Recently, due to the excitation of unique electronic and optical properties of these nanostructure, GaAs/AlGaAs core/shell NWs have been studied in a broad range of optoelectronic devices such as light-emitting diodes,^[1] laser diodes,^[2,3] solar cells,^[4,5] and photodetectors^[6,7] working towards the energy conservation and environment protection. One important point is that as an alternative material to silicon, GaAs offers the possibility of higher radiation resistance^[8] and a better overall performance.^[9] However, despite these advances, there is still an uncertainty over the role of point defects and/or defect clusters that may be present at exceedingly high density, which could destroy semiconductor devices in the radiation environment, such as space and atom energy field. The materials used in the relevant units above must withstand long-term use in strong radiation fields. It is well known that nano-material and nano-devices, such as quantum dot^[10] and quantum well,^[11] present radiation hardness intrinsically.^[12–15] Although the proton irradiation-induced intermixing effect in GaAs-AlGAs quantum well has been investigated^[16–18] and ultra-thin GaAs solar cell with intrinsic radiation tolerance has been demonstrated,^[19] these results have not been comprehensively explained due to the lack of systemized theory and data study. In terms of ion irradiation on NWs, currently the research mainly focuses on the doping profile of NWs,^[20,21] or bending and alignment for shaping NWs.^[21,22] Therefore, it is significant to develop the radiation effect theory and conduct experiments on NWs to understand the radiation effect on NWs in this corresponding area.

There are two categories of defects created by radiation, and they have distinctively different characteristics.^[23] The first category, called point defects, includes at most a few atoms aggregated together to form a relatively stable and simple defect. The second category, called a defect cluster, involves a few hundred atoms in a large disordered region. A displacement damage effect occurs when displacement defects are created in the crystal lattice, thereby introducing additional states in the forbidden band. This displacement damage to optoelectronics devices in the Van Allen belt by electrons and protons tends to be of the simple defect variety.^[23] Apparently in this case, the transmission electron microscope (TEM) technique is unable to provide enough resolution to directly observe the point defect in the crystal. Other techniques such as temperature-dependent Hall-effect (TDH), deep-level transient spectroscopy (DLTS), positron annihilation spectroscopy (PAS), and photoluminescence (PL) spectroscopy, have been employed to characterize single crystal and their defect states. While the first three techniques are used for studying materials from different viewpoints, they are the known well-established tool for bulk solids and thin films.^[24,25] These methods show several limitations and technical challenges, because of the formation of Ohmic contact for instance. Besides, the displacement lattice damage in semiconductor material can have a significant influence on their electrical or optoelectronic properties, through creating more than one deep-level defect in the bandgap. Consequently, the important material parameters such as free carrier mobility and density, resistivity and generation and recombination lifetime will be affected by displacement damage. The relevant study on single NW is likely to be an extreme challenge, such as the stability of fabricating a single NW device. On the other hand, micro-photoluminescence spectroscopy (μ-PL) could largely overcome these difficulties. In addition, it can be applied to the case where there is a low defect concentration in single NW, provided that competitive non-radiative carrier recombination is not the overwhelming optical transition. Our previous study has demonstrated that the NWs with smaller diameter, which has a relatively high surface-to-volume ratio, could have greater radiation tolerance.^[26] This work highlights the potential of the photoluminescence technique to estimate the irradiation-induced defect concentration by measuring minority carrier lifetime of NWs.

All the samples and equipment used in this study were provided by the Semiconductor Optoelectronics and Nanotechnology Group at the Department of Electronic Material Engineering, Australian National University. The GaAs/AlGaAs core-shell NWs grown via the metal-seeded vapor–liquid–solid (VLS) growth method in a metal–organic vapor phase epitaxy (MOVPE) system. The growth of the NWs, structural characterization, and PL characterization methods have been described in earlier publications by Joyce et al. ^[27–29] Trimethylgallium (TMGa), trimethylaluminum (TMAl), and arsine were used as the precursors for Ga, Al, and As, respectively. Gold particle size ∼ 50 nm was used and GaAs core NWs were grown using the two-temperature procedure.^[30] An AlGaAs shell was then grown around the GaAs core at 750 °C with an Al mole fraction of 0.5 (ratio of TMAl/(TMAl+TMGa) ) in the vapor phase. The SEM image of the grown GaAs/AlGaAs NWs is shown in Fig. 1(a). After growth, the NWs were randomly dispersed (Fig. 1(b)) onto silicon dioxide on silicon (SiO -on-Si) substrates for PL and time-resolved photoluminescence (TRPL) measurements. The PL measurements were performed in conjunction with TRPL spectroscopy: the PLs at the peak emission wavelengths before and after irradiation were measured. A monochromator with a cooled charge coupled device (CCD) (for spectral information) and a single photon counting system (for lifetime data) with an instrument response of ∼ 50 ps at full width at half maximum (FWHM) and 20.8-MHz repetition rate pulsed laser were used for measuring PL and TRPL, respectively. The cluster/ensemble of NWs were excited by a 522-nm solid state pulsed laser with a ∼ 100-μm diameter focal spot through a objective lens at an average power of ∼ 3.57 W/cm . More importantly, the measuring platform was mounted in a three-dimensional (3D) stepping motor system with positioning resolution ∼ 200 nm, which offers a relatively precise position control. It should be pointed out that the measuring uncertainty of PL intensity was about %, estimated through multiple measurements.

Fig. 1.

	Figure Option ViewDownloadNew Window
	Fig. 1. (color online) (a) Image of the GaAs/AlGaAs core–shell NW with core diameter ∼ 50 nm, and AlGaAs shell thickness ∼ 30 nm (Scale bar ∼ 1 μm). (b) GaAs NWs dispersed as a cluster on SiO -on-Si substrate for irradiation. The inset shows the NWs shinning by laser spot.

All the samples were irradiated by using 1-MeV H ions at room temperature on a Pelletron® tandem accelerator. To be more specific, further details can be found in Ref. [26]. It should be pointed out that all the samples were not annealed after irradiation. The ion fluences from cm to cm , corresponding to total integrated doses of 570 krad and 17100 krad (Si), respectively.^[31]

The plots of the PL intensity as a function of wavelength at different irradiation fluences for a typical cluster of GaAs/AlGaAs core/shell NWs are shown in Fig. 2, which manifests that with increasing fluence the NWs exhibit a decrease of PL intensity. The plots of room-temperature TRPL decay spectra versus irradiation fluence are displayed in Fig. 3, which demonstrates that the TRPL spectra decay faster with fluence increasing. The plots of minority carrier lifetime versus irradiation fluence ( cm are also presented in the inset of Fig. 3. The corresponding minority carrier lifetime is extracted from fitting TRPL spectrum data by mono-exponential decay equation: , where τ is the minority carrier lifetime, and are the PL intensities at the moments of 0 and t, respectively.

Fig. 2.

	Figure Option ViewDownloadNew Window
	Fig. 2. (color online) Plots of room-temperature PL spectra of GaAs/AlGaAs NWs irradiated at fluences ranging from cm to cm , excited by 522-nm laser. Data for the as-grown samples are also included for reference.

Fig. 3.

	Figure Option ViewDownloadNew Window
	Fig. 3. (color online) Plots of typical photoluminescence decay spectra at room temperature from NWs cluster with irradiation fluences ranging from cm to cm . Data for the as-grown samples are also included for reference. The inset shows the variations of measured minority carrier lifetime with irradiated fluence and fitting results.

Considering the fact that the minority carrier lifetime is the reciprocal of the recombination rate before irradiation, it can be written as . Accordingly, after irradiation the minority carrier lifetime in NW can be given by^[23] 1 where R is the post-irradiated recombination rate of the minority carrier in NW, k is the radiation damage constant, and is the proton fluence. According to Eq. (1), the data and fitting are plotted and shown in the inset of Fig. 3 (where ns).

Then the corresponding carrier lifetime damage coefficient can be defined as^[32–36] 2 where and (here ) are the post and pre-irradiated lifetime, respectively. The degradation of carrier lifetime τ is the utmost concern on the semiconductor device operation against irradiation.^[37]

To directly see the reduction of minority carrier lifetime and degradation of the PL intensity versus radiation fluence, here we define the normalized carrier lifetime as 3 and normalized PL intensity as 4 and are the post and pre-irradiated PL peak intensity, respectively. The normalized PL intensity and carrier lifetime calculated from Eqs. (3) and (4) are plotted in the inset of Fig. 4, revealing that both of them have the same trend as a function of irradiation fluence.

Fig. 4.

	Figure Option ViewDownloadNew Window
	Fig. 4. (color online) Variations of normalized PL intensity and normalized lifetime of GaAs NWs with core diameter of 50 nm with irradiation fluence mainly ranging from p/cm to p/cm . Both fitting data are also included.

Under low level injection conditions, the dependence of PL intensity on external injection and minority lifetime in NWs can be expressed as:^[26] 5 where is the minority carrier spontaneous radiation rate and U is the external injection. Apparently, equation (5) shows that the normalized PL intensity and normalized lifetime versus fluence have similar trends. Hereby, the linear relationship between carrier lifetime and PL intensity in NWs could be confirmed. Furthermore, for comparing with the normalized carrier lifetime, the PL damage coefficient can be defined and the corresponding normalized PL intensity Eq. (4) can be expressed as 6 According to Eqs. (6) and (3), the measured normalized PL intensity and carrier lifetime could be fitted and the results are shown in Fig. 4, respectively. The degradation of PL intensity and carrier lifetime reflect both decrease dramatically once the irradiation fluence cm , and PL intensity decreases faster than carrier lifetime.

Carrier lifetime is extremely important from a device-physics standpoint, because it governs the operation of semiconductor p–n junctions, one of the basic device structures, to a large extent.^[32] In contrast to the bulk recombination lifetime, surface recombination is more sensitive to ionization damage, and only marginally to displacements.^[32] Studies have also found that the AlGaAs shell acts as a passivation layer which could eliminate the surface states for GaAs NWs core.^[29,38] One important point, however, is that the surface recombination is determined by the density of interface traps and the surface potential. Hence this study takes it into account that the surface recombination remains constant after irradiation. On the other hand, Shockley–Read–Hall (SRH) recombination mechanism is typically used to discuss the recombination of electron-hole pairs at defect levels within the bandgap.^[39] It has been usually described in the frame of the SRH theory,^[37] in which the thermal (i.e., phonon-assisted) carrier transitions to and from the bands are considered. In principle, it can be extended to taking into account the contribution of different recombination centers, radiative lifetime, and the surface recombination lifetime. Hence, there is more than one deep-level defect present, and the total SRH lifetime of the NW at the low-injection level can be described by . Therefore, the effective minority carrier lifetime can be written as 7 where and d are the surface recombination velocity and NW core diameter respectively, is the SRH lifetime for type-i defect, and is the radiative lifetime of minority carrier. In NWs, the minority carrier lifetime is dominated by the non-radiative lifetime, which is mainly determined by SRH. The reason is that the radiative lifetime is much longer than the non-radiative lifetime.^[40,41] Then equation (7) can be given as 8 where is treated as a constant principally for NWs with AlGaAs shell passivation. In addition, following the definition of SRH lifetime, one can write , where is the electron (hole) capture cross-section, is the volume concentration of deep-level defects, and is the thermal velocity of the electron (hole). Consequently, the effective minority carrier lifetime can be approximated in the following simple way^[26] 9 which renders the analysis simpler when SRH recombination predominates. Accordingly, we could define a defect concentration creation rate , thus and are the corresponding post- and pre-irradiated defect concentration and can be related by 10 which indicates that the post-irradiated defect concentration varies linearly with irradiation proton fluence. The pre-irradiated defect concentration is a certain concentration of point defects of the starting NWs. Equation (10) in combination with Eq. (9), can be extended to defining a normalized defect concentration as 11 From the above assumptions, the normalized defect concentration will be governed by both irradiation fluence and creation rate. It is also proportional to the reciprocal of normalized carrier lifetime. In a simple case, the linear relation coefficient between the normalized defect concentration and normalized carrier lifetime is 1. The fitting normalized defect concentration versus proton fluence is shown in Fig. 5. This result accords well with the previously reported result on the semi-insulating SI GaAs sample, which has shown that the defect concentration increases linearly with proton fluence.^[42]

Fig. 5.

	Figure Option ViewDownloadNew Window
	Fig. 5. (color online) Representative normalized defect concentration from Eq. (11) which is fitted through the data from measured normalized lifetime (blue dotted line). The fitting based on normalized defect concentration is directly proportional to the reciprocal of normalized lifetime on the assumption that the proportionality factor is 1 which renders the fitting simpler.

In summary, the radiation effect on GaAs/AlGaAs core-shell NWs with core diameter ∼ 50 nm is studied by proton irradiation in combination with the PL measurement technique. The model of relation between PL intensity degradation and minority carrier lifetime decline is obtained and explained under low photon injection and low proton irradiation fluence. Furthermore, the radiation effects on surface recombination and radiative recombination could be eliminated in the GaAs/AlGaAs core-shell NWs, and the irradiation-induced defect concentration can be extracted from the measurement of minority carrier lifetime. The results and their understanding achieved from this work will help further investigate irradiation-induced defects in nano-materials for space or atomic energy applications.