Spectral properties and luminescence mechanism of red emitting BCNO phosphors

Abstract

Red emitting (λ_em = 620 nm) BCNO phosphors were synthesized at 650 °C with solid state reaction method using boric acid and hexamethy lenetetramine as raw materials. The BCNO phosphors have turbostratic boron nitride structure and particle sizes are in micro scale. Carbon and oxygen elements were bonded to boron and nitrogen to form BCNO phosphors. The emission peaks were shifted from blue light (420–470 nm) to red light (590–620 nm) with increasing sintering temperature, heating time and the ratio of boric acid to hexamethy lenetetramine, which was induced by partially formed BCNO and completely formed BCNO phosphors. The decay curves and emission spectra indicated that the red emission was induced by two luminescence centers, corresponding to longer lifetime τ₁ and short lifetime τ₂. The ultraviolet visible absorption spectra disclosed that the optical band gap was changed from 1.75 eV to 2.0 eV with different preparation conditions. The high temperature emission spectra suggested that the nitrogen defects levels served as electron traps and attended the red emission. The luminescence mechanism of BCNO phosphors was stated by a simplified energy level diagram. The red emission BCNO phosphors have good thermal stability and great potential application on lighting, display, solar cell and biomedical fields.

---

Introduction

Red emitting phosphors have been paid much attention due to their ability to improve the color rendering index (CRI) and produce warm white light for white light emitting diodes (LEDs). Up to now, different red emitting phosphors have been developed such as Eu doped nitrogen compound (Sr₂Si₅N₈:Eu²⁺)^1a and sulfides (CaS:Eu²⁺),^1b Mn doped compounds (CaMg₂Al₁₆O₄:Mn⁴⁺)^1c and some quantum dot materials (CdSe/CdS).^1d It is still a challenge to develop new red emitting phosphors with pollution free, economize and energy saving. Hexagonal born nitride (BN) have been attracted much attention since the first observation of an intense far ultraviolet (FUV) excitation emission.² Owing to the wide band gap of BN (∼5.9 eV), its luminescence was limited in short wavelength (∼215 nm). A great deal of work has been done for broadening the emission spectra of BN in visible light.³ A recent development of boron nitride based phosphors was performed by T. Ogi with preparing carbon and oxygen codoped boron nitride (BCNO) phosphors via urea combustion method.^4a The BCNO phosphors can be synthesized from inexpensive and environmentally friendly raw materials with wide range of excitation from short ultraviolet to blue light (250–430 nm) and emission spectra from violet to near-red regions (387–571 nm), and high quantum yields.

BCNO phosphors have been focused on owing to its great potential application in white light emitting diodes (LEDs), display, biology and medical imaging areas.^4b At present, the preparation, spectral properties and luminescence mechanism of BCNO phosphors have been investigated preliminarily. For preparation, different methods were adopted to synthesize BCNO phosphors with various morphologies. Singh et al.^5a prepared BCNO nanophosphor by solid state reaction method at 1400 °C with boric acid and melamine (λ_em = 357 nm). Lei et al.^5b synthesized BCNO nanoparticles in a eutectic LiCl/KCl salt melt using sodium borohydride and urea as raw materials (λ_em = 440–528 nm). Our group used chemical method^5c–e to prepare BCNO phosphors in micrometer size. In addition, BCNO nanofibers and porous microbelts were also synthesized by electrospinning^5f and one-step template-free method, respectively.^5g For spectral properties, Kaihatsu et al.^6a investigated the effects of carbon source, sintering time and temperature, ratio of raw materials on luminescence spectra of BCNO phosphors. The emission spectra were modulated from 387 nm to 570 nm with increasing ratio of carbon source to boron source (0–1.02), and the emission spectra were shifted from 420 nm to 550 nm with decreasing ratio of nitrogen source to boron source (9.96–1.0). Ogi et al.^6b disclosed the effects of polymer (polyethyleneimine, polyallylamine, tetraethylene glycol) on the quantum yields and photoluminescence properties of BCNO phosphors, and the emission of BCNO can be tuned from 380 to 490 nm by changing the reaction and polymer concentration. Lu et al.^6c studied the influence of annealing temperature and ambient atmosphere on the photoluminescence properties of BCNO phosphors. Chu et al.^6d tuned emission spectra of BCNO phosphor from blue to yellow by doping grapheme oxide into BCNO phosphors. For luminescence mechanism, there are mainly two points of view to clarify the luminescence origin of BCNO phosphors. One argument is that the luminescence of BCNO phosphor was induced by the closed shell BO₂⁻ and BO⁻ species.^7a,b Another suggestion^7c is defects energy level transition which led to the luminescence of BCNO phosphors. At present, the mainly method to modulate the emission spectra was selected proper carbon source and increasing content of carbon source. However, it is difficult to modulate the emission peak to red light because the emission peak will shift back to short wavelength with too much carbon source. In addition, it was often adopted three sources (boron, nitrogen, carbon sources) to synthesize BCNO phosphor. The selection of carbon source is important and it is hard to control the carbon source during the preparation of BCNO phosphor. Up to now, there is no report to synthesize red emission BCNO phosphors with two sources, much less to investigate the spectral properties and luminescence mechanism of red emission BCNO phosphors. In addition, it is easier to control the reaction process and modulating spectral properties by changing the two sources, and solid state reaction method is favourable for realizing quantity production and industrialization.

In this paper, we prepared red emitting BCNO phosphors with solid state reaction method using boric acid and hexamethy lenetetramine as raw materials. The chemical bond status was investigated by X-ray photoelectrical spectra (XPS) and infrared spectra. The excitation spectra, emission spectra, and decay process were systematically measured to investigate the luminescence properties of BCNO phosphors. The optical band gap were studied by ultraviolet visible (UV-vis) absorption spectra and the high temperature emission spectra were used to investigate the thermal stability of BCNO phosphors and the electron transition process under heating situation. In addition, the luminescence mechanism of red emitting BCNO phosphors was also investigated.

Experimental procedures

The BCNO phosphors were prepared by a solid state reaction method using boric acid and hexamethy lenetetramine as raw materials. The proper carbon, nitrogen concentration and the inherent C–N bonds with heterocyclic structure was favourable for hexamethy lenetetramine using as a good choice of carbon and nitrogen source. Firstly, the boric acid (H₃BO₃, B source) and hexamethy lenetetramine (C₆H₁₂N₄, C and N source) were weighed and ground sufficiently in an agate mortar. Then, the ground raw materials were pressed into a disc with 10 mm in diameter (2 mm thick) and sintered at 600–700 °C for 4–24 h in a ceramic crucible under ambient atmospheric pressure. The air condition was helpful for synthesizing BCNO phosphors and modulating the carbon and oxygen concentration in BCNO phosphors. Three series of BCNO samples were prepared with various sintering temperatures, heating times and different mole ratios of boron acid and hexamethy lenetetramine. For the first series of BCNO phosphor, the mole ratio of boron acid to hexamethy lenetetramine (R_B/H) was fixed at R_B/H = 1 : 1, and the sintering temperatures (T_S) were changed from T_S = 600 °C to T_S = 700 °C for t = 12 h. For the second series of BCNO phosphors, the sintering temperature was fixed at 650 °C and R_B/H = 1 : 1, while the heating times were varied from t = 4 h to t = 24 h. For the third series of BCNO phosphors, the sintering temperature T_S = 650 °C and heating time t = 12 h, and mole ratios of boric acid (fixed at 0.02 mol) and hexamethy lenetetramine were changed from R_B/H = 1 : 0.1 to R_B/H = 1 : 5. In addition, the high temperature (T = 300–500 K) emission spectra of the BCNO phosphor synthesized with R_B/H = 1 : 1, T_S = 650 °C, and t = 12 h were studied.

The phase structure of BCNO phosphors was characterized by powder X-ray diffraction (XRD) (Bruker Model D8). The morphology of BCNO phosphors was measured by a scanning electronic microscope (Hitachi, S-4800). X-Ray photoelectron spectroscopy (XPS, PHI1600EXCA) was used to characterize the chemical state of BCNO phosphors. The chemical bonding status was measured by Fourier transform infrared (FTIR) spectrometer (Bruker, WQF-410). The ultraviolet-visible (UV-vis) absorption spectra were measured by a spectrophotometer (Hitchi, U-3900H). The excitation spectra, emission spectra and decay curves of BCNO phosphors were measured by a steady and transient state spectrophotometer (Horiba, FL-3-22). The high temperature emission spectra in the range of T = 300–500 K were also measured by steady and transient state spectrophotometer with a high temperature attachment.

Results and discussions

Fig. S1† shows the XRD pattern and SEM image of the BCNO phosphor sintered at 650 °C for 12 h with R_B/H = 1 : 1. There are two broad peaks centered at 22.2° and 43.1°, which is induced by the BCNO phosphor with turbostratic boron nitride (t-BN) structure. The obvious peak at 22.2° was related to (002) reflection, and the peak around 43.1° was originated from (100) and (101) reflections, which indicated the formation of BCNO phosphors.^8a In addition, there were two sharp peaks centered at 14.6° and 27.9° corresponding to (111) and (310) reflections of B₂O₃ with cubic structure, respectively. The cubic B₂O₃ was produced by the thermal decomposition of unreacted boric acid. The SEM image (inset of Fig. S1(a)†) displayed the BCNO phosphor was irregular in shape, and the particle size was in the range of several micrometers to one hundred micrometers. From the XRD and SEM results, it suggested that the prepared samples consisted of BCNO phosphor and B₂O₃, and the particle size was in micro scale.

Fig. S2† displays XPS spectra and FTIR spectrum in the range of 400–2500 cm⁻¹ for BCNO phosphor prepared at 650 °C for 12 h with R_B/H = 1 : 1. The XPS spectra indicated the prepared phosphors were consist of B, C, N, and O elements.^8b For B1s spectrum, it could be fitted well with two Gaussian curves centered at 193.0 eV and 190.9 eV, corresponding to B–O and B–N bonds, respectively,^5a and no B–C signals could be observed in B1s spectrum. The C1s spectrum was fitted with three components centered at 286.2, 284.5, and 281.7 eV which were related to C≡N, C–C and C=O bonds, respectively.^8b–e In the N1s spectrum, the peak at 398.4 eV was originated from N–B bonds, while the shoulder peak at 399.9 eV was owing to the formation of C≡N bonds.^8c,d For the O1s spectrum, it was fitted well with only one Gaussian curve centered at 532.8 eV which was assigned to O–B bonds.^8f The chemical bond states of BCNO phosphors can also be verified by FTIR spectra, as shown in Fig. S2(e).† The absorption at 785 and 1377 cm⁻¹ was related to B–N–B vibration and B–N stretching mode, which indicated the formation of B–N linkage in BCNO phosphor.^7a,8g The weak B–N stretching bands and B–N–B bands were induced by the partially crystalline BCNO phosphors.^8h The B–C vibration peak was observed at 1196 cm⁻¹, and the absorption peaks centered at 1654, 2264, 2362 cm⁻¹ were assigned to C=C, C=O and C≡N bonds, respectively.^5f,8i The vibration at 544, 644, 734, and 1448 cm⁻¹ were originated from B–O bonds, and the absorption at 885 cm⁻¹ corresponded to stretching vibration of tetrahedral BO₄⁻ units in B₂O₃ crystal.^8j The N–B–O stretching bands in 900–1100 cm⁻¹ were also observed, which was induced by the oxidation of N–B bonds.^7a From the XPS and FTIR, it could be concluded that C and O elements were involved in the structure of prepared samples and bonded to B and N to form BCNO phosphors.

The influence of heating temperature and time on excitation and emission spectra for BCNO phosphors were investigated. Fig. 1(a) shows the normalized excitation (λ_em ∼ 610 nm) and emission spectra (λ_ex = 370 nm) of BCNO phosphors synthesized with R_B/H = 1 : 1 at different sintering temperatures for 12 h. The BCNO phosphors had wide excitation spectra range which was between 300 nm and 570 nm, and the excitation spectra range was not sensitive to the sintering temperature. There were three excitation peaks centered at 370 nm, 470 nm and 550–570 nm. The 370 nm excitation was induced by the transition from the valence band to nitrogen defects levels.^9a The 470 nm and 550–570 nm excitation was originated from the transition from carbon and oxygen impurity levels to conduction bands and nitrogen defects levels, respectively. The 550–570 nm excitation peaks shifted from 550 nm to 570 nm first with increasing heating temperature from T_S = 600 °C to 650 °C and then the excitation peaks shifted back to 554 nm with further increasing sintering temperature. The shift of excitation peaks was related to the energy levels' position which was originated form carbon and oxygen impurity defects. With increasing temperature, the position of carbon and oxygen related impurity levels was changed and led to the shift of excitation peaks, which could be reflected by UV-vis absorption spectra. In addition, the relative excitation peak intensity of 550–570 nm was increased first and then decreased with the enhancement of sintering temperature, which might be resulted from the varied concentration of carbon and oxygen impurity. For the emission spectra, the emission peak position shifted from 465 nm for T_S = 600 °C to 620 nm for T_S = 650 °C, and then the emission peak shifted to 602 nm with further increasing temperature. For the low sintering temperature (T_S = 600 °C), BCNO phosphors were not completely formed, and the luminescence is dominated by blue emission centered at 470 nm. However, a shoulder emission peak in red emission range was also observed on the emission spectra for the BCNO phosphor. With increasing temperature, the blue emission intensity was decreased rapidly and the red emission became the dominate luminescence center in BCNO phosphors. The red emission intensity of BCNO phosphors was increased with heating temperature first and reached a maximum value at 675 °C and then was decreased, which may be related to the crystallinity of BCNO phosphors. The slightly emission peak shift from 620 nm to 602 nm was induced by the change of carbon and oxygen impurity concentration and the shift of the impurity levels with increasing temperature.^5c The luminescence spectra of BCNO phosphors were also affected by heating time. Fig. 1(b) displays the normalized excitation and emission spectra of BCNO phosphors with R_B/H = 1 : 1 sintered at 650 °C for different heating times. For t = 4 h, there were two emission peaks centered at 473 nm and 605 nm, which was induced by the partially crystal BCNO phosphor. With increasing time, the amount of BCNO was increased and the crystallinity of BCNO phosphor was improved, which led to the disappearance of blue emission and the appearance of dominated red emission. The intensity of red emission peak was enhanced with increasing heating time from t = 4 h to t = 12 h and then it was decreased with further extending heating time. The red emission peak was slightly shifted from 622 nm for t = 8 h to 611 nm for t = 24 h, which indicated the carbon and oxygen impurity levels' position was not very sensitive to longer heating time. The blue shift of emission peak might be induced by the change of optical band gap, which could be reflected from absorption spectra and optical band gap value. The range and tendency of excitation spectra for BCNO phosphor with different sintering times were very similar to that of BCNO phosphors prepared with different sintering temperatures.

Fig. 1 Normalized excitation and emission spectra of BCNO phosphors prepared with (a) different sintering temperatures and (b) various heating times with R_B/H = 1 : 1.

Fig. 2 shows the normalized excitation and emission spectra of BCNO phosphors synthesized at 650 °C for 12 h with different ratios of boric acid and hexamethy lenetetramine (R_B/H = 1 : 0.1–1 : 5). For R_B/H = 1 : 0.1, the emission of BCNO phosphor was dominated by blue emitting centered at 410 nm and 445 nm, which was induced by the carbon doped B₂O₃ that was similar to carbon doped Al₂O₃.^9b In addition, the excitation spectra was in the range of ultraviolet and centered at 275 nm, which was induced by the transition of carbon impurity levels to conduction band. With increasing content of hexamethy lenetetramine (R_B/H = 1 : 0.3), the boric acid become not too much and more BCNO phosphors were formed, which led to the decrease of blue emission and increase of red emission intensity. With further increasing hexamethy lenetetramine (R_B/H ≥ 1 : 0.5), the blue emission was disappeared and there was only red emission of BCNO phosphors. The red emission peak position was shifted from 597 nm for R_B/H = 1 : 0.5 to 620 nm for R_B/H = 1 : 1, and then the red emission peak took a blue shift to 590 nm for R_B/H = 1 : 5 sample. The red emission intensity was increased first and then it started to decrease at R_B/H = 1 : 3, which might be related to the concentration and band structure of BCNO phosphors. The shift of emission spectra with increasing R_B/H was induced by the change of carbon related impurity levels' position. For R_B/H ≤ 1 : 1, the carbon impurity levels would shift towards to conduction bands with increasing carbon impurity concentration, which led to the red shift of emission spectra. For R_B/H > 1 : 1, there were too much carbon impurity concentration and formation of carbon residue on the product might cause the blue shift of emission spectra.^6a The excess carbon was not effectively doped in BCNO phosphors and the floating carbon impurity on surface of BCNO phosphors had negative effects on red emission of BCNO phosphors, which might lead to the blue shift of emission spectra for too much carbon. For excitation spectra, there were also three excitation peaks centered at 370, 470 and 533–570 nm for R_B/H ≥ 1 : 0.3. The relative excitation peak intensity for 470 nm and 533–570 nm were increased from R_B/H = 1 : 0.3 to R_B/H = 1 : 1, and then the relative excitation intensity was reduced with the further increase of R_B/H. The excitation peak position at 533–570 nm was also related to the change of carbon related impurity levels' position.

Fig. 2 Normalized excitation spectra and emission spectra of BCNO phosphors prepared at 650 C for 12 h with (a) R_B/H = 1 : 0.1–1 : 0.9 and (b) R_B/H = 1 : 1–1 : 5.

We also investigated the effects of preparation conditions on life times for BCNO phosphors. Fig. 3 displays the red emission peak decay curves of BCNO phosphors prepared with different sintering temperatures (T_S = 600–700 °C) and times (t = 4–24 h), and various ratios of boric acid and hexamethy lenetetramine (R_B/H = 1 : 0.5–1 : 5). All the decay curves can be fitted well by dual exponential function as the formula:^9c

I(t) = I₁ exp(−t/τ₁) + I₂ exp(−t/τ₂) (1)

where I₁ and I₂ are constants and their values are emission intensities measured at t = 0. τ₁ and τ₂ are the lifetimes of the two channels responsible for the decay process. The inset of Fig. 3(a)–(c) show the τ₁ and τ₂ as a function of sintering temperature, heating time and 1/R_B/H, respectively. For samples sintered at T_S = 600–700 °C, the τ₁ of BCNO phosphors was in the range of 6.4–51.1 ns, and τ₂ was in the range of 0.9–3.4 ns. The general tendency of τ₁ was increased first and then decreased with enhancement of sintering temperature, and the τ₂ was not very sensitive to the sintering temperature. For BCNO phosphors with t = 4–24 h, the τ₁ of BCNO phosphors were changed between 15.6 ns and 51.9 ns, and τ₂ was in the range of 1.8–3.6 ns. The tendency of τ₁ and τ₂ for BCNO phosphors with different heating times was similar to that of BCNO phosphors sintered at various sintering temperatures. For the BCNO phosphors with R_B/H = 1 : 0.5–1 : 5, the τ₁ of BCNO phosphors was changed between 13.7 ns and 66.5 ns, and τ₂ was varied in the range of 2.3–3.3 ns. There were two lifetime peaks appeared at τ₁ = 34.4 ns with R_B/H = 1 : 0.5 and τ₁ = 66.5 ns with R_B/H = 1 : 4, respectively. The change of life times τ₁ and τ₂ with preparing condition might be related to the carbon and oxygen impurity concentration during the formation of BCNO phosphors.^5c From the decay curves and fitting results of τ₁ and τ₂ for BCNO phosphors, it could be concluded that the red emission was induced by two luminescence centers, corresponding to longer lifetime τ₁ and short lifetime τ₂. The two luminescence centers could also be verified by red emission spectra of BCNO phosphors, there was an obvious shoulder emission peak on red emission spectra and the red emission spectra could be fitted well by two Gaussian curves (not shown).

Fig. 3 Decay curves of BCNO phosphors prepared with (a) T_S = 600–700 °C, (b) t = 4–24 h and (c) R_B/H = 1 : 0.5–1 : 5, the inset of (a)–(c) gives lifetimes versus sintering temperature, heating time and ratios of hexamethy lenetetramine and boric acid, respectively.

The UV-vis absorption spectra and optical band gaps of BCNO phosphors prepared with different sintering temperatures and times, and various ratios of boric acid and hexamethy lenetetramine were shown in Fig. 4. The optical band gap values were calculated by the relation as below:^9d

αhν = A(hν − E_g)^1/2 (2)

where α is the absorbance, A is a constant and E_g is the direct band gap energy. From Fig. 4(a), it could be seen that the absorption range was between 210 nm and 700 nm, and the absorption spectra range were increased first and then decreased with the increase of sintering temperature. The optical band gap (shown in Fig. 4(b)) was changed from 1.84 eV to 1.97 eV as sintering temperature was in the range of 600–700 °C, which was induced by the absorption between carbon related defects' levels and nitrogen defects' levels.^8h The change of the optical band gap value was related to the shift of carbon related defects' levels. For Fig. 4(c), the absorption spectra range was improved first and then reduced with the enhancement of heating time, and the optical band gap value was increased from 1.75 eV for t = 4 h to 1.92 eV for t = 8 h, and then E_g was decreased to 1.88 eV for t = 24 h. The first increase of band gap value was related to the partially formation of BCNO phosphors for t = 4 h to completely formation of BCNO phosphor for t = 8 h, and the slightly decrease of band gap with further increasing time was originated from the change of carbon related defects concentration and shift of their levels towards to conduction band. From Fig. 4(e) and (f), one could see that the absorption range was also in the range of 210–700 nm. However, the tendency of optical band gap was different with Fig. 4(b) and (d). The band gap value was increased from 1.76 eV for R_B/H = 1 : 0.5 to 1.95 eV for R_B/H = 1 : 0.7, and then it was increased slowly to 2.01 eV for R_B/H = 1 : 5. The first quickly enhancement of band gap was related to the increase content of BCNO phosphors, and the slowly increase of band gap with R_B/H was resulted from the change of carbon and nitrogen related defects' levels position. For smaller ratio of boric acid to hexamethy lenetetramine, the carbon impurity concentration was higher and their related levels were shifted towards conduction band, which resulted in a smaller band gap value. With increasing R_B/H, the relative carbon concentration was decreased and their impurity levels were shifted away from conduction band, which led to the increase of optical ban gap value. The broad absorption of BCNO phosphors in visible light range might be used in solar cell to improve its light absorption and conversion efficiency.

Fig. 4 UV-vis absorption spectra of BCNO phosphors synthesized with (a) T_S = 600–700 °C, (c) t = 4–24 h and (e) R_B/H = 1 : 0.5–1 : 5, the optical band gap values as a function of (b) sintering temperature, (d) heating time, and (f) ratio of hexamethy lenetetramine to boric acid.

To further investigated the luminescence properties and thermal stability of red emitting BCNO phosphors, the high temperature emission spectra of BCNO phosphor were measured between T = 300 K and T = 500 K, as shown in Fig. 5. Fig. 5(a) shows the high temperature emission spectra excited by 370 nm for BCNO phosphor synthesized at 650 °C for 12 h with R_B/H = 1 : 1. The red emission peak position was not influenced by the enhancement of temperature, which indicated that the transition levels' position was not changed by the sample temperature. However, the emission peak intensity was affected by the sample temperature, as shown in Fig. 5(b). The emission peak intensity was slightly decreased for T ≤ 400 K. With enhancing sample temperature, the emission peak intensity was increased and reached a maximum value at T = 440 K (∼170 °C), and then the emission peak intensity started to reduce with increasing sample temperature. The tendency of red emission peak intensity under various sample temperatures was correlated with the luminescence process of the BCNO phosphors. In BCNO phosphors, carbon related impurity was mainly served as donor impurity levels which provided electrons in the transition process.^2c,9d,10a The carbon related impurity levels might shift towards to conduction band with increasing carbon related impurity concentration. In addition, there were lots of nitrogen defects such as nitrogen vacancies in BCNO phosphors. The nitrogen vacancies could be served as electron traps and their defect levels were located at 0.7–1.0 eV below the conduction band.^10b,c For 370 nm (∼3.4 eV) excitation, the electrons would be excited from carbon related impurity levels to conduction band. Then the electrons would jump to nitrogen vacancy levels. Some electrons would be trapped by nitrogen vacancies, and some electrons would jump from the nitrogen vacancy levels to carbon related impurity levels. The red emission of BCNO phosphors was induced by the transition between nitrogen vacancy levels and carbon related impurity levels (∼2.0 eV). For low heating temperatures (T < 410 K), the electron trapping and electron transition kept a fine equilibrium. With increasing temperature (410 K ≤ T ≤ 440 K), the equilibrium was destructed and the trapped electrons would also attend the transition from nitrogen vacancy levels to carbon related impurity levels, which resulted in the enhancement of emission intensity. With further increasing temperature (T > 440 K), the nitrogen vacancy was increased and there was no enough trapped electrons to attend the transition, then the excited electrons would be trapped first and less electrons would jump from nitrogen vacancy levels to carbon related impurity levels, which led to the decrease of emission intensity. In addition, the lattice relaxation and non radiation process would increase under higher temperature, which would also reduce the red emission intensity of BCNO phosphor.^10d From Fig. 5(b), it can also be seen that the BCNO phosphor keep high emission intensity without decreasing for T ≤ 480 K (∼210 °C), which indicates the BCNO phosphors have good thermal stability and it would be beneficial for application.^10e

Fig. 5 (a) High temperature emission spectra of BCNO phosphors sintered at 650 °C for 12 h with R_B/H = 1 : 1. (b) Emission peak intensity as a function of sample temperature with T = 300–500 K.

Based on the above results and our previous research, a simplified energy level diagram was tentatively constructed to explain the luminescence mechanism of BCNO phosphors, as shown in Fig. 6. With no C and O doping, BN has a large band gap with about 5.9 eV. The nitrogen vacancy levels called three boron center (V_N3) and one-boron center (V_N1) will be formed below conduction band 1.0 and 0.7 eV, respectively.^9c,d For less C and O doping and partially formed BCNO, carbon and oxygen levels will appear in the band gap below the conduction band 4.1 and 3.5 eV, respectively.^6d,8h For more C and O doping and completely formed BCNO phosphor, new carbon related levels such as C_B (C substituted B) or C_N (C substituted N) levels will appear below the conduction band 3.0 eV.^5g,8d For partially formed BCNO phosphors, the excitation spectra and absorption spectra were induced by the transition from valence band to conduction band (λ_ex = 210 nm to 5.9 eV), valance band to nitrogen vacancy levels (λ_ex = 270 nm to 4.6 eV), carbon impurity levels (such as C substituted O levels) to conduction band (λ_ex = 270 nm to 4.1 eV), oxygen energy levels to conduction band (λ_ex = 354 nm to 3.5 eV). After relaxation, the photon electrons will be trapped by nitrogen vacancy levels, and the emission spectra were mainly in blue light range and originated from the transition from nitrogen vacancy levels to carbon impurity levels (λ_em = 400 nm to 3.1 eV) and oxygen levels (λ_em = 460 nm to 2.7 eV), respectively. In addition, the weak red emission was induced by the transition from vacancy levels to C_B or C_N levels (λ_em = 600 nm to 2.0 eV). For completely formed BCNO phosphors, the blue excitation around 460 nm was induced by the transition from C_B, C_N and C_I (interstitial carbon) levels to conduction band (λ_ex = 460 nm to 2.7 eV). The excitation around 550–580 nm was originated from the transition from C_B, C_N and C_I levels to nitrogen vacancy levels (λ_ex = 550–580 nm to 2.1–2.3 eV). The red emission of BCNO phosphors was dominated by the transition from nitrogen vacancy levels to C_B, C_N and C_I levels (λ_em = 600–620 nm to 2.0 eV). In addition, the broad absorption spectra between 210 and 700 nm (∼1.7–5.9 eV) were induced by the transition from valance band, carbon and oxygen related levels to conduction band and nitrogen vacancy levels.

Fig. 6 Schematic energy diagram of red emitting BCNO phosphors.

Conclusions

In summary, red emitting BCNO phosphors were successfully synthesized by solid state reaction method and the spectral properties and luminescence mechanism were systematically investigated. The red emission BCNO phosphors were turbostratic boron nitride structure and their size was in the range of several micrometers to one hundred micrometers. The XPS spectra and FTIR spectra indicated the carbon and oxygen elements were bonded with boron and nitrogen elements. The excitation spectra of BCNO phosphors were in the range of 300–570 nm, and the emission spectra were shifted from blue light to red light with increasing sintering temperature, time and the ratio of boric acid to hexamethy lenetetramine, corresponding with the partially formed BCNO to completely formed BCNO phosphor. The shift of emission spectra was resulted from the change of carbon and oxygen related impurity concentration and appearance of new carbon related impurity (C_B/C_N/C_I) levels. The decay curves suggested that the red emission process were induced by two luminescence centers with short and long lifetimes, respectively. The short lifetime was not sensitive to preparation conditions while the long lifetime was varied with sintering temperature, time and the mole ratio of raw materials. The UV-vis absorption spectra disclosed that the red emitting BCNO phosphors had wide absorption range (210–700 nm) which was originated from the transition from valence band, carbon and oxygen related levels to conduction band and nitrogen vacancy levels. In addition, the high temperature red emission spectra of BCNO phosphor implied that the nitrogen vacancies served as electron trapping which would participate in the red emitting process under certain temperatures (T > 410 K). The red emitting BCNO phosphors with good stability have great potential application on white LEDs, solar cell, display and biomedical fields.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (nos 51202055, 51402084, 51172060, 51171056, 51371075, 51272064) and Natural Science Foundation of Hebei Province of China (no. E2014202131), and Excellent Youth Foundation for Institution of Higher Learning of Hebei Province (no. YQ2014017). We also appreciate Dr Jianping Xu for the measurement and discussion of high temperature photoluminescence spectra.

Notes and references

1. (a) S. E. Brinkley, N. Pfaff, K. A. Denault, Z. Zhang, H. T. Hintzen, R. Seshadri, S. Nakamura and S. P. DenBaars, Appl. Phys. Lett., 2011, 99, 241106; (b) L. Yang, N. Zhang, R. Y. Zhang, B. Wen, H. L. Li and X. B. Bian, Mater. Lett., 2014, 129, 134; (c) B. Wang, H. Lin, J. Xu, H. Chen and Y. S. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 22905; (d) B. Mahler, N. Lequeux and B. Dubertret, J. Am. Chem. Soc., 2010, 132, 953.
2. (a) K. Watanabe, T. Taniguchi and H. Kanda, Nat. Mater., 2004, 3, 404; (b) Y. Kubota, K. Watanabe, O. Tsuda and T. Taniguchi, Science, 2007, 317, 932; (c) L. Museur, E. Feldbach and A. Kanaev, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 78, 155204.
3. (a) A. Rubio, J. L. Corkill and M. L. Cohen, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 49, 5081; (b) M. O. Watanabe, S. Itoh, T. Sasaki and K. Mizushima, Phys. Rev. Lett., 1996, 77, 187; (c) C. Y. Zhi, X. D. Bai and E. G. Wang, Appl. Phys. Lett., 2002, 80, 3590; (d) C. Y. Zhi, X. D. Bai and E. G. Wang, J. Nanosci. Nanotechnol., 2004, 4, 35.
4. (a) T. Ogi, Y. Kaihatsu, F. Iskandar, W. N. Wang and K. Okuyama, Adv. Mater., 2008, 20, 3235; (b) W. N. Wang, T. Ogi, Y. Kaihatsu, F. Iskandar and K. Okuyama, J. Mater. Chem., 2011, 21, 5183.
5. (a) L. Singh and V. Chopra, J. Mater. Sci. Technol., 2011, 27, 967; (b) W. W. Lei, D. Portehault, R. Dimova and M. Antonietti, J. Am. Chem. Soc., 2011, 133, 7121; (c) X. H. Zhang, Z. M. Lu, J. Lin, Y. Fan, L. L. Li, X. W. Xu, L. Hu, F. B. Meng, J. L. Zhao and C. C. Tang, ECS J. Solid State Sci. Technol., 2013, 2, R39; (d) C. R. Ren, X. H. Zhang, L. Y. Zhou, Z. M. Lu, J. Lin, X. W. Xu, L. L. Li, X. Zhang, Y. M. Xue, F. B. Meng, J. L. Zhao and C. C. Tang, J. Lumin., 2014, 153, 338; (e) X. H. Zhang, Z. M. Lu, J. Lin, L. L. Li, Y. Fan, L. Hu, X. W. Xu, F. B. Meng, J. L. Zhao and C. C. Tang, Mater. Lett., 2013, 94, 72; (f) A. B. Suryamas, M. M. Munir, T. Ogi, Khairurrijal and K. Okuyama, J. Mater. Chem., 2011, 21, 12629; (g) Q. H. Weng, X. B. Wang, X. Wang, D. Q. Liu, X. F. Jiang, C. Y. Zhi, Y. Bando and D. Golberg, Chem.–Asian J., 2013, 8, 2936.
6. (a) Y. Kaihatsu, W. N. Wang, F. Iskandar, T. Ogi and K. Okuyama, J. Electrochem. Soc., 2010, 157, J329; (b) T. Ogi, F. Iskandar, A. B. D. Nandiyanto, W. N. Wang and K. Okuyama, J. Chem. Eng. Jpn., 2012, 45, 995; (c) F. Lu, X. H. Zhang, Z. M. Lu, X. W. Xu and C. C. Tang, J. Lumin., 2013, 143, 343; (d) Z. Y. Chu, Y. Kang, Z. H. Jiang, G. Y. Li, T. J. Hu, J. Wang, Z. F. Zhou, Y. H. Li and X. J. Wang, RSC Adv., 2014, 4, 26855.
7. (a) C. C. Tang, Y. Bando, C. Y. Zhi and D. Golberg, Chem. Commun., 2007, 44, 4599; (b) Y. Kaihatsu, F. Iskandar, H. Widiyandari, W. N. Wang and K. Okuyama, Electrochem. Solid-State Lett., 2009, 12, J33; (c) X. F. Liu, S. Ye, G. P. Dong, Y. B. Qiao, J. Ruan, Y. X. Zhuang, Q. Zhang, G. Lin, D. P. Chen and J. R. Qiu, J. Phys. D: Appl. Phys., 2009, 42, 215409.
8. (a) X. F. Liu, S. Ye, Y. B. Qiao, G. P. Dong, Q. Zhang and J. R. Qiu, Chem. Commun., 2009, 27, 4073; (b) T. Ogi, H. Iwasaki, A. B. D. Nadiyanto, F. Iskandar, W. N. Wang and K. Okuyama, J. Mater. Chem. C, 2014, 2, 4297; (c) L. J. Ci, L. Song, C. H. Jin, D. Jariwala, D. X. Wu, Y. J. Li, A. Srivastava, Z. F. Wang, K. Storr, L. Balicas, F. Liu and P. M. Ajayan, Nat. Mater., 2010, 9, 430; (d) K. Raidongia, A. Nag, K. P. S. S. Hembram, U. V. Waghmare, R. Datta and C. N. R. Rao, Chem.–Eur. J., 2010, 16, 149; (e) Z. Bastl and S. Černý, J. Alloys Compd., 1991, 176, 159; (f) C. Y. Zhi, Y. Bando, C. C. Tang, D. Golberg, R. G. Xie and T. Sekigushi, Appl. Phys. Lett., 2005, 86, 213110; (g) X. Gouin, P. Grange, L. Bois, P. L'Haridon and Y. Laurent, J. Alloys Compd., 1995, 224, 22; (h) X. H. Zhang, S. Yan, Y. H. Cheng, K. H. Gao, Z. M. Lu, F. B. Meng, J. Lin, X. W. Xu, J. L. Zhao and C. C. Tang, Mater. Lett., 2013, 102–103, 102; (i) Z. M. El-Bahy, A. I. Hanafy, M. M. Ibrahim and M. Anpo, J. Mol. Catal. A: Chem., 2011, 344, 111; (j) D. Singh, K. Singh, G. Singh, Manupriya, S. Mohan, M. Arora and G. Sharma, J. Phys.: Condens. Matter, 2008, 20, 075228.
9. (a) X. H. Zhang, L. L. Li, Z. M. Lu, J. Lin, X. W. Xu, Y. H. Ma, X. J. Yang, F. B. Meng, J. L. Zhao and C. C. Tang, J. Am. Ceram. Soc., 2014, 97, 246; (b) Y. Kaihatsu, W. N. Wang, F. Iskandar and K. Okuyama, Mater. Lett., 2010, 64, 836; (c) H. F. Li, R. Zhao, Y. L. Jia, W. Z. Sun, J. P. Fu, L. H. Jiang, S. Zhang, R. Pang and C. Y. Li, ACS Appl. Mater. Interfaces, 2014, 6, 3163; (d) M. Girtan and G. Folcher, Surf. Coat. Technol., 2003, 172, 242.
10. (a) W. N. Wang, Y. Kaihatsu, F. Iskandar and K. Okuyama, Mater. Res. Bull., 2009, 44, 2099; (b) A. Katzir, J. T. Suss, A. Zunger and A. Halperin, Phys. Rev. B: Condens. Matter Mater. Phys., 1975, 11, 2370; (c) E. Y. Andrei, A. Katzir and J. T. Suss, Phys. Rev. B: Condens. Matter Mater. Phys., 1976, 13, 2831; (d) R. J. Xie, N. Hirosaki, N. Kimura, K. Sakuma and M. Mitomo, Appl. Phys. Lett., 2007, 90, 191101; (e) X. J. Zhang, L. Huang, F. J. Pan, M. M. Wu, J. Wang, Y. Chen and Q. Su, ACS Appl. Mater. Interfaces, 2014, 6, 2709.

---

Footnote

† Electronic supplementary information (ESI) available: XRD pattern, SEM, XPS and FTIR spectral of the BCNO phosphor. See DOI: 10.1039/c5ra07054f

---