The XRD patterns of the as-synthesized products using CTAB and SDS by hydrothermal method are shown in Figure 1. All diffraction peaks can be indexed as hexagonal wurtzite-structure (JCPDS card No. 36-1451). The sharp and narrow peaks also illustrate that ZnO particles enjoy high crystallinity and purity. Samples synthesized with constant amount of SDS followed by addition of different molar ratios of CTAB show higher intensity compared to the sample prepared with the present of SDS only. Similar results was observed by Tan et al. [18], in which reflections of the lamellar phase constructed by mixture of SDS and CTAB show significantly increase in intensities (three times) compared to those produced by the present of SDS only. In fact, CTAB interacts with the SDS-bilayer assemblies and affects their packing and degree of ordering. This is due to different degrees of preferred growth orientation along the c-axis of the ZnO phase [21–23]. In addition, in cases of samples prepared at SDS:CTAB = 1:2 and CTAB:SDS = 1:0.2, in which CTAB concentration is higher than SDS, higher intensity of X-ray reflections was observed, indicating high crystallinity of the resulting samples.

Figure 2a–f show field emission scanning electron microscopy (FESEM) images of as-synthesized ZnO nanoparticles at different molar ratios of CTAB to SDS with constant amount of SDS.

To investigate the effects of cationic/anionic surfactants on ZnO morphology, each surfactant was used individually. ZnO-SDS particles shown in Figure 2a are mostly nanoflakes and containing few nanorods similar to previous study [21]. Samaele et al. [10] obtained ZnO-SDS nanorods agglomorated with sphere-like particles by aqueos solution. Sun et al. [12] gained ZnO-SDS nanowires with high aspect ratio by microemulsion method. The as-synthesised ZnO-SDS nanoparticles calcined at 500 °C shows no significant difference compared with that of the uncalcined one.

Adding CTAB to the mentioned solution has noticeably affected the morphology of particles. Aggregated nanosheets can be observed with the existence of CTAB with molar ratio of 0.5 as seen in Figure 2b. Usui [19] have reported almost same morphology for ZnO crystals grown with the mixture of CTAB/SDS by precipitation method. Increasing the CTAB molar ratio from 0.5 to 1 and 1.5 alter the morphology to rose-like structure and hexagonal-shaped rods form spherical outline as shown in Figure 2c,d, respectively. Detailed view on the rose-like structure indicates that it is comprised of many nanoflakes and nanoparticles. The hexagonal-shape rods form spherical outline ZnO is clear in the inset of Figure 2d, which shows TEM image. When the molar ratio of SDS:CTAB = 1:2, some sphere outline consisting rods and flower-like (hyacinth-like) structures coexist in the product (see Figure 2e,f).

Samples prepared with no CTAB and low concentration of CTAB, the length of flakes and sheets were in the range of 80–500 nm while for sample prepared using SDS:CTAB = 1:1.5 rods with length of 2–3 μm and diameter of 50–250 nm were observed. With increasing the CTAB molar ratio to 2, length and diameter of the rods both decreased to 1 μm and 50–150 nm, respectively. The decrease in diameter of nanorods can be described as follow. The negatively charged SDS ions assembled and form spherical micelles, which encapsulate ZnO seeds. This encapsulation restricts the access of zinc ions to the ZnO seeds. This process leads to formation of nucleation sites of new ZnO rods growing randomly at the outer edge of micelles. When CTAB was added, the surfactant may cap the ZnO nanorods to inhibit further lateral growth. Therefore, the increased of CTAB concentration limits the lateral growth of ZnO rods and decreased the diameter and length of ZnO crystals. Similar results on lateral growth restriction of ZnO nanorods were previously reported by Maiti et al. [22].

Figure 3a–f show FESEM images of as-obtained ZnO nanostructures at different molar ratios of SDS to CTAB.

It was found that nanoflaked-shape obtained in presence of CTAB, while by adding little amount of SDS, hexagonal-shaped rods form spherical outline were observed. The length of nanorods are in the range of 100 nm to 2 μm and diameter varies, but below 200 nm. With increasing of SDS molar ratio from 0.2 to 0.36, another morpholgy has revealed which is branch rod-like structure. The length and diameter decreased dramatically to the range of 100–300 nm and 15–50 nm, respectively. Most of rods are agglomorated and seems thicker. The nano- and micro-sized structures can be observed in Figure 3c.

Figure 3d–f show ZnO synthesized at different molar ratios of SDS; 0.5, 1 and 1.5, respectively. The morphology is based on the branch rod-like nanostructures but more agglomoration can be easily observed. Rod’s shape is not clear and nanoparticles look like a triangular structure.

The origin of different morphology can be elaborated by a chemical interaction between the ZnO seed surface and ions produced by the ionization of the SDS and CTAB [19]. The growth mechanism of ZnO crystal is suggested as follows.

The results showed that by adjusting the ratio of the surfactants, SDS to CTAB, we are able to control the physico-chemical properties of the resulting samples. In the case of constant amount of SDS followed by addition of CTAB, the interactions between the zinc species, ([Zn(OH)₄]²⁻) and SDS in the alkaline media are mediated by CTAB cationic surfactant, in which an increase in cationic concentration can significantly promote the interactions between zinc species and SDS [23]. As it can be seen from Figure 2a–f, the morphology of ZnO nanostructures changed from disordered flake-shapes to rod-, flower- and sphere-like structures.

In the case of the synthesis using a constant amount of CTAB, the addition of SDS as a second surfactant would control the electrostatic interaction between zinc species ([Zn(OH)₄]²⁻) and cationic part of CTAB on the surface of formed micelles. SDS due to its sulfate ion can change the geometry of CTAB micelles and suppresses the growth of ZnO crystals [24]. Thus ZnO nanoparticles with smaller size are obtained. As seen in Figure 3a–f, the increase of the SDS amounts in the mother liquor, resulted in different environments of interaction between CTA⁺ and zinc species, and subsequently resulted in different morphologies of the ZnO, from rods-spherical outline and branched rod-like to nanoparticles shapes.

Figure 4 shows TEM images of ZnO nanostructure synthesized at molar ratios of CTAB:SDS = 1:0.5, 1:1 and 1:1.5. The average size of particles decreased from 300 nm to 150 nm and then 100 nm with increasing the amount of SDS. Therefore at specific molar ratios of SDS to CTAB, ZnO nanoparticles were observed.

Thermal analysis of the as-synthesized products is shown in Figure 5 for samples synthesized at different molar ratios of CTAB and SDS. (The rest of Thermogravimetric and differential thermogravimetric analysis (TGA-DTG) plots can be found in the supplementary information).

Most of the TGA/DTG curves present similar behavior with one main weight loss step in the temperature range of 30–500 °C including the dehydration of physically adsorbed water from 30–200 °C and decomposition of surfactants from 200 to 500 °C [25]. Thermal degradation of pure SDS and CTAB were seperately recorded, showing a loss of between 160–380 °C and 180–340 °C, respectively. However it can be easily interpreted from the curves that the remaining weight did not level off even near to 1000 °C. Therefore there is possibility of gradually decomposition of surfactants and shifting to higher temperature. As seen for sample CTAB:SDS = 1:0.36 in Table 1, the offset temperature of surfactants decomposition has shifted to 934 °C.

As shown in the Table 1, two weight loss steps were obeserved in ZnO samples prepared at SDS:CTAB = 1:0.5 and 1:2. The temperature range of decomposition for sample prepared at molar ratio of 0.5 is about 60–500 °C which is very close to the rest of samples. For the sample with the molar ratio of 2, the first weight loss of 1.7% is described as removal of bonded water and the second weight loss of 5.3% observed between 424 and 948 °C due to the surfactants removal. Samaele et al. [10] reported almost similar temperature (577–900 °C) for SDS-modified ZnO particles via the precipitation method. The higher temperature for surfactant degradation can be attributed to relatively high heating rate and the quality of the material. Denoyel et al. [26] claimed that reducing the heating rate from 10 to 1 °C/min can definitely shift the TGA curves to lower temperatures.

The only three steps of weight loss was assigned for sample prepared at CTAB:SDS = 1:0.5 with temperature range of 30–349 °C, 355–464 °C and 464–599 °C which are corresponding to two weight losses as discussed above (30–500 °C). On the basis of a little temperature differences, there is another possibility of partial degradation of surfactant, meaning that in the first step, the surfactant started to degrade right after dehydration of water but incomplete, and again in second step the residual surfactants decomposed and so on until entire degradation taken place [26].

To study the adsorption of surfactants on the surface of ZnO nanostructures, FTIR spectrum was recorded in the range of 4000–280 cm⁻¹ (Figure 6A,B). It is notable that most of FTIR spectra show no characteristic peaks of CTAB or SDS which is in a good agreement with the XRD results previously discussed. Samples prepared using SDS:CTAB = 1:0.5 (Figure 6A) and CTAB:SDS = 1:1 (Figure 6B) are exceptional which demonstrate quite similar bands at 3229, 2918, 1468, 1224, 1027 and 860 cm⁻¹ corresponds to water O–H stretching vibration, C–H stretching and bending, S=O stretching vibration of SO₄ from SDS molecule and C–H stretching, respectively [27–29]. Indeed these peaks are assigned to chemical groups at the particle surface [27], which cannot be easily removed during the washing process, but can be easily removed through calcinations. As seen in Figure. 7, the characteristic peaks of surfactants obviously disappeared after calcinations at 500 °C for 5 h.

The results from TGA show total weight loss from 2% to 12.5% including decomposition of adsorbed surfactants. Therefore, the absence of the adsorption of CTAB and SDS molecules in the corresponding FTIR spectra may be due to the trace content of CTAB and SDS molecules adsorbed on ZnO structures.

The characteristic peak of ZnO absorption can be clearly observed at 330–370 cm⁻¹ which is less than that of reported in literatures. Wu et al. [30] reported two maximum characteristic absorption peaks for ZnO nanorods using CTAB as surfactant at 512 cm⁻¹ (blue-shifted) and another with wavenumber shorter than 500 cm⁻¹. Samaele et al. [10] and Prasad et al. [27] also reported absorption peak of 470 cm⁻¹ and 422 cm⁻¹ for SDS-modified ZnO particles and nanorods, respectively. It seems that mixture of these cationic-anionic surfactants shifted the characteristic peak of ZnO to lower wavenumbers (blue-shifted) as seen in Figure 6A,B. It is worth to mention that calcination treatment could shift characteristic peaks of ZnO back to higher wavenumbers (Figure 7).

All chemicals used in this work were of analytical reagent grade and used as received without any further purification. All the aqueous solutions were prepared using deionized water.

Powder X-ray diffraction (PXRD) analysis was performed on a Shimadzu diffractometer, XRD-6000 (Tokyo, Japan) equipped with CuKα radiation. The morphology of the micro and nanostructures were characterized by field emission scanning electron microscopy (FESEM) JOEL JSM-6400 (Tokyo, Japan) and transmission electron microscopy (TEM) Hitachi H7100. Surface characterization of the material was carried out using nitrogen gas adsorption–desorption technique at 77 K using a Micromeritics ASAP 2000 (Norcross, GA, USA). Thermogravimetric and differential thermogravimetric analysis (TGA-DTG) were carried out using a Mettler Toledo instrument (Greifensee, Switzerland) using heating rate of 10 °C/min, in the range of 25–1000 °C under a nitrogen atmosphere. Fourier transform infrared spectra were recorded over the 280–4000 cm⁻¹ range using a Perkin-Elmer 100 spectrophotometer (Waltham, MA, USA) under standard conditions. Ultraviolet-visible spectra were used to determine the optical properties.