1. Introduction

Zinc oxide (ZnO) is an important wide-band-gap semiconducting ceramic material with many useful properties. It has been extensively investigated for wide applications in luminescence, ultraviolet (UV) light emitters or light emitting diodes (LEDs), spin functional devices, solar cells, surface acoustic coatings, microsensors and so on [1–6]. In order to induce new interesting properties doping different elements has been attempted [7–13]. Alumina doped zinc oxide (ZAO) is one of the most widely reported [14–16] transparent conducting oxide (TCO) for its high stability, low cost and non-toxicity. These substantial advantages make ZAO an important candidate for multifunctional photoelectric materials.

Though ZAO has been treated as a versatile material, serving as a LPE material has never been tried. Here we first report a large and stable LPE observed in this focal film based on ZAO/SiO₂/Si homo-heterostructure under a 532 nm laser illumination. In fact since the LPV in response to spot illumination was first discovered by Schottky [17] and later expanded by Wallmark in floating Ge p⁺-n junctions [18], different systems have been reported such as Ti/Si amorphous-superlattices [19], modulation-doped AlGaAs/GaAs heterostructure [20], hydrogenated amorphous silicon Schottky barrier structures [21], perovskite materials [22] and metal–semiconductor (MS) like or metal-oxide-semiconductor (MOS) structures [23–25]. Different physical mechanisms have also been proposed including Dember effect [26], p-n junction mechanism [27] and Schottky barrier mechanism [17].

In this report we present the relation between the LPV and laser position on ZAO film of different thickness and find that all samples thickness ranging from 20 to 200 nm can output effective LPV. We also put forward a new mechanism in terms of quasi-Fermi level that can well explain the LPE. All data and analyses demonstrate ZAO may be a candidate of priority for ZnO-related multifunctional devices.

2. Experimental details

The ZAO films (composited of 2% Al₂O₃, 98% ZnO) were deposited on n-type Si (1 1 1) substrate at room temperature by DC magnetron reactive sputtering. The substrate was covered with a native SiO₂ layer of 1.2 nm measured by transmission electron microscopy (TEM). The thickness of the Si wafers is around 0.3 mm and the resistivity is in the range of 50-80 Ωcm. The base pressure of the vacuum system prior to deposition was better than 6.0 × 10⁻⁵ Pa. High purity ZAO (>99.9%)) target (60 mm diameter) was used. An argon gas pressure of 0.68Pa was maintained during deposition. The deposition rate, determined by stylus profile meter on thick calibration samples is 1.23Å/s.

All the samples were scanned spatially with a Green Diode laser (5 mW and 532 nm) focused on a roughly 50-μm diameter spot at the ZAO film surface without any spurious illumination (e.g. background light) reaching the samples. All the contacts (less than 1 mm in diameter) to the films were formed by alloying indium and showed no measurable rectifying behavior (perfect ohmic contact). The schematic picture of the experimental set-up for the LPV measurement is shown in the inset of the second figure in this paper. The optical transmittance spectra of the ZAO film was determined by UV-Vis-NIR spectrophotometer and the wavelength ranged from 200 to 1000 nm. All measurements were taken within 24 hours after the samples taken out of vacuum environment. The distance is 4 mm between two alloying indium contacts.

3. Experimental results

Figure 1 presents the optical transmittance spectra of a 100nm thick ZAO film corrected for the attenuation of a glass substrate. The film is highly transparent in the Vis-IR region with a transmittance between 60% and 90%. It shows that ZAO film is a good transparent conducting oxide in Vis-IR region.

 

Fig. 1 Optical transmission spectra of 100 nm thick ZAO film

Download Full Size | PDF

The dependence of the LPV on the laser spot position along the y = 0 line at room temperature with different ZAO thickness is shown in Fig. 2 .The value of LPV gets largest when the illumination spot is closest to the measurement electrodes and shows a monotonic linear decrease as the spot scanned away from the contacts. The largest open-circuit position sensitivity is 41.85mV/mm for sample 7(172.2 nm) at a wider measurement range (4 mm) compared with previous studies [24,25,28. Even the smallest one gets 6.90 mV/mm for sample 2 (49.2nm). The correlation coefficients, which measure the linearity of the device output, are very close to1.000 for all samples. It indicates a perfect linearity for our samples.

 

Fig. 2 LPV as a function of the laser position (x) observed on ZAO film surface with different thickness for y = 0 line. The bottom inset displays the schematic illustration of LPV measurement.

Download Full Size | PDF

According to extensive works [20,29,30], there are three main criteria to judge whether a device is suitable for a PSD. They are the position sensitivity, nonlinearity and spatial resolution. Table 1 shows a summery result of the three main criteria on ZAO films with different thickness. As can be seen, all the samples output satisfying sensitivities and show good to excellent nonlinearities for 100μm spatial resolution. The nonlinearity of sample 3 (73.8 nm) gets as low as 3.22% for 100μm spatial resolution. Other samples are also controlled in 6.50% while the usual acceptable nonlinearity is less than 15.00%. All data and analyses demonstrate ZAO may be a candidate for PSDs.

Table 1. Results of ZAO-SiO₂-Si Structures with Different ZAO Film Thickness

View Table

Furthermore, in usual MS or MOS structures there always presents a “thickness effect” between position sensitivity and film thickness. That is an optimum film thickness for the largest position sensitivity always existing within an appropriate thickness range. When film thickness departs away from the optimum point, the LPV will decay monotonously. ZAO films shows different. The attenuation of LPV is not monotonic to film thickness apart from the optimum point instead the outputs keep effective in a wide range. However, the basic mechanism of this anomalous phenomenon is not clear now and needs a further investigation.

Besides it has been well known that ZAO film can act as an antireflection coating with stable physical properties, such as good electrical conductivity and high optical transmittance. This property will help increase stability and service life of devices. From Table 1 and Fig. 2 we find sample 3 gets the least nonlinearity and sample 7gets the largest position sensitivity. Choosing these two as typical ones, we re-measured the LPV 4 weeks later.

Figure 3 (a) is the comparison results, showing this structure retains stable in LPV output. This again proves ZAO film is quite qualified as candidate for PSD. The measurement results also confirm our latest report [31], an enhanced LPE can be observed by coating a thin oxide layer on the metal surface of MS structure.

 

Fig. 3 (a) Comparison of experimental results of sample 3 and 7, 3 and 7 are results of measurement taken within 24 hours out of vacuum environment, 3′ and 7' are results of measurement taken 4 weeks later (b) AFM images (1µm × 1µ m) of sample 3 and 7 in view of flatten and 3d.

Download Full Size | PDF

Sample 3 and 7 were also measured using atomic force microscopy (AFM) in tapping mode as shown in Fig. 3 (b). We can see clearly crystal size and surface roughness of sample 7 is larger than sample 3. In 3d view we find obviously that the crystal stacking style is quite different. Sample 3 presents individual nanorod separately while sample 7 presents a cluster of nanorods gather round. All these may contribute to different surface architecture related to film thickness. As most film properties are affected significantly by surface architecture, the unusual thickness phenomenon mentioned above may be attributed to the differences of surface architecture.

4. Physical mechanism

To explain the LPE observed in ZAO film based on ZAO/SiO₂/Si homo-heterostructure, we propose the following physical model.

Figure 4 (a) shows the energy band diagram of the ZAO/SiO2/Si system in equilibrium state existing under uniform environment (e.g. background temperature, light and so on).A barrier is formed to integrate the two Fermi levels by energy band bending. The oxide layer has a tunneling thickness(1.2nm) according to extensive works [24,30].When ZAO film is illuminated by the 532 nm laser spot, energy is mainly absorbed in Si substrate where generating electron-hole pairs. The generated electrons tunnel through SiO₂ into ZAO layer while the holes are left in Si substrate. These excess carriers generate a concentration gradient between the illuminated spot and nun-illuminated zone. Due to the concentration gradient these excess carriers move laterally away from the illuminated spot. Noticeable factor is ZAO film is quite thin (nano scale) and must be described by surface concentration. As a result, with the same number carrier injected, ZAO film concentration is much more influenced.

 

Fig. 4 (a) Schematic simple equilibrium energy-band diagram of the ZAO/SiO₂/Si homo-hererostucture, the native SiO₂ layer has a tunneling thickness (1.2 nm) (b) Schematic profile diagram of the excess carrier diffusion on ZAO surface under a spot illumination.

Download Full Size | PDF

For better investigation, a quantitative explanation is given in ideal one-dimensional model. According to the diffusion equation in the semiconductor, the distribution of the light induced electrons can be calculated as following:

Here r is the distance from the laser spot and${\lambda }_z$ is the electron diffusion length in ZAO film. ${\lambda }_z$ can be written as:

Here $D_Z=k_{0}T\sigma /n_0{q}^2$(according to Einstein relation) is the diffusion constant and ${\tau }_Z$ is the lifetime of the non-equilibrium electrons of ZAO layer separately. ${\sigma }_Z$ is the conductivity of the ZAO film and n₀ is the area density of electrons at equilibrium state.

Due to the diffusion of the excess electrons, two quasi-Fermi levels were produced at the contact electrode as shown in Fig. 4 (b). The quasi-Fermi level is related to the excess carrier density and the relationship can be written as:

Here $\Delta n$and n are the excess electron density and the equilibrium state electron density.

The LPV can be obtained by calculating the difference of the quasi-Fermi level between the two contact electrodes position A and B in Fig. 4 (b).

Substituted $\Delta n(B)$and $\Delta n(A)$ in Eq. (4) by$\Delta n(A,B)=\Delta N\exp \left[(-\left|x\pm L\right|)/{\lambda }_z\right]$, L is the half distance between A and B while x is the laser spot position shown in Fig. 4 (b) Eq. (4) can be written as following:

When satisfying the requirement of $2L\le {\lambda }_Z$ the LPV value will be proportional to the laser spot position. Besides it depends on carrier diffusion length λ (or film conductivity σ) significantly. In fact λ and σ are auto-correlation and both related to film thickness and microstructures of the materials including crystal size, orientation and so on. In doping materials the impact is more complex. So the cause of the anomalous thickness phenomenon is still hard to give an exact explanation.

5. Conclusion

In summary a large LPE effect with both large sensitivity and good linearity is first observed on ZAO nano film based on ZAO/SiO₂/Si homo-heterostructure. A new physical mechanism based on quasi-Fermi level is given. ZAO is more competent in increasing stability and service life of devices. More detailed investigation on microstructures of doping materials may stimulate researches on ZnO-related multifunctional devices.