(SO4)2 followed by heating Trace quantity of nitrogen doped 52 Ammonium carbonate & bicarbonate

Hydrolysis of TiCl4 followed by heating at moderate temp.

Less than 0.01 % nitrogen doped 53

TiC Oxidative annealing 0.32 % of carbon doped with respect to oxygen in TiO2

54

Titanium isopropoxide modified with different amine type ligands

Sol-gel method, single source precursor 55

Amines Direct reaction with TiO2 nanoparticles 8 % 58 HF Sol-gel method F doped TiO2 67 TiN Controlled oxidation by heating at

450, 550 and 650°C Mixture of rutile and anatase N-doped TiO2

71

INDIAN J CHEM, SEC A, JAN-FEB 2012

150

N-doped TiO2 anatase were obtained62 by manual grinding with urea and thiourea, respectively and annealing at 400 ºC. During the hydrolysis of titanium tetraisopropoxide, urea-water-alcohol mixture was added to effect anion doped TiO2

63. Electrochemical methods were employed to exactly control anion doping of N and F-ions taking electrolyte solutions of titanium salt and ammonium ions, nitrate ions and fluoride ions64. F-doping in to TiO2 is found to be effective for enhancing the photocatalytic activity, as in the case of nitrogen and carbon doped systems65-67.

While in the products from the wet chemical methods, substantial portion of the doped nitrogen is not directly bound to Ti68, in the N, F co-doped TiO2 powders obtained by the spray pyrolysis, the N-atoms form a localized energy state above the valence band of TiO2 suggesting bonding between the dopants and Ti. F-doping in TiO2 had no influence on the band structure69. Band structure calculations carried out for the F-doped TiO2 indicate that the F-2p levels do not mix with the existing bands and hence, are not expected to contribute to the optical absorption spectra70. F-doping modifies the DOS near the conduction band minimum so that the resulting effect is similar to the creation of oxygen vacancies.

Yellow colored carbon-doped anatase TiO2 powders were fabricated by oxidative annealing of TiC. The amount of carbon doped by this method was 0.32 % with respect to oxygen. Carbon-doped TiO2 showed photocatalytic activities for the decomposition of IPA (isopropyl alcohol) to CO2 via acetone under visible light (400–530 nm) irradiation71. The dopant carbon was introduced into the TiO2 nano arrays by annealing them under flowing CO gas at higher temperatures of 500-800 °C. TiO2-xCx nanotube arrays showed much higher photocurrent densities and more efficient water splitting under visible-light illumination (> 420 nm) than pure TiO2 nanotube arrays. The total photocurrent was more than 20 times higher than that with a P-25 nano particulate film under white-light illumination72.

Presently, the question of whether anionic doping by nitrogen and/or carbon resulting in band gap narrowing in TiO2, the extent of such narrowing, and the utility of the resultant materials, is a matter of much debate in the scientific literature. On the other hand, so far as the photoelectrochemical response by the anionic doped TiO2 is concerned, the improvement is at best modest in majority of the research reports70. Anion doping in ZnO

Table 3 summarizes the dopant sources and the synthetic methodology employed for the fluoride ion doping in ZnO. F-doping in ZnO films and powders was achieved by CVD method, sputtering procedures, spray pyrolysis and precipitation methods. Liang & Gordon73 fabricated ZnO:F films by the APCVD procedure using tetramethylethylenediamine adduct of diethyl zinc. This process had the advantage of using a stable chelated precursor, paving way for a better control over doping and reproducible growth of uniform films. Tsai et al.74 made ZnO:F films from ZnF2 followed by post-annealing under vacuum. Kim & co-workers75 used a mixture of Ar and CF4 to achieve F-doped ZnO thin films from ZnO target.

Spray pyrolysis method presents some noticeable advantages, such as a wide possibility of varying the films properties by changing the composition of the starting solution and also that of low cost when large-scale production is needed. Many reports are available for the spray pyrolysis method of making F-doped ZnO films76-78. Kumar et al.79 deposited ZnO:F thin films using Zn(OOCCH3)2 and NH4F by spray pyrolysis technique. A similar procedure was adopted by Martinez et al.80 with the modification of ZnCl2 and trifluoroacetic acid.

Yakuphanoglu et al.81 prepared fluoride doped ZnO thin films by spin coating method using a gel obtained from Zn(OOCCH3)2 and NH4F with monoethanolamine as stabilizer. Fujihara et al.82

Table 3 – Summary of the synthetic conditions and methodology adopted for the successful F-doping in ZnO

Source employed Methodology Dopant Remarks Ref. Benzoyl fluoride CVD (thin film) ~ 0.2 % Film growth rate decreased with fluoride 73 ZnF2 RF magnetron sputtering (thin film) 1-10 % Single Source precursor 74, 75 NH4F Spray pyrolysis (thin film) 1-5 % Film thickness controlled procedure 76-80 NH4F/ZnF2 Sol gel (thin film) 5-15 % Low equipment cost 81, 82 NH4F Precipitation method (powder) 5 % Time and energy saving 83 KZnF3 Decomposition via oxidation (powder) ~ 6 % Air stable inorganic precursor, No

handling of harmful HF or fluorine gas generation, readily scalable, reproducible

84

NAGARAJAN et al.: FABRICATION & PROPERTIES OF ANION DOPED BINARY OXIDES OF Sn, Ti & Zn

151

employed ZnF2 precursor for the synthesis of ZnO:F by the hydrolysis procedure similar to the one employed for sol-gel synthesis. Martinez & co-workers83 obtained F-doped ZnO by heating the mixture of oxalic acid, zinc acetate solutions in ethanol and NH4F.

Our research group successfully synthesized the ZnO:F nanocrystals by a novel synthetic approach in which the fluoride-doped zinc peroxide was first obtained followed by its decomposition84. ZnO2:F was obtained by a simple low-temperature oxidation of the cubic perovskite, KZnF3, with H2O2. The following mechanism involving the sequences of chemical reactions could be conceived for the formation of inherently F- doped ZnO: KZnF3 + H2O2 → KZnF3-x (OH)x KZnF3-x (OH) x → Zn(OH)2 + HF + KOH Zn(OH)2 + HF → ZnO2: F ZnO2: F → ZnO: F

With H2O2 as the oxidizing agent, KZnF3 was believed to be dissociated resulting in Zn(OH)2 and KOH. Zinc hydroxide on refluxing at 120 ºC, under highly alkaline conditions provided by the generated KOH, could possibly have resulted in ZnO2. The highly reactive HF, presumed to be generated in situ, might have replaced some of the hydroxyl groups, thus resulting in ZnO2:F. Yellow colored ZnO:F nanocrystals were obtained by just simple thermal decomposition of ZnO2:F at 450 °C.

We extensively studied the structure, optical, and photocatalytic properties of the ZnO:F nanocrystals. Core level XPS measurements provided conclusive evidence for the doping of fluorine (6.1 %) in ZnO. A red shift in the excitonic absorption and photoluminescence spectrum were observed on F doping in ZnO. The presence of higher oxygen vacancies in ZnO:F was inferred from the Raman spectroscopic analysis. The band gap value of ZnO:F, estimated from the diffuse reflectance spectrum, was 3.0 eV, and showed broad visible emission. As a consequence of higher oxygen vacancies, ZnO:F exhibited efficient photocatalytic activity under visible irradiation for the degradation of aqueous MB dye solution.

The low electrical resistivity of the ZnO:F films has been attributed to the high electron mobility caused by the fluorine doping in oxygen sites,

perturbing mainly the valence band, thereby leaving the conduction band relatively free of scattering. Recently, Xu et al.86 reported fluoride doped ZnO films with low electrical resistivity and good optical transparency. The refractive index of the ZnO:F films were found to vary with increase in the fluorine concentration76. Fluorine doped ZnO thin films were demonstrated to exhibit cathodoluminescence charac-teristics86. A giant linear electro-optic (Pockels) effect was observed in the ZnO:F films due to substantial non-centrosymmetric charge density distribution between the wurtzite ZnO films and the additional charge density polarization caused by fluoride doping87. Kumar et al.79 showed that fluoride doping leads to considerable reduction in electrical resistivity, enhancement of optical transmission and reduction in photosensitivity. Fluoride doping in ZnO presents the advantage of improving the transport properties; however, it adversely affects the growth, setting a compromise between these two facts, which limits the performance of F-doped ZnO as a transparent electrode.

The p-type doping in ZnO may be possible by substituting either group-I elements (Li, Na, K) for Zn sites or group-V elements (N, P, As) for the O sites. Theoretically, nitrogen is the best dopant for p-type doping, since it has nearly the same radius as oxygen, low ionization energy, ease of handling, low material toxicity, and source abundance and is the shallowest acceptor in ZnO. The p-type ZnO films were deposited by magnetron sputtering88, chemical vapour deposition89, pulsed laser deposition90 and molecular beam epitaxy91 techniques. The N incorporated into ZnO was found to be compensated severely by both intrinsic and extrinsic defects, such as oxygen vacancies92, hydrogen impurities93 and nitrogen molecules94, making N-doped ZnO to remain as n-type. Several nitrogen-doping sources, e.g. N2, NH4NO3, NO, NH3, CH3COONH4 and N2O88-94, were examined to fabricate p-type ZnO samples. Although nitrogen is not very soluble in ZnO, its solubility can be greatly enhanced by forming NO–H complexes in ZnO co-doped with H impurities95. Two acceptor states, N-on-O substitution (NO) and zinc vacancy (VZn) were identified in N doped ZnO, which may have contributed towards the observed p-type conductivity96.

Though oxygen has many physical and chemical properties similar to those of sulfur, due to a similar structure of the electronic shell, S has rarely been doped in ZnO due to problems in fabrication owing to

INDIAN J CHEM, SEC A, JAN-FEB 2012

152

the stability of sulfur at the ZnO film growth temperature. The fabrication of ZnO film usually carried out at growth temperatures above 400 °C. On the other hand, sulfur is stable only at a low temperature, and vaporizes readily at 100 °C to an appreciable amount and its vapor pressure reaches 1 atm at 447 °C. Moreover, sulfur is highly reactive in the presence of oxygen to form intermediate phases such as a ZnSO4 easily. These appear to be some of the reasons for the non-availability of many studies. Bae et al.97 used a two-step catalyst assisted chemical vapor deposition process to get S-doped ZnO nanowires. Yoo and co-workers98 obtained S-doped ZnO thin film by the laser ablation of ZnS target. In an ecofriendly and novel preocedure, S-doped ZnO was obtained by simple mechanochemical synthesis followed by thermal decomposition of bisthiourea zinc oxalate powders99. Conclusions and Future Prospects

It is evident from this review that the wet chemical routes for the anion doped SnO2, TiO2 and ZnO are superior to other methods as they produce anion doped specimens reproducibly and eliminate the secondary impurity phase formation. Doping the anions from single source precursors appears to yield homogeneously doped products as compared to heating the oxide with the volatile and gaseous dopant sources. However, the control of the anion concentration in the lattice and designing new precursors for the auto doping of these anions are the indispensable challenges to the synthetic chemists. Critical and additional investigation in to band gap narrowing taking in to account the actual composition of the samples, especially when heavily doped, the growth of high quality anion doped single crystals and their structural determinations to establish the exact positions of the dopant atoms in the lattice, viz., substitutional versus interstitials are some of the other key issues that remain open. Since all the three oxides discussed in this review show extensive applications based on their efficient use under the solar spectrum, issues related to their photostability must also be addressed. Acknowledgement

Authors thank DST (Nanomission) Govt. of India, New Delhi and University of Delhi, Delhi, for funding this research. VK and SA thank UGC, New Delhi and CSIR, New Delhi, respectively for their research fellowship.

References 1 (a) Smith A M & Nie S, Acc Chem Res, 43 (2010) 190;

(b) Alivasatos A P, Science, 271 (1996) 933. 2 Rao C N R & Gopalakrishnan J, New Directions in Solid

State Chemistry, (Cambridge University Press) 1997. 3 Ha H-W, Kim K, Borniol M & Toupance T, J Solid State

Chem, 179 (2006) 702. 4 Gordon R G, Mat Res Soc Bull, 25 (2000) 52. 5 Chopra K L, Major S & Pandya D K, Thin Solid Films, 102

(1983) 1. 6 Han C-H, Hong D-U, Gwak J & Han S-D, Korean J Chem

Eng, 24 (2007) 927. 7 Yang F & Forrest S R, Adv Mater, 18 (2006) 2018. 8 Anderson A, Johansson N, Bröms P, Yu N, Lupo D &

Salaneck W R, Adv Mater, 11 (1998) 10. 9 Hu Z, Zhang J, Hao Z, Hao Q, Geng X & Zhao Y, Appl Phys

Lett, 98 (2011) 123302. 10 Liao H-C, Lin C-C, Chen Y-W, Liu T-C & Chen S-Y,

J Mater Chem, 20 (2010) 5429. 11 Thomson T L & Yates J T, Chem Rev, 106 (2006) 4428 12 Lewis B G & Paine D C, Mat Res Soc Bull, 25 (2000) 22. 13 Hu J & Gordon R G, Solar Cells, 30 (1991) 437. 14 (a) Freeman A J, Poeppelmeier K R, Mason T O, Chang R P

H & Marks T J, Mat Res Soc Bull, 25 (2000) 45; (b) Kawazoe H, Yanagi H, Ueda K & Hosono H, Mat Res Soc Bull, 25 (2000) 28.

15 Talaty N N, Beck K, Citeau H, Kirschbaum K & Giolando D M, Z Anorg Allg Chem, 635 (2009) 53.

16 Ray S C, Karanjai M K & DasGupta D, Surf Coat Int, 102 (1998) 73.

17 Shanthi E, Banerjee A, Dutta V & Chopra K L, J Appl Phys, 53 (1982) 1615.

18 Yadav A A, Masumdar E U, Moholkar A V, Spallart M N, Rajpure K Y & Bhosale V H, J Alloys Compd, 488 (2009) 350.

19 Zhang B, Tian Y & Cai W, Mater Lett, 64 (2010) 2707. 20 Miao D, Zhao Q, Wu S, Wang Z, Zhang X & Zhao X, J

Non-Crys Solids, 356 (2010) 2557. 21 (a) Elangovan E & Ramamurthi K, Thin Solid Films, 476 (2005)

231; (b) Thangaraju B, Thin Solid Films, 402 (2002) 71. 22 Canestraro C D, Oliveira M M, Valaski R, da Silva M V S,

David D G F, Pepe I, da Silva A F, Roman L S & Persson C, Appl Surf Sci, 255 (2008) 1874.

23 (a) Bilgin V, Akyuz I, Ketenci E, Kose S & Atay F, Appl Surf Sci, 256 (2010) 6586; (b) Zhang B, Tian Y, Zhang J X & Cai W, Appl Phys Lett, 98 (2011) 021906.

24 Arefi-Khonsari F, Bauduin N, Donsanti F & Amouroux J, Thin Solid Films, 427 (2003) 208.

25 Kim H, Auyeung R C Y & Piqué A, Thin Solid Films, 516 (2008) 5052.

26 Bae J W, Lee S W & Yeom G Y, J Electrochem Soc, 154 (2007) D34.

27 Cachet H, Gamard A, Campet G, Jousseaume B & Toupance T, Thin Solid Films, 388 (2001) 41.

28 Zhao H, Liu Q, Cai Y & Zhang F, Mater Lett, 62 (2008) 1294.

29 Gottlieb B, Koropecki R, Arce R, Crisalle R & Ferron J, Thin Solid Films, 199 (1991) 13.

30 Molloy K C, J Chem Res, 2008, 549. 31 Tudela D, Gutierrez-Puebla E & Monge A, J Chem Soc

Dalton Trans, (1992) 1069.

NAGARAJAN et al.: FABRICATION & PROPERTIES OF ANION DOPED BINARY OXIDES OF Sn, Ti & Zn

153

32 Maruyama T & Tabata K, J Appl Phys, 68 (1990) 4282. 33 (a) Suh S, Zhang Z, Chu W-K & Hoffman D M, Thin Solid

Films, 354 (1999) 240; (b) Suh S, Hoffman D M, Atagi L M, Smith D C, Liu J-R & Chu W-K, Chem Mater, 9 (1997) 730.

34 (a) Molloy K C & Stanley J E, Appl Organomet Chem, 23 (2009) 62; (b) Stanley J E, Swain A C, Molloy K C, Rankin D W H, Robertson H E & Johnston B F, Appl Organomet Chem, 19 (2005) 644.

35 Mahon M F, Molloy K C, Stanley J E, Rankin D W H, Robertson H E & Johnston B F, Appl Organomet Chem, 19 (2005) 658.

36 Wu S, Yuan S, Shi L, Zhao Y & Fang J, J Colloid Interface Sci, 346 (2010) 12.

37 Han C-H, Han S-D, Gwak J & Khatkar S P, Mater Lett, 61 (2007) 1701.

38 Suffner J, Agoston P, Kling J & Hahn H, J Nanopart Res, 12 (2010) 2579.

39 (a) Gamard A, Jousseaume B, Toupance T & Campet G, Inorg Chem, 38 (1999) 4671; (b) Franc C, Jousseaume B, Linker M & Toupance T, Chem Mater, 12 (2000) 3100; (c) Gamard A, Babot O, Jousseaume B, Rascle M-C, Toupance T & Campet G, Chem Mater, 12 (2000) 3419.

40 (a) Boutet S, Gamard A, Jousseaume B, Toupance T, Campet G & Cachet H, Main Group Met Chem, 25 (2002) 59; (b) Boegeat D, Jousseaume B, Toupance T, Campet G & Fournes L, Inorg Chem, 39 (2000) 3924.

41 Esteves M C, Gouvea D & Sumodjo P T A, Appl Surf Sci, 229 (2004) 24.

42 Avadhut Y S, Weber J, Hammarberg E, Feldmann C, Schellenberg I, Pottgen R & auf der Gunne J S, Chem Mater, 23 (2011) 1526.

43 Kumar V, Govind A & Nagarajan R, Inorg Chem, 50 (2011) 5637.

44 (a) Abram A, Rees G J & Wilson B L H, Adv Phys, 27 (1978) 799; (b) Mocatta D, Cohen G, Schattner J, Millo O, Rabani E & Babin U, Science, 332 (2011) 77.

45 (a) Ji Z, Zhao L, He Z, Zhou Q & Chen C, Mater Lett, 60 (2006) 1387; (b) Bagheri-Mohagheghi M-M & Shokooh-Saremi M, J Phys D Appl Phys, 37 (2004) 1248.

46 Yan Y, Zhang S B & Pantelides S T, Phys Rev Lett, 86 (2001) 5723.

47 Pan S S, Ye C, Teng X M, Fan H T & Li G H, Appl Phys A, 85 (2006) 21.

48 (a) Lee D H, Cho, Y S, Yi W I, Kim T S, Lee J K & Jung H, Appl Phys Lett, 66 (1995) 815; (b) Saha N C & Tompkins H G, J Appl Phys, 72 (1992) 3072.

49 Asahi R, Morikawa T, Ohwaki T, Aoki K & Taga Y, Science, 293 (2001) 269.

50 (a) Sepone N, J Phys Chem B, 110 (2006) 24287; (b) Zhao J, Chen C & Ma W, Top Catal, 35 (2005) 267; (c) Emeline A V, Kuznetsov V N, Rybakutv U K & Serpone N, Int J Photoenergy, 2008, article no.258394.

51 (a) Mor G K, Varghese O K, Paulose M, Shankar K & Grimes C A, Sol Energ Mat Sol C, 90 (2006) 2011; (b) Shankar K, Basham J I, Allam N K, Varghese O K, Mor G K, Feng X, Paulose M, Seabold J A, Choi K S & Grimes C A, J Phys Chem C, 113 (2009) 6327.

52 Ihara T, Miyoshi M, Iriyama Y, Matsumoto O & Sugihara S, Appl Catal B, 42 (2003) 403.

53 Sakthivel S & Kisch H, Chem Phys Chem, 4 (2003) 487.

54 (a) Reddy K M, Baruwati B, Jayalakshmi M, Rao M M & Manorama S V, J Solid State Chem,178 (2005) 3033; (b) Stewart S J, Fernández-García M, Belver C, Mun B S & Requejo F G, J Phys Chem B, 110 (2006) 16482.

55 Belver C, Bellord A, Fuerte M & Fernandez-Garcia, Appl Catal B, 65 (2006) 301.

56 Sano T, Negishi N, Koike K, Takeuchi S & Matsuzawa, J Mater Chem, 14 (2004) 380.

57 Li G, Chen L, Dimitrijevic N M & Gray K A, Chem Phys Lett, 451 (2008) 75.

58 Burda C, Lou Y B, Chen X B, Samia A C S, Stout J & Gole J L, Nano Lett, 3 (2003) 1049.

59 Wu Z B, Dong F, Zhao W R & Guo S, J Hazard Mater, 157 (2008) 57.

60 Diwald O, Thompson T L, Goralski E G, Walck S D & Yates J T, J Phys Chem B, 108 (2004) 52.

61 Colon G, Sampedro P, Fernandez-Garcia M, Chen H, Hanson J C & Rodriguez J A, Langmuir, 24 (2008) 11111.

62 Rengifo-Herrera J A & Pulgarin C, Solar Energy, 84 (2010) 37.

63 Silija P, Yaakob Z, Yarmo M A, Sugunan S & Binitha N N, J Sol-Gel Sci Tech, 59 (2011) 252.

64 Shankar K, Tep K C, Mor G K & Grimes C A, J Phys D Appl Phys, 39 (2006) 2361.

65 Yu J C, Yu J, Ho W, Jiang Z & Zhang L, Chem Mater, 14 (2002) 3808.

66 Hattori A, Shimoda K, Tada H & Ito S, Langmuir, 15 (1999) 5422.

67 Li D, Haneda H, Hishita S, Ohashi N & Labhsetwar N K, J Fluorine Chem, 126 (2005) 69.

68 Sato S, Nakamura R & Abe S, Appl Catal A, 284 (2005) 131. 69 (a) Li D, Ohashi N, Hishita S, Kolodiazhnyi T & Haneda H,

J Solid State Chem, 178 (2005) 3293; (b) Li D, Haneda H, Hishita S & Ohashi N, Chem Mater, 17 (2005) 2596.

70 Yamaki T, Umebayashi T, Sumita T, Yamamoto S, Maekawa M, Kawasuso A & Itoh H, Nucl Instrum Methods Phys Res B, 206 (2003) 254.

71 Irie H, Watanabe Y& Hashimoto K, Chem Lett, 32 (2003) 772.

72 Park J H, Kim S & Bard A J, Nano Lett, 6 (2006) 24. 73 Liang H & Gordon R G, J Mater Sci, 42 (2007) 6388. 74 Tsai Y-Z, Wang N-F & Tsai C-L, Thin Solid Films, 518

(2010) 4955. 75 (a) Yoon H S, Lee K S, Lee T S, Cheong Ba, Choi D K, Kim

D H & Kim W M, Sol Energ Mat Sol C, 92 (2008) 1366; (b) Choi B G, Kim I H, Kim D H, Lee K S, Lee T S, Cheong B, Baik Y J & Kim W M, J Eur Ceram Soc, 25 (2005) 2161; (c) Kim I, Lee K-S, Lee T S, Jeong J-H, Cheong B-K, Baik Y-J & Kim W M, J Appl Phys, 100 (2006) 063701.

76 (a) Sanchez-Juarez A, Tiburcio-Silver A & Ortiz A, Sol Energ Mat Sol C, 52 (1998) 301; (b) Sanchez-Juarez A, Tiburcio-Silver A, Ortiz A, Zironi E P & Rickards J, Thin Solid Films, 333 (1998) 196.

77 Pawar B N, Ham D-H, Mane R S, Ganesh T, Cho B-W & Han S-H, Appl Surf Sci, 254 (2008) 6294.

78 (a) Rodríguez-Báez J, Maldonado A, Castañeda L, Delgado G T, Castanedo-Pérez R & Olvera M de la L, Thin Solid Films, 515 (2007) 8689; (b) Olvera M de la L, Maldonado A & Asomoza R, Sol Energ Mat Sol C, 73 (2002) 425.

79 Kumar P M R, Kartha C S, Vijayakumar K P, Singh F & Avasthi D K, Mater Sci Engg B, 117 (2005) 307.

INDIAN J CHEM, SEC A, JAN-FEB 2012

154

80 Martinez A I & Acosta D R, Thin Solid Films, 483 (2005) 107.

81 Yakuphanoglu F, Caglar Y, Ilican S & Caglar M, Physica B, 394 (2007) 86.

82 Fujihara S, Kusakado J & Kimura T, J Mater Sci Lett, 17 (1998) 781.

83 Gonzalez-Hernandez R, Martinez A I, Falcony C, Lopez A A, Pech-Canul M I & Hdz-Garcia H M, Mater Lett, 64 (2010) 1493.

84 Ahmad S, Kharkwal M, Govind & Nagarajan R, J Phys Chem C, 115 (2011) 10131.

85 Xu H Y, Liu Y C, Mu R, Shao C L, Lu Y M, Shen D Z & Fan X W, Appl Phys Lett, 86 (2005) 123107-1.

86 Hichou A El, Bougrine A, Bubendorff J L, Ebothe J, Addou M & Troyon M, Semicond Sci Technol, 17 (2002) 607.

87 Kityk I V, Ebothe J, Hichou A El, Addou M, Bougrine A & Sahraoui B, J Phys Cond Matt, 14 (2002) 5407.

88 (a) Lu J G, Ye Z Z, Zhuge F, Zeng Y J, Zhao B H & Zhu L P, Appl Phys Lett, 85 (2004) 3134; (b) Ye Z Z, Zhuge F, Lu J G, Zhang Z H, Zhu L P, Zhao B H & Huang J Y, J Cryst Growth, 265 (2004) 127; (c) Lu J G, Zhu L P, Ye Z Z, Zhuge F, Zhao B H, Ma D W, Wang L & Huang J Y, J Mater Sci, 41 (2006) 467. (d) Lu J G, Zhang Y Z, Ye Z Z, Zhu L P, Wang L, Zhao B H & Liang Q L, Appl Phys Lett, 88 (2006) 222114.

89 (a) Liang H W, Lu Y M, Shen D Z, Liu Y C, Yan J F, Shan C X, Li B H, Zhang Z Z, Zhang J Y & Fan X W, Phys Status Solidi A, 202 (2005) 1060; (b) Ye Z Z, Lu J G, Chen H H,

Zhang Y Z, Wang L, Zhao B H & Huang J Y, J Cryst Growth, 253 (2003) 258.

90 Lu J G, Zhang Y Z, Ye Z Z, Wang L, Zhao B H & Huang J Y, Mater Lett, 57 (2003) 3311.

91 (a) Lu J G, Ye Z Z, Wang L, Huang J Y & Zhao B H, Mater Sci Semicond Process, 5 (2002) 491; (b) Rommeluere J F, Svob L, Jomard F, Mimila-Arroyo J, Lusson A, Sallet V & Marfaing Y, Appl Phys Lett, 83 (2003) 287.

92 (a) Lee E C, Kim Y S, Jin Y G & Chang K J, Phys Rev B, 64 (2001) 085120; (b) Marfaing Y & Lusson A, Superlatt Microstruct, 38 (2005) 385.

93 (a) Walle C G V, Phys Rev Lett, 85 (2000) 1012; (b) Hofmann D M, Meyer B K, Schmidt J & Baranov P G, Phys Rev Lett, 88 (2002) 045504.

94 Soudi A, Khan E H, Dickinson J T & Gu Y, Nano Lett, 9 (2009) 1844.

95 (a) Garces N Y, Wang L J, Giles N C, Halliburton L E, Cantwell G & Eason D B, J Appl Phys, 94 (2003) 519; (b) Keyes B M, Gedvilas L M, Li X & Coutts T J, J Cryst Growth, 281(2005) 297.

96 Lu J, Liang Q, Zhang Y, Ye Z & Fujita S, J Phys D Appl Phys, 40 (2007) 3177.

97 Bae S Y, Seo H W & Park J, J Phys Chem B, 108 (2004) 5206.

98 Yoo Y-Z, Jin Z-W, Chikyow T, Fukumura T, Kawasaki M & Koinuma H, Appl Phys Lett, 81 (2002) 3798.

99 Patil A B, Patil K R & Pardeshi S K, J Hazard Mater, 183 (2010) 315.