ORIGINAL PAPER

Tunable white light-emission of a CaW12xMoxO4:Tm3+, Tb3+,Eu3+ phosphor prepared by a Pechini sol–gel method

Matthew Mickens • Zerihun Assefa •

Dhananjay Kumar

Received: 12 February 2012 / Accepted: 17 April 2012 / Published online: 27 April 2012

� Springer Science+Business Media, LLC 2012

Abstract A white light-emitting CaW1-xMoxO4:Tm3?,

Tb3?, Eu3? phosphor was prepared by a Pechini sol–gel

method. The incorporation of Mo6? into the CaWO4 host

matrix can broaden its excitation range and promote tun-

ability to its emission. When the CaW1-xMoxO4 system is

triply-doped with Tm3?, Tb3?, and Eu3? ions, energy

transfer occurs from both WO42- and MoO4

2- groups to

Tm3? and Tb3? ions. A significant red-shift in the excita-

tion of Eu3? allows the resulting emission to be tunable

between cool, natural, and warm white light by varying the

excitation wavelength. The undoped and triply-doped

CaW1-xMoxO4 phosphors were characterized by X-ray

diffraction, scanning electron microscopy, photolumines-

cence excitation and emission spectra, and CIE chroma-

ticity (x, y) coordinates.

Keywords Phosphor � White light � Tungstate �Molybdate � Pechini sol–gel � Energy transfer

1 Introduction

Since the commercialization of the first white light-emit-

ting diode (WLED) by Nichia Corporation in 1996 [1],

tremendous progress has been achieved for solid-state

lighting (SSL) as indicated by recent reviews [2, 3]. As the

development of new inorganic phosphors steadily improves

the efficiency of WLEDs, their use for general illumination

is projected to become price-competitive with fluorescent

lamps by the year 2020 [4]. To optimize the color ren-

dering properties of WLEDs, the current focus of their

fabrication is gradually shifting from YAG:Ce3?-based to

single-phased RGB-emitting/full color phosphors excited

by UV-LEDs [5–8].

Numerous reports show that crystalline calcium tung-

state/molybdate (CaWO4/MoO4) systems with scheelite

structure serve as excellent single-phased hosts for Ln3?

ions [9–17]. In addition to their chemical and thermal

stability, the relatively low vibrational energies permit

efficient energy transfer to emitting Ln3? centers [18].

Among the alkaline-earth metal (Ca2?, Sr2?, Ba2?) tung-

states, CaWO4 was shown as the most effective host for

transferring its excitation energy to Tm3? [19] and Sm3?

[20] dopant ions. In terms of observing white light,

Nazarov et al. [21] reports on varying the luminescence

color of CaWO4:Eu3?,Tb3? through the entire visible

spectrum, but the emission of white light via energy

transfer required high-energy VUV (147 nm) excitation.

Similarly, in CaMoO4 systems, warm white light emission

was also reported in a multivalent CaMoO4:Pr3?, Pr4?

phosphor [22], while cool white light was observed in

CaMoO4:Eu3? nanofibers studied by Hou et al. [9] as

confirmed by chromaticity and photographs, but the

observation was not emphasized in the report. One unique

route to convert CaWO4/MoO4 hosts into white light-

M. Mickens

Energy and Environmental Systems, North Carolina A&T State

University, Greensboro, NC 27411, USA

e-mail: mamicken@ncat.edu

Z. Assefa (&)

Department of Chemistry, North Carolina A&T State University,

Greensboro, NC 27411, USA

e-mail: zassefa@ncat.edu

D. Kumar

Department of Mechanical Engineering, North Carolina A&T

State University, Greensboro, NC 27411, USA

e-mail: dkumar@ncat.edu

123

J Sol-Gel Sci Technol (2012) 63:153–161

DOI 10.1007/s10971-012-2780-0

emitting phosphors is by combining the two systems.

Zheng and co-workers [23] recently demonstrated the

capability of extending the excitation range of CaWO4 by

the addition Mo, where cool white light was observed by

varying the concentrations doped Eu3? and Tb3? ions.

However, to the best of our knowledge, no investigation

has reported utilizing a single-phased Ca(W,Mo)O4 system

where various hues of white light are observed by tuning

the excitation wavelength.

In this work, a CaWO4 host doped with a small amount

of Mo6? (CaW1-xMoxO4) was prepared by a Pechini sol–

gel method, where its excitation properties are extended

and wavelength-tunability of its emission is attained.

Moreover, the extended excitation range is nonradiatively

transferred to triply-doped RGB-emitting Ln3? ions

(Ln = Tm3?, Tb3?, Eu3?) with the resulting emission

being tunable between cool, natural, and warm white light.

2 Experimental

2.1 Sample synthesis

All chemicals and reagents were purchased from Sigma-

Aldrich with the exception of ammonium molybdate,

which was obtained from Baker & Adamson. The phos-

phors were prepared by the Pechini sol–gel method [24,

25]. The CaWO4:Mo host was prepared by using the for-

mula CaW0.99Mo0.01O4, while the synthesis of the triply-

doped phosphor samples adhered to the general formula

Ca1–2x–2y–2zTmxTbyEuzNax?y?zW0.99Mo0.01O4. In propor-

tion to Ca2?, the doping concentrations of Ln3? and Na?

ions were 1 mol% Tm3?, 0.5 mol% Tb3?, 0.5 mol% Eu3?,

and 2 mol% Na?. To obtain 0.5 g of doped phosphor

product, a typical experiment involved dissolving 438 mg

of ammonium tungstate ((NH4)10W12O41�5H2O) and 3 mg

of ammonium molybdate ((NH4)6Mo7O24�4H2O) in a

water–ethanol (V/V = 3:1) solution containing 667 mg of

citric acid (C6H8O7) as the chelating agent for the metal

ions. The molar ratio of metal ions to citric acid was 1:2. In

a separate beaker, 166 mg of calcium carbonate (99.99 %

CaCO3), 1.8 mg of sodium carbonate (99.5 % Na2CO3),

3.3 mg of thulium oxide (99.99 % Tm2O3), 1.6 mg of

terbium oxide (99.99 % Tb4O7), and 1.5 mg of europium

oxide (99.99 % Eu2O3) were dissolved in 5 mL of dilute

nitric acid (70 % HNO3) under heated agitation until the

solution became clear. The excess HNO3 was removed by

heat treatment and then 4 mL of deionized water was

added to dissolve the nitrate salts. The resulting metal-

nitrate solution was adjusted to a pH of 2–3, and then

added to the flask containing the metal-citrate solution.

Polyethylene glycol (PEG; molecular weight = 200) was

added as a cross-linking agent with a final concentration of

0.20 g mL-1. The solution mixture was covered and stirred

magnetically for 1 h at 70 �C, and ultrasonically for

20 min to yield a highly transparent sol. The sol was

transferred to an open beaker and placed in a water bath at

65 �C for 8 h where it was concentrated to a viscous gel.

The gel was immediately placed in an oven at 100 �C for

30 min to drive off remaining solvents until a color change

to transparent yellow was observed. While hot, the reduced

viscosity of the gel allowed easy transfer to a porcelain

crucible where annealing took place at 900 �C for 2 h with

a heating rate of 1 �C min-1. The phosphor powders were

allowed to cool naturally in the furnace and were charac-

terized immediately.

2.2 Structural characterizations

The X-ray powder diffraction (XRD) patterns of the syn-

thesized phosphors were collected in the range of

108\ 2h\ 908 using a Bruker D8 Advance X-ray dif-

fractometer with a Cu Ka radiation source (k = 0.15418

A). The voltage and current were maintained at 40 kV and

40 mA, respectively. All obtained data were compared to

the theoretically calculated XRD patterns provided by the

Joint Commission on Powder Diffraction Standards

(JCPDS). The particle size and morphology were inspected

using a scanning electron microscope (SEM, Hitachi

S-3,000 N).

2.3 Photoluminescence and chromaticity

The PL studies were conducted on the basis of excitation and

emission spectra using a Shimadzu RF-5301 spectrofluoro-

photometer equipped with a 150 W xenon lamp excitation

source. The phosphor powders were loosely placed on a solid

platform and were excited directly. All spectroscopic mea-

surements were performed at room temperature. The spec-

troscopic data were acquired and managed by HyperRF

software (Version 1.57). The data were later exported to

Microsoft Excel for manipulation and figure preparation.

The chromaticity coordinates and correlated color temper-

ature (Tc) were measured by a Gigahertz-Optik HCT99D

optometer. The detachable optometer detector was inserted

inside the sample compartment of the spectrofluoropho-

tometer and chromaticity measurements were obtained as

samples were excited by the xenon lamp source.

3 Results and discussion

3.1 X-ray diffraction analysis

Figure 1a shows the XRD profiles of undoped CaWO4 and

Ca0.96W0.99Mo0.01O4:0.01Tm3?, 0.005Tb3?, 0.005Eu3?,

154 J Sol-Gel Sci Technol (2012) 63:153–161

123

0.02Na? both annealed at 900 �C. For reference, the

JCPDS Card 41-1431 for CaWO4 is shown in Fig. 1b. It

can be seen that all diffraction peaks of the doped and

undoped samples agree well with JCPDS 41-1431 standard

data of scheelite phase CaWO4. The peaks in the XRD

spectra are sharp and intense confirming that a highly

crystalline single-phase CaWO4 phosphor has been suc-

cessfully synthesized by the Pechini sol–gel method. Since

there are no impurity peaks or notable peak shifts, it can be

assumed that the substitution of Mo6?, Ln3?, and Na? ions

has been accomplished without causing any structural

change to the CaWO4 lattice. Although the smaller ionic

radii of Tm3?, Tb3?, and Eu3? ions (0.87–1.11 A) at Ca2?

sites (1.12 A) causes distortions due to sub-lattice con-

traction, the simultaneous doping of slightly larger Na?

ions (1.18 A) was used in this work to compensate for both

excess charge and ionic radii distribution within the

CaWO4 lattice. Previous reports [26–29] also reveal that

replacing 2Ca2? ions with 1Ln3? ?1M? (M? = alkaline

ion) optimizes the emission intensity of Ln3? ions within

CaWO4/CaMoO4 hosts.

3.2 Morphology analysis

For comparison purposes, undoped CaWO4 and CaMoO4

phosphors were both prepared by the Pechini sol–gel

method. The morphologies of the phosphor powders were

examined by SEM. The typical morphological photographs

are shown in Figs. 2 and 3. In Fig. 2a, the undoped CaWO4

consists of well-defined cubical shaped particles with mod-

erate agglomeration. Under higher magnification (Fig. 2b),

it can also be seen that the crystallites exhibit a variety of

particle sizes ranging from 0.3 to 1.5 lm. On the other hand,

Fig. 2c, d show that the undoped CaMoO4 exhibits signifi-

cantly large grains (the largest of all samples) with a high

degree of agglomeration. Even though particles as small as

1 lm are seen attached to the particle surface, the images

reveal dominant formations of large closely-packed grains

with sizes ranging from 5 to 10 lm. The dramatic increase in

grain size is likely due to the melting point of CaMoO4

(1,468 �C) being considerably lower than that of CaWO4

(1,620 �C) [30]. The CaWO4:Mo (CaW1-xMoxO4) powders

in Fig. 3a exhibit the most uniform particle size distribution

with the average size between 2 and 3 lm. The textural

properties appear to be slightly more spherical than undoped

CaWO4, while also resembling the larger CaMoO4 particles

by their pronounced agglomeration and smooth crystallite

edges (Fig. 3b). As seen in Fig. 3c, the CaWO4:Mo, Tm3?,

Tb3?, Eu3? powder exhibits very similar particle sizes to

CaWO4:Mo in the range of 2–3 lm. However, under higher

magnification in Fig. 3d, it is evident that their morphologies

exhibit sharp and well-defined crystallite edges similar to the

undoped CaWO4 powders.

3.3 Photoluminescent properties

of CaW0.99Mo0.01O4 host

The excitation spectra shown in Fig. 4a, c correspond to that

of CaWO4 and CaMoO4 monitored at 430 and 495 nm,

respectively. The comparison of these figures clearly indi-

cates that doping MoO42- into the CaWO4 lattice could

expand the excitation range of the host, as well as broaden its

emission profile. Figure 5 shows the emission spectra of

CaWO4 doped with 1 mol% Mo (CaW0.99Mo0.01O4). As the

excitation wavelength increases systematically from 250 to

275 nm, the emission maximum shifts from 439 to 485 nm,

corresponding to an overall shift of *2,160 cm-1. The shift

in the emission maximum is a result of the varying contri-

bution of the two emitting WO42-/MoO4

2- groups under

different excitation wavelengths. The inset of Fig. 5 shows

the excitation spectra monitored at 430, 450, and 495 nm.

When monitored at 430 nm, the WO42- CT band overlaps

with the MoO42- CT band which exhibits a well-defined

shoulder at *280 nm. The overall spectral profile is

extended to *315 nm. In contrast, when monitored at

Fig. 1 X-ray diffraction patterns for undoped CaWO4, Ca0.96W0.99-

Mo0.01O4:0.01Tm3?, 0.005 Tb3?, 0.005Eu3?, 0.02Na? (a) and

JCPDS Card 41-1431 data for CaWO4 (b)

J Sol-Gel Sci Technol (2012) 63:153–161 155

123

CaWO4 – Magnification: 10 k CaWO4 – Magnification: 20 k

CaMoO4 – Magnification: 10 k CaMoO4 – Magnification: 5 k

(a) (b)

(c) (d)

Fig. 2 SEM images of CaWO4 (a, b) and CaMoO4 (c, d) obtained at different magnifications

CaWO4:Mo – Magnification 5 k CaWO4:Mo – Magnification 10 k

CaWO4:Mo,Tm, Tb, Eu – Magnification 20 k CaWO4:Mo, Tm, Tb, Eu – Magnification 10 k

(a) (b)

(c) (d)

Fig. 3 SEM images of CaWO4:Mo (a, b) and CaWO4:Mo, Tm3?, Tb3?, Eu3? (c, d) obtained at different magnifications

156 J Sol-Gel Sci Technol (2012) 63:153–161

123

450 nm the excitation bands of the two groups exhibit

comparable intensities. As the monitoring emission wave-

length moves closer to the MoO42- maximum at 495 nm, the

MoO42- excitation band becomes dominant. A characteris-

tic feature in this system is that the usable excitation range of

the WO42- CT band is accompanied by the contributing

excitation band of the MoO42- CT, which extends the overall

excitation range of the CaWO4 host.

The potential for WO42- to transfer energy to MoO4

2-

within the current system has been noted, but the extent of

energy migration was not specifically evaluated in this

work. It is important to remark that earlier reports on a

similar mixed scheelite-powellite (CaWO4-CaMoO4)

series indicated the possibility of nonradiative energy

transfer from WO42- to MoO4

2-. However, the observance

of a significant enhancement in the MoO42- CT band at the

expense of attenuated emission of WO42-, required a

concentration of at least 10 mol% Mo [31]. The MoO42-

CT band was also reported to be dominant in CaWO4:Mo

microspheres when varied concentrations exceeded

10 mol% Mo [23].

3.4 Emission spectra of CaW0.99Mo0.01O4:Tm3?,

Tb3?, Eu3?

Figure 6 shows the emission spectra of the Ca0.96W0.99

Mo0.01O4:0.01Tm3?, 0.005 Tb3?, 0.005Eu3?, 0.02Na?

phosphor under excitation at (a) 250, (b) 260, and

(c) 280 nm. When compared with that of Figs. 4 and 5, the

emission intensity of WO42- in Fig. 6a, is significantly

quenched resulting in dominating f–f transitions. The sharp

bands at 455 and 476 nm are characteristic of the Tm3?

1D2 ? 3F4 and 1G4 ? 3H6 transitions, respectively. Sim-

ilarly, the Tb3? emissions are observed at 488 and 545 nm

corresponding to the 5D4 ? 7F6 and 5D4 ? 7F5 transi-

tions, respectively. In addition, rarely exhibited higher

level energy bands are observed at 382, 414, and 439 nm

that correspond to the 5D3 ? 7FJ (J = 6, 5, 4) transitions,

respectively. These transitions are observed due to the low

phonon energies of the host unlikely to bridge the Tb3? 5D3

and 5D4 energy gap efficiently.

The 580–595 nm region exhibits a combination of weak

bands that belong to the 5D4 ? 7F4 transition of Tb3?, and

the 5D0 ? 7F0–1 magnetic dipole transitions of Eu3?. The

Fig. 4 Room temperature PL spectra of undoped CaWO4 and

CaMoO4; (a, b, and c, d, respectively). Excitation spectra were

obtained by monitoring the emission at (a) 430 nm and (c) 495 nm.

Emission spectra were obtained by excitation at (b) 260 nm and

(d) 290 nm

Fig. 5 Room temperature emission spectra of CaW0.99Mo0.01O4

under excitation wavelengths (a) 250 nm, (b) 255 nm, (c) 260 nm,

and (d) 275 nm (to avoid 2nd order effect to the PMT tube, spectral

regions between 495 and 560 nm were bypassed due to the lack of

appropriate interference filter suitable to the PL instrument)

Fig. 6 Room-temperature emission spectra of Ca0.96W0.99

Mo0.01O4:0.01Tm3?, 0.005Tb3?, 0.005Eu3?, 0.02Na? under excita-

tion wavelengths of: (a) 250 nm, (b) 260 nm, and (c) 280 nm (to

avoid 2nd order effect to the PMT tube, spectral regions between 500

and 575 nm were bypassed)

J Sol-Gel Sci Technol (2012) 63:153–161 157

123

red emission of Eu3? observed at 616 nm is due to the

hypersensitive 5D0 ? 7F2 electric dipole transition. Since

this transition of Eu3? is dominant, it is indicative that

Eu3? is substituted at Ca2? sites that lack inversion sym-

metry. As shown in Fig. 7a, the Ca2? ions have eight

oxygen neighbors; four with bond lengths of 0.244 nm and

the other four at 0.248 nm, forming a distorted cube [32].

Since the tetragonal structure of CaWO4 corresponds to

space group I41/a, the substituted Mo6?, Ln3?, and charge

compensating Na? ions are expected to occupy sites with

S4 point symmetry [32]. In the illustration of Fig. 7b, the

crystallographic assignments of the dopants are in agree-

ment with the XRD patterns of Fig. 1a, which previously

confirmed that their inclusion did not alter the phase of the

CaWO4 host.

Figure 6b shows the emission spectrum of the same

sample under excitation at 260 nm. The prominent increase

in the red and green emissions of Eu3? and Tb3? shifts the

visual appearance from bluish white to cool white as shown

in the inset of Fig. 6b. Furthermore, the quenched broad

emission of the WO42- host is now accompanied by weak

emission from the MoO42- groups. This is evident in the

baseline region around the Tb3? transition at 545 nm

which is slightly elevated in comparison to Fig. 6a. Upon

increasing the excitation to 280 nm, the quenched WO42-

emission is dominated by the broad MoO42- emission

band. Likewise, the red emission of Eu3? is further

enhanced shifting the visual color from cool white to warm

white as shown in the inset of Fig. 6c. Concomitantly, the

intensities of the major emission bands of Tb3? (488 and

545 nm) and Tm3? (476 nm) are reduced. Moreover,

emission from the higher excited states of Tb3?

(5D3 ? 7FJ, J = 6, 5, 4) and Tm3? (1D2 ? 3F4) are no

longer visible. It can, thus, be assumed that the MoO42- CT

is not as efficient as that of the WO42- CT in generating

emission from the higher 5D3 and 1D2 excited states of the

Tb3? and Tm3? ions, respectively. As a result, a lower

contribution from the blue region and a higher from the red

region induces a warmer white color appearance. The

overall energy exchange pathways are discussed more in

subsequent sections.

3.5 Excitation spectra of CaW0.99Mo0.01O4:Tm3?,

Tb3?, Eu3?

Figure 8 shows the excitation spectra of the Ln3?-doped

CaW0.99Mo0.01O4 host monitored at 455, 550, and 615 nm

which were the selected transitions of Tm3? (1D2 ? 3F4),

Tb3? (5D4 ? 7F5), and Eu3? (5D0 ? 7F2), respectively.

The excitation spectra monitored at the Tm3? and Tb3?

emissions, consist of broad bands assignable to the host

species and provide direct evidence of energy transfer from

the WO42- and MoO4

2-groups to both Ln3? ions. A

unique situation of dual donor processes are exhibited

where both host groups are involved in the energy transfer

to Tm3? and Tb3? ions within the Ca0.96W0.99Mo0.01O4

host.

In contrast, the excitation spectra collected by moni-

toring the Eu3? emission at 615 nm consist of a broad band

at *282 nm, which may imply that energy transfer takes

place only from the MoO42- group. Prior reports of a

CaMoO4:Eu3?,Sm3? system [33] showed a spectral over-

lap between the MoO42- CT and a Eu3?–O2- CT band.

The excitation within a Eu3?–O2- CT could permit an

electronic transition from a 2p orbital of oxygen to an

empty 4f orbital of Eu3?, resulting in the observance of a

broad excitation band [12, 27, 34]. In considering the Eu3?

Fig. 7 Crystal structure of

undoped CaWO4 (a) and

CaWO4:Mo, Tm, Tb, Eu

(b) (charge compensation in the

positions of Na? and Ca-

vacancies is also suggested)

158 J Sol-Gel Sci Technol (2012) 63:153–161

123

red-shift in Fig. 8, and the absence of the WO42- CT; it can

be deduced that the overlapping Eu3?–O2- and

MoO42-CT transitions in this system favor energy transfer

to Eu3? more than WO42-. Nevertheless, in some instances

the WO42- and Eu3?–O2- CT bands have been reported to

exist concurrently [24, 27, 29].

Additional evidence of the prevalent Eu3?–O2- CT is

supported by the inset of Fig. 8, which shows the excita-

tion spectra of CaWO4:0.01Tm3?, 0.005 Tb3?, 0.005Eu3?

without the inclusion of the Mo dopant. While energy

transfer from WO42- to Tm3? and Tb3? is clearly evident,

the Eu3? excitation band still exhibits a shifted maximum

(259 vs 274 nm), revealing the Eu3?–O2- CT state to be

dominant while the WO42- contribution is negligible. The

overall spectral profile of the Fig. 8 inset appears to

exhibit two independent excitation processes covering two

regions. In the *235–260 nm region the dominant feature

is the CT within the WO42- group, while in

the *265–300 nm region the dominant process is the

Eu3?–O2- CT. Importantly, without the presence of Mo in

this system, white light is only observed under one exci-

tation wavelength (265 nm), while under longer wave-

lengths the red emission of Eu3? is visually dominant.

In the mixed CaW0.99Mo0.01O4:Tm3?, Tb3?, Eu3? system,

the presence of Mo extends the excitation region to where

the blue and green emissions of Tm3? and Tb3? can still

be blended with the red emission of Eu3?, allowing the

white light appearance to be tuned under longer excitation

wavelengths.

3.6 Energy transfer pathways

Figure 9 illustrates the proposed energy transfer scheme

based on the features of the emission and excitation spectra

shown in Figs. 6 and 8. The energy gap between the

WO42- excitation band and the 3P2 (38,315 cm-1) excited

state of Tm3? [19] is estimated to be *445 cm-1. Since

the maximum phonon energy of the WO42- host lattice

is *900 cm-1 [24, 35], this condition creates an ideal

donor/acceptor energy match-up where the gap is easily

bridged via one phonon-assisted process. The nonradiative

(NR) energy transfer from WO42- to the 3P2 excited level

of Tm3? is efficient enough to generate emission from both

the 1D2 and 1G4 excited states, albeit the emission from the1D2 level is much weaker (Fig. 6a, b).

The energy gap comparison of the MoO42- CT

(34,480 cm-1) indicates that the donor level matches well

with the 3P0 (33,710 cm-1) excited level of Tm3?; thereby,

the energy gap of *770 cm-1 can also promote energy

transfer via one phonon-assisted process. However, the

overall efficiency of the energy transfer from MoO42-

appears to be reduced in comparison to WO42-. As a result,

emission from the higher 1D2 level is absent (Fig. 6c), and

emission from the lower 1G4 level is weak when compared

to the emission profiles in Fig. 6a, b.

A similar excitation mechanism is observed in the

energy transfer process to the Tb3? ion. In Fig. 9, it can be

seen that the WO42- CT has a good match with the 5KJ

levels of Tb3? located at *37,880 cm-1. Hence, energy

transfer is likely to these levels from the WO42- CT state.

The donor/acceptor gap is estimated to be *880 cm-1

which can easily be bridged by the assistance of one

phonon. The analysis of Fig. 6 suggests that a highly

efficient energy transfer process takes place from WO42-

since the acceptor Tb3? ion emits from both the 5D3 and5D4 levels. Moreover, the 5D4 ? 7F5 (545 nm) transition

of Tb3? exhibits the overall most intense band in Fig. 6b

when energy is pumped from both the WO42- and MoO4

2-

CT states.

Comparatively, the 5HJ levels of Tb3? are also in

vicinity to be pumped by the MoO42- CT. However, the

overall efficiency of the energy transfer is lower than

WO42-, similar to that observed in the Tm3? scenario. This

conclusion is based on the following observations from

Fig. 6c: (1) the decreased emission intensity of the5D4 ? 7F6 (488 nm) and 5D4 ? 7F5 (545 nm) transitions,

(2) the slight increase of the MoO42- emission intensity,

and (3) the absence of the high-energy transitions from5D3 ? 7FJ (J = 6, 5, 4). Unfavorably, instrumental limi-

tations restricted the ability to conduct lifetime measure-

ments in the short wavelength UV regions (\300 nm),

which would have provided a clear understanding of the

energy transfer efficiency.

Fig. 8 Room temperature excitation spectra of Ca0.96W0.99

Mo0.01O4:0.01Tm3?, 0.005Tb3?, 0.005Eu3?, 0.02Na? monitored at

455, 550, and 615 nm. The inset shows the excitation spectra of

CaWO4:0.01Tm3?, 0.005Tb3?, 0.005Eu3?, 0.02Na? without the

inclusion of Mo dopant. (to avoid 2nd order effect to the PMT tube,

spectral regions between 270 and 280 nm (Tb3?) and 290–315 nm

(Eu3?) were bypassed)

J Sol-Gel Sci Technol (2012) 63:153–161 159

123

As mentioned in earlier sections, the energy transfer

from WO42- to Eu3? appears to be negligible. The broad

excitation band observed in Fig. 8 when monitoring the

Eu3? f–f emission is assignable to contributions from both

the Eu3?–O2- and the MoO42- CT transitions. As indi-

cated in Fig. 9, both of these transitions lead to the char-

acteristic emissions of Eu3?. The absence of the WO42-

CT band in the excitation spectra is indicative that the

longer-wavelength excitation band of Fig. 8 can be

assigned to the dominant Eu3?–O2- and MoO42- CT

states. Hence, warm white light is observed due to the

enhanced red emission of Eu3? under longer excitation

wavelengths.

3.7 CIE chromaticity

Figure 10 shows the Commission Internationale de

I’Eclairage (CIE) chromaticity coordinates (x, y) and cor-

related color temperatures (Tc) of the Ca0.96W0.99

Mo0.01O4:0.01Tm3?, 0.005Tb3?, 0.005Eu3?, 0.02Na?

phosphor under varied excitation wavelengths. It can be

clearly seen that under excitation at 255 and 260 nm, the

respective chromaticity coordinates (0.3003, 0.3281) and

(0.3316, 0.3520) exhibit an excellent closeness to the D65

sunlight white point (0.3127, 0.3290). Moreover, the high

Tc values under these excitation wavelengths make the

phosphor suitable for cool and natural white light. When

the excitation wavelengths are changed to 265 and 280 nm,

the Tc values are lowered significantly to color tempera-

tures appropriate for warm white light. Ultimately, the

simultaneous emission of the red, green, and blue-emitting

Ln3? ions in the CaWO4 system is improved by the

inclusion Mo. The MoO42- groups not only contribute

additional emission, but also extend the excitation wave-

lengths under which the white light emission is perceived,

promoting tunable levels of chromaticity. Since warm

white light appeals to human emotion for use in home

lighting, and cool white light is preferred for public light-

ing [36], the tunability of this single-phase emitting phos-

phor exhibits promising applications for SSL.

WO42- CT

MoO42- CT

WO42- CT

MoO42- CT

NR NR3P2

3P0

1D2

1G4

3H4

3H5

3F4

NR

3H6

5L6

5D2

5D0

6

4

7F0

5D4

5D3

5D1

4f75d1

Tm3+ Eu3+ Tb3+

NR

616n

m

592n

m

476n

m

476n

m

455n

m

382n

m

414n

m43

9nm

48

8nm

545n

m

545n

m

488n

m

260n

m

260n

m

290n

m

290n

m

290n

m

290n

m

285n

m

5F45HJ

5KJ

7F6

0

4

5

CTEu3+ O2-

Fig. 9 Energy transfer pathways in Ca0.96W0.99Mo0.01O4:0.01Tm3?, 0.005Tb3?, 0.005Eu3?

x y0.3127 0.3290

0.2864 0.3042

0.3003 0.3281

0.3316 0.3520

0.3547 0.3742

0.3917 0.3949

D65 White Point

λex 250nm

λex 255nm

λex 260nm

λex 265nm

λex 280nm

λex 250nm = 8661 °K

λex 255nm = 7207 °K

λex 260nm = 5529 °K

λex 265nm = 4745 °K

λex 280nm = 3840 °K

y

x

Correlated Color Temperature (Tc)D65 = 6504°K

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Fig. 10 CIE chromaticity diagram for Ca0.96W0.99-

Mo0.01O4:0.01Tm3?, 0.005Tb3?, 0.005Eu3? showing (x, y) coordi-

nates, correlated color temperature, and photographs of phosphor

powder under various excitation wavelengths

160 J Sol-Gel Sci Technol (2012) 63:153–161

123

4 Conclusions

A single phase white light-emitting CaW1-xMoxO4:Tm3?,

Tb3?, Eu3? phosphor has been successfully synthesized via

a Pechini sol–gel process. The Mo6?, Ln3?, and compen-

sating Na? ions were uniformly incorporated into the

scheelite phase CaWO4 host. The excitation and emission

spectra reveal that introducing Mo into the CaWO4 lattice

extends the excitation range, and induces a dual energy

transfer from WO42- and MoO4

2- groups to Tm3? and

Tb3? ions. The energy transfer schemes have been pro-

posed and further experiments examining the temperature

dependent transfer behaviors and the dynamics of the

excited states could help to verify these proposed mecha-

nisms. Due to the different excitation mechanisms of Eu3?,

the resulting white light emission is made tunable from

cool white to warm white by changing the excitation

wavelength as confirmed by chromaticity coordinates. By

application to shortwave UV-LEDs, the CaW1-xMoxO4:

Tm3?, Tb3?, Eu3? phosphor is a suitable candidate for

both cool and warm WLEDs.

Acknowledgments This research is supported by the NASA Har-

riett G. Jenkins Predoctoral Fellowship and the NOAA-Education

Partnership Program (Grant no. NA06OAR4810187). The authors

extend thanks to Dr. Arona Diouf for initial advice in this research,

Dr. Zhigang Xu and Kwadwo Mensah–Darkwa for SEM analysis, and

Dr. Sergey Yarmolenko for XRD assistance.

References

1. Nakamura S, Fasol G (1997) The blue laser diode: GaN based

light emitters and lasers. Springer, Berlin

2. Ye S, Xiao F, Pan YX, Ma YY, Zhang QY (2010) Mater Sci Eng,

R 71(1):1–34. doi:10.1016/j.mser.2010.07.001

3. Chen L, Lin C–C, Yeh C-W, Liu R-S (2010) Mater 3(3):

2172–2195. doi:10.3390/ma3032172

4. Krames MR, Shchekin OB, Mueller-Mach R, Mueller GO, Ling

Z, Harbers G, Craford MG (2007) J Disp Technol 3(2):160–175.

doi:10.1109/JDT.2007.895339

5. Wang B, Sun L, Ju H (2010) J Sol-Gel Sci Technol

53(2):454–458. doi:10.1007/s10971-009-2120-1

6. Kim JS, Jeon PE, Park YH, Choi JC, Park HL, Kim GC, Kim TW

(2004) Appl Phys Lett 85(17):3696–3698. doi:10.1063/

1.1808501

7. Kim JS, Lim KT, Jeong YS, Jeon PE, Choi JC, Park HL (2005)

Solid State Commun 135:21–24. doi:10.1016/j.ssc.2005.03.068

8. Wenbo M, Shi Z, Wang R (2010) J Alloys Compd 503(1):

118–121. doi:10.1016/j.jallcom.2010.04.213

9. Hou Z, Chai R, Zhang M, Zhang C, Chong P, Xu Z, Li G, Lin J

(2009) Langmuir 25(20):12340–12348. doi:10.1021/la9016189

10. Cavalli E, Boutinaud P, Mahiou R, Bettinelli M, Dorenbos P

(2010) Inorg Chem 49(11):4916–4921. doi:10.1021/ic902445c

11. Zalga A, Sazinas R, Garskaite E, Kareiva A, Bareika T, Tamu-

laitis G, Juskenas R, Ramanauskas R (2009) Chemija 20(3):

169–174

12. Wang W, Yang P, Cheng Z, Hou Z, Li C, Lin J (2011) ACS Appl

Mater Interfaces. doi:10.1021/am2008008

13. Lei F, Yan B (2008) J Solid State Chem 181(4):855–862. doi:

10.1016/j.jssc.2008.01.033

14. Gao D, Li Y, Lai X, Wei Y, Bi J, Li Y, Liu M (2011) Mater Chem

Phys 126(1–2):391–397. doi:10.1016/j.matchemphys.2010.10.

053

15. Wang W, Yang P, Gai S, Niu N, He F, Lin J (2010) J Nanopart

Res. doi:10.1007/s11051-010-9850-4

16. Grobelna B, Lipowska B, Klonkowski AM (2006) J Alloys

Compd 419(1–2):191–196. doi:10.1016/j.jallcom.2005.07.078

17. Zhu F, Xiao Z, Yan L, Zhang F, Huang A (2010) Appl Phys A

101(4):689–693. doi:10.1007/s00339-010-5950-3

18. Blasse G, Grabmarier BC (1994) Luminescent materials. Springer,

Berlin, p 103

19. Liao J, Qiu B, Wen H, Chen J, You W, Liu L (2009) J Alloys

Compd 487(1–2):758–762. doi:10.1016/j.jallcom.2009.08.068

20. Liao J, Liu L, You H, Huang H, You W (2011) Optik-Int J Light

Electron Opt. doi:10.1016/j.ijleo.2011.07.002

21. Nazarov MV, Jeon DY, Kang JH, Popovici E-J, Muresan L-E,

Zamoryanskaya MV, Tsukerblat BS (2004) Solid State Commun

131:307–311. doi:10.1016/j.ssc.2004.05.025

22. Zhu F, Xiao Z, Zhang F, Yan L, Huang A (2011) J Lumin

131(1):22–24. doi:10.1016/j.jlumin.2010.08.019

23. Zheng Y, Huang Y, Yang M, Guo N, Qiao H, Jia Y, You H

(2012) J Lumin 132(2):362–367. doi:10.1016/j.jlumin.2011.

09.010

24. Pang ML, Lin J, Wang SB, Yu M, Zhou YH, Han XM (2003)

J Phys: Condens Matter 15:5157–5169. doi:10.1088/0953-8984/

15/29/327

25. Yang P, Quan Z, Li C, Lian H, Huang S, Lin J (2008) Micro-

porous Mesoporous Mater 116(1–3):524–531. doi:10.1016/

j.micromeso.2008.05.016

26. Li X, Yang Z, Guan L, Guo J, Wang Y, Guo Q (2009) J Alloys

Compd 478(1–2):684–686. doi:10.1016/j.jallcom.2008.11.109

27. Shi S, Gao J, Zhou J (2008) Opt Mater 30(10):1616–1620. doi:

10.1016/j.optmat.2007.10.007

28. Xie A, Yuan X, Hai S, Wang J, Wang F, Li L (2009) J Phys D

Appl Phys 42(10):105107. doi:10.1088/0022-3727/42/10/105107

29. Su Y, Li L, Li G (2008) Chem Mater 20(19):6060–6067. doi:

10.1021/cm8014435

30. Teshima K, Yubuta K, Sugiura S, Fujita Y, Suzuki T, Endo M,

Shishido T, Oishi S (2006) Cryst Growth Des 6(7):1598–1601.

doi:10.1021/cg050673z

31. Tyson RM, Hemphill WR, Theisen AF (1988) Am Min 73:

1145–1154

32. Nazarov MV, Tsukerblat BS, Popovici EJ, Jeon DY (2004) Phys

Lett A 330(3–4):291–298. doi:10.1016/j.physleta.2004.07.071

33. Jin Y, Hao Z, Zhang X, Luo Y, Wang X, Zhang J (2011) Opt

Mater 33(11):1591–1594. doi:10.1016/j.optmat.2011.04.00934. Yu Q, Liu Y, Wu S, Lu X, Huang X, Li X (2008) J Rare Earths

26(6):783–786. doi:10.1016/S1002-0721(09)60005-3

35. Thongtem T, Phuruangrat A, Thongtem S (2008) Appl Surf Sci

254(23):7581–7585. doi:10.1016/j.apsusc.2008.01.092

36. Lakshminarayana G, Yang R, Qiu JR, Brik MG, Kumar GA,

Kityk IV (2009) J Phys D Appl Phys 42(1):015414. doi:10.1088/

0022-3727/42/1/015414

J Sol-Gel Sci Technol (2012) 63:153–161 161

123