Thermolysis-derived copper sulphide nanoparticles as visible-light photocatalysts for the degradation of methylene blue Authors: Jocelyn N Kuate, Adrien P Yepseu, Awawou G Paboudam, Walter N Kun, Mariappan Mariappan, Monisha Arumugam, Viorel Cîrcu, Sunder N Dhuri, Peter T Ndifon Published: 2026 ABSTRACT AbstractCopper sulphide (Cu2−xS) photocatalyst have been synthesized by thermolysis from [Cu(L)2(H2O)2](NO3)2.H2O (1) and [Cu(L)(SCN)2(H2O)2]. 6H2O (2) complexes as single source precursors (SSPs) using olive oil (OO) and castor oil (CO) as eco-friendly capping agents at 190 and 230 °C. The as-prepared Cu2−xS nanoparticles (Nps) were characterized using powder X-ray diffraction (p-XRD), scanning electron microscopy (SEM), Transmission Electron Microscopy (TEM), energy-dispersive X-ray (EDX), Fourier transform infrared (FT-IR), UV–Visible, and Raman spectroscopy, and employed for the visible-light-driven degradation of cationic methylene blue (MB) dyes. p-XRD revealed a monoclinic Djurleite (Cu1.94S) phase at 190 °C when complex (2), (irrespective of the capping agent) was used as precursor, while orthorhombic Djurleite (Cu1.97S) phase was observed when the reaction temperature and precursor type were varied. EDX confirmed the elemental composition of Cu2−xS nanoparticles. TEM images showed the formation of nanoparticles of various shapes including spherical, elliptical, and irregularly shape, with particle sizes ranging from 5 to 142 nm. Optical properties of Cu2−xS nanoparticles showed band gaps ranging between 2.17 and 2.48 eV and 2.20–2.46 eV for Castor and Olive oil-capped NPs respectively, (larger than the bulk material value of 1.3 eV), due to quantum size confinement effect. Raman spectroscopy confirms the formation of castor and olive oil-capped Cu2−xS nanoparticles. Under optimized conditions, Cu2−xS demonstrated excellent photocatalytic performance, achieving 56.7 and 59.8% degradation efficiency of methylene blue (MB) within 120 min under visible light for CuS2 and CuS6 derived from complex (2) at 190 °C in olive oil or castor oil respectively. These results clearly highlight the effect of reaction parameters on the properties of Copper Sulphide NPs which render them potential candidates for environmental remediation applications.Supplementary InformationThe online version contains supplementary material available at 10.1186/s11671-026-04675-7.Keywords: Thiosemicarbazone, Thermolysis, Single source precursors, Copper sulphide nanoparticles, Optimization and photocatalysis study Introduction Over the past few decades, the discharge of hazardous chemical substances into the environment by many industries such as textile, paints, printing inks, paper, food, drug and cosmetics, have generated many emerging pollutants including dyes [1, 2]. Dyes are an important class of synthetic, coloured organic compounds which are the primary agents that cause this pollution due to their large-scale use in many industries. Among these dyes, methylene blue (MB) is widely used in industries such as textiles, pharmaceuticals, and food processing. Their release into aquatic ecosystems poses serious environmental and health risks because of their persistence, toxicity, and resistance to conventional treatment methods [3, 4]. In response to these challenges, photocatalysis represent one of the recent oxidation techniques (Advanced Oxidation Processes) which has proven to be of interest due, not only to its cost-effectiveness, but also to its efficiency in degrading recalcitrant organic compounds into harmless products (CO2 and H2O) [5, 6]. Photocatalysis has therefore emerged as a committed avenue for the degradation of organic pollutants, offering an environmentally friendly and cost-effective solution to mitigate water contamination [7]. The considerable interest in nanomaterials as photocatalyst is triggered by their remarkably nanoscale properties such as large surface area, high reactive sites and efficient light absorption capacity relative to those of bulk materials. Furthermore, semiconductor-based photocatalysis harnesses solar or artificial light to initiate oxidative reactions, leading to the breakdown of organic pollutants into non-toxic by-products [5, 6]. The ongoing development of efficient visible light-responsive photocatalysts is therefore, of great significance, making full use of solar energy [8]. Semiconductor transition metal sulphide nanomaterials have been extensively investigated as typical photocatalyst due to their high surface area, moderate band gap and high efficiency [9]. Among these, copper sulphide (Cu2−xS) has garnered worthy concern as traditional photocatalysts due to its non-toxic, cost-effective, and eco-friendly properties and its composition of earth-abundant elements, ideal for environmental remediation. Furthermore, its tuneable electronic structure and lower recombination rate of photogenerated electron–hole pairs have been demonstrated to significantly improve photocatalytic efficiency. Yepseu et al. reported the degradation of MB under UV irradiation with degradation efficiency ranging from 47.1 to 80% after 90 min using digenite (Cu9S5) photocatalyst [10]. Another interesting property of copper sulphide is its existence in variable stoichiometries (Cu2−xS) and crystalline phases from the covellite “Copper-poor” phase (x = 1) to the chalcocite “Copper-rich” phase (x = 0) [11, 12]. Furthermore, the covellite copper-poor phase has been known to exist in two forms namely, brown and green CuS. While the Green CuS has been found to be the true crystalline form of covellite (composed of 1/3 Cu(II) and 2/3 Cu(I)), the brown CuS is a poor crystalline or amorphous type of covellite [13]. On the contrary, the chalcocite “Copper-rich” phase exhibit three different forms: low chalcocite (α-Cu2S), high chalcocite (β-Cu2S) and cubic chalcocite phase (γ-Cu2S). Among these three structures, the low chalcocite phase is the most stable at room temperature [14]. Copper sulphides exhibit intrinsic semi-conductivities, and solar energy-controlled characteristics depending on its stoichiometry [15]. Several methods based on chemical approach have been reported for the synthesis of copper sulphides nanoparticles and these include solvothermal/hydrothermal, sonochemical and thermolysis of single-source precursors in high boiling point solvents that act as surface passivating agents [16]. Among these synthetic methods, the decomposition of single-source precursor (SSPs) derived from metal complexes, has been progressively employed since the preformed Cu–S bond might provide a better control method over the stoichiometry of the resulting copper sulphide phase [17]. Several studies have reported on the use of organic ligands such as dithiocarbamates, xanthates, thioureas, and thiosemicarbazones for the preparation of complexes as SSPs for the synthesis of metal sulphide semiconductor nanomaterials [17, 18]. Some of these studies have focused on the use of synthetic organic solvents as stabilising agents [19]. However, the toxic nature of some of the solvents used in this approach still remains a problem. The synthesis of less toxic and well-defined pure-phase copper sulphide is a challenge, and the very close relationship that exists between optical properties and shapes means that tailoring the morphology of nanostructures is essential in the synthesis of desired materials with specific properties. Moreover, the photocatalytic efficiency of copper sulphide nanoparticles depends on the as-obtained crystalline phases. In this work, we address these gaps by using olive oil and castor oil as green capping agents for synthesizing CuS nanoparticles, replacing the toxic chemicals such as alkylamines or phosphines [20, 21]. The role of capping or stabilising agents is critical to preventing aggregation and control particle size, shape, and surface morphology. The use of plant extracts as capping agents, represents an economically viable and environmentally friendly alternative to conventional stabilizing agents in synthesis. Barrios-Navarro et al. reported the synthesis of zinc oxide (ZnO) nanoparticles using an eco-friendly methodology with a natural extract from fruit of tejocote (Crataegus mexicana) as a stabilizing agent. The as-prepared nanoparticles were used as a photocatalyst and they exhibited excellent photocatalytic properties [22]. In this paper, we report on the preparation of copper sulphide (Cu2−xS) nanoparticles by the thermolysis of copper complexes of thiosemicarbazone and thiocyanate co-ligands as SSPs using green solvents (olive and castor oil) at 190 and 230 °C. The use of copper thiosemicarbazones complexes as single-source precursors to prepare copper sulphide nanoparticles has proven to be an efficient route and gives rise to good-quality copper sulphides. In addition, the effect of reaction parameters (capping agent, precursor type and reaction temperature) on the optical and photocatalytic properties of the as-obtained copper sulphide nanoparticles are also reported. Experimental Chemicals Furan-2-carboxaldehyde (99%, SRL), Thiosemicarbazone (99%, PANREAC), Ammonium thiocyanate (99%, PROBUS), Copper nitrate (Cu(NO3)2.5H2O; 99%, Riedel-De-Haën), Ethanol (99.9%, Jiangsu), Hexane (98%, GPR), Olive Oil (local supermarket), Castor Oil (JHD), Hydrochloric Acid (34.5%, Avra), Sodium hydroxide (99%, Avondale laboratories), Methylene Blue (96%, Alfa Aesar) and toluene (99%, GPR), were obtained from commercial sources and used without any further purification. Physical measurements Electronic spectra (340–800 nm) were recorded at room temperature using OCEAN OPTICs UV–vis Spectrophotometer. FTIR spectroscopy data was collected over the range from 4000 to 400 cm−1 on FT-IR Perkin Elmer 400 spectrometer. The elemental analysis (CHNS) was carried out in a Vario micro-CUBE elemental analyser. Thermal analysis (DT-TGA) was carried out under an inert environment from 30 to 945 °C with a heating rate of 10 °C/min1 using a SCAN 600 TGA. The powder XRD measurements of the samples were carried out using Rigaku smartlab SE X-ray diffractometer with Cu-Kα radiation (λ = 1.54056 Å) and 2θ range = 20–60°. The morphology and elemental composition of the nanoparticles was characterized using a Carl Zeiss Gemini SEM 500 scanning electron microscopy (SEM) at an accelerating voltage of 20 kV at different magnifications. TEM was used to determine the sizes and morphologies of the nanoparticles. Analyses were done on a LEO JEM 912 electron microscope with an acceleration voltage of 120 kV and a tungsten wire filament. The copper sulphide samples were prepared by placing a drop of dilute solution of sample in toluene onto a copper grid and the samples were allowed to dry completely at room temperature before the analysis. Synthesis of precursors Synthesis of Schiff base ligand The thiosemicarbazone ligand (L) was synthesised and characterised using the method reported elsewhere [20]. Synthesis of complexes The synthesis of thiosemicarbazone complexes employed a previously reported procedure with some modification [23, 24]. Preparation of thiosemicarbazone complexes (1) Preparation of thiosemicarbazone complexes (1) Copper nitrate (5 mmol, 1.40 g) dissolved in ethanol (20 mL) was added drop-wise to a solution (20 mL) of furan-2-carbaldehyde thiosemicarbazone ligand (10 mmol, 1.69 g). The reaction mixture was then refluxed for 3 h. The instantly formed precipitate was filtered, washed thoroughly with ethanol and dried at room temperature. Analytical Data are presented on Tables 1–2. (b)Preparation of mixed thiosemicarbazone complexes (2) Preparation of mixed thiosemicarbazone complexes (2) Table 1Some physical properties of the prepared complexesCompoundColorm.p (°C)Yield(%)Elemental analysis Found (calc.)%C%H%N%SC6H7N3OS (Ligand, L)Brown155 ± 28642.59 (43.34)3.61 (4.17)24.83 (24.93)18.95 (18.33)[Cu(L)2(H2O)2]. (NO3)2. H2OOlive green> 300 °C26.424.23 (24.85)2.65 (3.48)18.15 (19.32)11.23(11.06)[Cu(L)(SCN)2(H2O)2]. 6H2OYellowish> 300 °C24.418.32(19.53)3.07(4.51)13.65(14.23)18.57(19.55) Some physical properties of the prepared complexes Brown 155 ± 2 Olive green > 300 °C 11.23 (11.06) Yellowish > 300 °C 18.32 (19.53) 3.07 (4.51) 13.65 (14.23) 18.57 (19.55) Table 2Selected infrared band (wavenumber) cm−1 of the ligand and its complexesCompoundυ(N–H)υ(O–H)υ(SCN)υ(C = N)υ(C = S)υ(M–N)υ(M–S)L3130––1606840––CuL (1)Overlap3083–1622823588463CuLSCN (2)Overlap328721721591824590449 Selected infrared band (wavenumber) cm−1 of the ligand and its complexes A hot solution of furan-2-carbaldehyde thiosemicarbazone ligand (5 mmol, 0. 85 g) in ethanol (20 mL) was added to a stirred solution of copper nitrate (5 mmol, 1.40 g) in ethanol (20 mL). After 15 min, a solution of NaN3 (0.65 g, 10 mmol) in water (5 ml) The resulting mixture was stirred for 3 h. Then the mixture was filtered, washed with an excess of ethanol and dried at room temperature. Analytical Data are presented on Table 1, Table 2 (Scheme 1). Reaction equation and chemical structure of copper complex (1) and (2) Synthesis of vegetable oil-capped copper sulphide nanoparticles The synthesis of metal sulphide nanoparticles was carried out using the protocols described in the literature, with a slight modification [25, 26]. For this purpose, 0.5 g of each precursor, [Cu(L)2(H2O)2](NO3)2. H2O (1) or [Cu(L)(SCN)2(H2O)2]. 6H2O (2) was dispersed in 6.0 mL of vegetable oil (olive oil or castor oil). The resulting solution was then injected into a three-necked flask containing 10 mL of vegetable oil heated to the desired temperature (190 or 230 °C). A slight decrease in the temperature of the solution of about 20 °C was observed. Subsequently, the solution was allowed to stabilize at the desired temperature. After 1 h of reaction, the heating was stopped and the mixture allowed to cool. Once the temperature was drop down to around 70 °C, ethanol was then added to the reaction mixture which resulted in the formation of a flocculent precipitate. This precipitate was washed several times with hexane, then with ethanol and the particles separated by centrifugation to give vegetable oil capped copper sulphide nanoparticles. The resulting nanoparticles are labelled as CuS1 and CuS3 for Cu(2−x)S prepared from complex (1) at 190 and 230 °C respectively using olive oil as capping agent and CuS2 and CuS4 for Cu(2−x)S obtained from complex (2) at the same temperatures, using olive as capping agent respectively. The samples labels CuS5; CuS7 and CuS6; CuS8 were used for castor oil-capped NPs obtained using complexes (1) and (2) respectively at 190 and 230 °C, as shown in Scheme 2. Schematic representation of the preparation of castor oil-capped and olive oil-capped NPs Point of zero charge determination (PZC) The point of zero charge of the as-obtained nanoparticles was determined by the pH drift method [27]. Thus, the PZC values for Cu(2−x)S were determined using a 10–1 M NaCl solution at a temperature between 303–333 °K. In this respect, a mixture consisting of 5 mg of each sample and 20 mL of the NaCl solution were prepared in different reaction vessels. The pH of each suspension was then adjusted to an initial pH value (pHi) between 3 and 9 using 0.1 M H2SO4 or 0.1 M NaOH solutions. Each flask was then shaken vigorously in a shaker bath for 24 h. After settling, the final pH (pHf) of each suspension was measured very carefully using a DIGITAL pH CUM potentiometer HQ-636. The pH at the point of zero charge (pHpzc) was determined by the intersection of the curves pHf = f(pHi) and pHf = pHi. Photocatalytic studies of nanoparticles The photocatalytic activity of the as-prepared nanoparticles was evaluated using an aqueous solution of methylene blue (MB) as a test contaminant in the presence of visible light. For this purpose, 20 mg of photocatalyst (Cu(2−x)S) was introduced into the photocatalytic reactor containing 100 ml of 10 ppm (10 mg/L) aqueous MB solution. The pH of the solution was adjusted to about 9 by adding 0.1 M NaOH or 0.1 M HCl solution. Before the degradation test, MB solution containing 20 mg of samples was stirred for about 40 min in the dark to achieve an adsorption–desorption equilibrium between MB molecules and the photocatalyst [28, 29]. The reactor was then illuminated using visible light from a xenon lamp for 120 min. Aliquots were collected at 30 min interval and centrifuged to obtain the resulting solution whose absorbance was recorded using UV-Vis-NIR spectrophotometer at maximum wavelength for MB (λ = 662 nm). The photodegradation efficiency (R) was calculated using Eq. 1 [28]. The residual concentration (η) of methylene blue was monitored using Eq. 2 [30] where At and A0 are the absorbance values of the residual and initial MB solution recorded at the wavelength of λ = 663 nm. Results and discussion Ligand and complexes characterization All the as-obtained complexes are coloured powder, non-hygroscopic, stable in air, and are soluble in Dimethylformamide (DMF) and Dimethylsulphoxide (DMSO). The ligand and their Copper complexes obtained are confirmed by IR and UV–Visible spectroscopy, elemental analysis and TG-DTG. The molar conductance values (Table S1) of the complexes (1) and (2) recorded in the DMSO are consistence with electrolytic and molecular nature respectively for complex (1) and (2) [31]. Analytical data of compounds are presented in Table 1. FT-IR and UV–visible spectroscopy The FT-IR spectra of both the complexes and the ligand (Figure S1a) show that the band at 1606 cm−1 attributed to the azomethine group ν(C=N) of the ligand underwent a shift towards a higher wavenumber of 1622 cm−1 for complex (1) and shifted to a lower wavenumbers 1591 cm−1 for complex (2), showing the coordination of the nitrogen of the azomethine. This is justified by the appearance of new bands around 590 cm−1 attributed to ν(M–N) [32]. Furthermore, the band at 840 cm−1 attributed to the ν(C=S) bond in the ligand spectrum shifted to low wavenumbers in the spectrum of the complexes around 824 cm−1, probably due to the coordination of the sulphur atom in the thione form. This is confirmed by the non-ligand-bound bands around 449–463 cm−1 attributed to the ν(M-S) vibration [33]. Furthermore, the spectrum of the mixed-ligand complex shows an intense band at 2172 cm−1 due to the thiocyanate (SCN) asymmetric vibration, evidence of the coordination of thiocyanate through the sulphur atom [34]. In sum, it is evident that the ligand acts in a bidentate manner, coordinating via the sulphur of the thione function and the nitrogen of the azomethine function, thus forming a five-membered ring chelate. The IR data of the compounds are listed in Table 3. Table 3Electronic spectral assignments for Cu (II) complexesCompoundsElectronic bands: nm (cm−1)ε (L mol−1 cm−1)Band assignmentL354 (28,329)397 (25,189)CuL (1)358 (27,933)411 (24,331)665 (15,038)575.3589.3143.8Intra ligand transitionIntra ligand transition2B1g → 2A1gCuLSCN (2)361 (27,701)396 (25,253)665 (15,038)468.4470.0111.8Intra ligand transitionIntra ligand transition2B1g → 2A1g Electronic spectral assignments for Cu (II) complexes 354 (28,329) 397 (25,189) 358 (27,933) 411 (24,331) 665 (15,038) 575.3 589.3 143.8 Intra ligand transition Intra ligand transition 2B1g → 2A1g 361 (27,701) 396 (25,253) 665 (15,038) 468.4 470.0 111.8 Intra ligand transition Intra ligand transition 2B1g → 2A1g The electronic spectra of the thiosemicarbazone Schiff base and its copper (II) complexes are shown in Figure S1b. The Schiff base exhibits two absorption bands at 354 and 400 nm respectively attributed to the π → π* transition of the furan ring moiety and n → π* on the imine nitrogen atom [35]. The electronic spectra of the complexes exhibit, in addition to the bands due to the Schiff base ligand (intra-ligand) around 358 nm and 408 nm, a weak and broad band around 665 nm (15,038 cm−1) attributed to the d-d transition 2Blg→2Alg in a tetragonally distorted octahedral geometry with four short and two long metal–ligand bonds due to the Jahn–Teller distortion which generally absorbed around 16,000 cm−1 [36]. Thermal study Figure S2 shows the thermal analysis carried out between 30 and 945 °C under an inert atmosphere at a heating rate of 10 °C/min. The thermograms of each of complexes (1) ([Cu(L)2(H2O)2] (NO3)2. H2O (1)) and complex (2) ([Cu(L)(SCN)2(H2O)2]. 6H2O (2)) undergo decomposition in two stages. For complex (1), the loss of lattice water molecules (green curve) is observed between 30 and 90 °C representing the first decomposition step with a mass loss observed between 100 and 250 °C followed by an intense endothermic peak attributed to the departure of coordinated water molecules corresponding to a mass loss of 33.21% (calc = 32.44%). The second mass loss corresponding to 33.45% (calc = 32.10%) observed between 250 and 945 °C is attributed to the complete decomposition of the Schiff base ligands. The thermogram of complex (2) shows a first stage of decomposition between 30 and 400 °C, with a mass loss of 29.10% (calc = 29.23%), is due to the progressive loss of water molecules (from lattice to coordinated water molecule). The second stage is characterized by a mass loss of 18.40% (calc = 18.07%), attributed to the decomposition of ligands. The residues obtained from both decompositions are attributed to a mixture of copper sulphide and carbon, representing 32.87% (calc = 32.30%) and 53.4% (calc = 53.6%) of the initial weight of complexes (1) and (2) respectively. The thermal data are presented in Table S2. Characterization of nanoparticles The copper sulphide nanoparticles were synthesized by thermolysis of the single source molecular precursor and the study of the reaction parameters such as the treatment temperature (190 and 230 °C), capping agent (Olive oil and Castor oil) and precursor (1 and 2) on the size, morphology and properties of compounds are investigated. Powder-XRD P-XRD of the nanoparticles obtained at 190 and 230 °C using castor oil for both complex (1) and (2) are shown in Fig. 1. When complex (1) is used as precursor at 190 °C and 230 °C respectively the as-obtained samples CuS5 and CuS7 exhibited similar diffraction patterns with four (04) diffraction peaks at 2θ values of 24.7°, 26.2°, 35.0° and 42.6° with a slight shift (⁓0.3°) originating from (0 0 4), (8 0 0), (1 5 3) and (11 3 2) lattice planes which could be assigned to orthorhombic djurleite (Cu1.97S) phase (ICDD N°: 00–020-0365; a = 26.900 Å, b = 15.720 Å, c = 13.570 Å) [37–40]. This shows that temperature has no effect on the obtained crystalline phase with complex (1). When complex (2) is used as precursor, instead of complex (1), diffraction pattern obtained at 230 °C (CuS8) is also ascribed to orthorhombic djurleite. whereas the sample obtained 190 °C (CuS6) exhibit diffraction peaks at 2θ values 21.3°, 24.5, 24.7°, 27.1° and 46.1° assigned to (4 3 0), (6 1 2), (7 1 1), (− 5 3 2), and (− 3 7 3) which can all be ascribed to monoclinic Djurleite (Cu1.94S) phase (ICDD N°: 00-034-0660; a = 26.897, b = 15.745, c = 13.465 and β = 90.13°) [11]. This clearly shows that temperature has significant effect on the as-obtained crystalline phase. In fact, we can observe that when temperature increase (CuS6 → CuS8). the diffraction peaks at 2θ values at 21.3° and 50.7° attributed to djurleite (Cu1.94S) disappear, and the intensity of the 2θ peaks at 42.6° assigned to djurleite (Cu1.97S) significantly increase. This increase in peak intensity with increase in temperature at 230 °C as well as the absence of some diffraction peaks in the djurleite (Cu1.94S) phase could be explained by a slight desulphurization of monoclinic djurleite leading to the formation of orthorhombic djurleite (Cu1.94S → Cu1.97S). Similar results found by Kundu et al. show the formation of chalcocite (Cu2S) through desulphurization of covellite (CuS) with increase of treatment temperature [41]. In summary, the aforementioned clearly highlights the influence of the precursor as well as the reaction temperature on the composition and the crystal lattice’ parameters of the as-prepared Cu(2−x)S nanoparticles. Similar studies involving the thermal decomposition of Bis(O-alkylxanthato)copper(II) in oleylamine showed the dependence of the factors such as precursor type, growth temperature, and synthetic method used on the as-obtained copper sulphide phases [42]. p-XRD patterns of Cu2-xS NPs; Orthorhombic Djurleite (ICDD: 00-020-0365) and monoclinic djurleite (ICDD: 00-034-0660) The capping agents (olive oil and castol oil) were varied in an attempt to study the effect of capping agent on the crystalline phase and crystallinity of copper sulphide nanoparticles. When olive oil was used instead of castor oil (Fig. 1), the olive oil capped copper sulphide nanoparticles obtained from complex (1) at 230 °C (CuS3) is amorphous whereas the as-obtained sample at 190 °C (CuS1) is crystalline and can be ascribed to orthorhombic djurleite phase (Cu1.97S). In the same vein, when complex (2) is used at 230 °C the sample CuS4 is also amorphous. On the contrary, the sample CuS2 obtained at 190 °C is crystalline and can be ascribed to monoclinic djurleite phase (Cu1.94S). This show that irrespective of the complex used as precursor, the crystallinity of the as-obtained samples decreases with the increase in temperature. Nonetheless, it is worth mentioning that samples CuS1 and CuS2 obtained at 190 °C respectively from complex (1) and complex (2) appear as orthorhombic and monoclinic djurleite phases. This highlights the impact of both temperature and precursor on the crystallinity of copper sulphide nanoparticles. When both sets of olive oil capped (CuS1, CuS2, CuS3 and CuS4) and castor oil (CuS5, CuS6, CuS7 and CuS8) NPs were analysed, we observe that the samples obtained at 230 °C with castor oil as capping agent (CuS7 and CuS8) are crystalline whereas those obtained with olive oil (CuS3 and CuS4) are amorphous. This show that the crystallinity of copper sulphide nanoparticles obtained under the same reaction conditions using both complexes as precursor is affected by the type of stabilizing agent used. These results agreed with the work reported by Shen et al. on the influence of capping agent on the crystallinity of the obtained phase [43]. This could be explained by the thermal stability of the main constituents of the vegetables oil. Indeed, unlike oleic acid (HO), ricinoleic acid (HR) with an additional –OH group on the 12th carbon atom, could form hydrogen bonds and therefore become more thermally stable [44]. Moreover, the spectra of samples obtained from both complexes at 190 °C with olive oil (CuS1 and CuS2) are respectively nearly similar to those obtained with castor oil (CuS5 and CuS6) showing that at this temperature the variation of capping agent does not affect the phase and crystallinity of the as-obtained nanoparticles. In summary, the diffractograms obtained with complex (2) at 190 °C with OO (CuS2) or CO (CuS6) are ascribed to the monoclinic djurleite (Cu1.94S) phase and could potentially be more efficient as photocatalyst compared to those with Cu1.97S, due to increased concentration of copper vacancies within the monoclinic phase than the orthorhombic phase. The crystalline phase of the as-obtained NPs are resumed in the Table 4. Table 4Crystalline phase of Cu(2-x)S NPsTemperature (°C)PrecursorsOlive oilCastor oilSamplesCrystalline phaseSamplesCrystalline phase190Complex (1)CuS1Cu1.97SCuS5Cu1.97SComplex (2)CuS2Cu1.94SCuS6Cu1.94S230Complex (1)CuS3AmorphousCuS7Cu1.97SComplex (2)CuS4Cu1.97SCuS8Cu1.97S Crystalline phase of Cu(2-x)S NPs The average crystalline size was estimated using Debye– Scherrer equation [45]: where D is the crystallite size, λ is the wavelength of the incident radiation, k is the geometrical factor, considered to be 0.90, θ is the Bragg angle taken in radians, and β is the full width at half maximum (FWHM) in radians. Utilizing the FWHM of the deconvoluted diffraction band, the average crystallite size in the sample was estimated to be 2.48, 6.76, 25.73, 7.36, 25.73 and 25.75 nm respectively for CuS1, CuS2, CuS5, CuS6, CuS7 and CuS8 samples. However, we notice that the crystallite size of the CuS2 and CuS6 monoclinic djurleite samples obtained from complex (2) increase with change of capping agent, from olive oil capped NPs (6.76 nm) to castor oil capped NPs (7.36 nm). This could be explained by the structure of the main constituents (oleic acid and ricinoleic acid) of OO and CO respectively. Indeed, although they possess the same hydrocarbon chain, ricinoleic has an additional –OH group (on the 12th carbon) in its structure which can create intermolecular interaction (hydrogen bond) between the capping agent molecules themselves and thus weakening their binding action with the crystal surface [44]. This leads to a less ordered protective layer, allowing precursors to reach the crystal surface more easily during growth. The effects of reaction temperature and precursors on the morphology were investigated by scanning electron microscopy and transmission electron microscopy. For this purpose, when complex (1) was used with olive oil, the sample treated at 190 °C (CuS1) exhibited monodisperse spherical particles with average diameter of about 5.4 nm with spherical shape (Fig. 2a1). When the sample was treated at 230 °C (CuS3), the as-prepared NPs showed disk-like island of 105 nm diameter made of aggregation of small irregular shapes particles of 6.2 nm average size (Fig. 2a3) which clearly show an intermediate from small NPs to nano-disks during the growth process. Similar results were reported by Chen et al. [46]. Aggregation observed could lead to weak photocatalytic activity of the sample CuS3. Since the nanoparticles in this process, strongly cluster together to form larger aggregates (often hundreds of nanometers) reducing the surface sites necessary for adsorption of pollutant and therefore limiting the degradation rate. This clearly shows the impact of temperature on the size and shape of the as-obtained particles. However, when complex (1) is replaced by complex (2), the sample treated at 230 °C (CuS4) show elliptical-like shape with 110.5 nm average size. We have noticed that the particles sizes became larger when precursor change, this shows the effect of precursor composition on the size and shape of particle. In the same vein, sample treated at 190 °C (CuS2) exhibit agglomerated particles of nearly rectangular shapes with average size equal to 38.5 nm (Fig. 2a2). The agglomeration observed may be due to high surface area-to-volume ratio as size decrease. Agglomeration could also lead to reduce photocatalytic activity. However, the particles cluster together with weak bond which can be disperse with enhanced activity. The HRTEM micrographs and selected area diffraction pattern (SAED) of representative copper sulphide (Fig. 2b1–b4) reveals, the well resolved lattice fringes and diffraction spots, due to the diffraction from different planes of the as-prepare NPs. All the interlayer spacing (dhkl) 0.256, 0.228, 0.452 and 0.397 nm determines by HRTEM respectively for CuS1, CuS2, CuS3 et CuS4 are attributed to (8 4 0), (7 5 2), (6 0 0) and (0 2 3) planes of djurleite phase, in respect with p-XRD. On the other hand, these interlayer spacing are observed not only to decreases with decreasing reaction temperature but also with the incorporation of thiocyanate ion within the precursor. a TEM micrographs; b HRTEM and SAED images of OO capped Cu(2−x)S NPs The influence of capping agent used during synthesis on the formation, growth, and morphology of the nanoparticles was also verified through the replacement of olive oil by castor oil under same reaction conditions. TEM image in Figure S3 depict elongated nanocrystals as large as 175.1 and 183.3 nm derived from complex (1) at 190 and 230 °C (CuS5 and CuS7). The mean value is estimated to be 107.9 and 86.9 nm respectively. Furthermore, when complex (1) is replaced by complex (2) as precursor, TEM micrographs of the samples CuS6 and CuS8 obtained at 190 and 230 °C exhibited nearly spherical-like particles with average size of 51.7 and 141.8 nm respectively. These differences in the morphology and size of the as-prepared NPs obtained at the same temperature could be explained by the number of coordinating atoms of the main constituents of the vegetable oils [47]. It was observed that ricinoleic acid, the main constituent of CO could coordinate through two sites (COOH and OH functions) whereas, oleic acid (the main constituent of OO), could coordinate through one site (COOH) which greatly affects the as-obtained morphology and size of the nanoparticles. Restrepo et al. reported the action of these constotuents using Araza fruit extracts as reducing and capping agents [48] (Scheme 3). Surface’s charge of copper sulphide NPs regarding PZC pH HRTEM and SAED clearly show lattice fringes and bright spots respectively confirming their crystalline nature. The interlayer spacing determined from HRTEM are 0.396, 0.240, 0.258 and 0.440 nm respectively for CuS5, CuS6, CuS7 and CuS8 are observed to decrease with the incorporation of thiocyanate ion when treated at 190 °C whereas the opposite trend is observed when treated at 230 °C. This behaviour is different from that obtained from olive oil-capped NPs and clearly highlights the influence of both capping agent and temperature. The influence of capping agent on the morphology and the size of the nanoparticles could be related to the number of reacting sites present within the main constituents of capping agent (Scheme 4) respectively for castor and olive oil which form a coordinate bond with Cu2+ ion and effectively modify the size and morphology of the as-prepared nanoparticles during crystal growth. Moreover, the impact of temperature is mainly related to the thermal stability of these main constituents (ricinoleic and oleic acid) [20, 49]. Structure of ricinoleic (1) and oleic (2) Acid In summary, the particle size of the as-prepared samples ranges from 5.4 to 142.1 nm. The reason for the broad particle size was that the number of crystal nuclei in the reaction and the surface energy of the nanostructures were different due to different reaction conditions used. In fact, a lower number of nuclei formed during the reaction might lead to a larger crystal size [50]. However, the average particle size of the samples CuS2 (38.5 nm), CuS6 (51.7 nm) was among the smallest, which might indicate that the quantum size effect originating from these samples could not be ignored. This show that variation in reaction parameters (reaction temperature, type of precursor and capping agent) clearly influences the morphology and size of the nanoparticles. SEM and EDX The surface morphology of the as-obtained nanoparticles obtained at constant magnification is presented by SEM micrographs (Fig. 3 and S4). The aggregation of the nanoparticles is clearly observed in the SEM micrographs. Thus, it can be seen from the images that major samples are formed of spherical structures [51]. Nonetheless, all the samples obtained from complex (2) as precursor (CuS2, CuS4 and CuS6, CuS8) have spherical morphology with different diameters [52]. While samples obtained from complex (1) have different morphologies ranging from spherical (CuS1) to cubic (CuS3) with olive oil and from elongated (CuS5) to compact agglomerate (CuS7) for castor oil. SEM micrographs of Cu(2−x)S NPs EDS spectra depicted by Figs. 4 and S5 was used to determine the chemical composition of the as-obtained Cu2−xS sample. From the EDX spectra, the formation of copper sulphide could be confirmed by the presence of the main elemental constituents (copper and sulphur) of the nanoparticles. Moreover, the presence of other element such as carbon and oxygen observed on the spectra, could be attributed to the main constituents of the vegetable oil (ricinoleic, oleic, linoleic and palmitoleic acid etc.) acting as capping agent. The absence of other elements indicated the purity of the reported samples. Furthermore, Tables 5 and S3 indicates the percentage content of copper and sulphur. The calculation demonstrated that the molar ratio of the two elements (Cu/S) range from 3.44:1 to 5.77:1. However, Cu/S molar ratio of the samples CuS7, CuS5 and CuS6 exhibit a slight enhancement of copper content than the expected (4.0:1 and 1.15:1) for djurleite phases sometimes occurring as mixtures [38]. This slight increase of copper content specifically from sample CuS7 could be due to pronounced desulphurization (sulphur deficiency) at 230 °C treatment temperature. The formation of S-vacancies in metal sulphides resulted in an enhanced photocatalytic performance by modulating structural and electronic properties [53]. On the contrary, a slight decrease of copper content of the as-obtained samples at 190 °C from CuS5 to CuS6 could be due, not only to the enhanced sulphur content through the incorporation of thiocyanate moieties within the precursor (CuS6) but also to the weaker treatment temperature use for the preparation of the NPs (CuS5). EDX spectra of castor oil capped copper sulphide NPs Table 5EDX analysis of castor oil capped CuxSy NPs at 190 °C and 230 °CCuS5CuS7CuS6CuS8ElementWeight%ElementWeight%ElementWeight%ElementWeight%Cu46.98Cu46.20Cu50.99Cu52.61S10.45S8.00S12.29S15.29C9.22C13.39C9.65C30.14O33.35O32.41O27.68O1.96Cu/S4.49/1Cu/S5.77/1Cu/S4.15/1Cu/S3.44/1 EDX analysis of castor oil capped CuxSy NPs at 190 °C and 230 °C In summary, we have synthesized different djurleite phases and Cu/S molar ratio of copper sulphides by changing the synthesis parameters (temperature, precursor and capping agent). The difference in stoichiometry resulted in different morphologies and optical properties of the products. Optical study of CuS nanoparticles The band gap of CuS was obtained using the Tauc plot Eq. 4, where α is the adsorption coefficient, hν is the incident photon energy; A is a constant, Eg is the band gap energy for direct transitions, and the exponent n (which can take values of 2, ½ for direct and indirect transition respectively) indicates the nature of the band-gap [51, 54]. The optical bandgap of the as-prepared copper sulphide nanoparticles was determined using Tauc plot by plotting (αhυ)2 against (hυ) and extrapolating the slope of the band edge against zero as shown in Fig. 5 [51] The as-obtained band gaps of the Cu(2−x)S NPs range between 2.17 and 2.48 eV and 2.20–2.46 eV respectively for CO and OO capped NPs (Figs. 5 and S6). The variation of optical band gap energy of the synthesized Cu2-xS NPs shows a blue shift in the absorption compared to bulk, due to quantum size effect and Moss-Burstein effect caused by copper deficiency in Cu2−xS which lead to an increase of hole formation in the valence band and effectively lower the energy of the lowest occupied energy level [40, 55, 56]. In fact, copper deficiency or copper vacancies in the crystal lattice lead to a stoichiometry dependent with corresponding bandgap range of 1.2–2.0 eV from copper poor (CuS) to copper rich (Cu2S) sulphide compounds [54]. Moreover, the variation of optical properties could be assigned to the change of lattice strain and interlayer spacing [57]. Furthermore, the variation of the band gap energy also observed when varying the capping agent could be due to the reaction sites available for each capping agent. In fact, the presence of specific reactive sites such as hydroxyl groups and double bond within ricinoleic acid and linoleic acid dictates the strength of the bond to metal, affecting the surfaces states and quantum confinement through the size and shape [58]. In addition, the results revealed that when complex (1) is used as precursor the energy band gap of the samples increase with treatment temperature whether CO or OO are used as capping agent. However, the opposite trends are observed when complex (1) is replaced by complex (2). Similar results reported by Molla et al. [2] and Okechukwua et al. [59], show the influence of precursor’s composition on the energy band gap value. Furthermore, the influence of the Sulphur content of the as-prepared NPs on the variation of the energy band gap was also highlighted. In fact, for a precursor type, the energy band gap decreases with the increase amount of sulphur content whatever the capping agent used as shown by EDX data (Table 5 and S3). This could be ascribed to the change in gain boundary and mid-gap defect states due to sulphur deficiency. In suphide-based compounds the amount of Sulphur content plays a significant role in energy band structure formation since it induces the shifting of the valence band edge leading to the change in band gap value. Hence, the increase amount of Sulphur tends to decrease the band gap by reducing possible defects. Similar results have been reported by Khadka et al. [60]. Moreover, we can also notice that, regardless of whether the complex and the capping agent are used at the same reaction temperature, a slight shift (~ 0.1–0.3 eV) is observed when treatment temperature is 190 °C, while the shift becomes larger (~ 0.9–0.12 eV) when the treatment temperature is 230 °C. This clearly shows the influence of the reaction temperature on the optical properties. This study shows that the variation of the thermolysis temperature as well as the type of precursor used affect the optical properties of the prepared nanoparticles, which could have a beneficial effect on the photocatalytic properties of CuS nanoparticles. Table 6 summarizes the as-obtained band. The absorption spectra and tauc’s plot of olive oil capped Cu(2−x)S NPs Table 6Energy band gap of Cu(2−x)S NPsTemperature (°)PrecursorsOlive oilCastor oilSamplesEg (eV)SamplesEg (eV)190Complex (1)CuS12.20CuS52.17Complex (2)CuS22.46CuS62.45230Complex (1)CuS32.36CuS72.48Complex (2)CuS42.21CuS82.30 Energy band gap of Cu(2−x)S NPs Raman study Figures 6 and S7 depicts the Raman spectrum of copper sulphide nanoparticles taken at room temperature in the range of 100–2000 cm−1 using a light beam with a wavelength of 632.8 nm. For this purpose, the Raman studies were divided into two regions 600 and 2000 cm−1 and 100–600 cm−1. In the first region between 600 and 2000 cm−1, a broad band ranging from 1280 to 1800 cm−1 is observed. However, the deconvolution of the latter represented by Fig. 6b has highlighted four bands of different intensity around 1565, 1446, 1344 and 1217 cm−1 respectively attributed to the Raman modes G, A, D and D4 of carbon [61]. The D-band is attributed to the sp3 bonded carbons chain, whereas the G-band is attributed to the first-order scattering of the stretching vibration E2g mode of the sp2 carbon atoms chain of the main constituents (oleic, ricinoleic and linoleic acids) of the solvent used, confirming the action of vegetable oils as a styling agent [20]. The D4− and A-bands, could be due to the presence of disordered graphite or structured amorphous carbon [62, 63]. On the other hand, Raman spectra provided more details about the lattice defect. As shown in Fig. 6a highlight the Raman modes related to metal sulphides. The characteristic vibrational spectra of different copper sulphides samples were carried out and compared. The low intensity of the Raman mode as well as the absence of some vibration modes could be related to the quantum confinement effect [64]. These weak bands observed around 242, 293 and 362 cm−1 were respectively assigned to longitudinal acoustic (LA) phonon, transverse optical (TO) mode and longitudinal optical (LO) phonon of metal-Sulphur vibrational modes. In addition, the broad character of these peaks could be associated with lattice defects emerged. [65–67]. Moreover, the bands observed around 435, 437, 438 and 455 cm−1 respectively for CuS5, CuS7, CuS8 and CuS6 could be ascribed to stretching mode of the S–S bond characteristic of the lattice [45]. This could be explained by the loss of copper ions, leaving behind undercoordinated S−. The accumulation of S− upon further loss of Cu+ leads to formation of disulphide (S–S) bonds and increased hole concentration in the valence band, which influence the optical properties and potentially impact the photocatalytic activity [68]. Nonetheless, it should be noted that their intensities are weaker than those of the Raman spectra of CuS reported by Ishii et al. [69]. This might be due to the reduced Sulphur content in djurleite (Cu1.94S) relative to the covellite (CuS) and therefore the low concentration of disuphide bonds in djurleite (Cu1.94S) as reported by Holtz et al. with silicon concentration [70]. Moreover, the degree of disulphide bond formation is also an indicator of the mobility of copper ions (copper vacancies formation). Furthermore, the Raman shift of the vibration mode of disulphide bond in Raman spectra, could be attributed to Defects related Raman scatterings. Note that the Raman shift of CuS6 was relatively greater compared to the other samples. The blue-shift effect caused by lattice strain implied that lattice defects in CuS6 were more significant, which matched well with the results from XRD analysis [67, 71]. In conclusion, due to the composition dependence of the Raman spectra, the similarity between the as-obtained Raman spectra could be evidence of the same chemical composition [70]. The Raman bands of the compounds are listed in Table 7. Raman shift spectra of castor oil capped Cu(2−x)S Table 7Raman data of castor oil capped Cu(2−x)SSamplesRaman assignments of latticeRaman assignments of carbon productsCu–SS–SD4-bandD-bandA-bandG-bandCuS62412833574551219134114441558CuS52422943624351219133814411563CuS82442943624381216134114351563CuS72422933554371217134414461564 Raman data of castor oil capped Cu(2−x)S Determination of point of zero charge (PZC) The Point of Zero Charge (PZC) is an electrokinetic property which defines the nature of the charge on the material’s surface as a function of the pH of the solution in which it is found. In this perspective, the determination of the pHPZC was carried out using the salt addition method, where the intersection between the curves pHf = f(pHi) and pHf = pHi curves is set as pH of zero-point charge (pHzpc). The experimental results are presented in Table 8 and Fig. S8. Since the photocatalysis occurs on the surface of nanoparticles, the determination of the PZC is essential to predict the charge on the nanoparticle’s surface during the photodegradation process [72]. Therefore, the efficiency of the photocatalyst is greatly influenced by factors such as: the solution pH, the pollutant type and the surface ability of catalyst to adsorb the pollutant [73, 74]. Herein, we report the determination of the PZC values of eight samples of copper sulphide obtained under different reaction’s conditions to determine the pH range suitable for the adsorption of methylene blue (MB) by electrostatic attraction. During this process, the ionic charge on the nanoparticle’s surface is affected by the pH of the bath in which they are immersed. Therefore, the distribution of surface’s charge was determined according to the pH of the PZC (Scheme 3). Table 8Point of zero charge of CuxSy NpsTemperature (°C)PrecursorsOlive oilCastor oilSamplesPZCSamplesPZC190Complex (1)CuS15.41CuS55.40Complex (2)CuS24.75CuS66.05230Complex (1)CuS35.14CuS75.58Complex (2)CuS45.31CuS85.53 Point of zero charge of CuxSy Nps The PZC values of the as-obtained samples ranged from 4.5–6.1. The values of the CO capped NPs (CuS5, CuS7, CuS8) and OO capped Nps (CuS1, CuS3, CuS4) fell within a narrow range of 5.40–5.58 and 5.41–5.14 respectively. In contrast, the values of the prepared samples CuS6 and CuS2 derived from complex (2) as precursor respectively in CO and OO are 6.05 and 4.75. These two values, appear far away from the others, might result from the change in crystal structure of nanoparticles ascribed to the transformation of orthorhombic djurleite phase (Cu1.97S) to monoclinic djurleite (Cu1.94S) phase through the diffusion of copper ion out of the crystal such as defect’s formation which can act as charge sites and therefore modify the PZC value. This could be indicated by the decrease of diffraction peak intensity assigned to Cu1.97S at 2θ = 42.6° (*). These values are suitable with cationic dyes such as MB, because the pH of the solution could easily be adjusted above the pHPZC. When pH > pHPZC the surface of the NPs remains negatively charged and therefore favors the adsorption of cation. Furthermore, the samples (CuS6) characterized by high pHPZC value (6.05) could be easily used for the adsorption of anionic dyes as the surface of the NPs were positively charged when the pH of the solution is lower than the pHPZC values. On the other hand, a slight increase in the pHPZC values was observed when olive oil was replaced by castor oil as stabilizing agent. This may be attributed to the existence of an additional hydroxyl functional group located on carbon number 12 (C12) of the most abundant constituent of castor oil: ricinoleic acid (Scheme 4) which increase the basicity of the compound. Nonetheless, the change is more significant for the sample treated at 190 °C using complex (2) as precursor (CuS6 → CuS2). This variation could be explained by a relative thermal stability of ricinoleic acid and their proportions in both capping agents used at 190 °C and therefore, greatly influence the surface properties [75, 76]. Photocatalytic studies of nanoparticles Photodegradation of methylene blue The photocatalytic degradation potential of the copper sulphide nanoparticles was investigated for the degradation of methylene blue dye under visible light irradiation. The experiment was firstly carried out in the dark for 40 min in view to ensuring a high concentration of MB near the active sites of the adsorbent. The dark reaction is a crucial, often fundamental step in methylene blue removal. It constitutes the initial step that concentrates the dye molecules at the catalysts’ surface. Dark adsorption accelerates photocatalytic degradation by ensuring a high concentration of the dye near the active sites of the adsorbent. The experiment was then exposed to visible light irradiation in the presence of the photocatalyst. At various time intervals of 30, 60, 90, and 120 min, the aliquots of 5 mL was taken and the absorbance was measured. The spectra as a function of irradiation time showed a decrease of the absorbance’s ratio (At/A0) (Fig. 7). The photocatalytic degradation efficiency plots of olive capped NPs obtained after 120 min (Fig. 8) show that CuS1, CuS2, CuS3 and CuS4 degraded 36.2, 57.4, 49.8 and 41.7% respectively of the initial concentration of the methylene blue dye. The sample CuS2 with monoclinic lattice (Cu1.94S) shows the highest photocatalytic efficiency which could not only be ascribed to a largest band gap energy (2.46 eV) relates to the oxidative capacity when subject to irradiation thus separating the photogenerated electron–hole pair into methylene blue dye [77] but also to the higher concentration copper vacancies resulting from the transformation of Cu1.97S to Cu1.94S. An attempt made by replacing OO with CO. The as-obtained samples CuS5, CuS6, CuS7 and CuS8 (Figure S9) exhibit under the same conditions, the degradation efficiency of 17.1, 59.8, 46.6 and 40.21% respectively. Nonetheless, the highest degradation efficiency of the sample CuS6 could be ascribed to the association of the presence of copper vacancies and his highest adsorption properties showing that adsorption properties can affect the degradation efficiency. However, in the case of the samples CuS5, CuS7 and CuS8 where crystallite sizes are quite similar (25.80, 25.81 and 2.88 nm respectively), the differences in photocatalytic performance could be explained by factors such as surface reactivity (point of zero charge which govern the photocatalytic activity) where density of surface groups can differ based on the capping agent or on the synthesis temperature. Moreover, the morphologies of the as-prepared products related to Cu/S ratios and band gap energy could also influence. This aforementioned, clearly highlight the significant impact of synthesis’ parameters on surface properties and charge separation efficiency in copper sulphide nanoparticles [10]. Absorption spectra of methylene blue with CO-capped copper sulphide a kinetic analysis and b degradation efficiency bar chart of MB dyes with OO-capped Cu(2−x)S NPs Effect of isoelectric pH of the samples To assess the influence of catalyst surface charges on copper sulphide photoactivity in 10 ppm MB solution, PZC of different samples was determined and the obtained values range from 5.4 to 6.05 and 4.75–5.41 respectively for castor and olive oil capped NPs. Since MB is cationic, this show that there was no significant adsorption in acidic pH (below the PZC). This is because the adsorption of cationic dye at the surface of catalyst is more productive above the PZC where the catalyst exists as M-S− form [39, 78]. So, the capacity of the photocatalyst to absorb contaminants (pollutants) on its surface is significantly impacted by both the pH of the solution and the PZC of the samples which further influence the efficiency of degradation. To study the evolution of the degradation efficiency of the samples in function of their PZC (Fig. 9) when the pH of the MB solution is maintained constant above the PZC (or isoelectric pH) value of the samples. The results show that the degradation efficiency and the PZC of castor oil capped NPs evolve in opposite ways (Fig. 9b). This observation can be explained by the fact that, the more the PZC of the samples and the working pH (pH = 9) are far apart, the higher the concentration of hydroxyl ion (HO−) is in solution, leading to an accumulation of metal hydroxide precipitates on the surface of the NPs thus deactivating the catalyst [78]. However, when castor oil is replaced with olive oil as stabilizing agent, the degradation efficiency evolves in opposite ways to isoelectric pH of the as-obtained samples (Fig. 9a). This could be explained by the fact that the greater the difference between both pH (working and isoelectric pH), the more is the hydroxyl ions available in solution to activate the surface of the nanocatalysts. This study clearly highlights the influence of surface charge as a consequence of the capping agent use. Nonetheless, the higher degradation efficiency achieved with both CuS2 and CuS6 might be due to its higher surface adsorption property as shown by the experiment performed in the dark (Fig. 8 and S8) compared to the remaining samples. Evolution degradation efficiency with PZC of the NPs Effect of energy band gap of the samples The optimum catalysts (therefore the optimum reaction condition) were determined by drawing the correlation among different band gap and the efficiency of degradation of the as-prepared NPs (Fig. 10). To compare the effectiveness of the catalysts, an experiment carried out without the use of the nanocatalysts (blank control or photolysis) under visible-light irradiation led to about 4.8% MB degradation of the initial concentration after 120 min (black curve). While the use of CuS1, CuS2, CuS3 and CuS4 nano-catalysts samples show that it rises to 36.2, 57.4, 49.8 and 41.7% respectively as compared to that obtained by photolysis, showing the degradation properties of the samples (Fig. 8). Moreover, it could be observed that degradation efficiency and energy band gap (Eg) evolve together showing its effect on the photocatalytic activity (Fig. 10a). Nonetheless, the sample CuS2 obtained from complex (2) at 190 °C with energy band gap value equal to 2.45 eV, exhibit the highest degradation efficiency. This finding could be justified by a rapid electron recombination rate when the band gap energy decreases lowering the photocatalytic efficiency. An attempt was also made by replacing OO as the capping agent with CO and the obtained samples CuS5, CuS6, CuS7 and CuS8 exhibit degradation efficiency equal to 17.1, 59.8, 46.6 and 40.21% respectively, quite different from blank experiment (Figure S8). The conclusions are almost similar to those obtained with olive oil capped NPs except the fact that the sample with the highest degradation efficiency CuS6 is not linked to the sample with the greatest energy band gap CuS7. This reveal that the optimum band gap above which the charge separation (electron–hole) is not favorable could be assigned to Eg = 2.46 eV. Evolution degradation efficiency with energy band gap (Eg) of the NPs In conclusion, the samples CuS2 and CuS6 derived both from complex (2) as precursor treated at 190 °C using OO and CO as stabilizing agent respectively could be better candidates for the complete removal of MB from water as they show degradation efficiency above 50% of the initial dye concentration (57.4 and 59.8%). This could be the result of a change of the specific area of the nanoparticle, due to the polymorphism of the djurleite phase in respect with powder XRD [38]. These results show the significance of both the synthesis temperature, type of precursor and the use of green capping agents on the optical and photocatalytic properties of the as-prepared copper sulphide nanoparticles for effective photocatalytic applications. Photocatalytic degradation mechanism Based on literature, the general mechanism of photocatalysis (Fig. 11) involved in the degradation of methylene blue under visible light can be explained [79]. When the nanocatalyst (Cu(2−x)S) is irradiated with visible light, the electrons (e−) are moved up from the valence band (VB) to the conduction band (CB) with the simultaneous creation of holes (h+) in the VB: where hV is the energy essential to transfer the electron from valence band to conduction band. Schematic illustration of the photocatalytic mechanism of degradation The generated electrons react with atmospheric oxygen (O2) absorbed on the photocatalyst surface and expedites the formation of superoxide radicals (O2−·) as indicated by Eq. (6): O2−· could react with H2O to form hydroxyl group (HO−) and hydroperoxyl radical (HO2·), which are strong oxidizing agents to decompose the organic molecule: On the other hand, the presence of the H2O further facilitates the photocatalytic reaction by the interaction of the hydroxyl ions (HO−) with the h+ and results in the formation of the hydroxyl free radicals (HO·) in the reaction medium as shown by Eqs. 8 and 9: Finally, the MB will be degraded through redox reaction leading to carbon dioxide and water: The degradation was monitored by the changes in the maximum absorption peak at 664 nm using UV–Vis spectrophotometer, this peak corresponds to the hetero-polyaromatic linkage of methylene blue, the spectra also showed another peak around 300 nm corresponding to the benzene ring of methylene blue [77]. Hence, the as-prepared copper sulphide showed better degradation efficiency for methylene blue dye when compared with previous studies, as presented in Table 8. It can be seen that the greater degradation efficiency recorded required longer irradiation time than that used in this investigation. The average photocatalytic efficiency of vegetable oil capped Cu(2−x)S NPs derived from complex (2) as precursor at 190 °C (CuS2 and CuS6), exhibit high photocatalytic degradation efficiency 57.4 and 59.8% respectively which could be mainly ascribed to optical properties such as low recombination electron–hole rate enhance by structural properties such as surface defects. In fact, surfaces defects (copper vacancies) introduce localized energy states within the band gap which act as recombination centres for photogenerated electron–hole (e−/h+) pairs, enhancing the charge separation efficiency and therefore impact on optical properties [40]. Table 9 summarizes the results of the photocatalytic performance of Cu2−xS materials against recently reported CuS-based systems under visible light of MB dyes. The table shows that copper composite all exhibit photodegradation efficiency higher than 90%. Whereas coper sulphide NPs present catalytic activity ranges from 31 to 100%. Based on the moderate photocatalytic efficiency of Cu2−xS NPs coated with vegetable oils (CuS2 and CuS6), the as-obtained copper sulphide samples can be considered as potential visible-light photocatalysts under adequate conditions for the degradation of organic pollutants from water effluents. Table 9Comparative photocatalytic performance of some Cu(2−x)S catalysts for the photodegradation of some organic pollutantsPollutantsCatalystLight sourceIrradiation time (mins)Degradation efficiency (%)ReferencesMethylene blueCuS7W, mercury12031–100[16]Methylene blueCu9S5UV irradiation9047.1–80.0[10]Methylene blueCuS NPsVis light irradiation12085.03[80]Methylene blueCdS/CuSVis light irradiation24095%[81]Methylene blueBi2S3/Cu2SHalogen lamp (λ = 420 nm, 86 W)9090.2[82]Methylene BlueCu(2−x)SVis (250 W)12059.8Current work Comparative photocatalytic performance of some Cu(2−x)S catalysts for the photodegradation of some organic pollutants Conclusion CuS nanoparticles were successfully synthesized through a simple, low-cost thermolysis of Cu(II) thiosemicarbazone complex and its thiocyanate adducts as single-source precursor under different reaction conditions (precursor type and temperature) by utilizing eco-friendly olive and castor oil as sustainable alternative to toxic synthesis solvents. To this end, Powder X-ray diffraction confirmed the formation of djurleite phase, with samples resulting from complex (2) at 190 °C in vegetable oils yielding nearly pure monoclinic structure (Cu1.94S). The EDX spectra and Raman spectroscopy confirmed the formation of copper sulphide. Raman analysis also provided more details about the lattice defect. The as-obtained NPs exhibited a significant blue shift in optical band gaps (2.17–2.48 eV) compared to bulk material, as a result of size reduction. TEM analysis revealed morphologies ranging from spherical to rectangular, with average particles size ranging from 5.41 to 141.7 nm. Among the as-prepared copper sulphides, the monoclinic samples (CuS2 and CuS6) achieved the highest methylene blue (MB) degradation efficiencies of 57.4 and 59.8% respectively under visible light. This performance is attributed to a combination of lower electron–hole recombination rate and a higher copper density that act as active catalytic sites within monoclinic structure compared to orthorhombic structure (Cu1.97S). The result of this study has significant implications for environmental remediation by identifying the optimal conditions for the degradation of recalcitrant dyes. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1. Acknowledgements The authors acknowledge The World Academy of Sciences (T.W.A.S.), India, for a DBT-UNESCO-TWAS Sandwich Postgraduate Fellowship and the accompanying financial support for JNK. The authors also acknowledge financial assistance through the ‘Fond de modernization et d’appui a la recherche’ allocation to Higher Education Teachers of Cameroon (AGP and PTN) Author contributions Jocelyn N. Kuate: Methodology, Formal analysis, Data curation, Visualization, writing—original draft, Validation, Adrien P. Yepseu.: Methodology, Formal analysis, Data curation, writing—original draft, Kun Walter N: Formal analysis, Data curation, Mariappan Mariappan: Formal analysis, data curation, Monisha Arumugam: Formal analysis, writing—review and editing, Awawou G. Paboudam: Validation, Supervision, review, and editing, Viorel Cîrcu: writing review, editing, Sunder N. Dhuri: Funding acquisition, Resources, Validation, Supervision, review, editing, Peter T. Ndifon: Conceptualization, Funding acquisition, Resources, Project administration, Validation, Supervision, review and editing. Data availability The data used to support the findings of this study are included within the article and the supplementary materials section. Any other data are available from the corresponding author upon request. Declarations Ethical approval and consent to participate Not applicable. Consent for publication All authors give consent for the publication of the manuscript. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The data used to support the findings of this study are included within the article and the supplementary materials section. Any other data are available from the corresponding author upon request.