Introduction

Around 1/10th million types of dyes are produced annually and utilized in several industries such as paper, rubber, leather, textile, printing, paint, pigments and plastic (Chowdhury and Saha 2011; Shaban et al. 2018). Among these dyes, only about 10–15% of MB is discharged directly into water bodies and surrounding environment causing cancer, skin irritations, allergy, malfunctioning of liver, kidneys and reproductive system in humans (Al-Degs et al. 2008). Moreover, MB-contaminated wastewater caused severe harmful environmental issues to aquatic life with higher chemical oxygen demand (COD) (Al-Kdasi et al. 2004).

For dye removal, several techniques such as electrolysis, dialysis, ion exchange, adsorption, and photocatalytic degradation have been employed (Wang et al. 2016). However, photocatalytic degradation being environment friendly, cost effective and efficient has been utilized widely (Sivakumar et al. 2010). Semiconductor photocatalyst is preferred in dye degradation due to sustained and fundamentally applied research in environmental remediation (Chauhan et al. 2013). Fujishima and Honda (1972) reported hydrogen splitting using TiO2 at water expense (Su et al. 2016; Deng et al. 2017b). Transition metal sulfides (Kriegel et al. 2012; Kundu and Pradhan 2014; Ma et al. 2014; Jiang et al. 2015), metal oxides (Park et al. 2009; Xu et al. 2012; Basnet and Zhao 2016; Deng et al. 2017a), doping materials (Zhao et al. 2015; Wan et al. 2015; Deng et al. 2015), and composite heterojunctions (Dong et al. 2013; Xu et al. 2016; He et al. 2017) have been prepared to achieve efficient photocatalyst. However, metal chalcogenides have shown improved photocatalysis (Hong et al. 2014; He et al. 2016). Interestingly, cadmium sulphide (CdS) served as promising photocatalyst due to optimal band gap, ideal electronic band position, chemical and thermal stability on behalf of quantum confinement (Thambidurai et al. 2010; Wang et al. 2015). Moreover, CdS having absorbance in visible regime and its conduction band is more negative than reduction potential of H+/H2 (Li et al. 2008; Ertis and Boz 2017).

The CdS efficacy for photodegradation depends on the integration with suitable impurities. Transition metals (TM: Ni, Sb, Mn, Co, Fe, etc.)-doped CdS showed electrical, optical and magnetic properties as single material (Rathore et al. 2010; Kumar and Sharma 2017). Various TM-doped CdS have been synthesized but Fe doped remained less focused. Therefore, we aimed to synthesize Fe-doped CdS having all functional attributes required for efficient MB degradation. CdS size-tunable nanocrystals have been synthesized using chemical bath deposition, spray pyrolysis, laser deposition, sol–gel and successive ion layer adsorption (Mercy et al. 2014) that had some limitations. Co-precipitation being cost effective, with high doping of impurities at room temperature, adapted chemistry of doping and size control using capping agents has been utilized widely (Mercy et al. 2014). Therefore, in the current research, we aimed to prepare Fe-doped CdS using co-precipitation.

Experimental details

Materials

Cadmium chloride 2,5 hydrate (CdCl2. 2,5H2O) and sodium sulfide pentahydrate (Na2S.5H2O) were acquired from “Panreac PRS” and “Daejung Chemicals and Metals Co. Ltd”, respectively. For capping agent, thioglycolic acid (TGA, 80%) was purchased from “Merck”. Finally, iron (III) nitrate 9-hydrate (Fe(NO3)3.9H2O) was purchased from “UNI CHEM” and all chemicals were utilized without further processing.

Preparation of CdS and Fe doping

Chemical precipitation was adopted to synthesize CdS; 0.5 M solutions of each of CdCl2, Na2S and TGA were prepared in deionized water (DIW) under vigorous stirring for 30 min. Afterwards, TGA was added dropwise in CdCl2 solution at 65 °C to control the size of nanoparticles for 15 min. Subsequently, Na2S solution was added dropwise in CdCl2 and TGA solution. The yellowish precipitates of CdS which started to form slowly were centrifuged, washed and filtered to remove all sorts of impurities. Finally, precipitates were dried at 100 °C to obtain fine powder of CdS. Later, Fe (5, 10 and 15%) was mixed in CdCl2 solution for doping using the above-mentioned procedure (Fig. 1).

Fig. 1
figure 1

Schematic representation of Fe-doped CdS preparation

Full size image

Photocatalytic activity process

The photodegradation activity of iron (Fe)-doped CdS NPs was evaluated in terms of photocatalytic reduction of MB (10 mg/L). A mercury (Hg) lamp (400 W) as the visible light source was used with the principal wavelength of 400–700 nm. MB (60 mL) was mixed with 10 mg suspension of the prepared photocatalyst under stirring for 5 min to achieve equilibrium between MB and nanocomposites before illumination. After exposure of visible light for specific time intervals, 5 mL suspension was collected for UV–Vis absorption to measure MB concentration. The changes in dye concentration during photodegradation were a measure of intensity of peak (665 nm) absorption with irradiation time.

Catalysis

Freshly prepared (400 µL) 0.1M sodium borohydride (NaBH4) solution was mixed with 3 mL aqueous MB (10 ppm). Then 400 µL of CdS and Fe-doped CdS of certain concentration was added under agitation. The decolorization indicates reduction of dyes. However, decolorization of MB in the presence of NaBH4 represents reduction of MB to leucomethylene blue (LMB) as shown in Eq. 1. The reaction without nanocatalyst was referred as blank (Fig. 2) and the absorption spectrum was determined using UV–Vis spectrophotometer (200–800 nm).

(1)
Fig. 2
figure 2

Graphical representation of catalysis process

Full size image

Characterization

Fe:CdS was analyzed using Fourier transform infrared (FTIR) spectroscopy, Raman spectrometer, X-ray photoelectron spectroscopy (XPS), field emission scanning electron microscopy (FESEM) and UV–Vis spectroscopy. The crystal structure of Fe-doped CdS and phase information were collected using PANalytical Xpert PRO X-ray diffraction (XRD) with Cu Kα radiation (λ ~ 0.154 nm) by varying 2θ from 20° to 70°. The presence of functional groups was confirmed using FTIR Perkin Elmer spectrometer. Raman spectra were acquired with DXR Raman microscope (Thermo Scientific) with 532-nm (6 mW) laser. The morphological characteristics and microstructures of products were investigated using JSM-6460LV FE-SEM coupled with EDX spectrometer. The optical properties of products were obtained from UV–visible Genesys 10S spectrophotometer.

Results and discussion

Figure 3 represents the XRD pattern of Cd1−xFexS (where x = 0, 0.05, 0.10 and 0.15) nanopowders that showed single cubic phase having peaks at 26.5°,43.9° and 51.9° indexed to be scattering from (111), (220) and (311) planes, respectively, and well matched with JCPDS card no. 00-001-0647. Interestingly, no extra peak appeared due to dopant material which suggests proper incorporation of Fe into CdS. The observed peaks have identical position and increased peak broadening with doping indicates surface defects and confirmed miniaturization in nanometric range for Fe–CdS NPs (Thambidurai et al. 2010). The measured average crystallite sizes were 11.6–4.1 nm using Debye–Scherrer formula \(\left( {D=\frac{{k\lambda }}{{\beta \cos \theta }}} \right)\) from full width half maxima (FWHM) of XRD peaks, where k is Scherrer constant, λ is incident wavelength of X-rays, β is full width half maxima, θ is Bragg’s angle of diffraction and D is crystallite size. The calculated crystallite size values 11.6, 11.4, 7.0 and 4.1 correspond to x = 0, 0.05, 0.10 and 0.15, respectively, which shows that crystallite size decreased upon Fe doping.

Fig. 3
figure 3

XRD pattern of pristine and Fe-doped CdS

Full size image

According to Heisenberg uncertainty principle, ∆X∆P ≥ h2/4, which provides the relationship between particle size (∆X) and phonon distribution momentum (∆P), where ћ is the reduced Planck’ constant. Decrease in particle size leads to increase in phonon momentum distribution that resulted in Raman band shift as well as broader peak (Chauhan et al. 2013). Raman spectra of pristine and doped CdS nanoparticles are shown in Fig. 4. Peaks at 287 and 589 cm−1 correspond to 1 LO (longitudinal optical) and 2 LO phonon modes which were consistent with previously reported values (Thambidurai et al. 2010). The peak intensity reduction upon doping attributed to the smaller difference in ionic radii of Fe2+ than Cd2+ which results in slight redshift of peaks (Chauhan et al. 2013) (Fig. 5).

Fig. 4
figure 4

Raman spectra of pure and Fe-doped CdS

Full size image
Fig. 5
figure 5

FTIR spectra of CdS and Fe-doped CdS

Full size image

The presence of functional groups was confirmed using FTIR for CdS and Fe:CdS which exhibited stretching and bending vibrations of O–H, C–H, C=C and C–O functional groups. However, broad peak at 3432 cm−1 attributed to O–H exhibits affinity of water towards CdS (Seoudi et al. 2015; Qutub et al. 2016), whereas peaks at 1627 cm− 1 and 1544 cm−1 correspond to C=C and C–O stretching modes of carboxyl and carbonyl groups (Kumar et al. 2012; Abdolahzadeh Ziabari and Ghodsi 2012; Elevathoor Vikraman et al. 2015). Interestingly, adsorption of water and CO2 is common owing to exposure in atmosphere, particularly for NPs (Qutub et al. 2016). SH stretching vibrations were associated with an absorption band at 2351 cm−1 which confirmed the presence of the capping agent (TGA). However, shoulder at 665 cm−1 confirmed the presence of CdS stretching modes (Kumar et al. 2012; Abdolahzadeh Ziabari and Ghodsi 2012; Elevathoor Vikraman et al. 2015; Seoudi et al. 2015).

Surface morphology and elemental analysis of synthesized nanoparticles, FESEM and EDX were employed. FESEM images of synthesized CdS and Fe-doped CdS (5, 10 and 15%) NPs are shown in Fig. 6a–d, respectively. CdS had agglomerated nanoparticles (6a) that resulted in nanoclusters (Thambidurai et al. 2010). Upon doping, NPs become spherical with irregular growth due to Ostwald ripening (Devi et al. 2015; Desai et al. 2017; Waly et al. 2017). However, EDAX confirmed Cd, S and Fe (Fig. 7a–c), and the relative ratio between elements was stoichiometric (Thambidurai et al. 2010).

Fig. 6
figure 6

FESEM images of pristine CdS (a) and Fe (5, 10, 15%)-doped CdS (b, c and d)

Full size image
Fig. 7
figure 7

EDX spectra of pristine CdS (a) and Fe-doped CdS (b and c)

Full size image

Optical properties after Fe incorporation into CdS (5, 10 and 15%) were investigated using UV–Vis spectroscopy (Fig. 8a). Absorption peak for CdS was found around 475 nm (Thambidurai et al. 2010; Seoudi et al. 2015). However, with mixing of Fe in CdS, absorption intensity increased 280–480 nm accompanied blueshift with doping amount. This shift was attributed to quantum confinement as evident from XRD (Thambidurai et al. 2010; Elevathoor Vikraman et al. 2015; Seoudi et al. 2015). On the other hand, band gap energy was measured between (αhυ) vs (hυ) graphs plotted in Fig. 8b and longer wavelength of absorption band decreased band gap energy.

Fig. 8
figure 8

Absorption and corresponding band gap spectra using Tauc plot method for pristine and Fe (5, 10 and 15%)-doped CdS

Full size image

TGA analysis confirmed initial mass loss (2–10%) up to 400 °C ascribed to adsorbed water removal on nanocrystals surface for undoped and doped NPs suggesting endothermic transition as shown in Fig. 9a–d (Dhage et al. 2013; Patel et al. 2017). However, mass loss (10–30%) in doped NPs corresponds to exothermic peak that starts from 400 °C providing evidence of cubic crystal collapse and evolution of impurity phases as well (Dhage et al. 2013). Furthermore, another endothermic peak after 800 °C represented mass loss corresponding to sublimation of CdS. It is noteworthy to infer from the increase in weight loss upon Fe doping that control sample is more stable than Fe-incorporated CdS (Dhage et al. 2013) (Table 1).

Fig. 9
figure 9

DSC/TGA graph of pristine CdS (a) and Fe (5, 10, 15%)-doped CdS (b, c and d)

Full size image
Table 1 DSC/TGA result extracted from Fig. 9
Full size table

The photocatalytic activity (PCA) of control and Fe-doped (5, 10 and 15%) NPs for MB degradation has been presented in Fig. 10. The successive decrease in MB with Fe-doped NPs attributed to crystal defects that acted as recombination centers to reduced photocatalytic performance. However, control sample CdS exhibited inverse degradation in contrast to doped samples (Chauhan et al. 2013).

Fig. 10
figure 10

Photocatalytic activity phenomena. (a) Degradation of MB in the presence of control and Fe-doped (5, 10 and 15%) (b)

Full size image

The catalytic reduction of MB was investigated using NaBH4 as the reducing agent with doped and undoped CdS as nano-catalysts. Reducing capacity of NaBH4 and undoped CdS was not significant (Fig. 11a) while successive decrease in MB was observed with doping concentration of Fe in CdS and maximum catalytic efficiency was recorded for 15% Fe-doped CdS NPs (Fig. 11d).

Fig. 11
figure 11

a–d Time-dependent UV–Vis spectra for the reduction of dyes. Dye with NaBH4 + CdS (a), dye with NaBH4 + 5% Fe-doped CdS (b), dye with NaBH4 + 10% Fe-doped CdS (c) and dye with NaBH4 + 15% Fe-doped CdS (d)

Full size image

Pure CdS gives incomplete reduction of MB within 40 min while 5% Fe-doped CdS degrades 70% MB within 25 min (Fig. 12). However, 10 and 15% Fe-doped CdS reduced MB within 15 and 10 min, respectively, and showed complete reduction of MB to leucomethylene blue (LMB) at room temperature.

Fig. 12
figure 12

Catalytic degradation of MB with CdS (a), 5% Fe-doped CdS (b), 10% Fe-doped CdS (c) and 15% Fe-doped CdS (d)

Full size image

The variations in absorption intensity of MB pointed to a rapid reaction rate over a certain period of time. In addition, reduction of MB was proximately completed at termination of reaction. Conversely, undoped CdS represented slower degradation of MB, suggesting superior catalytic function of Fe-doped CdS. These NPs were potential nanocatalysts with excellent catalytic potential to be employed in industries (Table 2).

Table 2 Comparison of current and reported studies of various ratios of Fe-doped CdS NPs
Full size table

Conclusion

CdS and Fe (5, 10 and 15%)-doped CdS NPs were prepared using chemical co-precipitation and their XRD pattern indicates gradual decrease in crystallite size from 11.6 to 4.1 nm in Fe-doped CdS NPs. Absorption increased (400–500 nm) with the amount of Fe doping and bandgap decreased from 2.34 to 2.2 eV (2.42 eV of bulk CdS). FESEM images displayed nanoclusters and surface defects with doping as evident with XRD pattern. However, FTIR confirmed the presence of Cd–S linkage at 619 (cm−1) and other functional groups involved during synthesis. The Fe-doped CdS NPs showed superior catalytic potential compared to undoped CdS that suggests their way to dye usage industries especially leather and tanneries. Additionally, NPs could not only provide superior catalytic activity but also help in cost reduction and complete removal of dyes for wastewater management.