Introduction

Short-wavelength optical devices are used in a variety of applications such as optical recording displays, optical storage devices, medical appliances, and measuring instruments [1]. Recently, lot of research has been focused on realization of practical application from cheap and efficient semiconductor based short-wavelength optoelectronic devices [2, 3]. Zinc oxide (ZnO) is a versatile semiconductor material and has attractive properties like nontoxicity, environment-friendly, wide band gap (3.2 eV), and a large exciton binding energy of 60 meV at room temperature [4, 5]. Zinc oxide is of great interest due to its strong UV photo response, established synthesis methods, chemical and thermal stability [6]. To obtain better ZnO-based optoelectronic devices, high quality n- and p-type ZnO is needed. Generally, ZnO is a native n-type semiconductor material and to acquire p-type conduction in ZnO is challenging [7]. The difficulty in p-type doping in ZnO is mainly due to self-compensation by donor like native defects, low dopant solubility, and less stability [8, 9].

p-type ZnO can be achieved by the substitution of group I element in the Zn site and group V element in oxygen site. Recently, group I elements such as Li, Na, and K have been found to be potential dopants in ZnO [10, 11]. However, when elements of smaller ionic radius like Li were doped with ZnO, a less stable p-type semiconductor was formed, because Li behaves like a shallow acceptor/reactive donor [12]. K and Na (bigger cationic radius) ions show stable p-type conductivity as they behave like deep acceptors [13]. Similarly cesium possess higher ionic radius (174 pm) when compared to other group I elements, therefore may provide a better p-type stability than other group I elements. On the contrary, very few reports on optical and electrical properties of Cs-doped ZnO NPs have been published so far [14, 15].

Also, a large number of physical and chemical methods are available to synthesize undoped and doped ZnO NPs [16, 17]. Compared with chemical and physical methods, green synthesis methods have received a lot of attention due to low cost, simplicity, versatility, and scalability [18]. “Green synthesis” of nanoparticles (NPs) makes use of environmental friendly, non-toxic, and safe reagents and therefore neither high temperature, pressure nor toxic chemicals are required [19]. Hence, in the present paper, a green approach is adopted for the synthesis of ZnO NPs and its effect on p-type conduction in Cs-doped ZnO NPs have been reported for the first time in lieu of our knowledge. In this manuscript effort has been made to demonstrate the structural, optical, and electrical properties of Cs-doped ZnO NPs.

Experimental

Zinc nitrate hexahydrate (98 % Sigma Aldrich) and cesium nitrate (99.99 %, Alfa Aesar) were used as the precursors for synthesis of ZnO NPs. Green-solvent was extracted from citrus × sinensis peel. The brief synthesis of bio-extract was mentioned in the previous work [20]. For the synthesis of undoped ZnO and Cs-doped ZnO (1 and 3 mol%) NPs, appropriate amounts of precursors were dissolved in double distilled water. The bio-extract was added to the precursor solution drop by drop over a period of 6 h with constant stirring. The precursor solution was stirred overnight at constant speed, and then the solution was heated to 60 °C for 2 h to get a homogeneous and transparent sol. The transparent gel was dried at 150 °C for 12 h. The dried white precipitate was calcined at 500 °C for 2 h to obtain ZnO and Cs-doped ZnO NPs. As grown Cs:ZnO NPs results were reported earlier [21]. In the present study in order to enhance crystallinity, all the samples were annealed at 500 °C for 2 h under oxygen atmosphere.

Characterization

X-ray diffraction (XRD) pattern of samples were recorded using X-ray diffractometer (SEIFERT—2002 Model, DYEFLAX, Germany) with Cu Kα radiation. Morphology of the samples was analyzed by FESEM S-4800 Hitachi model scanning electron microscopy (SEM). The optical measurements were carried out using Shimadzu UV 2450 spectrophotometer equipped with an integrated sphere assembly ISR-240A. Room temperature photoluminescence (PL) measurements were carried out on a Fluorolog-3 spectrometer. Hall effect measurements were done using an Ecopia Hall effect (HMS 5000) system. To measure electrical properties of the samples, thin pellets (thickness—0.2 mm, pressure—10 ton) were made from powder samples using hydraulic press. Ohmic contacts were made on the pellets with silver paste at the edges of four corners to maintain Vander Pauw configuration.

Results and discussion

Structural analysis

The structural parameters and phase purity of the undoped and the Cs-doped ZnO NPs at different concentration were estimated using XRD (Fig. 1a). All the diffraction peaks are indexed with standard ZnO data (JCPDS 36-1451). Prominent diffraction peaks such as (100), (002), (101) patterns from all annealed samples correspond to polycrystalline ZnO with hexagonal wurtzite structure [12]. No secondary peak corresponding to Cs was observed. XRD results confirmed that Cs 1 mol% doped ZnO showed enhanced crystallinity, which indicates inclusion of Cs in ZnO matrix is more effective in Cs 1 mol%:ZnO NPs and with further increase in concentration the crystallinity is observed to decrease, which may be due to agglomeration. In order to observe the effect of Cs doping, an enlarged version of XRD patterns is shown in Fig. 1b. It is obviously seen from the figure (Fig. 1b) that the predominant peaks (100), (002), and (101) are shifted toward lower angle side, suggesting the inclusion of Cs ions into ZnO matrix in accordance with Vegard’s law. The ionic radius of Cs+ ion (167 pm) is much larger than Zn2+ ions (74 pm) and it is well known that if the ionic size of the dopant is larger than the host, the XRD peak tend to shift toward lower angle. A similar phenomenon has been observed elsewhere in ZnO doped with p-type dopant [12, 13].

Fig. 1
figure 1

a XRD spectra of undoped and Cs-doped ZnO NPS, b enlarged XRD pattern of undoped and Cs-doped ZnO NPs

Full size image

Crystallite size was calculated from the well-known Debye–Scherrer equation. Crystallite size, lattice parameter, and stain of the samples are summarized in Table 1. The estimated crystallite size as a function of Cs doping increases from 40 to 52 nm, due to the fact, that the larger ionic size of Cs cation induces grain growth. A slight reduction in lattice parameter value suggests heterogeneous nucleation as well as a decrease in the lattice strain is observed with inclusion of Cs ions (Table 1). Similar effects were observed by Kumar et al. [22] in Ni-doped CeO2 NPs. Thus the decrease in lattice parameter and lattice strain in the current study suggests substitution of Cs in ZnO matrix.

Table 1 Lattice parameter, crystallite size, and strain values for undoped and Cs-doped ZnO NPs
Full size table

Morphology analysis

Figure 2a shows the SEM images of undoped and Cs-doped ZnO NPs. In the earlier report, we found alcohol derived ZnO NPs showed pyramid shaped particles whereas green synthesized ZnO NPs showed triangular-shaped particles. It is observed that the particle size is reduced and morphology change is observed due to effect of “green-solvent” [20]. It was evidenced from the Fig. 2a, that triangular-shaped particles were observed in both ZnO- and Cs-doped ZnO samples and an increase in particles size with Cs concentration is noticed. It is also observed from the Fig. 2a, that increase in dopant concentration induces agglomeration which is in concurrence with reduction in crystallinity at higher dopant concentration due to agglomeration (XRD pattern Fig. 1b). Energy dispersive X-ray (EDX) analysis shown in Fig. 3 confirms inclusion of Cs ion in ZnO lattices.

Fig. 2
figure 2

SEM images of a undoped, b 1 mol% Cs: ZnO, and c 3 mol% Cs: ZnO NPs

Full size image
Fig. 3
figure 3

EDS images of a undoped b 1 mol% Cs:ZnO, and c 3 mol% Cs:ZnO NPs

Full size image

Optical analysis

Transmission spectra of the undoped and Cs-doped ZnO NPs are shown in Fig. 4a. From the spectra, it is clear that all NPs exhibit a transmission of about 85 % in the visible range and subsequently has increased with dopant concentration. The effect of Cs on band gap energy is analyzed using UV–visible absorption analysis. The characteristic adsorption band edge of ZnO is red shifted with increase in Cs concentration in ZnO:Cs samples (inset Fig. 4a). It may be due to the surface effect and quantum confinement effect [23]. The observed enhancement in blue shift band edge with increasing Cs content indicates the incorporation of Cs ions into the Zn site in the ZnO lattice.

Fig. 4
figure 4

a Transmittance spectra of undoped and Cs-doped ZnO NPs, inset figure enlarged view of band edge shift in absorbance spectra, b plot of (αhγ)2 versus hγ

Full size image

The energy band gap (E g) was determined by well-known Tauc equation [24]. Figure 4b shows the plot of (αhγ)2 versus hγ, where α is the optical absorption coefficient and hγ is the energy of the incident photon. The E g values are 3.12, 3.09, and 3.08 for the undoped, 1 and 3 %, Cs-doped ZnO NPs, respectively. Change in morphology and size due to green-solvent may be one of the reasons in reduction of bandgap. In this study, the reduction in band gap is attributed to effect of green-solvent, size effect, and oxygen vacancy. Band gap narrowing was observed for Cs-doped ZnO samples. Generally reduction in bandgap can be explained by two ways, one is size effect and another one is dopant concentration. It is well known that the increase in crystallite size will decrease the band gap [9, 24]. XRD and SEM results of present study indicate an increase in particle size with inclusion of Cs. Usually, a widening of the bandgap is observed for ZnO doped with donors; while a bandgap reduction is observed if doped with acceptors (i.e., p-type ZnO) due to the formation of impurity band inside the forbidden gap region [6]. Thus the band gap shrinkage observed in Cs-doped ZnO particles can be attributed to size effect as well as to dopant effect. Bandgap reduction also suggests the possibility of p-type behavior in the Cs-doped ZnO samples.

Photoluminescence studies

The photo luminescence study is a powerful tool for investigating effect of impurity doping in ZnO nanostructures. Figure 5 shows room temperature PL emission spectra of undoped and Cs-doped ZnO NPs under 320 nm excitation. The emission spectra consist of two main peaks, a sharp peak at 379 nm in the UV region and a broad peak around 450 nm in the visible region. The observed UV emission is attributed to excitonic recombination of near band edge emission in ZnO and the blue emission may be due to Zn interstitial/surface defects [25]. Kshirsagar et al. [26] reported the blue emission in ZnO due to acceptor bound exciton. Lü et al. [27] has reported that the blue emission is associated with Zn vacancy and Zn interstitial defects.

Fig. 5
figure 5

Room temperature PL spectra of undoped and Cs-doped ZnO NPs annealed at 500 °C, 2 h under O2 atmosphere (inset figure magnified view of PL peak shift)

Full size image

In the present study, reduction in PL emission intensity is observed for Cs-doped ZnO NPs. This can be explained from the net charge value, i.e., if the net charge of the slightly ionized oxygen vacancy in ZnO is negative, it creates an acceptor state by pushing the valance band levels higher into the bandgap [28]. Also doped cation provides competitive pathways for recombination, which results in quenching of emission intensity.

It is evident from the Fig. 5 (inset figure), that a blue shift is observed in the Cs-doped samples compared to the undoped ZnO. Thangavel et al. have reported blue shift with inclusion of Cs ions and attributed it to band filling effect of free carriers. Similar results were observed for Li-doped ZnO thin films deposited by PLD [28]. UV emission peak is observed to shift toward lower wave length side with decreasing Cs concentration. However, it is interesting to note in our experiment that the shift is more predominant in Cs 1 mol% and also the UV emission peak observed at 375 nm in Cs 1 mol% doped ZnO is less intense. Thus the results indicate that the Cs doping can slightly modify the band gap of ZnO and also illustrate that doping can influence the optical properties in ZnO NPs. Hence to analyze the annealing effect in Cs 1 mol% doped ZnO, the annealing temperature was increased from 500 to 700 °C under oxygen and air atmosphere were used for analysis. Figure 6 shows the characteristic PL spectrum Cs 1 mol% doped ZnO annealed at 700 °C for 2 h under O2 and air atmosphere. It is obvious from the Fig. 6 that the annealing of Cs 1 mol% doped ZnO NPs at 700 °C in oxygen atmosphere enhanced the UV emission, while decrease in UV emission is observed when the sample is annealed in air. In this O-rich condition, the amount of oxygen that diffuses into the sample is high and the concentration of oxygen vacancies decreases [29].

Fig. 6
figure 6

Room temperature PL spectra of Cs 1 mol%: ZnO NPs annealed at 700 °C, 2 h under O2 and air atmosphere

Full size image

Electrical measurements

The electrical properties of the undoped and Cs-doped samples are shown in Table 2. It was evidenced from the Table 2, undoped ZnO shows n-type conduction with a carrier concentration of −7.49 × 1018/cm3. Cs is a group I element, thus it has the potential to be a deep acceptor in Cs-doped ZnO. These results indicate, Cs act as a deep acceptor in the present Cs-doped ZnO samples and the hole concentration decreased with increase in Cs concentration. The data shows that the doping of 1 mol% Cs in ZnO shows higher carrier concentration than Cs 3 mol% doped ZnO. The reduction in hole concentration with increasing dopant concentration (more than 1 mol%) may be attributed to limitation in dopant solubility in ZnO lattices [15]. Thus the optimum concentration is found to be Cs 1 mol% with a high mobility of 7.3 cm2/V s, a maximum hole concentration of 1.3 × 1018/cm3 and a positive Hall coefficient of 2.18 × 102 cm3/C. These results confirm effective substitution of Cs in Zn sites and it act as a deep acceptor, which leading to p-type conductivity in Cs-doped ZnO NPs.

Table 2 Electrical properties of undoped ZnO and Cs-doped ZnO NPs
Full size table

Conclusion

In this work, Cs-doped ZnO of different Cs concentrations (1, 3 mol%) were synthesized by eco-friendly green route and the effect of Cs concentration was analyzed. XRD results confirmed that NPs have polycrystalline and hexagonal wurtzite structure. Better crystallinity was observed for Cs 1 mol%:ZnO NPs as well as crystallite size increases with Cs concentration. SEM result shows triangular-shaped NPs and the particles are highly agglomerated with dopant concentration. Optical analysis revealed, band edge shifted toward higher wavelength. Hence, from the results, it is concluded that there is a reduction in band gap with inclusion of Cs ions in ZnO matrix. The PL spectra showed characteristic UV emission at 379 nm and defect induced blue emission around 450 nm. The observed blue shift of the optical band gap and quenching of bandedge emission are attributed to dopant effect and the net charge carrier value (−ve). Enhanced UV emission for the annealed samples at higher temperature (700 °C) under oxygen atmosphere confirmed the charge compensation. Hall measurement results showed p-type conductivity in Cs-doped ZnO NPs, to the fact that Cs forms deep acceptor level in substitutional Zn site. From the above results, it may be concluded that the 1 mol% Cs in ZnO is an optimum concentration. The correlation between the conductivity type and near bandedge peak shift with doping induced bandgap changes suggest p-type conduction in Cs-doped ZnO NPs.