Cu2(OH)PO4/reduced graphene oxide nanocomposites for enhanced photocatalytic degradation of 2,4-dichlorophenol under infrared light irradiation

Sparked by the growing environmental crises, photocatalytic degradation of chlorophenols with inexhaustible solar energy is expected to be converted into actual applications. Here, we report the preparation of the nanocomposite of Cu2(OH)PO4 and reduced graphene oxide (Cu2(OH)PO4/rGO) through a one-step hydrothermal method and examined its infrared-light photocatalytic activity in the degradation of 2,4-dichlorophenol (2,4-DCP). As evidenced by the absorption spectra and the degradation of 2,4-DCP, Cu2(OH)PO4/rGO exhibited enhanced infrared light-driven photocatalytic activity compared to pure Cu2(OH)PO4 and was very stable even after repeated cycling. More importantly, the introduction of hydrogen peroxide (H2O2) could combine the photocatalytic and photo-Fenton effects into one reaction system and maximize the infrared light photocatalytic efficiency. Typically, the rate constant of Cu2(OH)PO4/rGO and H2O2 was more than 6.25 times higher than that of only Cu2(OH)PO4/rGO, and almost 10 times greater than the value for pure Cu2(OH)PO4. Further, a plausible mechanism for the enhanced photocatalytic properties of Cu2(OH)PO4/rGO has been discussed. These findings may help the development of novel hybrid photocatalysts with enhanced infrared light photocatalytic activity for applications in the treatment of chlorophenol-contaminated wastewater.


Introduction
Pesticides are chemical substances widely used in horticulture, forestry and public healthand, [1][2][3][4] of course, in agriculture where the unwanted pests that carry or transmit diseases can be repelled and killed. 57][8] For example, chlorophenols readily bio-accumulate in the human body, and subsequently cause disturbances in the structure of cellular bilayer phospholipids, nally causing carcinogenic effects. 9Therefore, a large number of methods have been developed to remove chlorophenols from water, including adsorption, 10 biological degradation 11 and electrochemical degradation. 12However, adsorption merely concentrates chlorophenols, but does not degrade them into less toxic compounds.Biological treatment suffers from the drawbacks of slow reaction rate and the need for strict control of suitable pH and temperature.For these reasons, an effective technique needs to be proposed for the removal of chlorophenols from different water systems.
Alternatively, semiconductor nanomaterials have been emerging as efficient photocatalysts for the degradation of chlorophenols, such as TiO 2 in the ultraviolet range (<400 nm) 13 and Ag 3 PO 4 in the visible range (400-800 nm). 14For the optimized use of solar energy, efficient and stable photocatalysts that are capable of harvesting infrared light, which accounts for ca.50% of solar energy, are required.Much effort has been under way so far to tentatively seek the efficient photocatalysts, including Bi 2 WO 6 (ref.15) and WS 2 , 16 for the degradation of organic pollutants, but not chlorophenols, under infrared irradiation.The limitation of photocatalysts for pesticide photocatalytic degradation under infrared light is essentially due to the insufficient photocatalytic activity that results from charge-carrier recombination as well as the low-photon energy of infrared light, and the inhibition of charge transfer because of the mismatched band energy alignment between each other.To overcome the above limitation, graphitic carbon nitride coupled with upconversion nanoparticles can extend the activity towards the infrared region for the photodegradation of chlorophenols. 17However, the efficiency of this photocatalyst is rather low due to the narrow absorption band of light at 980 nm.Therefore, infrared light responsive photocatalysts for the degradation of chlorophenols are still being actively pursued.
Herein, we synthesized the infrared-light active nanocomposites composing of copper hydroxide phosphate (Cu 2 (OH)PO 4 ) and reduced graphene oxide (rGO) by a one-step hydrothermal method (Cu 2 (OH)PO 4 /rGO) (Scheme 1).Cu 2 (OH) PO 4 , which consists of CuO 4 (OH) 2 octahedron and CuO 4 (OH) trigonal bipyramid, has been considered as a promising photocatalyst for the degradation of organic pollutants under visible light. 18,19With the presence of the distorted polyhedrons in the crystal structure, Cu 2 (OH)PO 4 is even responsive to infrared light and hence displays photocatalytic activity in the infrared range. 20In previous studies, it has been reported that the generated electrons at CuO 4 (OH) trigonal bipyramids under infrared light irradiation (forming Cu III sites) can be transferred to the neighboring CuO 4 (OH) 2 octahedra (forming Cu I sites). 20,21ubsequently, the produced Cu III sites are responsible for oxidizing chlorophenols (Fig. 1, route 1).Nevertheless, Cu 2 (OH) PO 4 , as an infrared-activated photocatalyst, suffers from the fast recombination of photogenerated electron-hole pairs.To overcome this limitation, graphene with the high-surface area and electrical conductivity should act as an avenue for driving photogenerated carriers away from the surface of Cu 2 (OH) PO 4 , 22,23 facilitating more efficient generation of Cu III sites which are applied to degrade chlorophenols and, as a result, become Cu II sites (Fig. 1, route 2).Moreover, to fully use the advantage of this photocatalyst, we use hydrogen peroxide (H 2 O 2 ) as an electron acceptor to react irreversibly with the produced Cu I sites (that is, photo-Fenton reaction, similar to Cu I -induced Fenton reaction 24 ) to further enhance the separation efficiency of photogenerated electron-hole pairs; while the produced Cu I can effectively promote the generation of highly active hydroxyl radicals (HOc) which are capable of oxidizing chlorophenols, and return to Cu II sites.More importantly, these two processes can complete the full photocatalytic circle and be occurred repeatedly, resulting in the combination of the photocatalytic and photo-Fenton effects in one reaction system and nally maximizing the photocatalytic activity of Cu 2 (OH)PO 4 / rGO for the mineralization of chlorophenols to CO 2 and H 2 O.As expected, our results show that, compared to pure Cu 2 (OH)PO 4 , the as-prepared Cu 2 (OH)PO 4 /rGO exhibits remarkably enhanced photocatalytic activity for the degradation of 2,4dichlorophenol (2,4-DCP, a typical type of chlorophenols) under infrared light irradiation (>800 nm).Typically, the photocatalytic rate constant of Cu 2 (OH)PO 4 /rGO and H 2 O 2 is almost 10 times higher than that of only Cu 2 (OH)PO 4 under the same condition.In addition, there is no appreciable loss of photocatalytic activity aer repeated cycles, and the morphology and structure of Cu 2 (OH)PO 4 /rGO remain nearly unchanged.Finally, mechanism of enhanced photocatalysis under infrared light is further proposed and discussed in detailed.

Synthesis of nanocomposites with different mass ratios of Cu 2 (OH)PO 4 to graphene oxide
A series of nanocomposites of Cu 2 (OH)PO 4 and rGO with various mass ratios of graphene oxide (GO) were synthesized by a simple one-step hydrothermal method (Cu 2 (OH)PO/rGO). 20riey, 4 mL of Cu(NO 3 ) 2 solution (1.0 M) and an appropriate amount of GO dispersion (2.0 mg mL À1 , ESI †) were mixed into 20 mL of deionized water under constantly stirring for 15 min.Then, 2 mL of Na 2 HPO 4 solution (1.0 M) was added to the above mixture.Aer stirring for another 1 h, the pH value of the obtained mixture was adjusted to $7.00 through gradually adding NaOH aqueous solution.The resulting suspension was then transferred into a 45 mL sealed teon-scaled autoclave and kept at 120 C for 6 h.Aer naturally cooling to room temperature, the products were collected by centrifuging, washed with deionized water several times, and nally dried overnight in a freezer dryer for further use.According to the mass ratios of Cu 2 (OH)PO 4 and GO in the original mixtures, nanocomposites were labelled as 1 : 0.001, 1 : 0.002, 1 : 0.005, 1 : 0.01, 1 : 0.02, 1 : 0.05 and 1 : 0.1, respectively.

Characterization
Transmission electron microscopy (TEM) images were obtained on a JEM-2100 microscope at an acceleration voltage of 200 kV.Morphologies of samples were characterized using scanning electron microscopes (FE-SEM, S-4800, an acceleration voltage of 10 kV, Hitachi High-technologies, Japan) with an energy-dispersive X-ray (EDX) acceleration voltage analyzer.X-ray diffraction (XRD) patterns was obtained from a Bruker D8 Advance X-ray diffractometer (Bruker, USA) with Cu-Ka radiation (l ¼ 1.5406 Å) at a scanning rate of 10 min À1 and a scanning range from 10 to 90 .X-ray photoelectron spectroscopy (XPS) measurement was carried out with an ESCALab220i-XL spectrometer using a twin-anode Al-Ka X-ray source (1486.6 eV).Micro-Raman spectra were achieved from a Raman Spectroscope (Renishaw inVia plus, United Kingdom) under ambient conditions with 514 nm excitation from an argon ion laser.Ultraviolet-visible (UV-vis) data were acquired with a U-3900 spectrophotometer (Hitachi, Ltd., Japan).Ultraviolet-visible-infrared (UV-vis-IR) diffuse reectance spectra were recorded at room temperature on an Agilent Cary 500 UV-vis-IR Spectrometer equipped with an integrating sphere using BaSO 4 as a reference.The photoluminescence spectra were obtained using a Horiba Jobin Yvon FluoroLog3 spectrometer.

Photocatalytic performance measurement
The photocatalytic activity of Cu 2 (OH)PO 4 /rGO was evaluated by the degradation of 2,4-DCP in the presence and absence of H 2 O 2 under infrared light irradiation.In addition, the photocatalytic activity of Cu 2 (OH)PO 4 and sample 1 : 0.005 to 2,4-DCP was explored under visible light irradiation.Typically, 60 mL mixture of sample 1 : 0.005 (60 mg) and 2,4-DCP (30 mg mL À1 ) was rstly stirred in the dark for 100 min to achieve an adsorption/desorption equilibrium between the photocatalyst and 2,4-DCP.Aerwards, an appropriate amount of H 2 O 2 aqueous solution was added into the obtained mixture prior to photo-irradiation if necessary.Then, the mixture was irradiated by infrared light (>800 nm) using a 300 W xenon lamp installed an 800 nm cut-off lter, where the transmission spectrum of the cut-off lter was shown in Fig. S1.† Then, 1.5 mL of mixture was collected at varied irradiation time, centrifuged, and nally analyzed by a UV-3900 UV-vis spectrophotometer to determine the concentration of 2,4-DCP in the absence of H 2 O 2 or by a Multi TOC Analyzer (2100, Analytik Jena AG Corporation) to detect total organic carbon (TOC) in the presence of H 2 O 2 .The photodegradation of 2,4-DCP by other Cu 2 (OH)PO/rGO were also performed under the similar condition.In addition, the stability of Cu 2 (OH)PO/rGO was studied by a separated photodegradation experiment which repeatedly re-employed the used samples for the next cycle under the identical conditions.Aer each photocatalytic process, Cu 2 (OH)PO/rGO was recovered by centrifugation, washed with deionized water, and then dried in the freezer dryer under vacuum for 24 h before until the subsequent reaction cycle.

Detection of hydroxyl radical
The generation of HOc by nanocomposites under infrared light irradiation was evaluated using terephthalic acid (TA). 25The concentration of hydroxyl radical (HOc) is determined via monitoring the uorescence of 2-hydroxy terephthalic acid (TAOH, the maximum uorescence peak at 435 nm) which is formed from the reaction of HOc with TA.Taking sample 1 : 0.005 as an example, ve groups were obtained: group I (TA and infrared light); group II (TA, H 2 O 2 and infrared light); group III (sample 1 : 0.005 and infrared light); group IV (TA, sample 1 : 0.005 and infrared light); group V (TA, H 2 O 2 , sample 1 : 0.005 and infrared light).The nal working concentrations were 50 mg mL À1 , 100 mM and 500 mM for sample 1 : 0.005, H 2 O 2 and TA, respectively.Aer infrared light irradiation, the changes at the 435 nm uorescence emission peak were recorded.

Cytotoxicity assay for Cu 2 (OH)PO 4 /rGO nanocomposite
The cytotoxicity of Cu 2 (OH)PO 4 /rGO nanocomposites with 1 : 0.005 ratio to HUVECs (human umbilical vein endothelial cells) was assessed by using the CCK-8 assay.HUVECs were placed in 96-well plates at a density of 8 Â 10 3 cells per well, where HUVECs were maintained in the DMEM medium with 10% FBS, 1% penicillin/streptomycin for 24 h at 37 C in 5% CO 2 .Then, the cells were incubated with different concentrations of sample 1 : 0.005 (5, 10, 20, 40, 60, 80, and 90 mg mL À1 ).Aer incubation for another 24 h, the cell medium was removed and replaced with 100 mL of fresh culture medium containing 10 mL of CCK-8 solution per well for detecting the cell viability.Aer 1 h of incubation, cell viability on HUVECs was assayed by measuring the absorbance at 450 nm using a microplate reader (Thermo Scientic, Multiscan MNK3).

Results and discussion
XRD patterns of pure Cu 2 (OH)PO 4 and Cu 2 (OH)PO 4 /rGO with different mass ratios of Cu 2 (OH)PO 4 and GO nanosheets with the thickness of $1.2 nm (Fig. S2 †) are displayed and compared in Fig. 2. According to JCPDS card no.360404, all the diffraction peaks are associated with the orthorhombic phase of Cu 2 (OH) PO 4 . 26Moreover, it is clear that the position of the diffraction peaks of Cu 2 (OH)PO 4 /rGO basically keep unchanged, which indicates that the crystalline structure of Cu 2 (OH)PO 4 is not affected aer the introduction of GO.Additionally, no stackingrelated (002) diffraction peaks of graphene (at $26 for graphite and $13 for graphite oxide) are detected, suggesting that the dispersion of graphene is probably close to the single-sheet level in all the nanocomposites. 27he morphology of pure Cu 2 (OH)PO 4 and Cu 2 (OH)PO 4 /rGO were characterized by SEM and TEM measurement.As shown in Fig. 3, pure Cu 2 (OH)PO 4 are ellipsoid-shaped with an average length of 3-4 mm and an aspect ratio of $3, and there are ravines on their surface that are short and narrow.In contrast, with the addition of GO, Cu 2 (OH)PO 4 are tightly encapsulated by (rGO) nanosheets, which is consistent with the TEM images shown in Fig. S3.† This indicates the presence of strong van der Waals force between graphene and Cu 2 (OH)PO 4 . 28In addition, on increasing the mass ratio of GO, more Cu 2 (OH)PO 4 is encapsulated by graphene, exhibiting more obviously crinkled and rough textures on the surface of Cu 2 (OH)PO 4 /rGO.Remarkably, graphene nanosheets not only are adsorbed onto the surface of Cu 2 (OH)PO 4 /rGO tightly, but also are connected or even overlapped between the adjacent microcrystals, building interconnected conductive pathways for electron transfer.
The structural and chemical information of Cu 2 (OH)PO 4 / rGO was further studied using Raman spectroscopy and XPS measurement.Raman spectra of GO, rGO, pure Cu 2 (OH)PO 4 and Cu 2 (OH)PO 4 /rGO are displayed in Fig. S4a.† From Raman spectra of pure Cu 2 (OH)PO 4 and Cu 2 (OH)PO 4 /rGO, vibration peak at $972.5 cm À1 is the characteristics of Cu 2 (OH)PO 4 .Raman spectra of rGO exhibits two characteristic peaks corresponding to D band at around 1350.4 cm À1 (involving the disorder and defect) and G band at about 1599.6 cm À1 (involving rst order scattering of the tangential stretching phonon mode), respectively. 29,30In comparison to rGO, Cu 2 (OH) PO 4 /rGO also shows the typical features of graphene with the presence of D band and G band, indicating the successful combination of Cu 2 (OH)PO 4 with rGO.Interestingly, it can be obviously found from Fig. S4b † that, on progressively increasing the proportion of GO, there is signicant red-shi in the G band; while the D band rstly blue-shis to 1361.14 cm À1 from samples 1 : 0.001 to 1 : 0.005, and then red-shis to 1351.95 cm À1 from samples 1 : 0.005 to 1 : 0.1.This indicates the presence of van der Waals interaction alone with charge transfer between rGO and Cu 2 (OH)PO 4 , and the interaction is the strongest for sample 1 : 0.005. 31Moreover, the D/G intensity ratios of Cu 2 (OH)PO 4 /rGO are larger than that of GO (I D /I G ¼ 0.806), which suggests a decrease in the average size of the sp 2 domains upon reduction of GO, as well as an increase of edge planes and the degree of disorder. 32The full-scale XPS spectra of Cu 2 (OH)PO 4 /rGO shown in Fig. 4a and S5 † reveal the presence of P, O, and Cu elements, which is consistent with the EDS results shown in Fig. S6.† For the XPS spectra of pure Cu 2 (OH) PO 4 , two main peaks are observed at about 936.61 and 955.86 eV, which can be attributed to Cu 2p 3/2 and Cu 2p 1/2 of copper ions, respectively (Fig. 4b). 33More importantly, the peak position of Cu 2p 3/2 in the Cu 2 (OH)PO 4 /rGO rstly decreases from 935.76 eV to 935.46 eV as the mass ratio increases from 1 : 0.001 to 1 : 0.005, and then increases to 936.56 eV on further increasing the proportion of GO.This may be due to the screening effect via charge transfer between rGO and pure Cu 2 (OH)PO 4 . 34With low mass ratio of GO, the electron of rGO can effectively transfer to pure Cu 2 (OH)PO 4 , causing the shi of both Cu 2p 3/2 and Cu 2p 1/2 towards lower binding energy; whereas, above the mass ratio of 1 : 0.005, rGO begins to aggregate, thereby leading to inhibit electron transfer.Therefore, we predict that sample 1 : 0.005 may be the optimal mass   35,36 There is also three peaks at the same position for Cu 2 (OH)PO 4 / rGO (ESI, Fig. S7 †); while the peak areas of C-OH and C]O signicantly decrease compared to GO, suggesting the oxygen functional groups are removed during the hydrothermal treatment.This is good agreement with the result obtained by the Raman spectroscopy.These demonstrate that graphene is successfully combined with Cu 2 (OH)PO 4 and simultaneously is reduced effectively during the hydrothermal reaction.Therefore, this structure of Cu 2 (OH)PO 4 /rGO can enhance their photocatalytic activity under infrared light via promoting charge separation of photocarriers. 37,38he optical properties of pure Cu 2 (OH)PO 4 and Cu 2 (OH)PO 4 / rGO are investigated by UV-vis-NIR diffuse reectance spectroscopy (DRS).As shown in Fig. S8a, † pure Cu 2 (OH)PO 4 displays a strong absorption in the infrared region, which can be further tted with four Gaussian peaks centered at 662, 777, 965 and 1237 nm (1.55 eV, 1.32 eV, 1.06 eV and 0.88 eV, respectively, Fig. S8b †).According to the previous report, the absorption peak at $777 nm is mainly attributed to the 2 E g -B 1g transition for Cu II sites that exists in axially elongated CuO 4 (OH) 2 octahedra, and the absorption peaks located at $662, 965, and 1237 nm are largely associated with d-d transitions for Cu II sites that exist in axially compressed CuO 4 (OH) trigonal bipyramids. 20,39,40On the introduction of GO into Cu 2 (OH)PO 4 , Cu 2 (OH)PO 4 /rGO exhibits enhanced optical absorption in infrared region over pure Cu 2 (OH)PO 4 , with red-shi in the maximum absorption peak.(Fig. S8c, d and S9 †) Note that the color of Cu 2 (OH)PO 4 /rGO becomes much darker with the increasing mass ratio of GO.Remarkably, when mass ratio of GO reaches up to 1 : 0.01, the absorption peak in the infrared region gradually decreases because that the excessive graphene is capable of shielding the infrared light and inhibits the photo absorption by Cu 2 (OH)PO 4 . 41,42o evaluate the photocatalytic properties of Cu 2 (OH)PO 4 /rGO under infrared light (l > 800 nm), 2,4-DCP, which is fairly stable under solar light irradiation, is used as a model pollutant.To minimize the loss of 2,4-DCP by evaporation, the temperature of the solution was maintained at 20-25 C during infrared light irradiation.Fig. 5a and S10a † shows time proles of C t /C 0 under infrared light irradiation in the presence of Cu 2 (OH)PO 4 /rGO, where C t is the concentration of 2,4-DCP at the irradiation time of t and C 0 is the concentration in the adsorption equilibrium of the photocatalysts before photo-irradiation.The results show that almost no photolysis is observed without photocatalysts aer 6 h of infrared light irradiation.2,4-DCP is slightly degraded in the presence of Cu 2 (OH)PO 4 ; while the degradation is remarkably accelerated with Cu 2 (OH)PO 4 /rGO.On progressively increasing the proportion of GO, the degradation rate initially increases for Cu 2 (OH)PO 4 /rGO, and then declines.Among all the nanocomposites, sample 1 : 0.005 shows the highest degradation rate (87.1%) aer irradiation with infrared light for 6 h, which is rather higher than that of pure Cu 2 (OH) PO 4 (66.4%).This enhancement clearly indicates effective recombination suppression, which can be attributed to the van der Waal heterojunction between Cu 2 (OH)PO 4 and rGO.This is consistent with the results of uorescence spectra of pure Cu 2 (OH)PO 4 , 1 : 0.001, 1 : 0.005 and 1 : 0.1, which is shown in Fig. S11.† However, from samples 1 : 0.01 to 1 : 0.1, the degradation rate decreases to 78.5% aer 6 h of infrared light  irradiation.This is largely attributed to the fact that the excessive graphene shields Cu 2 (OH)PO 4 , preventing absorption and blocking electron transfer between Cu 2 (OH)PO 4 and graphene, as demonstrated by the results of XPS, Raman spectrum and DRS measurements.It is worth noting that these values are still higher than that of pure Cu 2 (OH)PO 4 .Fig. 5b shows the evaluation of the absorption spectra of 2,4-DCP with reaction time in the presence of sample 1 : 0.005.
The pseudo-rst order kinetic model is then used for the determination of the photocatalytic degradation rate constant (k, h À1 ) which is expressed by eqn (1): 43,44 Fig. 5c and S10b † depict the ln(C 0 /C t ) versus t for pure Cu 2 (OH)PO 4 and Cu 2 (OH)PO 4 /rGO.Pure Cu 2 (OH)PO 4 presents an apparent photocatalytic rate constant of 0.190 h À1 under the irradiation of infrared light.However, the photocatalytic rate constants of Cu 2 (OH)PO 4 /rGO for 2,4-DCP are higher than that of pure Cu 2 (OH)PO 4 under the same condition.Among these Cu 2 (OH)PO 4 /rGO with various mass ratios, sample 1 : 0.005 shows the highest photocatalytic rate constant, which is 1.72 times than that of pure Cu 2 (OH)PO 4 .The photocatalytic rate constant of other samples is summarized in Table S1.† The above results clearly demonstrate that graphene plays a signicant role in the enhanced infrared light photocatalytic activity.A tentative photocatalytic mechanism of Cu 2 (OH)PO 4 / rGO for degradation of 2,4-DCP under infrared light irradiation is proposed and schematically illustrated in Fig. 1.Similar to pure Cu 2 (OH)PO 4 , 20 copper atoms of Cu 2 (OH)PO 4 /rGO have two nonequivalent crystallographic sites, including octahedral sites (rGO-OCT Cu II ) and trigonal bipyramidal sites (rGO-TBP Cu II ).In this system, the distorted polyhedrons can lead to a net dipole moment in such units, facilitating electron transfer from rGO-TBP Cu II to neighboring rGO-OCT Cu II .Beneting from the strong absorption in infrared region, Cu 2 (OH)PO 4 /rGO can be excited by infrared light and therefore generate electrons (forming rGO-OCT Cu I sites at the CuO 4 (OH) 2 octahedra) and holes (forming rGO-TBP Cu III sites at the CuO 4 (OH) trigonal bipyramids).Then, by taking advantage of the high-specic area surface and electrical conductivity, graphene nanosheets can provide more sites to absorb 2,4-DCP and promote photogenerated holes transfer from rGO-TBP Cu III to 2,4-DCP, which is responsible for degrading 2,4-DCP.In parallel, rGO-OCT Cu I may drive the half-reaction of oxygen reduction.The main processes in the photocatalytic degradation of 2,4-DCP could be summarized as follows: rGO-TBP Cu II  Furthermore, the introduction of H 2 O 2 as a sacricial electron acceptor can maximize the photocatalytic efficiency for the degradation of 2,4-DCP with Cu 2 (OH)PO 4 /rGO through the combination of the photocatalytic and photo-Fenton effects into one reaction system.Taking sample 1 : 0.005 as an example, it is found that the photocatalytic activity is gradually improved on the increase of the amount of H 2 O 2 (Fig. S12 †).As shown in Fig. 5d, when 2.4 mL H 2 O 2 is added, the photodegradation efficiency of 2,4-DCP increases from 33.6% to 74.1% aer 2 h of infrared light irradiation.According to the kinetic curves in Fig. S13, † the photocatalytic rate constant of sample 1 : 0.005 and 2.4 mL H 2 O 2 is more than 6.25 times higher than the corresponding value for only sample 1 : 0.005, and almost 10 times greater than the value for pure Cu 2 (OH) PO 4 .These results conrm that the addition of H 2 O 2 as a sacricial acceptor can remarkably enhance the photocatalytic efficiency of Cu 2 (OH)PO 4 /rGO for the degradation of 2,4-DCP under infrared light irradiation.In addition, it is found from Fig. S14 † that the photocatalytic properties of sample 1 : 0.005 is not much better than pure Cu 2 (OH)PO 4 under visible light irradiation.At the same time, sample 1 : 0.005 shows the higher degradation rate (87.1%) aer irradiation with infrared light for 6 h relative to the degradation rate (54.5%) with visible light irradiation, which it is possible that the redox potential of 2,4-DCP is not match the band position of Cu 2 (OH)PO It is generally accepted that H 2 O 2 is a better sacricial electron acceptor than O 2 . 45Therefore, in the presence of H 2 O 2 , rGO-OCT Cu I produced under infrared light irradiation can react irreversibly with H 2 O 2 to (a) further improve the electronhole separation and (b) the effective generation of powerful reactive species HOc radicals via the photo-Fenton reaction, which can further oxidize 2,4-DCP during the photocatalytic degradation reaction, 46 nally returning to Cu II sites.In addition to reaction (4), these two processes complete the full photocatalytic circle and are occurred repeatedly, nally maximizing the photocatalytic efficiency for the mineralization of 2,4-DCP to CO 2 and H 2 O.This is the unique advantage of these Cu 2 (OH)PO 4 /rGO as a novel infrared-light-active photocatalyst.As shown in Fig. S15, † it can be easily seen that sample 1 : 0.005 is capable of catalyzing H 2 O 2 efficiently to generate HOc only when irradiated with infrared light.The amount of the generated HOc for sample 1 : 0.005 under infrared light irradiation is over 7 times than the control groups.Remarkably, there is no generation of HOc without the irradiation of infrared light.These demonstrate that the infrared light photocatalytic activity is further signicantly enhanced when the photocatalytic and photo-Fenton effects are combined in one reaction system.
Finally, the stability of Cu 2 (OH)PO 4 /rGO was evaluated using cyclic experiments.Fig. 6a shows the degradation of 2,4-DCP for four runs of reactions.The photocatalytic efficiency of sample 1 : 0.005 does not decrease even aer several cycles.This indicates that sample 1 : 0.005 can be efficiently recycled and reused for repeated cycles without appreciable loss of activity.SEM images and XPS spectra of sample 1 : 0.005 before the photocatalytic reaction and aer the photocatalytic reaction cycles are presented in Fig. 6b and c

Conclusion
In summary, Cu 2 (OH)PO 4 /rGO nanocomposites with different mass ratios of GO are successfully synthesized through a onestep hydrothermal method.By coupling with graphene, the photocatalytic activity of pure Cu 2 (OH)PO 4 is remarkably enhanced under infrared light irradiation.With an optimal mass ratio of 0.5 wt% graphene oxide, the highest rate of 2,4-DCP degradation is achieved due to the effective hybridization between Cu 2 (OH)PO 4 .Moreover, the introduction of H 2 O 2 as the sacricial electron acceptor can maximize the infrared light photocatalytic activity via the combination of the photocatalytic and photo-Fenton effects into one reaction system.In this system, the reaction of H 2 O 2 with rGO-OCT Cu I sites, as well as the coupling of graphene with Cu 2 (OH)PO 4 , can improve the separation and transportation of photogenerated electrons and holes; while the generated HOc via the photo-Fenton reaction can further oxidize 2,4-DCP.These are the unique advantages of Cu 2 (OH)PO 4 /rGO as a novel infrared-light-active photocatalyst.Typically, the photocatalytic rate constant of Cu 2 (OH)PO 4 /rGO and H 2 O 2 is $6.25 times higher than the corresponding value for only Cu 2 (OH)PO 4 /rGO, and $10 times greater than the value for pure Cu 2 (OH)PO 4 .Moreover, Cu 2 (OH)PO 4 /rGO is very stable aer many photocatalytic cycles and can be reused without signicant loss of photocatalytic activity.In addition, a possible decomposition mechanism for the degradation of 2,4-DCP is further proposed.This work may help the development of a new strategy to search stable and effective photocatalysts with high infrared light photocatalytic activity and bring the promise to fullment of actual applications in the treatment of nonbiodegradable chlorophenols with lower costs and nonsecondary pollution to the environment.

Scheme 1
Scheme 1 Schematic illustration of the preparation of Cu 2 (OH)PO 4 / rGO.
4 corresponding to visible light.The mechanism of photocatalytic degradation of 2,4-DCP by Cu 2 (OH)PO 4 /rGO with H 2 O 2 as the sacricial electron acceptor is proposed as follows: rGO-OCT Cu I + H 2 O 2 / rGO-OCT Cu II + HOc + OH À (6) H + + OH À / H 2 O (7) 2,4-DCP + HOc / H 2 O + CO 2 + H + + Cl À , respectively.No differences have been found, either in the morphology of sample 1 : 0.005 or in their chemical structure aer photocatalytic degradation.Furthermore, XRD patterns shown in Fig. 6d indicate that the crystal structures of sample 1 : 0.005 remained the same aer repeated photocatalytic cycling.At the same time, SEM images, XPS spectra and XRD patterns shown in Fig. S16-S18, † respectively arrested that other Cu 2 (OH)PO 4 /rGO also possess nice photocatalytic stability.In addition, in order to evaluate the biosecurity of Cu 2 (OH)PO 4 /rGO nanocomposites, the cytotoxicity of sample 1 : 0.005 was assessed to HUVECs (human umbilical vein endothelial cells) by the CCK-8 assay.It can be seen from Fig. S19 † that no obvious cytotoxicity is induced in human umbilical vein endothelial cells, even at high concentrations up to 90 mg mL À1 , which exhibits Cu 2 (OH)PO 4 /rGO nanocomposites have good biosecurity.

Fig. 6
Fig. 6 (a) Cycles of the photocatalytic degradation of 2,4-DCP in the presence of sample 1 : 0.005.(b) SEM image of sample 1 : 0.005 after the photocatalytic reaction cycles.Inset: enlarged SEM image.(c and d) The Cu 2p XPS spectra and XRD patterns of sample 1 : 0.005 after the photocatalytic reaction cycles.