Evolution of Structural and Optical Properties of ZnO ...

0 downloads 0 Views 8MB Size Report
Jan 26, 2018 - Ji Hyun Kim 2 and Sam-Dong Kim 1,*. 1. Division of Electronics and Electrical Engineering, Dongguk University, Seoul 100-715, Korea;.
nanomaterials Article

Evolution of Structural and Optical Properties of ZnO Nanorods Grown on Vacuum Annealed Seed Crystallites Waqar Khan 1 ID , Fasihullah Khan 1 , Hafiz Muhammad Salman Ajmal 1 , Noor Ul Huda 1 , Ji Hyun Kim 2 and Sam-Dong Kim 1, * 1

2

*

Division of Electronics and Electrical Engineering, Dongguk University, Seoul 100-715, Korea; [email protected] (W.K.); [email protected] (F.K.); [email protected] (H.M.S.A.); [email protected] (N.U.H.) School of Advanced Materials Science and Engineering, Sungkyunkwan University, Suwon-Si, Gyeonggi-Do 16419, Korea; [email protected] Correspondence: [email protected]; Tel.: +82-2-2260-3800; Fax: +82-2-2277-8735

Received: 26 December 2017; Accepted: 23 January 2018; Published: 26 January 2018

Abstract: In this study, the ambient condition for the as-coated seed layer (SL) annealing at 350 ◦ C is varied from air or nitrogen to vacuum to examine the evolution of structural and optical properties of ZnO nanorods (NRs). The NR crystals of high surface density (~240 rods/µm2 ) and aspect ratio (~20.3) show greatly enhanced (002) degree of orientation and crystalline quality, when grown on the SLs annealed in vacuum, compared to those annealed in air or nitrogen ambient. This is due to the vacuum-annealed SL crystals of a highly preferred orientation toward (002) and large grain sizes. X-ray photoelectron spectroscopy also reveals that the highest O/Zn atomic ratio of 0.89 is obtained in the case of vacuum-annealed SL crystals, which is due to the effective desorption of hydroxyl groups and other contaminants adsorbed on the surface formed during aqueous solution-based growth process. Near band edge emission (ultra violet range of 360–400 nm) of the vacuum-annealed SLs is also enhanced by 44% and 33% as compared to those annealed in air and nitrogen ambient, respectively, in photoluminescence with significant suppression of visible light emission associated with deep level transition. Due to this improvement of SL optical crystalline quality, the NR crystals grown on the vacuum-annealed SLs produce ~3 times higher ultra violet emission intensity than the other samples. In summary, it is shown that the ZnO NRs preferentially grow along the wurtzite c-axis direction, thereby producing the high crystalline quality of nanostructures when they grow on the vacuum-annealed SLs of high crystalline quality with minimized impurities and excellent preferred orientation. The ZnO nanostructures of high crystalline quality achieved in this study can be utilized for a wide range of potential device applications such as laser diodes, light-emitting diodes, piezoelectric transducers and generators, gas sensors, and ultraviolet detectors. Keywords: ZnO nanorods; vacuum annealing; photoluminescence; hydrothermal process; surface defects

1. Introduction Semiconducting metal oxides have been intensively developed for optoelectronics and sensor material applications of various transduction platforms in last several decades due their unique material properties of optical transparency based on wide energy bandgap and chemiresistive behavior depending on specific gas adsorption on the surface [1]. Future microelectronics technology utilizing these new functions of metal oxide will therefore lead to a variety of challenging applications and their new consequent industrial market. Among many semiconducting metal oxides, ZnO has been

Nanomaterials 2018, 8, 68; doi:10.3390/nano8020068

www.mdpi.com/journal/nanomaterials

Nanomaterials 2018, 8, 68

2 of 17

highlighted and extensively investigated in recent years for its distinctive properties and potential applications in optoelectronics, gas sensors, and energy harvesting applications [1–5]. It has a direct band gap of ~3.37 eV at room temperature (RT), a large exciton binding energy of 60 meV, thermal and chemical stability, piezoelectricity, radiation hardness, and the option of wet chemical etching [4–7]. Numerous methods have been investigated for the synthesis of ZnO crystals thus far, but typical methods adopted for the fabrication of ZnO nanostructures include vapor phase epitaxy, pulsed laser deposition, spray pyrolysis, molecular beam epitaxy, and chemical vapor deposition techniques [3,8–11]. However, most of these process schemes require complicated facilities and high thermal budget which ultimately hinder low-cost and large scale fabrication on flexible substrates. Among various growth methods for the ZnO nanorods (NRs), a synthesis technique in aqueous solution such as hydrothermal method is very promising because they can proceed at relatively low thermal budget allowing low-cost and large-scale roll to roll fabrication [4,7,12,13]. ZnO also provide a wide variety of nanostructure morphologies such as NRs, nanowires, nano-belts, and nano-flowers for many practical applications [8–13]. Fabrication of nano-scale ZnO materials with special morphology and high crystalline quality is of great interest for materials science because of its significance in the scientific research and potential in the miniaturized technological applications. Drawback of hydrothermal technique is however the presence of huge volume of defects on the surface of grown nanostructures affecting the structural and optical properties of ZnO nanostructures. The optimization of optical and structural properties can be achieved by fully understanding the growth mechanism of nanostructures and how the defects in the crystals behave through the post-surface treatment. However, the morphological control and crystal structure evolution to a low defect-density nanostructures remain challenging to material scientists. In order to suppress the defects in ZnO nanostructures, great amount of research effort has been made by using post growth annealings; for example, Qui et al. post-treated the as-grown ZnO nanowires at 550 ◦ C in different ambient conditions and reported the enhancement of ZnO nanocrystals to the high crystalline quality through vacuum annealing [6], however sufficient understanding how this method contributed to the change in crystalline quality was not achieved. In this study, we investigate how the post-annealing ambient for the ZnO seed layer (SL), which serves as a platform for the subsequent nanocrystal nucleation, affects the crystalline quality of ZnO NR growth. To develop hydrothermal method to synthesize high crystalline quality ZnO microstructures with a simple post-annealing process at relatively low temperature will be highly desirable due to its easily controllable condition. In addition, the possible defect annihilation mechanism in ZnO NR crystals during the post-annealing is expected to further investigate. If material properties can be precisely tuned by deploying defect chemistry, we can satisfy the precise requirements for applications through an understanding of the defect formation mechanism. For this purpose, we investigated in this study the role of annealing ambient and the evolution of defects in as-annealed SLs as well as ZnO NR crystals by using various surface characterization techniques including X-ray diffraction (XRD), photoluminescence (PL), X-ray photoelectron spectroscopy (XPS), field emission scanning electron microscopy (FE-SEM), Fourier transform infrared (FT-IR) spectroscopy, atomic force microscopy (AFM), and transmission electron microscopy (TEM). 2. Experimental Procedure ZnO NRs were grown on p-Si (100) substrate (2 × 2 cm2 , 0.01 Ω-cm resistivity of boron doping) using an aqueous solution method in the following way. After the surface cleaning by acetone and isopropyl alcohol sequentially to remove dust and organic contaminants, the substrates were rinsed by de-ionized (DI) water and dried with nitrogen (N2 ) purge followed by moisture removal on hot-plate at 120 ◦ C for 2 min. Aqueous solution route for the growths of ZnO NRs consists of the following two steps. First, a colloidal sol-gel was prepared by mixing zinc acetate dehydrate [Zn(CH3 COO)2 ·2H2 O] in an organic solvent of 1-propanol to form a 10 mM concentrate solution. After constantly sonicating the solution for 30 min, it was kept at room temperature longer than 5–6 h for the sol-gel stabilization.

Nanomaterials 2018, 8, 68 Nanomaterials 2018, 8, x FOR PEER REVIEW

3 of 17 3 of 17

the stabilization. This for the SL growth then spin-coated on the Thissol-gel solution for the ZnO SL solution growth was thenZnO spin-coated onwas the substrate at 3000 rpm forsubstrate 30 s and ◦ at 3000 at rpm s and baked 100 °C for was 1 min, and the10coating was repeated 10 times of to achieve baked 100forC30 for 1 min, andat the coating repeated times to achieve a thickness ~20 nm ◦ aafter thickness of ~20 nm after annealing. The SL-grown samples were post-annealed at 350 °C with in a annealing. The SL-grown samples were post-annealed at 350 C in a convection furnace convection furnace with three different of vacuum (50 air, and 2. three different ambients of vacuum (50 ambients mTorr), atmospheric air, mTorr), and N2 .atmospheric Post-annealings for N the Post-annealings for the as-coated SLsonly weretocarried outthe notorganic only to residuals remove the organic residuals and as-coated SLs were carried out not remove and unwanted reaction unwanted reaction by-products remaining in the but how also the to investigate the of crystalline by-products remaining in the crystals but also to crystals investigate crystallinehow quality the SL is quality of by thetheSL is influenced the annealing ambient in characteristics connection with material influenced annealing ambientby in connection with the material of thethe ZnO NRs to characteristics of the ZnO NRs to be grown afterwards. be grown afterwards. In In the the second second step, step, we we immersed immersed the the substrates substrates to to grow grow the the NR NR crystals crystalsin inan anequimolar equimolargrowth growth solution (25 zinc mM) zinc nitrate (Zn(NO hexahydrate (Zn(NO 3)2·6H 2O, 99%) and (25 mM) · solution of of (25 mM) nitrate hexahydrate ) 6H O, 99%) and (25 mM) hexamethylenetetramine 3 2 2 hexamethylenetetramine (HMT) (C 6 H 12 N 4 , 99.5%) in 250 ml DI water as illustrated in Figure 1. After (HMT) (C6 H12 N4 , 99.5%) in 250 ml DI water as illustrated in Figure 1. After complete stirring of the complete stirringthe of the growth solution, the seed-grown substrates placed for upside down in a growth solution, seed-grown substrates were placed upside downwere in a beaker the NR growth, ◦ C for on beaker the NR growth, was foil sealed an aluminum foil at and hot plate and thefor beaker was sealed and withthe anbeaker aluminum andwith placed on a hot plate 90placed 5–6ah. Finally, at °C forZnO 5–6 h. Finally, grown ZnO were the90grown NRs were the washed with DINRs water andwashed purged with with DI N2water gas. and purged with N2 gas. To explore the the evolved evolved morphologies morphologies and To explore and crystalline crystalline structures structures of of the the as-grown as-grown NRs, NRs, we we performed FE-SEM (Hitachi S-4800S, operated at 15 kV, Suwon, Korea) and cross-sectional TEM performed FE-SEM (Hitachi S-4800S, operated at 15 kV, Suwon, Korea) and cross-sectional TEM (Hitachi (Hitachi 9500 9500 at at 300 300 KV) KV) analysis analysis with with selected selected area area electron electron diffraction diffraction (SAED). (SAED). Crystalline Crystalline quality quality and orientationwere wereinvestigated investigatedby byXRD XRD(D8 (D8 Advance spectrometer of Bruker AXS with and preferred preferred orientation Advance spectrometer of Bruker AXS with Cu Cu Kα 1.540 Å radiation, Seoul, Korea) for each SL and NR samples prepared under different Kα 1.540 Å radiation, Seoul, Korea) for each SL and NR samples prepared under different conditions. conditions. PL spectroscopy (MFP-3D Bio,Research, Asylum Research, Suwon,excited Korea) by excited by a 325-nm line PL spectroscopy (MFP-3D Bio, Asylum Suwon, Korea) a 325-nm line HeCd HeCd laser was also carried out at RT to examine the optical properties of the samples. Chemical laser was also carried out at RT to examine the optical properties of the samples. Chemical bonding bonding and stoichiometric were by XPSPHI using PHI 5000Probe Versa(Ulvac-PHI, Probe (Ulvac-PHI, and stoichiometric analysis analysis were done bydone XPS using 5000 Versa Suwon, Suwon, Korea) spectrometer with monochromator Al Kα eV) (1486.6 eV)(25.0 anode W,The 15 kV). The Korea) spectrometer with monochromator Al Kα (1486.6 anode W, (25.0 15 kV). bonding bonding configurations of the ZnO SL crystals were investigated by FT-IR spectroscopy (IFS66v/s configurations of the ZnO SL crystals were investigated by FT-IR spectroscopy (IFS66v/s and Hyperion and AFMBruker) (N8-NEOS, Bruker) inmode non-contact mode was out alsotocarried outthe to 3000,Hyperion Bruker). 3000, AFMBruker). (N8-NEOS, in non-contact was also carried examine examine the surface morphologies of the SLs annealedambient in different ambient conditions. surface morphologies of the SLs annealed in different conditions.

Figure post-annealing ambient Figure 1. Process Process flow flow of of ZnO nanorods nanorods (NRs) (NRs) growth with three different post-annealing ambient conditions (vacuum, air and N 2 ) for the seed layers (SLs). conditions 2 ) for the seed layers (SLs).

3. 3. Results Results and and Discussion Discussion The The growth growth morphology morphology of of the the ZnO ZnO NRs NRs was was first first examined examined by by FE-SEM, FE-SEM, and and Figure Figure 2a–c 2a–c show show the top views of the NRs grown on SLs annealed under three different ambient conditions of vacuum, the top views of the NRs grown on SLs annealed under three different ambient conditions of vacuum, air, air, and and N N22,, respectively. respectively. Average Average lengths lengths of of the the NRs NRs were were 1~1.3 1~1.3 μm µm as as shown shown in in the the cross-sectional cross-sectional view in Figure 2d, and the NRs grown on the vacuum-annealed SLs exhibited the fastest growth rate. Shown in Figure 2e–g are the histograms of NR diameter distributions, and it was shown that the

Nanomaterials 2018, 8, 68

4 of 17

view in Figure andPEER the REVIEW NRs grown on the vacuum-annealed SLs exhibited the fastest growth Nanomaterials 2018, 2d, 8, x FOR 4 of 17 rate. Shown in Figure 2e–g are the histograms of NR diameter distributions, and it was shown that the statistics of NR diameter were also significantly affectedbybythe theSL SLannealing annealingcondition. condition. Table Table 1 statistics of NR diameter were also significantly affected summarized the morphology statistics of the NRs grown under three different conditions. conditions. The NRs grown on SLs annealed in vacuum showed the smallest mean diameter of 65 nm; on the other hand, the larger diameters diameters of of 80 80 and and 115 nm were were measured measured in in the the cases cases of of annealings annealings in in air air and and N N22 ambient condition, respectively. Aspect ratio of the NRs was decreased from 20.3 (vacuum) to 13.7 (air) and 8.8 (N22) depending on the annealing ambient. The NRs also exhibited the maximum surface density (~240 rods/µm of of vacuum-annealing, but but the density was significantly reducedreduced by 33–62% rods/μm22) in inthe thecase case vacuum-annealing, the density was significantly by in the cases of cases different ambient ambient conditions. 33–62% in the of different conditions.

(a)

(b)

(c)

(d)

(e)

(f) Figure 2. Cont.

Nanomaterials 2018, 8, 68 Nanomaterials 2018, 8, x FOR PEER REVIEW Nanomaterials 2018, 8, x FOR PEER REVIEW

5 of 17 5 of 17 5 of 17

(g) (g) Figure 2. Field emission scanningelectron electron microscopy(FE-SEM) (FE-SEM) (top view) of the ZnO NRs grown Figure 2. 2. Field emission ofof thethe ZnO NRs grown on Figure Field emission scanning scanning electronmicroscopy microscopy (FE-SEM)(top (topview) view) ZnO NRs grown on SLs annealed in (a) vacuum; (b) air; and (c) nitrogen. Cross-sectional views of the ZnO NRs are SLsSLs annealed in (a) (b) air; Cross-sectional viewsviews of the of ZnO shown on annealed in vacuum; (a) vacuum; (b) and air; (c) andnitrogen. (c) nitrogen. Cross-sectional theNRs ZnOare NRs are shown in (d). Histogram for the diameter distribution ofgrown NRs grown on SLs annealed in (e) vacuum; in (d). Histogram for the for diameter distribution of NRsof on SLson annealed in (e) vacuum; (f) air, shown in (d). Histogram the diameter distribution NRs grown SLs annealed in (e) vacuum; (f) air, and (g) nitrogen. andair, (g)and nitrogen. (f) (g) nitrogen. Table 1. Morphologies of the ZnO NRs (measured by FE-SEM) prepared under three different SL Table 1. Morphologies of of the the ZnO ZnO NRs NRs (measured (measured by by FE-SEM) FE-SEM) prepared prepared under under three three different different SL SL Table 1. Morphologies annealing ambient conditions. annealing ambient conditions. annealing ambient conditions. Annealing Diameter Mean Standard Deviation of NR Density Aspect Annealing Diameter Mean Standard Deviation of NRDensity Density Aspect Annealing Diameter Mean Standard NR Aspect Ratio 2) Ambient for SL Range (nm) Diameter (nm) DiameterDeviation (nm) (rods/μm Ratio of NR 2 )2 Ambient Range (nm) Diameter Diameter (nm) of Diameter (nm) (rods/µm of NRof NR Ambient forfor SLSL Range (nm) (nm) Diameter (nm) (rods/μm ) Ratio Vacuum 50–110 65 11.8 ~240 20.3 Vacuum 50–110 65 65 11.8 ~240 20.3 Vacuum 50–110 11.8 ~240 20.3 Air 60–130 80 12.0 ~160 13.7 60–130 12.0 ~160 13.7 AirAir 60–130 80 80 12.0 ~160 13.7 Nitrogen 70–190 115 20.4 ~90 8.8 Nitrogen 70–190 20.4 ~90 8.88.8 Nitrogen 70–190 115115 20.4 ~90

We obtained well aligned NRs normal to the substrates with pure wurtzite hexagonal faces as We obtained well aligned NRs normal to the substrates with pure wurtzite hexagonal hexagonal faces faces as as We shown in each micrograph of Figure 2a–c. The degree of NR c-axis alignment was more prominent shown in each micrograph was more prominent in shown micrograph of ofFigure Figure2a–c. 2a–c.The Thedegree degreeofofNR NRc-axis c-axisalignment alignment was more prominent in the case of vacuum-annealing than the other cases. Figure 3a,b respectively represent the θ-2θ XRD thethe case ofof vacuum-annealing the θ-2θ θ-2θ XRD in case vacuum-annealingthan thanthe theother othercases. cases.Figure Figure 3a,b 3a,b respectively represent the spectra obtained from the SLs annealed under three different ambient conditions and the NRs grown spectra obtained obtained from from the SLs SLs annealed annealed under under three three different different ambient ambient conditions conditions and and the the NRs NRs grown grown spectra atop each SL. atop each SL. atop

(a) (a)

(b) (b)

Figure 3. θ-2θ X-ray diffraction (XRD) patterns of (a) the SLs annealed in three different ambients; Figure 3. θ-2θ X-ray diffraction (XRD) patterns of (a) the SLs annealed in three different ambients; Figure diffraction (XRD) patterns and (b) 3. theθ-2θ ZnOX-ray NR grown on each different SL. of (a) the SLs annealed in three different ambients; and (b) the ZnO NR grown on each different SL. and (b) the ZnO NR grown on each different SL.

From the XRD patterns, it was shown that all reflections were in perfect agreement with the From the XRD patterns, it was shown that all reflections were in perfect agreement with the Fromindexes the XRD patterns, it was that all reflections were in perfect withcase, the reported of JCPDS files (cardshown number 36-1451) for the hexagonal phaseagreement ZnO. In each reported indexes of JCPDS files (card number 36-1451) for the hexagonal phase ZnO. In each case, reported indexes of JCPDS files (card number 36-1451) for the hexagonal phase ZnO. In each case, the most intense peaks along (002) orientation were observed from the SLs, and this represents that the most intense peaks along (002) orientation were observed from the SLs, and this represents that the most peaks along (002) orientationorientation were observed from the SLs, and thisthe represents that our our ZnO intense seed crystallites have a preferred along c-axis. Especially, degree of (002) our ZnO seed crystallites have a preferred orientation along c-axis. Especially, the degree of (002) preferred orientation was the most strong from the SLs annealed in vacuum (at 2θ = 34.44°) with no preferred orientation was the most strong from the SLs annealed in vacuum (at 2θ = 34.44°) with no visible reflection from other planes, while fairly significant (100) reflections were also observed from visible reflection from other planes, while fairly significant (100) reflections were also observed from

Nanomaterials 2018, 8, 68

6 of 17

ZnO seed crystallites have a preferred orientation along c-axis. Especially, the degree of (002) preferred orientation was the most strong from the SLs annealed in vacuum (at 2θ = 34.44◦ ) with no visible reflection from other planes, while fairly significant (100) reflections were also observed from the SLs when annealed in other ambient conditions [14]. The degree of orientation, F(hkl), can be given by the following equation: [15] ( P(hkl) − P0 (hkl)) F (hkl) = (1) (1 − P0 (hkl)) where P(hkl) = I(hkl)/ΣI(hkl), P0 (hkl) = I0 (hkl)/Σ(I0 (hkl), I(hkl) is the measured peak intensity from (hkl) plane, and I0 (hkl) is the reference peak intensity of (hkl) plane given by JCPDS card No. 36-1451. As summarized in Table 2, the highest value of F(002) was obtained from the SLs annealed in vacuum, whereas much lower degree of orientations were measured from the SL crystals annealed in air and N2 ambients. The excellent F(002) value in the case of vacuum annealing suggests that the gas molecules and/or contaminants present in the annealing ambient can play a very important role in the SL crystal growth and grain coalescence process [14–16]. The average grain size d was calculated from full width at half maximum (FWHM) of (002) peak by using Debye-Scherer formula: [15,16] d = (0.94 λ)/(β Cos θ)

(2)

where λ is the X-ray wavelength (0.154 nm), β is FWHM in radians, and θ is the Bragg’s diffraction angle. The grain sizes of SL crystals calculated by this method were 55.6 nm (vacuum), 30.4 nm (air), and 28.3 nm (N2 ) in each different annealing ambient condition as shown in Table 2. Annealing in vacuum can promote the grain growth and intergranular coalescence of the SL crystals through the grain boundary diffusion, thereby producing large grain sizes during the post-annealing. However, gas molecules and contaminants in the grain boundaries introduced during the annealing in air or N2 ambients can suppress the intergranular diffusion for the evolutionary SL crystal growth [12,15,16]. Table 2. Parameters of the ZnO SL crystals extracted by XRD analysis. Annealing Ambient for SL

(002) 2θ (◦ )

(002) FWHM (◦ )

Grain Size (nm)

(002) Spacing (Å)

Degree of (002) Orientation

Vacuum Air Nitrogen

34.44 34.53 34.54

0.156 0.303 0.304

55.6 30.4 28.3

2.600 2.594 2.593

0.98 0.66 0.61

The XRD patterns for the NRs grown on the SLs annealed under different environment were also shown in Figure 3b. Single strong peak (2θ = 34.47◦ ) along (002) orientation was observed from the NRs grown on the SLs annealed in vacuum, and no other peak linked to any different orientation was found. This reveals that the ZnO NR crystals, in this case, grow dominantly along c-axis in the vertical direction to the substrate. On the other hand, we obtained weak but visible additional peaks from (100) and (101) planes at 31.7◦ and 36.4◦ , respectively, with a dominant peak from (002) from the NRs grown on the SLs annealed in air and N2 . The intensity of (002) peak measured from the vacuum annealing was significantly reduced by 2.5 and 7.6 times in the cases of air and N2 ambient annealing, respectively. This depicts that the c-axis alignment of the ZnO NR growth was considerably effected by the annealing ambient condition for the SLs. As was discussed earlier, the SL crystals annealed in vacuum state have a highly preferred orientation toward (002) and large grain sizes. As suggested by many former researchers, each crystalline surface of the SL grain acts as a nuclei for the growth of NRs, and the ZnO NRs tend to dominantly grow along the [001] direction because of its lower surface free energy (1.6 J/m2 ) than those of (100) (3.4 J/m2 ) and (101) (2.0 J/m2 ) planes [17,18]. This can explain why we have a strong c-axis alignment along vertical direction to the substrate when the ZnO NRs grown on the vacuum-annealed SLs.

Nanomaterials 2018, 8, 68

7 of 17

TEM dark-field (DF) analyses were carried out for the NR structures grown under three different SL annealing conditions. Figure 4a–c shows the DF images of the NR crystals in the bottom rows, and clear differences in contrast were found from each NR crystal corresponding to their plane indexes diffracted the patterns. the Nanomaterialson 2018, 8, xSAED FOR PEER REVIEW As shown in Figure 4a, most of the NR crystals grown on 7 of 17 vacuum-annealed SLs were bright as observed under (002) diffraction condition, whereas only few NRs showed the bright bright diffraction diffraction contrast contrast under under (100) (100) or or (102) (102) diffraction diffraction condition. condition. The The samples samples prepared under different annealing ambients (such as air or N ) showed quite different contrast prepared under different annealing ambients (such as air or N22 showed different contrast distribution distribution in NRs depending depending upon upon the the SL SL annealing annealing condition condition as as shown shown in in Figure Figure 4b,c. 4b,c. Despite the limitation observation was in good agreement with the results limitation of of TEM TEM analysis analysis in very local areas, this observation of XRD analysis analysis of of higher higher statistical statistical reliability. reliability.

(a)

(b) Figure 4. Cont.

Nanomaterials 2018, 8, 68

8 of 17

Nanomaterials 2018, 8, x FOR PEER REVIEW Nanomaterials 2018, 8, x FOR PEER REVIEW

8 of 17 8 of 17

(c) (c)

Figure Figure 4. 4. Cross-sectional Cross-sectional TEM TEMmicrographs micrographsof ofthe theZnO ZnONRs NRsgrown grownon on SLs SLs annealed annealedin in(a) (a) vacuum, vacuum, Figure 4. Cross-sectional TEM micrographs of the ZnO and NRstheir grown on SLsarea annealed in (a) vacuum, (b) air, and (c) nitrogen ambients. Bright-field images selected electron diffraction (b) air, and (c) nitrogen ambients. Bright-field images and their selected area electron diffraction (SAED) (b) air, patterns and (c) nitrogen ambients. Bright-field images and their selected area electron diffraction (SAED) are respectively shown in (top-left) and (top-right). Dark field images are shown patterns are respectively shown in (top-left) and (top-right). Dark field images are shown according to (SAED) patterns are respectively shown in (top-left) and (top-right). Dark field images are shown according to each different diffraction condition of plane index in (bottom rows) with (top-center) each different diffraction condition of plane index in (bottom rows) with (top-center) their composite according to each different diffraction condition of plane index in (bottom rows) with (top-center) their composite views incolors. three different colors. views in three different their composite views in three different colors.

As shown in Figure 5, we performed AFM for each scanning area (20 × 20 μm22) of the SLs As in 5,5, we performed AFM area (20 µm As shown shown in Figure Figure we The performed AFM for for each eachscanning scanning (20×× 20 20 reduced μm2)) of of the the SLs SLs annealed in different ambients. AFM characterizations revealed area significantly surface annealed in different ambients. The AFM characterizations revealed significantly reduced surface annealed in different The AFM characterizations significantly surface roughness from the SL ambients. films annealed in vacuum compared torevealed those annealed in air reduced or N2. The root roughness from the annealed in vacuum compared to those annealed in or root 22.. The roughness the SL SLfilms films annealed inair air orNN Thefilms root mean squarefrom roughness of SL annealed annealed in in vacuum vacuum compared was foundtotothose be 1.34 nm, while those of SL mean square roughness of SL annealed in vacuum was found to be 1.34 nm, while those of SL films mean square roughness of SL annealed in vacuum was found to be 1.34 nm, while those of SL films annealed in air and N2 ambient were 1.84 and 2.05 nm, respectively. This enhancement surface annealed N ambient and 2.05 respectively. This enhancement of annealed in in air air and and ambient were were 1.84 1.84SL and 2.05 nm, nm, enhancement of surface surface smoothness from theN22vacuum-annealed crystals canrespectively. be due to This the higher (002) degree of smoothness from thethe vacuum-annealed SL crystals can becan duebe to the higher (002) degree of orientation smoothness from vacuum-annealed SL crystals due to the higher (002) degree of orientation as observed by our XRD analysis [19]. as observed as byobserved our XRD by analysis [19].analysis [19]. orientation our XRD

(a) (a) Figure 5. Cont.

Nanomaterials 2018, 8, 68

9 of 17

Nanomaterials 2018, 8, x FOR PEER REVIEW

9 of 17

(b)

(c) Figure 5. Atomic force microscopy (AFM) images of ZnO SLs annealed in (a) vacuum; (b) air, and (c) Figure 5. Atomic force microscopy (AFM) images of ZnO SLs annealed in (a) vacuum; (b) air, Nitrogen ambient. and (c) Nitrogen ambient.

Wide scan XPS spectra for the SLs annealed in different ambient conditions are shown in Figure scan XPS spectraeV. forMain the SLs annealed in ambient are shown inpeaks Figure 6aWide in a range of 0–1150 constituents ofdifferent Zn and O exhibitconditions various photoelectron of6a in adifferent range ofcore-levels 0–1150 eV.and Main constituents of Zn and O exhibit various photoelectron peaks of different spin-orbital splittings as well as Auger peaks, and some traces of carbon are core-levels and in spin-orbital splittings astwo Auger andZn some traces of(1044.4 carboneV) areand also also detected the SL crystals. Figure as 6b well shows clearpeaks, core-level peaks of 2p1/2 detected in theeV) SL separated crystals. Figure 6b shows two clear core-level Zn peaks of 2p (1044.4air, eV) and 2p3/2 (1021.3 by spin-orbital splitting for each SL crystal annealed in1/2 vacuum, and N2.(1021.3 The peak locations, symmetric shape ofsplitting the peaks, spin value (23.1 eV) air, of the 2p3/2 eV) separated by spin-orbital forand each SLorbital crystalsplitting annealed in vacuum, and 2+ chemical state in ZnO stoichiometry in all cases Zn 2p doublet confirms that Zn is present as Zn N2 . The peak locations, symmetric shape of the peaks, and spin orbital splitting value (23.1 eV) of the However, 2p1/2 peak annealed in2+vacuum was shifted by 0.17 eV to higherinenergy, and Zn [16]. 2p doublet confirms thatof Znthe is SL present as Zn chemical state in ZnO stoichiometry all cases [16]. this shift is attributed to a higher oxidization state of Zn element in the ZnO core matrix. However, 2p1/2 peak of the SL annealed in vacuum was shifted by 0.17 eV to higher energy, and this To examinetothe changeoxidization of chemical state crystals, we ZnO observed closely O1s peaks shift is attributed a higher state in ofSL ZnZnO element in the core more matrix. deconvoluted into three separate satellite peaks of O a, Ob, and Oc as shown in Figure 6c–e. Oa, which To examine the change of chemical state in SL ZnO crystals, we observed more closely O1s peaks 2− ion in the wurtzite structure of is the lowest energy peak at ~530.3 corresponds theO Oas deconvoluted into three separate satelliteeV, peaks of Oa , Ob ,to and shown in Figure 6c–e. Oa , which is c hexagonal ZnO. Because Oa is a good measure of stoichiometric oxygen presence in the wurtzite the lowest energy peak at ~530.3 eV, corresponds to the O2− ion in the wurtzite structure of hexagonal structure, we evaluated the stoichiometry of each ZnO SL crystals by using ∫Oa/∫Zn in each ZnO. Because Oa is a good measure of stoichiometric oxygen presence in the wurtzite structure, R corresponding annealing condition [16,20–22], where ∫Oa and ∫ZnRrespectively represent the peak we evaluated the stoichiometry of each ZnO SL crystals by using Oa / Zn in each corresponding R R Table 3 that the percentage contribution of Oa in Ot curve integration of Oa and Zn. It is observed in annealing condition [16,20–22], where Oa and Zn respectively represent the peak curve integration (Ot = Oa + Ob + Oc) is the highest in the case of vacuum-annealing. Each percentage value of Oa, Ob, of O Zn. It is3observed in Table 3kthat percentage of Oaofin Ot (Ot = O Ob + Oc ) a and a +satellite and Oc in Table was obtained by ∫O /∫Ot, the where ∫Ok is the contribution curve integration individual O1s is the highest in the case of vacuum-annealing. Each percentage value of O , O , and O in Table 3 was a c b peak (k = Ra, b, or R c), and ∫OtRis the curve integration of total O1s peak. Although ZnO is oxygenobtained by O / O , where O is the curve integration of individual O1s satellite peak (k = a, b, k deficient int nature, butk the highest O/Zn atomic ratio of 0.89 was found in the case of R material or c), and Ot is the curve integration of total O1s peak. Although is oxygen-deficient material vacuum-annealed SL crystals, as shown in Table 3, which is evenZnO higher than those of other asin nature, but the highest O/Zn atomic ratio of 0.89 was found in the case of vacuum-annealed annealed ZnO nanocrystals grown by hydrothermal methods [23]. Ob (~531.2 eV) is thought to beSL crystals, as shown in Table 3, which is even higher than those of other as-annealed ZnO nanocrystals

Nanomaterials 2018, 8, 68

10 of 17

grown by hydrothermal methods [23]. Ob (~531.2 eV) is thought to be originated from O2– ions in the oxygen deficient within the ZnO matrix [20,21]. The higher binding energy O10c of component Nanomaterials 2018, 8,region x FOR PEER REVIEW 17 (~532.1 eV) is associated with the presence of loosely bound oxygen such as –CO3 , –OH species on the originated from O2– ions in the oxygen deficient region within the ZnO matrix [20,21]. The higher surface of ZnO crystals. One interesting observation is that the SL crystals annealed in vacuum show binding energy Oc component (~532.1 eV) is associated with the presence of loosely bound oxygen significant inspecies Oc contribution compared to those other ambients asSL shown in such reduction as –CO3, –OH on the surface of ZnO crystals. Oneannealed interestingin observation is that the Figure crystals 6c–e and Table 3.in This highshow suppression Oc percentage of the vacuum-annealed samples is annealed vacuum significantofreduction in Oc contribution compared to those annealed ambientsdesorption as shown inofFigure 6c–e and Table 3. This high suppression of Oc formed most likely due in to other the effective O2 , CO , or OH groups adsorbed on the surface 3 percentage of the vacuum-annealed samples is most likely due to the effective desorption of O 2 , CO 3, during our wet organic solution-based process [23]. As summarized in Table 3, the Oc percentage of or OH groups adsorbed on the surface formed during our wet organic solution-based process [23]. the vacuum-annealed samples have almost halved in value compared to those of the samples annealed As summarized in Table 3, the Oc percentage of the vacuum-annealed samples have almost halved in different ambients. in value compared to those of the samples annealed in different ambients.

(a)

(b)

(c)

(d)

(e) Figure 6. (a) Wide scan X-ray photoelectron spectroscopy (XPS) spectra of ZnO SLs annealed in three

Figure 6. (a) Wide scan X-ray photoelectron spectroscopy (XPS) spectra of ZnO SLs annealed in three different ambients; (b) High resolution spectra of Zn-2p of the SLs; O1s core level spectra obtained different ambients; (b) High resolution spectra of Zn-2p of the SLs; O1s core level spectra obtained from the ZnO SLs annealed in (c) vacuum; (d) air, and (e) nitrogen ambient are de-convoluted into three distant satellite peaks.

Nanomaterials 2018, 8, x FOR PEER REVIEW

11 of 17

Nanomaterials 2018, 8, 68

11 of 17 from the ZnO SLs annealed in (c) vacuum; (d) air, and (e) nitrogen ambient are de-convoluted into three distant satellite peaks.

Table 3. Percentage values of each O1s satellite peak for the SLs and NRs calculated by dividing the Table 3. Percentage values of each O1speak satellite peak for integration the SLs andofNRs by atomic dividing the curve integration of individual satellite by the curve totalcalculated peak. O/Zn ratios curve integration of individual satellite peak by the curve integration of total peak. O/Zn atomic ratios extracted from each SL and NR crystal are also shown. extracted from each SL and NR crystal are also shown. Crystal Crystal

Annealing forfor SL SL OO a a(%) AnnealingAmbient Ambient (%)

O Obb (%) (%)

SLSL

Vacuum Vacuum Air Air Nitrogen

63.5 63.5 54.1 54.1 53.9

32.6 32.6 38.8 38.8 38.3

NR

Vacuum Vacuum Air Air Nitrogen

72.3 72.3 68.1 68.1 67.3

25.1 25.1 26.9 26.9 28.6

Nitrogen

67.3

28.6

Nitrogen

NR

53.9

38.3

(%) O/Zn O/Zn AtomicRatio Ratio OOcc(%) Atomic 3.7 0.89 3.7 0.89 6.9 0.76 6.9 0.76 7.7 0.74 7.7 0.74 2.4 0.72 2.4 0.72 4.9 0.66 4.9 0.66 4.0 0.65 4.0 0.65

Shown in Figure 7a,b are the wide scan XPS spectra and Zn 2p core-level patterns obtained from Shown in Figure 7a,b are the wide scan XPS spectra and Zn 2p core-level patterns obtained from the NRs grown on the SLs annealed in different ambient conditions. Similar trend was shown in the NRs grown on the SLs annealed in different ambient conditions. Similar trend was shown in the the spectra compared to those of the SLs, and all of the peaks can be attributed only to Zn, O, and spectra compared to those of the SLs, and all of the peaks can be attributed only to Zn, O, and C C elements. Zn 2p doublet has peaks at 1021.3 and 1044.4 eV with spin orbital splitting of 23.1 eV elements. Zn 2p doublet has peaks at 1021.3 and 1044.4 eV with spin orbital splitting of 23.1 eV for for Zn 2p1/3 and Zn 2p1/2 . From the deconvolution analysis of O1s peaks, as shown in Figure 7c–e, R R of O1s peaks, as shown in Figure 7c–e, the Zn 2p1/3 and Zn 2p1/2. From the deconvolution analysis the O/Zn atomic ratio estimated by calculating Oa / Zn was 0.72 for the NR crystals grown on the O/Zn atomic ratio estimated by calculating ∫Oa/∫Zn was 0.72 for the NR crystals grown on the vacuum-annealed SLs, and this value is 9~11% higher than those of the NRs grown on the SLs in vacuum-annealed SLs, and this value is 9~11% higher than those of the NRs grown on the SLs in different ambients as shown in the Table 3. We also observed significant reduction of O percentage different ambients as shown in the Table 3. We also observed significant reduction of Occ percentage in the NR crystals when grown on the SLs annealed in vacuum. From these XPS results, it can be in the NR crystals when grown on the SLs annealed in vacuum. From these XPS results, it can be concluded that the reason why the orientation of the ZnO NR crystal is strongly enhanced by the SL concluded that the reason why the orientation of the ZnO NR crystal is strongly enhanced by the SL vacuum-annealing is most likely based on the following phenomena. Vacuum annealing can remove vacuum-annealing is most likely based on the following phenomena. Vacuum annealing can remove very effectively the loosely bound oxygen species such as absorbed H O, O , and other oxygenated very effectively the loosely bound oxygen species such as absorbed H22O, O22, and other oxygenated impurities present on the surface or grain boundaries of the SL crystals formed during the sol-gel impurities present on the surface or grain boundaries of the SL crystals formed during the sol-gel based growth. Therefore, through more vigorous grain boundary diffusion process, the SL crystals can based growth. Therefore, through more vigorous grain boundary diffusion process, the SL crystals grow by merging into larger grains, and at the same time, they can have a more stable (002) preferential can grow by merging into larger grains, and at the same time, they can have a more stable (002) orientation in the annealing process. When the NR crystals grow atop the SLs of high crystalline preferential orientation in the annealing process. When the NR crystals grow atop the SLs of high quality with minimized impurities and excellent preferred orientation, they tend to preferentially grow crystalline quality with minimized impurities and excellent preferred orientation, they tend to perpendicularly along the c-axis direction, which is the growth process of the lowest activation energy, preferentially grow perpendicularly along the c-axis direction, which is the growth process of the thereby producing the high crystalline quality of nanostructures. lowest activation energy, thereby producing the high crystalline quality of nanostructures.

(a)

(b) Figure 7. Cont.

Nanomaterials 2018, 8, 68

12 of 17

Nanomaterials 2018, 8, x FOR PEER REVIEW

12 of 17

(c)

(d)

(e) Figure 7. (a) Wide scan XPS spectra of ZnO NRs grown on SLs annealed in three different ambients; Figure 7. (a) Wide scan XPS spectra of ZnO NRs grown on SLs annealed in three different ambients; (b) High resolution spectra of Zn-2p of the NRs; O1s core level spectra obtained from the ZnO NRs (b) High resolution spectra of Zn-2p of the NRs; O1s core level spectra obtained from the ZnO NRs grown on SLs annealed in (c) vacuum; (d) air, and (e) nitrogen ambient are de-convoluted into three grown on SLs annealed in (c) vacuum; (d) air, and (e) nitrogen ambient are de-convoluted into three distant distant satellite satellite peaks. peaks.

Figure 8a,b present the RT PL spesctra for the ZnO SLs annealed in different ambients and the Figure 8a,b present the RT PL spesctra for the ZnO SLs annealed in different ambients and the NR crystals grown afterwards, and two prominent emission regions were observed in all spectra. The NR crystals grown afterwards, and two prominent emission regions were observed in all spectra. first strong ultra violet (UV) emission in a range of 360–400 nm is due to a near-band-edge emission The first strong ultra violet (UV) emission in a range of 360–400 nm is due to a near-band-edge emission (NBE) associated with band-to-band excitonic recombination in ZnO, where the highest peak position (NBE) associated with band-to-band excitonic recombination in ZnO, where the highest peak position approximates the band gap of ZnO [4,5,15]. The UV peak position of the SL crystals annealed in approximates the band gap of ZnO [4,5,15]. The UV peak position of the SL crystals annealed in vacuum is centered at 374.3 nm (3.312 eV), while the crystals annealed in air and N2 show the peaks vacuum is centered at 374.3 nm (3.312 eV), the(8crystals annealed in N air2 annealings and N2 show the at at 375.2 nm (3.304 eV). A slight red shift ofwhile 0.9 nm meV) due to air or can bepeaks related 375.2 (3.304 stress eV). Acaused slight red shift of 0.9imperfections nm (8 meV) due to air or N canconsequent be related 2 annealings to thenm residual by structural in the ZnO crystals and the to the residual stress caused by structural imperfections in the ZnO crystals and the consequent bandgap modulation [15]. The intensity of NBE peak is related to the crystallinity quality of the ZnO bandgap modulation The intensity is related to the crystallinity qualityexhibit of the SLs [4,14,15,24]. The [15]. reason behind thisofisNBE that peak the vacuum-annealed seed crystallites ZnO SLs [4,14,15,24]. The reason behind this is that the vacuum-annealed seed crystallites exhibit reduction of shallow defects near the band edge and the non-radiative defects are passivated due to reduction of shallow bandannealing edge andinthe defectscan are distort passivated due to annealing effect [25]. defects On the near otherthe hand, N2non-radiative or air environment the crystal annealing effect [25]. Ondefective the othersurface hand, annealing in the N2 impurities or air environment can distort crystal lattice by creating more states when are adsorbed on the the surface or lattice by creating more defective surface states when the impurities are adsorbed on the surface or grain boundaries of ZnO crystallites [14,26]. The vacuum-annealed SLs showed 33~44% greater UV grainintensity boundaries ofthe ZnO crystallites [14,26]. The SLsresult showed UV peak than samples annealed in air orvacuum-annealed N2 ambient, and this was33~44% in good greater agreement peak intensity than the samples annealed in air or N ambient, and this result was in good agreement 2 with our earlier XRD analysis. with our earlier XRD analysis.

Nanomaterials 2018, 8, 68 Nanomaterials 2018, 8, x FOR PEER REVIEW

13 of 17 13 of 17

(a)

(b)

(c)

(d)

(e)

(f)

Figure ZnO SLs annealed in in three different ambients, andand (b) (b) the the NRsNRs grown on Figure8.8.PL PLspectra spectrafor for(a) (a)the the ZnO SLs annealed three different ambients, grown each different SL; De-convoluted visible emission spectra of theofSLs in (c)in vacuum; (d) air, on each different SL; De-convoluted visible emission spectra theannealed SLs annealed (c) vacuum; (d)and air, (e) nitrogen ambients. Relative intensities of each de-convoluted satellite peak are shown in (f). and (e) nitrogen ambients. Relative intensities of each de-convoluted satellite peak are shown in (f).

Visible light emission ranging from 500 to 700 nm is the second prominent peak as observed in Visible light emission ranging from 500 to 700 nm is the second prominent peak as observed in all PL spectra. This emission is known to be originated by deep-level emissions (DLE) caused by the all PL spectra. This emission is known to be originated by deep-level emissions (DLE) caused by the recombination of photo-generated charges with various kinds of intrinsic crystal defects such as recombination of photo-generated charges with various kinds of intrinsic crystal defects such as neutral neutral or charged O/Zn vacancies, interstitials and anti-sites. The exact nature of each defect state or charged O/Zn vacancies, interstitials and anti-sites. The exact nature of each defect state associated associated with the location of DLE is still a debatable topic; however, the DLE in sol-gel method exists in general due to various kinds of structural defects which are practically difficult to eradicate [4,14,15,24–27]. As shown in Figure 8a–c, Gaussian-deconvolution analysis was performed on the

Nanomaterials 2018, 8, 68

14 of 17

with the location of DLE is still a debatable topic; however, the DLE in sol-gel method exists in general due to various kinds of structural defects which are practically difficult to eradicate [4,14,15,24–27]. As shown in Figure 8a–c, Gaussian-deconvolution analysis was performed on the visible emissions of our PL spectra in order to understand the origin of the DLE. For example, as depicted in Figure 8a, the SL crystals annealed in vacuum exhibit three kinds of DLE, centered at 2.25 eV (green emission), 2.14 eV (yellow emission), and 1.94 eV (red emission). The green emissions is known to be associated with oxygen interstitial (Oint ), oxygen anti-site (Ozn ), and zinc vacancies [25–28]. Green emission originates from the recombination of electrons from conduction band with holes which are trapped in the deep levels. The yellow emissions are formed in aqueous solution method due to the presence of hydroxyl group. This type of emission is present due to both extrinsic and intrinsic defects. The extrinsic defects like Li or presence of –OH group has been reported, however the intrinsic defect which is the probable cause for yellow emissions is Oint [29]. The red emissions are still unknown, and however they can be connected to the transitions related to Zn vacancy complexes. Oxygen and Zn interstitial are also reported to be the primary origins of red emissions [15,26,29]. By changing the annealing environment from air or N2 to vacuum, the green emissions were suppressed by 23% or 63% as shown in Figure 8f; therefore, it can be reasonably assumed that many intrinsic defects related to O/Zn vacancies, interstitial, and anti-sites have been reduced in the SL crystals. More significant change was observed in yellow emission, which showed intensity reduction of 39–70% from the vacuum-annealed SLs compared to the emissions in the cases of air and N2 annealings. As confirmed by XPS analysis, the contribution of Oc percentage decreased most significantly in the case of vacuum annealing, which is closely related to the change in the hydroxyl group in the ZnO crystal. The same trend is observed in the yellow emission which is also closely associated with the hydroxyl group most probably formed during our aqueous solution-based growth process. Shown in Figure 8b is the PL spectra for the ZnO NRs grown on the SLs annealed under different ambient conditions. In this spectra, more intense UV emissions are observed than from the SLs, which is due to the high crystalline quality of NR crystals, whereas the intensities of UV and visible emission follow the same trend according to the SL annealing ambient conditions. The NR crystals grown on vacuum-annealed SLs produced ~3 times higher emission intensity in UV range than the other samples, which shows a good agreement with our XRD and XPS analysis. Moreover, broadband visible emission from the NR crystals was also effectively suppressed in the case of vacuum annealing for the SLs. Therefore, we can expect greatly reduced intrinsic defects in this NR crystalline state. The seed crystal quality of enhanced c-axis preferential orientation, when annealed in vacuum, can suppress the density of intrinsic defects because most of defect formation in ZnO NRs is deeply related with the distortion of crystalline structure and the interface with underlying seed-crystal grain boundaries. The FT-IR characterization for the SLs was carried out in a wave number range of 400–4000 cm−1 as shown in Figure 9. A strong absorption in the range 960–1080 cm−1 is associated with the stretching vibrations of C–O due zinc acetate dehydrate source [30], and the absorption spectra from 650 to 800 cm−1 also corresponds to acetate anion (–COO− ) twisting and scissoring mode [31]. All of these absorptions from the oxygenated carbons are supposed to be caused by our solution-based SL growth, however they were significantly suppressed in the case of vacuum-annealed SLs [30,31]. A weak absorption from the –CO–H symmetric stretching mode of vibration at ~1230 cm−1 was also suppressed in vacuum-annealed SL samples. A broad peak at 3200~3600 cm−1 is due to the OH stretch [30,31] and it was shown that this absorption was effectively reduced from the SL crystals when annealed in vacuum or N2 ambients of minimized humidity. All the results of this FT-IR analysis are in good agreement with our PL or XPS results, and it can be conclude that the annealing in vacuum is the most effective ambient condition for decontaminating OH and oxygenated carbons in the SL crystals.

Nanomaterials 2018, 8, 68 Nanomaterials 2018, 8, x FOR PEER REVIEW

15 of 17 15 of 17

Figure Figure9.9.Fourier Fouriertransform transforminfrared infrared(FT-IR) (FT-IR)spectra spectraofofZnO ZnOSLs SLsannealed annealedininvacuum, vacuum,air, air,and andNN2 2 ambients. ambients.Inset Insetshows showsananenlarged enlargedview viewofofthe thespectra spectrainina awave wavenumber numberrange rangeofof650–850, 650–850,800–1250, 800–1250, −1− and . 1. and3250–3650 3250–3650cm cm

4.4.Conclusions Conclusions Structural Structuraland andoptical opticalproperties propertiesofof the the ZnO ZnO NRs NRs grown grown by by hydrothermal hydrothermal method method were were investigated under different post-annealing ambients for the SLs. Vacuum annealing was investigated under different post-annealing ambients for the SLs. Vacuum annealing wasthe themost most effective effectiveininproducing producing(002) (002)preferred preferredorientation orientationand anddefect defectsuppression suppressionininSL SLcrystals crystalsamong amongall all annealing conditions. This was verified by XRD analysis by showing a much higher (002) degree of annealing conditions. This was verified by XRD analysis by showing a much higher (002) degree of orientation orientation(~0.98) (~0.98)and andnarrower narrower(002) (002)FWHM FWHM(~0.156°) (~0.156◦ )from fromthe theSL SLcrystals crystalsannealed annealedininvacuum vacuumthan than those annealed in different ambients. The greatest contribution of deconvoluted O a (63.5%) from O1s those annealed in different ambients. The greatest contribution of deconvoluted Oa (63.5%) from O1s XPS XPSpeak peakwas wasalso alsoobserved observedfrom fromthe theSL SLcrystals crystalsannealed annealedininvacuum vacuumamong amongall allsamples samplesannealed annealed under underdifferent differentambient ambientconditions conditionsused usedininthis thisstudy. study.This Thiscan canbe beachieved achievedby byeffectively effectivelyremoving removing the thehydroxyl hydroxylororoxygenated oxygenatedcarbon carbongroups groupsintroduced introducedby bythe theaqueous aqueoussolution-based solution-basedgrowth growthprocess process during the vacuum annealing and by blocking any further impurities introduced from other during the vacuum annealing and by blocking any further impurities introduced from other annealing annealing atmospheres, as investigated FT-IR spectroscopy and deconvolution of our XPS atmospheres, as investigated by FT-IRby spectroscopy and deconvolution analysisanalysis of our XPS and PL. and PL. ZnO NR crystals grown on the SLs of high crystalline quality and (002) preferred orientation, ZnO NR crystals grown on the SLs of high crystalline quality and (002) preferred orientation, when when annealed in vacuum, also showed excellent (002) orientation lowofdensity of annealed in vacuum, also showed excellent degree degree of (002)of orientation with lowwith density structural structural and optical defects. As shown in our XRD analysis, 2.5–7.6 times enhanced intensity of and optical defects. As shown in our XRD analysis, 2.5–7.6 times enhanced intensity of (002) peak (002) peak was obtained from the NR crystals grown on vacuum annealed SLs compared to those was obtained from the NR crystals grown on vacuum annealed SLs compared to those grown on grown on SLs annealed in2 air or N2 ambient annealing. NR crystals grown on vacuum-annealed SLs annealed in air or N ambient annealing. The NR The crystals grown on vacuum-annealed SLs also SLs also produced ~3 times higher PL emission intensity in UV range than the other samples. produced ~3 times higher PL emission intensity in UV range than the other samples. Research effort Research effort on the fabrication of high sensitivity and gas sensors is currently under the on the fabrication of high sensitivity ultraviolet andultraviolet gas sensors is currently under the work by using work by usingmethod the synthesis method of high crystalline quality ZnO NRs examined the synthesis of high crystalline quality ZnO NRs examined in this study. in this study. Acknowledgments: This work (project no. 2016006533) was supported by Mid-career Researcher Program Acknowledgments: This work (project no. 2016006533) was supported by Mid-career Researcher Program through National Research Foundation grant funded by the Ministry of Education, Science and Technology, through National Research Foundation grant funded by the Ministry of Education, Science and Technology, Korea. Korea. Author Contributions: The presented work was carried out in the collaboration of all authors which was Author Contributions: The presented work planned was carried out in the while collaboration of Khan, all authors which was supervised by Sam-Dong Kim. Waqar Khan the experiment Fasihullah Hafiz Muhammad Salman Ajmal, Noor Ul Huda, and Ji Hyun Kim in performing the experiment, analysis, supervised by Sam-Dong Kim. Waqar Khan planned thehelped experiment while Fasihullah Khan, Hafiz data Muhammad and interpretation. Khanand wrote manuscript which edited bythe Sam-Dong Kim.data analysis, and Salman Ajmal, NoorWaqar Ul Huda, Ji the Hyun Kim helped in was performing experiment, interpretation. Waqar Khan wrote the manuscript which was edited by Sam-Dong Kim. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest.

Nanomaterials 2018, 8, 68

16 of 17

References 1.

2.

3.

4. 5. 6.

7. 8. 9. 10. 11.

12.

13. 14. 15. 16.

17. 18.

19. 20.

Fortunato, E.; Gonçalves, A.; Pimentel, A.; Barquinha, P.; Gonçalves, G.; Pereira, L.; Ferreira, I.; Martins, R. Zinc oxide, a multifunctional material: From material to device applications. Appl. Phys. A 2009, 96, 197–205. [CrossRef] Fan, J.; Hao, Y.; Munuera, C.; García-Hernández, M.; Güell, F.; Johansson, E.M.J.; Boschloo, G.; Hagfeldt, A.; Cabot, A. Influence of the annealing atmosphere on the performance of ZnO nanowire dye-sensitized solar cells. J. Phys. Chem. C 2013, 117, 16349–16356. [CrossRef] Lupan, O.; Chai, G.; Chow, L.; Emelchenko, G.A.; Heinrich, H.; Ursaki, V.V.; Gruzintsev, A.N.; Tiginyanu, I.M.; Redkin, A.N. Ultraviolet photoconductive sensor based on single ZnO nanowire. Phys. Status Solidi 2010, 207, 1735–1740. [CrossRef] Khan, W.; Kim, S.D. Ultra-violet photo-response characteristics of p-Si/i-SiO2 /n-ZnO heterojunctions based on hydrothermal ZnO nanorods. Mater. Sci. Semicond. Process. 2017, 66, 232–240. [CrossRef] Dogar, S.; Khan, W.; Kim, S.D. Ultraviolet photoresponse of ZnO nanostructured AlGaN/GaN HEMTs. Mater. Sci. Semicond. Process. 2016, 44, 71–77. [CrossRef] Qiu, J.; Li, X.; He, W.; Park, S.-J.; Kim, H.-K.; Hwang, Y.-H.; Lee, J.-H.; Kim, Y.-D. The growth mechanism and optical properties of ultralong ZnO nanorod arrays with a high aspect ratio by a preheating hydrothermal method. Nanotechnology 2009, 20, 155603. [CrossRef] [PubMed] Vayssieres, L.; Keis, K.; Lindquist, S.-E.; Hagfeldt, A. Purpose-Built Anisotropic Metal Oxide Material: 3D Highly Oriented Microrod Array of ZnO. J. Phys. Chem. B 2001, 105, 3350–3352. [CrossRef] Park, W.I.; Yi, G.C.; Kim, M.; Pennycook, S.J. ZnO nanoneedles grown vertically on Si substrates by non-catalytic vapor-phase epitaxy. Adv. Mater. 2002, 14, 1841–1843. [CrossRef] Jin, B.J.; Im, S.; Lee, S.Y. Violet and UV luminescence emitted from ZnO thin films grown on sapphire by pulsed laser deposition. Thin Solid Films 2000, 366, 107–110. [CrossRef] Dedova, T.; Krunks, M.; Grossberg, M.; Volobujeva, O.; Oja Acik, I. A novel deposition method to grow ZnO nanorods: Spray pyrolysis. Superlattices Microstruct. 2007, 42, 444–450. [CrossRef] Heo, Y.W.; Varadarajan, V.; Kaufman, M.; Kim, K.; Norton, D.P.; Ren, F.; Fleming, P.H. Site-specific growth of Zno nanorods using catalysis-driven molecular-beam epitaxy. Appl. Phys. Lett. 2002, 81, 3046–3048. [CrossRef] Yoon, Y.-C.; Park, K.-S.; Kim, S.-D. Effects of low preheating temperature for ZnO seed layer deposited by sol–gel spin coating on the structural properties of hydrothermal ZnO nanorods. Thin Solid Films 2015, 597, 125–130. [CrossRef] Vayssieres, L. Growth of arrayed nanorods and nanowires of ZnO from aqueous solutions. Adv. Mater. 2003, 15, 464–466. [CrossRef] Sahu, D.R.; Huang, J.L. The properties of ZnO/Cu/ZnO multilayer films before and after annealing in the different atmosphere. Thin Solid Films 2007, 516, 208–211. [CrossRef] Kaur, N.; Sharma, S.K.; Young Kim, D. Stress relaxation and transitions in optical bandgap of yttrium doped zinc oxide (YZO) thin films. Curr. Appl. Phys. 2016, 16, 231–239. [CrossRef] György, E.; Logofatu, C.; Pérez del Pino, Á.; Datcu, A.; Pascu, O.; Ivan, R. Enhanced UV- and visible-light driven photocatalytic performances and recycling properties of graphene oxide/ZnO hybrid layers. Ceram. Int. 2018, 44, 1826–1835. [CrossRef] Singh, P.; Kumar, A.; Deepak; Kaur, D. Growth and characterization of ZnO nanocrystalline thin films and nanopowder via low-cost ultrasonic spray pyrolysis. J. Cryst. Growth 2007, 306, 303–310. [CrossRef] Chen, X.L.; Geng, X.H.; Xue, J.M.; Zhang, D.K.; Hou, G.F.; Zhao, Y. Temperature-dependent growth of zinc oxide thin films grown by metal organic chemical vapor deposition. J. Cryst. Growth 2006, 296, 43–50. [CrossRef] Lee, J.B.; Kwak, S.H.; Kim, H.J. Effects of surface roughness of substrates on the c-axis preferred orientation of ZnO films deposited by r.f. magnetron sputtering. Thin Solid Films 2003, 423, 262–266. [CrossRef] Liu, Q.; Gong, M.; Cook, B.; Ewing, D.; Casper, M.; Stramel, A.; Wu, J. Fused Nanojunctions of Electron-Depleted ZnO Nanoparticles for Extraordinary Performance in Ultraviolet Detection. Adv. Mater. Interfaces 2017, 4, 1601064. [CrossRef]

Nanomaterials 2018, 8, 68

21.

22.

23.

24. 25. 26.

27.

28.

29.

30. 31.

17 of 17

Lv, J.; Zhu, Q.; Zeng, Z.; Zhang, M.; Yang, J.; Zhao, M.; Wang, W.; Cheng, Y.; He, G.; Sun, Z. Enhanced photocurrent and photocatalytic properties of porous ZnO thin film by Ag nanoparticles. J. Phys. Chem. Solids 2017, 111, 104–109. [CrossRef] Chandraiahgari, C.R.; De Bellis, G.; Ballirano, P.; Balijepalli, S.K.; Kaciulis, S.; Caneve, L.; Sarto, F.; Sarto, M.S. Synthesis and characterization of ZnO nanorods with a narrow size distribution. RSC Adv. 2015, 5, 49861–49870. [CrossRef] Tam, K.H.; Cheung, C.K.; Leung, Y.H.; Djuriši´c, A.B.; Ling, C.C.; Beling, C.D.; Fung, S.; Kwok, W.M.; Chan, W.K.; Phillips, D.L.; et al. Defects in ZnO nanorods prepared by a hydrothermal method. J. Phys. Chem. B 2006, 110, 20865–20871. [CrossRef] [PubMed] Shi, S.; Dang, Z.M.; Xu, J.; Li, L.; Hu, G.H. Surface-related emissions and ferromagnetism in undoped ZnO nanorods. Superlattices Microstruct. 2013, 64, 375–387. [CrossRef] Singh, T.; Lehnen, T.; Leuning, T.; Sahu, D.; Mathur, S. Thickness dependence of optoelectronic properties in ALD grown ZnO thin films. Appl. Surf. Sci. 2014, 289, 27–32. [CrossRef] Kegel, J.; Laffir, F.; Povey, I.M.; Pemble, M.E. Defect-promoted photo-electrochemical performance enhancement of orange-luminescent ZnO nanorod-arrays. Phys. Chem. Chem. Phys. 2017, 19, 12255–12268. [CrossRef] [PubMed] Pimentel, A.; Nunes, D.; Duarte, P.; Rodrigues, J.; Costa, F.M.; Monteiro, T.; Martins, R.; Fortunato, E. Synthesis of long ZnO nanorods under microwave irradiation or conventional heating. J. Phys. Chem. C 2014, 118, 14629–14639. [CrossRef] Willander, M.; Nur, O.; Sadaf, J.R.; Qadir, M.I.; Zaman, S.; Zainelabdin, A.; Bano, N.; Hussain, I. Luminescence from Zinc Oxide Nanostructures and Polymers and their Hybrid Devices. Materials 2010, 3, 2643–2667. [CrossRef] Li, Y.R.; Wan, C.Y.; Chang, C.T.; Huang, Y.C.; Tsai, W.L.; Chou, C.H.; Wang, K.Y.; Cheng, H.C. Annealing effect on the photoluminescence characteristics of ZnO-nanowires and the improved optoelectronic characteristics of p-NiO/n-ZnO nanowire UV detectors. Jpn. J. Appl. Phys. 2015, 54. [CrossRef] Hosseini Vajargah, P.; Abdizadeh, H.; Ebrahimifard, R.; Golobostanfard, M.R. Sol-gel derived ZnO thin films: Effect of amino-additives. Appl. Surf. Sci. 2013, 285, 732–743. [CrossRef] Phoohinkong, W.; Foophow, T.; Pecharapa, W. Synthesis and characterization of copper zinc oxide nanoparticles obtained via metathesis process. Adv. Nat. Sci. Nanosci. Nanotechnol. 2017, 8, 35003. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).