Introduction

In recent years, semiconductor photocatalysis has received increasing research attention as a promising solution for solving the worldwide environmental pollution and energy crisis [1–4]. Amongst various semiconductor materials, TiO2 has been widely studied because of its excellent optical and electronic properties, low cost, chemical stability, and nontoxicity [4–6]. However, pristine TiO2 can only be excited by UV light because of its wide band gap (E g, anatase = 3.2 eV, E g, rutile = 3.0 eV), leading to the low utilization efficiency of the solar spectrum that reaches the earth (3–5 %) [7, 8], which has severely hindered the practical applications of TiO2 in the environmental and energy fields.

To solve this problem, considerable efforts have been devoted to exploiting more highly efficient visible light-active photocatalysts. So far, the typical strategies can be classified into two categories. The first involves doping with metals [9–11] or nonmetals [12, 13] and anchoring organic sensitizers [14, 15] for the wide-band-gap semiconductors such as TiO2 and ZnO; the second is to exploit new semiconductor materials with narrow band gaps such as CdS [16, 17], Bi2WO6 [18], Cu2O [19, 20], and so on. However, these photocatalysts are still not usable for practical applications because of a limited visible-light response and low stability [21–23].

Noble-metal nanoparticles (NPs), such as Au, Ag, Pt, and Cu, show strong UV–Vis light absorption because of their surface plasmon resonance (SPR) [24–27]. Consequently, noble-metal NPs can be employed as an alternative type of sensitizers to extend the visible light response of semiconductor materials without the problem of degradation encountered with organic sensitizers [28, 29]. Recently, silver/silver halide (Ag@AgX, X = Cl, Br, I) plasmonic photocatalysts have received much attention owing to their extraordinary visible light photocatalytic activity in terms of organic contaminant degradation under visible light irradiation [26, 27, 30–38]. As for photo-stability, these researches also proved that Ag@AgX can be reused under certain experimental conditions [30–38]. Unfortunately, although these researches proved that Ag@AgX photocatalysts are stable under certain experimental conditions [30–38], systematic studies on the stability of Ag@AgX under different application conditions have rarely been reported.

In this work, we synthesized Ag@AgBr plasmonic photocatalyst by a controllable double-jet precipitation technique, followed by light reduction. More importantly, several factors that may influence the stability of Ag@AgBr plasmonic photocatalyst, such as the type of light resource, electron/hole quenchers, dye sensitizer, and atmosphere environment, were systematically investigated for the first time.

Experimental

Synthesis of photocatalysts

Cubic AgBr grains were synthesized by a controllable double-jet precipitation method, similar to our previous report [32]. Under vigorous stirring, 150 mL of AgNO3 aqueous solution (0.4 M) and 150 mL of KBr aqueous solution (0.4 M) were simultaneously injected into a flask containing 150 mL of water and 0.8 g of gelatin. Throughout the reaction process, the pAg value of the reaction solution was controlled in the range of 6.2 ± 0.1, while the temperature was kept at 90 °C. After the two solutions were totally injected, the suspension was sequentially stirred for 30 min to eliminate the small particles. Finally, the sample was washed four times with warm water to eliminate the gelatin, and freeze-dried for use. AgBr powder was dispersed in water and irradiated with a 300-W high-pressure mercury lamp for 30 h to obtain the Ag@AgBr plasmonic photocatalyst.

Characterization

X-ray diffraction (XRD) measurements were carried out with a Rigaku D/max 2550 VB/PC X-ray diffractometer using Cu Kα radiation (λ = 0.154056 nm). The morphology of the samples was observed on a JEOL JSM-6360 LV scanning electron microscopy (SEM), operated at 15 kV.

Photo-stability evaluation

In the photocatalytic reaction, if Ag@AgBr is unstable, the Ag+ ions will react with photo-generated electrons to form metallic Ag, which can further grow into Ag nanoparticles (clusters). It is well known that XRD is a useful technique to analyze the information of different phases and components in a sample [39–42]. Here, we evaluate the photo-stability of Ag@AgBr in different experimental conditions by comparing the intensity ratios of the AgBr (2 0 0) peak and the Ag (1 1 1) peak in the corresponding samples (I AgBr (200)/I Ag (111)).

Factors influencing Ag@AgBr photo-stability

Light irradiation time

An amount of 25 mg of the as-prepared AgBr powder was dispersed in 25 mL of water, and then irradiated by a high-pressure mercury lamp for different times (5–40 h). Finally, the sample was collected by centrifugation.

Light source

An amount of 25 mg of Ag@AgBr was dispersed in 25 mL of water, and irradiated for 5 h by:

  1. (a)

    500-W halogen tungsten lamp with an UV cut-off filter (λ = 420 nm);

  2. (b)

    300-W high-pressure mercury lamp (λ max = 365 nm) fixed in a quartz cylindrical jacket and cooled with flowing water;

  3. (c)

    300-W xenon lamp with a UV cut-off filter (λ = 420 nm).

The distance between the light and the reaction tube was fixed at 10 cm.

Electron quencher: oxygen

Oxygen can react with electrons to produce superoxide anion radicals so that it can be regarded as a kind of electron quencher. For each test, 25 mg of Ag@AgBr was dispersed in 25 mL of water, and irradiated for 5 h by different lamps (high-pressure mercury lamp, halogen tungsten lamp, xenon lamp) under different atmospheres (air, vacuum, or oxygen-rich).

Hole quencher: methanol

Methanol was employed as a representative candidate to evaluate the influence of a hole quencher on Ag@AgBr photo-stability. For each test, 25 mg of Ag@AgBr were dispersed in a mixture of water (20 mL) and methanol (5 mL), and irradiated by a mercury lamp for 5 h under different atmospheres (air or vacuum).

Dye sensitizer

An amount of 25 mg of Ag@AgBr was dispersed in 25 mL of acid orange 7 (AO7) solution (10 mg/L). Prior to light irradiation, the suspension was stirred in the dark for 1 h to attain the adsorption–desorption equilibrium for AO7 dye. Then, the suspension was exposed by a halogen tungsten lamp for 25 min under different atmospherea (air or vacuum). At a given time interval, about 1 mL of the suspension was withdrawn and the concentration of the residual AO7 was measured with a UV–Vis spectrophotometer.

Results and discussion

SEM and XRD analysis

Figure 1 shows the SEM image and XRD pattern of the as-prepared AgBr. The sample presents diffraction peaks at 2θ = 26.7°, 30.9°, 44.4°, 52.5°, 55.0°, 64.5° and 73.3°, corresponding to (111), (200), (220), (311), (222), (400), (420) plane reflections, respectively (JCPDS: 06-0438). From the inset of Fig. 1, it can be seen that the as-prepared AgBr grains exhibit regular cubic morphology with an edge length of ~0.6 μm. These results confirm that regular cubic AgBr grains can be successfully fabricated by the double-jet precipitation technique.

Fig. 1
figure 1

XRD pattern of the as-prepared AgBr. Inset the SEM image of the as-prepared AgBr

Full size image

Influence of photo-reduction time on Ag@AgX photo-stability

For the preparation of Ag@AgX plasmonic photocatalyst, photo-reduction is one of the competent techniques which in situ can reduce Ag+ ions into metallic Ag. The photo-stability of Ag@AgX photocatalyst can be examined by analyzing the intensity changes of the Ag and AgBr peaks in XRD patterns. Figure 2 shows the XRD patterns of AgBr irradiated under a high-pressure mercury lamp for different times. The peaks located at 2θ = 38.1°, 44.4°, 64.4° and 77.4° should be ascribed to Ag (1 1 0), (2 0 0), (2 2 0) and (3 1 1) characteristic diffraction peaks, respectively (JCPDS: 65-2871). We investigated the photo-stability of Ag@AgBr by comparing the intensity ratios of the AgBr (200) peak and the Ag (111) peak (I AgBr (200)/I Ag (111)). Figure 3 illustrates the I AgBr (200)/I Ag (111) variation as a function of UV light irradiation time. During the first 10 h, I AgBr (200)/I Ag (111) decreases steeply, which is due to the rapid formation of metallic Ag. As the irradiation time lengthens, I AgBr (200)/I Ag (111) decreases more and more slowly. This is because the larger amounts of Ag clusters can more effectively capture photo-generated electrons, which can effectively block the reaction between the electrons and interstitial Ag+ ions to form metallic Ag. The value of I AgBr (200)/I Ag (111) remains almost unchanged from 30 to 40 h, implying that Ag@AgBr has become stable after high-pressure mercury lamp irradiation for 30 h. Therefore, all Ag@AgBr samples used for the following experiments were prepared by high-pressure mercury lamp irradiation for 30 h.

Fig. 2
figure 2

XRD patterns of AgBr after exposed under UV light for different times: a 5 h; b 10 h; c 15 h; d 20 h; e 25 h; f 30 h; g 35 h; h 40 h

Full size image
Fig. 3
figure 3

I AgBr (200)/I Ag (111) variation as a function of UV light irradiation time

Full size image

Influence of light sources on Ag@AgX photo-stability

Photocatalytic reactions may be conducted using different light sources (wavelength, composition, intensity, and so on). We investigated the influence of light sources on the photo-stability of the Ag@AgBr photocatalyst. The three kinds of lamps used for testing include a high-pressure mercury lamp (300 W), a halogen tungsten lamp (500 W, UV cut-off filter), and a xenon lamp (300 W, UV cut-off filter). Figure 4 shows the XRD patterns of the Ag@AgBr samples after irradiation for 5 h with these different light sources. As shown in the inset of Fig. 4, the values of I AgBr (200)/I Ag (111) corresponding to the high-pressure mercury lamp and halogen tungsten lamp irradiation are 14.2 and 14.4, respectively, which are similar to the value of the as-prepared Ag@AgBr. This result revealed that Ag@AgBr is stable under high-pressure mercury lamp and halogen tungsten lamp irradiation. In contrast, I AgBr(200)/I Ag (111) decreases to 8.9 after xenon lamp irradiation for 5 h, indicating that Ag@AgBr becomes unstable under xenon lamp irradiation. The light intensity of the xenon lamp (80.5 mW/cm2) is almost twice that of the halogen tungsten lamp (40.9 mW/cm2). The more powerful the light, the more photo-induced electrons appear. Before the photo-generated electrons are transformed to Ag clusters, some of them will react with Ag+ ions to produce metallic Ag, resulting in the deterioration of photo-stability.

Fig. 4
figure 4

XRD patterns of Ag@AgBr after exposed under different light sources: a halogen tungsten lamp; b high-pressure mercury lamp; c xenon lamp

Full size image

Influence of an electron quencher: oxygen on Ag@AgX photo-stability

In most cases, oxygen takes part in the photocatalytic reactions by trapping electrons and further forming reactive oxygen species such as superoxide free radicals (·O2 −). Here, we discuss the influence of oxygen on the stability of Ag@AgBr. Figure 5 shows the XRD patterns of Ag@AgBr irradiated by the high-pressure mercury lamp under air, vacuum, and oxygen-rich atmospheres. From the inset of Fig. 5, it can be seen that Ag@AgBr is stable under both air and oxygen-rich atmospheres while it becomes unstable under vacuum. In the presence of oxygen, the photo-generated electrons can be trapped by oxygen to form ·O2 −, which can decrease the density of electrons on Ag@AgBr, which is favorable for improving the stability of Ag@AgBr. In contrast, in the absence of oxygen, the large amounts of photo-generated electrons will concentrate on Ag@AgBr and further react with Ag+ ions to form metallic Ag, leading to the deterioration of photo-stability.

Fig. 5
figure 5

XRD patterns of Ag@AgBr irradiated by high-pressure mercury lamp under a air atmosphere, b vacuum atmosphere, and c oxygen-rich atmosphere

Full size image

The stability of Ag@AgBr under air, vacuum, and oxygen-rich atmospheres was also compared using the halogen tungsten lamp as light source. As shown in the inset of Fig. 6, Ag@AgBr shows much lower stability under vacuum than under air and oxygen-rich atmospheres, similar to the result under high-pressure mercury lamp irradiation. By comparing the I AgBr (200)/I Ag values in Figs. 5b, 6b (in vacuum atmosphere), it can be seen that Ag@AgBr showed higher stability under the halogen tungsten lamp than under the high-pressure mercury lamp.

Fig. 6
figure 6

XRD patterns of Ag@AgBr irradiated by halogen tungsten lamp under a air atmosphere, b vacuum atmosphere (HT-12), and c oxygen-rich atmosphere

Full size image

Figure 7 reveals that Ag@AgBr exhibits higher stability in an oxygen-rich atmosphere than in vacuum and air atmospheres. This result confirmed that the existence of oxygen is beneficial for the improvement of stability. By further comparing the I AgBr (200)/I Ag values in Fig. 7 with those in Figs. 5, 6, it can be seen that, under the same circumstance (air, vacuum, or oxygen-rich atmosphere), the stability of Ag@AgBr under xenon lamp irradiation is much lower than that under halogen tungsten lamp and high-pressure mercury lamp irradiation. The higher light illumination of the xenon lamp can increase the intensity of electrons, decreasing the photo-stability of Ag@AgBr.

Fig. 7
figure 7

XRD patterns of Ag@AgBr irradiated by xenon lamp under a air atmosphere, b vacuum atmosphere, and c oxygen-rich atmosphere

Full size image

Influence of hole quencher: methanol on Ag@AgX photo-stability

Figure 8 shows the XRD patterns of Ag@AgBr under different conditions (with or without methanol; air or vacuum atmosphere). In the presence of methanol, I AgBr (200)/I Ag value decreases from 14.2 (Fig. 8a) to 8.9 (Fig. 8b), implying that adding methanol to the solution observably decreases the stability of Ag@AgBr. When methanol is present, it can capture photo-induced holes and consequently reduce the recombination rate of electrons and holes. As a result, more electrons will react with Ag+ ions, resulting in the decrease of photo-stability. Under vacuum atmosphere, adding methanol makes I AgBr (200)/I Ag value further decrease to 4.3 (Fig. 8c), revealing that the synergistic effect caused by the presence of a hole quencher and the absence of an electron quencher.

Fig. 8
figure 8

XRD patterns of Ag@AgBr irradiated by high pressure mercury lamp under different conditions: a air atmosphere, no methanol; b air atmosphere, methanol; c vacuum atmosphere, methanol

Full size image

Influence of dye sensitizer on Ag@AgX photo-stability

Ag@AgX photocatalysts have been successfully used in the degradation of organic contaminants, especially organic dyes. However, the influence of these organics on the stability of Ag@AgX has scarcely been studied. We selected AO7 as a representative dye to examine the influence of dyes on the stability of the Ag@AgBr photocatalyst. Figure 9 shows the XRD patterns of Ag@AgBr after halogen tungsten lamp irradiation in the presence/absence of AO7 dye. In the presence of AO7, I AgBr (200)/I Ag decreases from 14.2 (Fig. 9a) to 6.1 (Fig. 9b), revealing that the presence of AO7 would drastically decrease the stability of the Ag@AgBr photocatalyst. After absorbing light, the dye molecules adsorbed on the Ag@AgBr surface can be excited and subsequently inject the electrons to Ag@AgBr, leading to the reduction of Ag+ ions to metallic Ag. Under vacuum atmosphere, I AgBr (200)/I Ag further decreases to 4.0, indicating the further decrease of stability. In the presence of oxygen (air atmosphere), some electrons can be transferred to the adsorbed oxygen and only some of the electrons take part in the reduction of Ag+ ions, while in the absence of oxygen (vacuum atmosphere), more electrons can be transferred to Ag@AgBr, favorable to forming more metallic Ag. Under high pressure mercury lamp irradiation, a similar result was found (not shown).

Fig. 9
figure 9

XRD patterns of Ag@AgBr irradiated by halogen tungsten lamp under different conditions: a air atmosphere, no AO7 dye; b air atmosphere, AO7 dye; c vacuum atmosphere, AO7 dye

Full size image

From the above experimental results, it can be concluded that the photo-stability of Ag@AgBr is relative to the photocatalytic reaction conditions, such as light source, the presence/absence of electron/hole quenchers, and the atmospheric environment. In the photocatalytic reactions, a weak light source, the presence of an electron quencher, and oxygen-rich atmosphere are favorable to the attainment of high photo-stability. In contrast, a strong light source, the presence of a hole quencher, and a vacuum atmosphere have detrimental effects on photo-stability.

Conclusions

In this work, we systematically investigated the factors that influence the photo-stability of Ag@AgBr plasmonic photocatalysts, such as light sources (halogen tungsten lamp, high-pressure mercury lamp, and xenon lamp), electron and hole quenchers (oxygen and methanol), photosensitizer, and atmospheric environment (air, vacuum, and oxygen-rich). By comparing the peak intensity ratio of AgBr (2 0 0) to Ag (1 1 1) in XRD patterns after light irradiation, we obtained following conclusions: (1) after a certain time of light irradiation, the content of Ag and AgBr in Ag@AgBr will remain unchanged, proving that Ag@AgBr is a stable photocatalyst; (2) the stability of Ag@AgBr is relative as Ag@AgBr is stable under weak light sources (halogen tungsten lamp, high-pressure mercury lamp), while it can become unstable under strong light sources (xenon lamp); (3) the presence of electron quenchers (e.g. oxygen) is favorable for improving the stability of Ag@AgBr, while hole quenchers (e.g. methanol) do harm to the stability of Ag@AgBr; (4) a dye sensitizer usually decreases the stability of Ag@AgBr; and (5) air and oxygen-rich atmospheres are more favorable for maintaining the stability of Ag@AgBr than a vacuum atmosphere.