EDTA, benzoquinone, isopropyl alcohol and terephthalic acid were purchased from Acros Corp (Belgium). RhB and MO were obtained from Shanghai Aladdin Chemistry Corp (China). All other chemicals were purchased from Tianjin Chemical Company (China) and used as received.

In a typical process, 0.006 mol AgNO₃ was dissolved in 20 mL H₂O, and 40 mL of an aqueous Na₂HPO₄ (0.002 mol) solution was then added dropwise into the solution under vigorous stirring. After stirring for 30 min, the yellow precipitate was collected by centrifugation, washed with distilled water and absolute ethanol, and then dried in air.

The phase of the as-obtained product was characterized by X-ray diffraction (XRD) with a Rigaku D/Max-2500 X-ray diffractometer using Cu K_α radiation, λ = 0.154056 nm. The morphology and size of the as-prepared product were characterized by field emission scanning electron microscopy (FESEM, JEOL JSM-6700). The Brunauer-Emmett-Teller (BET) surface area of the sample was measured on a Micromeritics Tristar 3000 analyzer. UV-Vis diffuse reflectance spectra (DRS) were recorded in the range of 200-800 nm with a UV-Vis spectrophotometer (Shimadzu, UV-3600) using BaSO₄ as the reference. Fourier transform infrared (FT-IR) spectra were recorded using a Nicolet AVATAR360 infrared Fourier spectrometer with the KBr wafer technique.

The photocatalytic activity of the as-obtained sample was evaluated using the degradation of RhB and MO under visible light. The photodegradation reactions were carried out in a simple photocatalytic reactor without a water circulation facility (Fig. 1). A 10 W LED lamp was employed as the visible light source. The distance between the lamp and solution was about 6 cm. The Ag₃PO₄ photocatalyst (0.05 g) was added to aqueous RhB or MO solution (5 mg/L, 100 mL) and was magnetically stirred in the dark for 30 min. 3 mL of the suspension was taken at given time intervals and separated through centrifugation. The concentrations of RhB and MO in the supernatant were analyzed at 553 nm and 464 nm, respectively, on a UV-Vis spectrophotometer (Shimadzu, UV2550). The active species generated during the photocatalytic process were detected using in situ capture experiments.

Fig. 1. Photocatalytic apparatus. (1) Magnetic stirrer; (2) Ag3PO4 photocatalyst; (3) Magnet; (4) Dye solution; (5) LED lamp; (6) Sampler.

The terephthalic acid fluorescence probe technique was employed to detect OH^• radicals in the system. We added 0.05 g of the Ag₃PO₄ photocatalyst to 100 mL of an aqueous solution containing 2 mmol/L NaOH and 0.5 mmol/L terephthalic acid. Prior to irradiation, the suspension was stirred in the dark for 30 min. Three milliliters of the suspension was collected at predetermined time intervals and centrifuged. The fluorescence spectrum of the remaining clear supernatant was measured using a Hitachi F4500 spectrometer.

The phase structure and purity of the product were characterized by XRD. Figure 2(a) shows the XRD pattern of the as-obtained product. All of the diffraction peaks can be indexed to the cubic structure of Ag₃PO₄ (JCPDS card No.06-0505). No peaks assignable to other phases were observed, indicating that high-purity Ag₃PO₄ was obtained. The sharp and strong diffraction peaks indicated that the Ag₃PO₄ product was well crystallized. The shape and size of the as-prepared Ag₃PO₄ product were investigated by FESEM. The SEM image revealed that the as-obtained product was composed of quasi-spherical particles having diameters in the range of 150-500 nm (Fig. 2(b)). The BET surface area of the as-obtained Ag₃PO₄ product was 2.6 m²/g.

Fig. 2. XRD pattern (a) and SEM image (b) of the Ag3PO4 sample.

The UV-Vis diffuse reflectance spectrum of the as-obtained Ag₃PO₄ is shown in Fig. 3(a), which shows that the product exhibited absorption in the visible range as well as the UV range. The band gap (E_g) of a semiconductor can be obtained from the following equation (Eq. 1):

Fig. 3. (a) UV-Vis diffuse reflectance spectrum of the Ag3PO4 product; (b) Plot of (αhν)1/2 versus hν for the Ag3PO4 product.

where α, ν, and A are absorption coefficient, light frequency, and proportionality constant, respectively. The band gap (E_g) of the Ag₃PO₄ was obtained from the plot of (αhν)^1/2 versus (hν) ^[15], as shown in Fig. 3(b), and was estimated to be 2.24 eV.

The FT-IR spectrum of the as-synthesized Ag₃PO₄ is shown in Fig. 4. The peaks at 1670 and 3200 cm⁻¹ were attributable to the bending vibration and the bending vibration of O-H ^[11], respectively. In addition, two strong peaks observed at 560 and 1014 cm⁻¹ were assignable to the molecular vibrations of phosphate (PO₄³⁻) ^[11].

Fig. 4. FT-IR spectra of the as-prepared Ag3PO4 product.

The photodegradation of RhB (a cationic dye) and MO (an anionic dye) by a suspension of the Ag₃PO₄ under irradiation with visible light was investigated. In the absence of Ag₃PO₄ catalyst, removal of the dyes was almost negligible (data not shown). Figure 5(a) shows the photodegradation of RhB and MO in Ag₃PO₄ suspensions as a function of irradiation time. After 30 min exposure to visible light, 98.8% of the RhB was degraded, but only 42.8% of the MO was removed. The zero point charge (pHzpc) for Ag₃PO₄ is about 5.4 ^[18]. At ranges lower or higher than the pHzpc, Ag₃PO₄ is positively charged or negatively charged, respectively. Therefore, in the neutral solution used (photodegradation of the dyes in water without pH adjustment), negatively charged Ag₃PO₄ surfaces were available for the adsorption of positively charged RhB (a cationic dye), which promoted degradation reaction. This was further confirmed with an adsorption experiment. Figure 5(b) displays the time-resolved adsorption of the dyes by the Ag₃PO₄ catalyst in the dark. After 60 min, the amount of RhB adsorbed by the Ag₃PO₄ catalyst in the dark was 24.5%, compared with only 3.1% adsorption of MO.

Fig. 5. (a) Photocatalytic degradation of RhB and MO in Ag3PO4 suspension under visible light without pH adjustment; (b) Time-resolved adsorption of dye by Ag3PO4 catalyst in the dark. Ag3PO4 dose, 0.5 g/L; initial concentration of dye, 5 mg/L.

The photodegradation efficiency for MO was still low when the Ag₃PO₄ loading was increased (from 0.5 g/L to 1.0 g/L) while keeping all the other conditions unchanged; only 58.8% of the MO was degraded (Fig. 6). This result further indicated that the as-synthesized Ag₃PO₄ catalyst was not good at degrading MO dye using visible light. However, the rate of MO photodegradation was greatly enhanced in the presence of RhB; about 81.0% of the MO in the aqueous suspension was decomposed by the Ag₃PO₄ catalyst under visible light irradiation after 30 min (Fig. 6). The reason for this phenomenon will be given in the following section.

Fig. 6. Photodegradation of MO in Ag3PO4 suspension under visible light (initial concentration of MO, 5 mg/L).

To investigate the photocatalytic mechanism of dye degradation in the Ag₃PO₄ system, we measured the reactive oxygen species present during the photocatalytic processes. It is well known that terephthalic acid reacts with OH^• radicals to form 2-hydroxy terephthalic acid, which emits fluorescence at 426 nm (λ_ex = 315 nm) ^{[19, 20]}. As shown in Fig. 7, no obvious fluorescence peak at 426 nm was observed, and the fluorescence intensity at 426 nm was not significantly changed with extended irradiation time, indicating that OH^• is not the main reactive oxygen species in the Ag₃PO₄ system ^[20]. From a theoretical viewpoint, the valence band edge potential of Ag₃PO₄ catalyst (2.45 eV) ^[16] is less positive than E^y(OH^•/H₂O) (2.68 eV) ^[21], which indicates that Ag₃PO₄ holes cannot oxidize adsorbed H₂O molecules to OH^• radicals. Next, reactive species trapping experiments were conducted to investigate the reactive oxygen species in the photocatalytic process of the Ag₃PO₄ system. Isopropyl alcohol, EDTA, and benzoquinone were introduced into the photocatalytic system as scavengers for OH^•, h⁺, and O₂^{•− [22, 23, 24]}, respectively. As shown in Fig. 8, the addition of isopropyl alcohol had little effect on the photocatalytic performance of the Ag₃PO₄ catalyst, which further indicates that OH^• does not play a key role in the degradation of dyes in the Ag₃PO₄ system. The addition of 1mM EDTA strongly inhibited the photocatalytic activity of the Ag₃PO₄ catalyst for the degradation of both RhB and MO. Inhibition of RhB and MO degradation was also observed when 1 mmol/L benzoquinone was added to the photocatalytic reaction systems, suggesting that O₂^•− radicals contribute to the photocatalytic performance of Ag₃PO₄ suspensions under visible light irradiation (Fig. 8). To confirm this result, the photocatalytic activities of Ag₃PO₄ for RhB and MO degradation were investigated with continuous N₂ sparging. Under anoxic conditions, both RhB and MO photodegradation were strongly inhibited (Fig. 9). Thus, it can be concluded that h⁺ and O₂^•− are the main reactive species in the Ag₃PO₄ system for RhB and MO degradation under visible light irradiation (Figs. 8 and 9).

Fig. 7. OH• trapping PL spectra for a terephthalic acid solution of the Ag3PO4 catalyst under visible light irradiation.

Fig. 8. Photodegradation of RhB (a) and MO (b) by Ag3PO4 catalyst under different conditions. (1) No scavengers; (2) 1 mmol/L isopropyl alcohol; (3) 1 mmol/L benzoquinone; (4) 1 mmol/L EDTA. Ag3PO4 loading, 0.5 g/L; initial concentration of dye, 5 mg/L.

Fig. 9. Photodegradation of RhB (a) and MO (b) by Ag3PO4 with and without N2. (1) No N2; (2) With continuous N2-sparging. Ag3PO4 loading, 0.5 g/L; initial concentration of dye, 5 mg/L.

On the basis of the above results, we propose a possible photocatalytic mechanism for RhB or MO degradation in Ag₃PO₄ suspension using visible light. Under irradiation, the electron-hole pairs of the Ag₃PO₄ semiconductor are separated, and the electrons at the valence band are excited to the conduction band, resulting in the formation of holes in the valence band. The valence band edge potential of Ag₃PO₄ is 2.45 eV ^[16], which indicates that holes formed in the valence band of the Ag₃PO₄ catalyst could directly oxidize RhB or MO adsorbed on its surface ^[25]. Much more RhB than MO could be adsorbed by the Ag₃PO₄ catalyst (Fig. 5(b)), resulting in a much higher rate of RhB photodegradation than that of MO by the as-synthesized Ag₃PO₄ catalyst. O₂ molecules adsorbed on the Ag₃PO₄ catalyst surface can be reduced by the photogenerated electrons to form O₂^•− radicals, which contribute to the photodegradation of dyes in the Ag₃PO₄ system ^[22]. Meanwhile, the dye molecules can be excited by the visible light, and the resulting photoelectrons are immediately injected into the conduction band of Ag₃PO₄ catalyst, further facilitating the removal of the dye ^{[26, 27]}. The photocatalytic mechanism of RhB or MO degradation in the present Ag₃PO₄ system under visible light irradiation is thus as follows:

According to the experimental results, the photodegradation of MO over the as-obtained Ag₃PO₄ catalyst under visible light irradiation was greatly enhanced with the assistance of RhB (Fig. 6). However, the rate of RhB photodegradation by the Ag₃PO₄ catalyst was decreased in the presence of MO (Fig. 5(a) and Fig. 10). As shown, RhB is easily adsorbed onto the surface of the Ag₃PO₄ catalyst. Under visible light irradiation, the adsorbed RhB molecules were excited by the visible light and the excited-state electrons are injected into the conduction band of Ag₃PO₄ catalyst. The injected electrons and the photoinduced electrons (Eq. 4) reacted with O₂ to generate O₂^•− radicals, which may have contributed to the degradation of MO via the photoreduction process ^[25], and thereby decreasing the efficiency of RhB photodegradation in the presence of MO. Hence, we deduced that the photodegradation of MO by Ag₃PO₄ was enhanced by a greater number of O₂^•− radicals produced with the assistance of RhB.

Fig. 10. Photodegradation of a mixture of RhB and MO solution by Ag3PO4. Ag3PO4 loading, 0.5 g/L; initial concentration of dye, 5 mg/L.

乙二胺四乙酸、苯醌、异丙醇和对苯二甲酸购于Acros化学试剂. 罗丹明B和甲基橙购于上海阿拉丁化学试剂. 其它试剂均购于天津化学试剂.

将0.006 mol AgNO₃和0.002 mol Na₂HPO₄分别溶于20和40 mL水中, 在强烈磁搅下将Na₂HPO₄溶液逐滴加入到AgNO₃溶液中, 滴加完后继续磁力搅拌30 min, 离心, 用蒸馏水和无水乙醇洗涤多次, 于空气中干燥备用.

采用X射线衍射仪(XRD, 日本理学, D/Max-2500型, Cu K_α, λ = 0.154056 nm)分析制备样品的晶相. 通过场发射扫描电镜(FESEM, JEOL, JSM-6700型)观察样品的形貌和尺寸. 采用氮吸附仪(Micromeritics, Tristar 3000)测定样品的BET比表面积. 通过紫外-可见(UV-Vis)分光光度计(Shimadzu, UV-3600)对样品进行紫外-可见漫反射光谱分析, 以BaSO₄为参比, 扫描范围为200-700nm. 采用傅里叶变换红外光谱仪(Nicolet, AVATAR360)测定样品的红外光谱(FT-IR).

通过降解水体中的罗丹明B和甲基橙评价样品的可见光催化性能. 光催化实验在自制无需循环水的反应装置中进行(见图1). 以10 W LED灯作为可见光光源, 灯和液面的垂直距离为6 cm. 将一定量的光催化剂(0.05 mg)加入到100 mL浓度为5 mg/L的罗丹明B溶液(甲基橙溶液)中, 光照前在黑暗条件下磁力搅拌30 min. 光照条件下, 每隔一定时间取样3 mL, 高速离心分离后采用紫外-可见分光光度计(Shimadzu, UV2550)测定罗丹明B(甲基橙)的浓度. 采用原位捕获实验检测光催化过程产生的反应活性物种.

对苯二甲酸(Acros)荧光探针技术用来检测光催化反应过程中产生的羟基自由基. 将0.05 g Ag₃PO₄催化剂加入到100 mL含有NaOH (2 mmol/L)和对苯二甲酸(0.5 mmol/L)的水溶液中. 光照射之前在黑暗条件下磁力搅拌30 min. 光照后每隔一定时间取样3 mL, 离心后取上层清液, 通过荧光分光光度计(Hitachi, F4500)获取上层清液的荧光光谱.

图2(a)为所制备样品的XRD谱. 由图可见, 该样品的衍射峰与立方相Ag₃PO₄一致(JCPDS 06-0505), 没有发现其它晶相衍射峰, 证实所制备样品为高纯立方相Ag₃PO₄, 且所有衍射峰峰形尖锐、强度高, 表明样品晶型完整、结晶度高. 如图2(b)所示, Ag₃PO₄样品呈现类球形颗粒状, 尺寸大小在150-500 nm范围内, 其比表面积为2.6 m²/g.

Ag₃PO₄样品的UV-Vis谱见图3(a). 可以看出, Ag₃PO₄样品对紫外光和可见光均有明显的吸收. 半导体的帯隙能(E_g)可由公式(1)求得:

式中α, ν和A分别为吸收系数、光频率以及比例常数. 如图3(b)所示, 通过以(αhν)^1/2对hν作图获取Ag₃PO₄的E_g为2.24 eV.

图4是Ag₃PO₄样品的FT-IR谱. 可以看出, 1670和3200 cm^-1处出现的吸收峰分别为O-H的伸缩振动峰和弯曲振动峰^[11], 而560和1014 cm^-1处吸收峰为PO₄^3-的分子振动峰^[11].

在没有Ag₃PO₄催化剂存在的条件下, 罗丹明B和甲基橙不降解(数据未给出). 图5(a)是在可见光催化系统中Ag₃PO₄催化降解罗丹明B和甲基橙的曲线图. 可以看出, 可见光照30 min后, 罗丹明B降解了98.8%, 而甲基橙仅降解了42.8%, 可见Ag₃PO₄催化剂对阳离子染料罗丹明B具有显著的可见光催化去除效果. Ag₃PO₄的零电荷点(pHzpc)为5.4左右^[18], 当pH值低于(高于) pHzpc时, Ag₃PO₄表面带正电(负电). 在本文中性条件下, Ag₃PO₄表面带负电, 容易吸附带正电的罗丹明B分子, 从而有利于降解反应的进行, 吸附实验也证实上述结论. 在黑暗条件下Ag₃PO₄催化剂吸附罗丹明B和甲基橙的曲线如图5(b)所示. 可以看出, 吸附60 min后, Ag₃PO₄催化剂对罗丹明B的吸附为24.5%, 而对甲基橙仅为3.1%.

当Ag₃PO₄催化剂的投入量由0.5 g/L增至1.0 g/L时, 甲基橙降解率仅为58.8% (图6), 进一步证实Ag₃PO₄催化剂不易降解甲基橙. 然而, 在上述反应体系中加入罗丹明B时, Ag₃PO₄可见光催化降解甲基橙的速率大大提高, 其降解率可达81.0%.

为阐述Ag₃PO₄系统可见光催化降解有机染料的机理, 对光催化反应过程中产生的活性氧物种进行了检测. 众所周知, 对苯二甲酸可与羟基自由基反应生成2-羟基对苯二甲酸, 当激发波长为315 nm时, 在波长426 nm处会产生荧光^{[19, 20]}. 由图7可见, 在Ag₃PO₄光催化反应系统中, 在426 nm处没有出现明显的荧光峰, 且随着光照时间的延长, 也未有明显的变化, 表明该反应系统中羟基自由基并不是主要的活性氧物种^[20], 这进一步在理论上证实了上述结论. Ag₃PO₄光催化剂的价带边电势为2.45 eV^[16], 低于使水转化成羟基自由基的标准电极电势(2.68 eV) ^[21], 表明光生空穴不能氧化Ag₃PO₄表面吸附的水分子成羟基自由基. 此外, 活性物种捕获实验用来间接检测Ag₃PO₄光催化反应系统中产生的活性氧物种. 异丙醇、乙二胺四乙酸和苯醌分别作为羟基自由基、空穴和超氧自由基的捕获剂, 加入到光催化体系中以检测这三种活性物种的存在^{[22, 23, 24]}. 由图8可知, 加入异丙醇对Ag₃PO₄催化剂的光催化活性影响不大, 证实羟基自由基对染料的降解贡献不大. 乙二胺四乙酸和苯醌(1 mmol/L)的加入均明显抑制了Ag₃PO₄催化剂对罗丹明B和甲基橙的降解, 表明空穴和超氧自由基在Ag₃PO₄可见光催化过程中起降解作用(图8). 正如图9所示, 在连续充氮厌氧条件下, 罗丹明B和甲基橙的降解均受到明显地抑制. 综上所述, 空穴和超氧自由基是Ag₃PO₄可见光催化系统中主要的反应活性物种.

基于以上分析, 我们提出Ag₃PO₄可见光催化降解罗丹明B和甲基橙的反应机理. 在光照条件下, Ag₃PO₄催化剂产生光生电子和空穴, 价带电子跃迁到导带上, 同时在价带上生成空穴. Ag₃PO₄价带边电势为2.45 eV^[16], 表明价带空穴能够直接氧化降解罗丹明B和甲基橙^[25]. 相比于甲基橙, 罗丹明B分子更易吸附到Ag₃PO₄催化剂表面, 其降解速率也更高. Ag₃PO₄催化剂表面吸附的溶解性氧气分子可被导带上的光生电子还原, 生成超氧自由基, 有利于染料分子的降解. 同时, 染料分子本身可被可见光激发, 产生的光生电子能够注入到Ag₃PO₄催化剂导带上, 从而促进了光催化降解反应的进行^{[17, 26]}. Ag₃PO₄可见光催化降解罗丹明B和甲基橙的反应机理如下所示:

如图6所示, 在罗丹明B协助下, Ag₃PO₄可见光催化降解甲基橙的活性明显增加, 但此时罗丹明B的降解速率下降(图5(a)和图10). 由于罗丹明B很容易吸附到Ag₃PO₄表面, 当受可见光激发而产生光生电子注入到Ag₃PO₄导带上, 可与Ag₃PO₄本身激发产生的光生电子一起吸附溶解氧, 反应生成超氧自由基, 通过还原反应从而促进了甲基橙的降解^[25]. 正是由于部分超氧自由基参与降解甲基橙, 从而抑制了罗丹明B的降解. 可见在罗丹明B的协助下可产生更多的超氧自由基, 从而促进甲基橙的降解.