Effects of Strong Band-Tail States on Exciton Recombination

Subscriber access provided by Kaohsiung Medical University

C: Physical Processes in Nanomaterials and Nanostructures

Effects of Strong Band-Tail States on Exciton Recombination Dynamics in Dilute Nitride GaP/GaNP Core/Shell Nanowires Shula Chen, Weimin M Chen, and Irina A. Buyanova J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b05199 • Publication Date (Web): 25 Jul 2018 Downloaded from http://pubs.acs.org on July 25, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Effects of Strong Band-tail States on Exciton Recombination Dynamics in Dilute Nitride GaP/GaNP Core/Shell Nanowires Shula Chen*, Weimin M. Chen, and Irina A. Buyanova Department of Physics, Chemistry, and Biology, Linköping University, 58183 Linköping, Sweden * Corresponding email: [email protected] ABSTRACT Exciton dynamics in dilute nitride GaP/GaNP core/shell nanowires (NWs) with pronounced band-tail states formed by nitrogen clusters is investigated using time-resolved photoluminescence (PL) spectroscopy. The emission of excitons localized at the N-related states in the GaNP shell is found to exhibit a stretched-exponential decay with the 1/e lifetime dramatically shortened with decreasing excitation wavelength and reduced shell thickness. The observed PL transient behavior is explained by markedly different exciton lifetimes between the surface and bulk regions of the GaNP shell, i.e. of ~ 20 ps vs ~ 10 ns, respectively. Despite being trapped at the deep localized N states, the photo-excited excitons are concluded to suffer from pronounced surface recombination via tunneling to the surface states within a distance of 10 nm from the surface, which results in the depth-dependent PL dynamics. The surface recombination rate is, however, lower than that previously reported for GaP, indicative of partial passivation of the surface states by nitrogen. From temperaturedependent PL measurements, characteristic thermal activation energies for the surface and bulk-related non-radiative recombination channels are deduced. The obtained results provide insight into the exciton/carrier dynamics in NW systems with strong localization or alloy disorder, which is important for future nanophotonic and photovoltaic applications of such structures.

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

INTRODUCTION Semiconductor nanowires (NWs) have been a subject of intense research as potential nano-building blocks for optoelectronic and spintronic devices with miniaturized footprints and a high power efficiency. The one-dimensional architecture of such structures provides both a superior photonic cavity for enhanced light-matter interaction and a waveguide for optical interconnection. The freedom in band engineering by designing radial or axial heterostructures allows tailoring of NW optical and electronic properties, enabling novel device functionalities that have already spurred a wealth of applications, such as NW light-emitting diodes (LEDs),1-3 lasers,4 5 sensors,6-8 and photovoltaic devices.9-12 In recent years, dilute nitride GaNP-based NWs are emerging as another material candidate for innovative device applications. A small amount, usually several percentages, of nitrogen (N) atoms incorporated into the GaP host lattice not only significantly reduces the band gap via the giant bowing effect,13 but also transforms it from an indirect character in the parental GaP to a quasidirect gap in GaNP alloys as nitrogen composition [N] exceeds 0.4 %.13-15 As a result, GaNP has been suggested as an active material in amber-red LEDs 16 and nano-sources of linearly-polarized light,17 as well as a two-photon mediated absorber for intermediate band solar cells.18 19 However, an in-depth understanding of exciton/carrier dynamics in GaNP NWs is so far lacking. Such understanding is especially important as non-equilibrium carriers in these structures are trapped at tail states formed due to clustering of N atoms even at room temperature.20

21

This makes carrier transport and recombination dynamics

distinctly different from that in conventional NWs, such as GaAs, InP, and CdS with sharper band edges and a higher carrier mobility. To clarify this important issue, in the


Page 2 of 30

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


present study we investigate the carrier recombination dynamics in GaP/GaNP core/shell NWs by employing time-resolved photoluminescence (PL) spectroscopy combined with temperature-dependent PL studies. We find that the room temperature emission from excitons trapped at N-related states exhibits a stretched exponential decay, which becomes faster with shorter excitation wavelengths or a reduced thickness of the GaNP shell. A model for exciton dynamics is proposed that takes into account the presence of surface and bulk non-radiative processes with appreciably different recombination rates. Their thermal behaviour is further analysed based on temperature-dependent PL transient measurements.

EXPERIMENTAL DETAILS The GaP/GaNP core/shell NWs were grown by gas-source molecular beam epitaxy (MBE) on (111)-oriented Si substrates. The growth details are described in detail in Ref. 22 and 23. Four sets of NW samples with the same GaP core of ~ 100 nm in diameter, but with different GaNP shell thicknesses (d) of 90 nm, 70 nm, 50 nm, and 20 nm, which are labelled as D1, D2, D3, and D4, respectively. All NW samples contain almost the same N content in the shell. Fig. 1(a) shows a representative scanning electron microscopy (SEM) image from the D1 NWs, which have clear and straight side facets confirming their good morphology. The schematic configuration of the GaP/GaNP core/shell NW structure is illustrated on the right side of Fig. 1(a). Time-resolved PL measurements were performed within the temperature (T) range of 4 – 300 K in a close-cycle cryostat. A wavelength tunable Ti: Sapphire pulsed laser was used as an excitation source, which has a repetition rate of 76 MHz and a temporal pulse width of 150 fs. The transient PL emission from the NWs was recorded by a streak camera assembled with a single-grating monochromator.



RESULTS AND DISCUSSION A representative time-integrated PL spectrum measured at room temperature (RT) from the studied NWs is shown in Fig.1b (symbols), taking as an example from the D1 structure. The spectrum was measured with the excitation wavelength, , of 400 nm. The PL emission is dominated by a broad asymmetric PL band centered at around 1.80 eV, with a long tail extending towards low energies. This PL band originates from the radiative recombination of localized excitons (LEs) trapped at the disordered N cluster states below the extended continuum,20 21 as illustrated by a random ensemble of solid lines in Fig. 1(c). The LE PL band exhibits a high-energy cut-off at around 2.06 eV, clearly defining the onset of an extended continuum and, thus, the alloy bandgap. The N content in the GaNP shell can then be derived using the bandanticrossing (BAC) model13 24 25 as being ~ 2.0 %. The exponential low energy tail of the GaNP emission, which reflects the density of states of the N clusters, can be approximated by the equation ~ (/ ),26-28 where E is the average localization potential. Fitting to the PL spectrum (shown by the dash-dotted line) yields ~ 55 meV. The derived localization energy is larger than the thermal energy of 26 meV at 300 K, thus warranting the observation of the LE PL emission at RT. The exciton dynamics at the N-related states was evaluated from temporal evolution of the PL spectra shown by the solid curves in Fig. 1b. Within 150 ps after the excitation pulse, the PL peak, marked by the dashed arrow, undergoes a sizable redshift of ~ 60 meV, indicating efficient energy transfer of the LEs to deeper N trap states. After 150 ps, the PL shift slows down and eventually stops when the LE dynamics is mainly determined by recombination lifetime at the involved N states. To further visualize the LE dynamics, the PL transients monitored at different LE energies are shown in Fig. 1(d). It is apparent that the PL decays become progressively slower from shallow to deep N


Page 4 of 30

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


states. The spectral variation of the PL decay is commonly observed in semiconductor materials with randomly fluctuating band edges, and is caused by competition between energy relaxation of excitons to deeper N-related states and recombination at localized states.31-34 We measured the effective lifetime, , which is taken at 1/e of the initial PL intensity and plot them in Fig. 2(a) as the solid dots. drastically increases from 20 ps at 2.01 eV to 2.55 ns at 1.64 eV, then saturates at lower detection energies, i.e. for the deepest localized levels. The spectral dependence of can be fitted with the equation 31-34

( ) =

(( )/ )

(1).

where is the PL lifetime of the localized excitons/carriers, denotes the mobility edge, where the exciton/carrier energy relaxation rate equals to that of recombination. The best fit to the experimental data is shown by the solid line in Fig. 2(a) and yields of 3 ns and of 1.75 eV. Since approaches the limit of below and reflects mainly the recombination process, we focus our following analysis on the PL transients integrated below 1.7 eV. From Fig.1 (d) it is noticeable that the RT PL decay in the studied NWs cannot be reproduced by a single exponential function even for the deepest N states. This is in contrast with the known dynamics of excitons bound to deep isoelectronic N-related centers in GaNP, where a single exponential decay is expected and was indeed observed experimentally.35 The most likely reason is the contribution of spectrally degenerate PL transient processes that decay with different lifetimes from spatiallyseparate regions.36 37 It is well known that bare NWs are subjected to intense surface recombination, usually yielding a single-exponential PL decay time determined by surface recombination time at the picosecond time scale.38-42 Surface passivation either



by growing an outer shell or coating with a layer of sulphide or nitride-based chemicals was employed in the past to increase carrier lifetime towards a bulk value within the nanosecond range.38-43 In our NWs with the untreated surface, surface related nonradiative recombination is expected to be rather severe, which may lead to a fast decay of excitons that are created in proximity to the surface. On the other hand, due to strong localization effects and a low mobility of the trapped LEs, their dynamics would be mainly determined by the bulk lifetime when the excitons reside within the volume region of the GaNP shell. To corroborate this model, we studied transient behaviour of the LE PL emission from the D1 NWs under the conditions where the LEs are generated closer to the surface, i.e. under optical excitation with shorter wavelengths. As is reduced, the PL decays become increasingly faster. drastically shortens from 2.55 ns under the 400 nm excitation down to 20 ps under the 266 nm excitation – see Fig. 2(b). Moreover, the exciton lifetime under the deep UV excitation becomes spectrally independent as plotted by the open circles in Fig. 2(a). This can be explained assuming that the rate of the surface-related non-radiative recombination exceeds that of the exciton relaxation between the N-related states. To evaluate the surface contribution to the PL dynamics at different laser wavelengths, we simulated distributions of the optical excitation intensity within the NW cross-section by using the Lumerical FDTD Solution Suite software. The corresponding results are given in Fig. 3(a)-(d), with the given as examples. The profiles of the excitation intensity calculated as a function of depth from the wire surface for these are plotted in Fig. 3(e). As can clearly be seen, excitons excited by the 400 nm light are generated across the whole NW, with a certain patterned distribution due to a cross-sectional cavity effect. Therefore, under these conditions the majority of the excitons are expected to recombine far from the surface, which would


Page 6 of 30

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


give rise to a long PL decay governed by the bulk lifetime. At shorter , the excitation profile becomes confined closer to the surface due to a higher absorption coefficient. This is especially apparent for = 266 nm, by which the excitation profile exhibits exponential distribution with most excitons being excited within a very thin surface layer of ~ 10 nm. Under these conditions, the detected PL emission predominantly experiences surface recombination. Based on the clarified mechanism for the LE recombination in the GaNP NWs, we can develop a model to quantitatively interpret the PL decays and to extract important recombination parameters. The model introduces two non-radiative channels from different spatial regions, i.e. surface- and bulk-related, with the corresponding lifetimes , and , respectively, as illustrated in Fig. 1(c). The bulk non-radiative lifetime in

NWs is known to be affected by the presence of impurities, as well as by structural and intrinsic point defects.44 45 Given their finite volume density, their contribution to the non-radiative recombination is usually less important in comparison with the contribution of surface recombination centres leading to ≪ . The non-radiative

centers on the NW surface can affect the carrier recombination within a certain distance from the surface due to (i) diffusion of mobile carriers or (ii) tunnelling of the localized carriers to the surface states.40 46 The second mechanism dominates in the systems with severe carrier localization and, therefore, is most relevant to our case, when the excitons are fast trapped at the N-related localized centres. The tunnelling probability depends on the extent of the exciton wavefunction at the monitored N centres and also their distance, !, from the surface. Therefore, the exciton lifetime associated with the surface recombination ( ) can be approximated by (!) =

exp (−!/&), where & is the effective tunnelling length determined by the exciton localization at the deep N-related states. The lifetime of the LE emission ('( ) at a



Page 8 of 30

specific position in the GaNP shell should be determined by combined contributions of all recombination channels via

) *+, ()

=

) *-

+

) 0 () */-

+

) 1 */-

. Here is the radiative

recombination time, which is an intrinsic material property and, therefore, is spatially independent. The detected PL transients 2'( (3) can, consequently, be calculated by performing spatial integration over the entire GaNP shell thickness (4) as 2'( (3) ∝ 6:→; 27 (! )exp (− *

8 ) 4! +, ()

(2).

Here 27 (! ) is the initial carrier generation profile defined by laser absorption, as already calculated in Fig. 3(e). = 100 ns can be derived from the PL transients measured at 5 K, where all non-radiative processes at the surface and in bulk are suppressed. Under the assumption ≪ , e.g. taking realistic values of = 20 ps and = 10 ns, our model predicts significant differences in the LE recombination

dynamics under different . This is illustrated in Fig. 3(f), where the exciton density is mapped versus the shell depth and the time delay after the excitation pulse for D1 NWs under of 400 nm. The corresponding results for the 266 nm excitation is given in the inset. As can be seen in the former case the exciton population close to the surface is depleted shortly after the laser pulse, while the excitons located deeper inside the shell display progressively slower dynamics. Under = 266 nm, the excitons are generated only within 10 nm from the surface and exist within a short time period of ~ 20 ps, which concurs with the experimental data shown in Fig. 2. The parameters , and & can be obtained by numerically fitting the

measured PL decays under different . The best fit to all PL transients shown in Fig. 2(b) yields a unique set of the fitting values, i.e. = 20 (±2) ps, = 10 (±1) ns, and

& = 10 (±0.5) nm. The results of the fitting are shown by the dashed lines in Fig. 2(b).


Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The surface states are expected to have an increasing impact on the PL dynamics in NWs with a reduced shell thickness due to increasing surface-to-volume ratio, which leaves a larger portion of photo-excited excitons exposed to the surface. This is directly visualized by simulating the excitation profiles under = 400 nm for the GaP/GaNP core/shell NWs with the shell thickness d varying between 90 and 20 nm, see Fig. 4(a)(d). The PL decays measured under these conditions from D1-D4 NW structures indeed accelerate with decreasing d - see Fig. 4(e). Remarkably, the measured PL decays can be fitted using Eq. (2) with the same set of the recombination parameters as that used in Fig.2(b). The results of the simulation are shown by the dashed lines in Fig. 4(e). The obtained excellent agreement between the experimental data and the simulation results further confirms the validity of the proposed model. We note that the ratio between the determined rates of the surface and volume nonradiative transitions, =>? = @>? /A , reaches ~ 103. The capture rate of excitons by defects can be expressed as />? = BCDE FD , where B is the capture cross-section, CDE is the thermal velocity of excitons and FD is the defect density. By assuming comparable values of B by the surface and bulk defects, =>? indicates three orders of magnitude higher FD at the surface than in the bulk, which is indeed reasonable for low-dimensional nanostructures favouring rich abundance of the surface states. The surface recombination velocity (SRV) can be estimated from A by integrating over the tunneling layer, G, to account for the sheet density of the surface states: G

H=I = 6

A

J =

G A

(3).

The so-obtained value, i.e. SRV ≈ 5×104 cm/s, is lower than that reported previously for GaP, where it ranges between 1×105 cm/s and 5×105 cm/s.47 48 The likely reason for this reduced SRV on the GaNP surface is a partial passivation of surface defects due to



Page 10 of 30

the presence of N which forms stronger bonding with Ga atoms and,43 to certain extent, suppresses the oxidization of NW surface and associated defects. Such improvement effect is found also in other dilute nitrides, e.g. in GaAs/GaNAs core/shell NWs36 49 and GaNAsP-based NWs.50 So far, we have quantitatively interpreted the PL dynamics at room temperature, where the surface and bulk-related non-radiative processes dominate in recombination. It is also important to know their individual temperature evolution and thermal activation energies. To single out the thermal behaviour of each non-radiative channel, we recorded the temperature dependent PL transients under = 266 nm and 400 nm. The corresponding results from the D1 NW are displayed in Fig. 5(a) by the open circles and solid lines, respectively. At 5 K, the PL decays are identical between the two and are determined by the radiative lifetime ? . As temperature rises above 150 K, the PL dynamics under = 266 nm starts to accelerate with respect to that under = 400 nm, implying a faster thermal depletion rate of LEs within the surface region. This trend also coincides with the thermal behaviour of the time-integrated PL intensity, as summarized against temperature in Fig. 5(b). In each case, the Tdependent PL intensity KLM (N) can be fitted by a two-component Arrhenius equation,

KLM (N) =

K O ( /PQ N)OR (R /PQ N)

(4).

Here K represents the PL intensity at 5 K. and R are the activation energies of two non-radiative channels, whose efficiencies are defined by the amplitudes O and OR . The best fit to the data obtained under = 400 nm (the dashed line) gives S

> (S

> ) of 26 meV (145 meV), and the OS

> /OS

> ~ 3×103 - see Table 1 R R for the summary of the deduced fitting parameters. The first thermal process with the smaller activation energy and relative intensity could be tentatively attributed to the


Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 1 Summary of Arrhenius fitting parameters from Eq. (4) (nm) 400 266

W) (meV) 26 18

X)

WY (meV)

XY

16 19

145 180

4.8×104 1.0×107

capture of hot carriers by defects during energy relaxation after the laser excitation or thermal activation and redistribution of excitons among the localized states. Independent of the exact mechanism, the PL quenching at this stage is not severe, judging from a limited value of O . The second process with a substantially higher efficiency, however, reflects the dominant bulk non-radiative recombination after carriers are thermalized at the localized N states. We denote hereafter S

> as @ for R the bulk non-radiative activation energy. For the case of = 266 nm, where only the surface region is probed, we obtain a similar RTT> , but a somewhat different RTT> R of 180 meV for the surface non-radiative channel, which we denote as A . It should be noted that ORTT> /ORTT> ratio now increases to ~ 5×105. This implies about three R orders of magnitude increase of the surface recombination efficiency in comparison with the bulk. This enhancement is consistent with the deduced ratio of nonradiative transition rates, =>? = @>? /A ≈ 103, which is ascribed to the higher density of the surface states. By employing the proposed carrier dynamic model, we also fit the Tdependent transient data in Fig. 5(a), as shown by the dashed lines, and plot the deduced thermal variation of the surface and bulk non-radiative lifetimes, A>? (N) and A(@)

@>? (N) in the inset in Fig. 5(b). >? is found to follow the temperature dependence defined by its corresponding activation energy, i.e. A(@) >? (V) ~ (A(@) /PQ N)– see the solid (dashed) lines in the inset in Fig. 5(b). The radiative lifetime, ? , is indicated by the dash-dotted line. Within the high temperature range, where non-radiative A(@)

recombination dominates, the exciton lifetime is mainly governed by >? , i.e.



A(@)

Page 12 of 30

A(@)

/LM ≈ />? . As temperature decreases below 120 K, >? becomes much longer A(@)

than ? . LM is no longer sensitive to >? , but is mainly determined by ? , i.e. A(@)

/LM ≈ /? , which makes an accurate derivation of >?

more difficult. The

correlation between the thermal quenching of the PL intensity and acceleration of A(@) >? (N) confirms the suggested model for the exciton dynamics in the GaNP NWs. CONCLUSIONS In conclusion, time-resolved PL measurements are performed to analyze dynamics of exciton recombination in GaP/GaNP core/shell NWs. The PL decay of excitons localized at the N-related states is found to critically depend on the excitation wavelength and the NW shell thickness. The obtained results are explained within the model, which considers bulk non-radiative recombination, as well as tunnelling of the localized excitons to the surface states acting as efficient non-radiative centres. The surface recombination velocity and the effective tunnelling length to the surface states are derived as being 5×104 cm/s and ~ 10 nm, respectively. From temperature dependent PL measurements, thermal activation energies of the individual nonradiative channels at the surface and in bulk are deduced as being 180 meV and 145 meV, respectively. This work, therefore, clarifies the recombination dynamics in dilute nitride GaNP-based NWs. The obtained results can also be generalized to other nanostructured semiconductors with a high degree of localization or disorder for material optimization and device implementation.

NOTES The authors declare no competing financal interest. ACKNOWLEDGEMENTS


Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The authors are grateful to Prof. C. W. Tu for providing the samples. Financial support by the Swedish Energy Agency (grant # P40119-1), the Swedish Research Council (grant # 2015-05532) and the Swedish Government Strategic Research Area in Materials Science on Functional Materials at Linköping University (Faculty Grant SFO-Mat-LiU No 2009 00971) is greatly appreciated.

REFERENCES [1] Guo, W.; Zhang, M.; Banerjee, A.; Bhattacharya, P. Catalyst-free InGaN/GaN nanowire light emitting diodes grown on (001) silicon by molecular beam epitaxy. Nano Lett. 2010, 10, 3355–3359. [2] Nguyen, H. P. T.; Zhang, S.; Connie, A. T.; Kibria, M. G.; Wang, Q.; Shih, I.; Mi, Z. Breaking the carrier injection bottleneck of phosphor-free nanowire white light-emitting diodes. Nano Lett. 2013, 13, 5437–5442. [3] Zhao, C.; Ng, T. K.; Elafandy, R. T.; Prabaswara, A.; Consiglio, G. B.; Ajia, I. A.; Roan, I. S.; Janjua, B.; Shen, C.; Eid, J.; Alyamani, A. Y.; et al. Droop-free, reliable, and highpower InGaN/GaN nanowire light-emitting diodes for monolithic metal-optoelectronics. Nano Lett. 2016, 16, 4616–4623. [4] Eaton, S. W.; Fu, A.; Wong, A. B.; Ning, C. Z.; Yang, P. D. Semiconductor nanowire lasers. Nat. Rev. Mater. 2016, 1, 16028. [5] Zimmler, M. A.; Capasso, F.; Müller, S.; Ronning, C. Optically pumped nanowire lasers: invited review. Semicond. Sci. Technol. 2010, 25, 024001. [6] Cui, Y.; Wei, Q. Q.; Park, H. K.; Lieber, C. M. Nanowire nanosensors for highly sensitive and selective detection of biological and chemical species. Science 2001, 293, 1289-1292.



[7] Xie, P.; Xiong, Q. H.; Fang, Y.; Qian, Q.; Lieber, C. M. Local electrical potential detection of DNA by nanowire–nanopore sensors. Nat. Nanotech. 2012, 7, 119-125. [8] Patolsky, F.; Timko, B. P.; Yu, G.; Fang, Y.; Greytak, A. B.; Zheng, G.; Lieber, C. M. Detection, stimulation, and inhibition of neuronal signals with high-density nanowire transistor arrays. Science 2006, 313, 1100-1104. [9] Krogstrup, P.; Jørgensen, H. I.; Heiss, M.; Demichel, O.; Holm, J. V.; Aagesen, M.; Nygard, J.; Morral, A. F. I. Single nanowire solar cells beyond the shockley-queisser limit. Nat. Photon. 2013, 7, 306-310. [10] Wallentin, J.; Anttu, N.; Asoli, D.; Huffman, M.; Åberg, I.; Magnusson, M. H.; Siefer, G.; Fuss-kailuweit, P.; Dimroth, F.; Witzigmann, B.; et al. InP nanowire array solar cells achieving 13.8% efficiency by exceeding the ray optics limit. Science 2013, 339, 10571060. [11] Law, M.; Greene, L. E.; Johnson, J. C.; Saykally, R.; Yang, P. D. Nanowire dyesensitized solar cells. Nat. Mater. 2005, 4, 455-459. [12] Tang, J. Y.; Huo, Z. Y.; Brittman, S.; Gao, H. W.; Yang, P. D. Solution-processed core– shell nanowires for efficient photovoltaic cells. Nat. Nanotech. 2011, 6, 568-572. [13] Shan, W.; Walukiewicz, W.; Yu, K. M.; Wu, J.; Ager III, J. W.; Haller, E. E.; Xin, H. P.; Hong, Y. G.; Tu, C. W. Nature of the fundamental band gap in GaNxP1-x alloys. Appl. Phys. Lett. 2000, 76, 3251. [14] Buyanova, I. A.; Pozina, G.; Bergman, J. P.; Chen, W. M.; Xin, H. P.; Tu, C. W. Timeresolved studies of photoluminescence in GaNxP1-x alloys: evidence for indirect-direct band gap crossover. Appl. Phys. Lett. 2002, 81, 52. [15] O’Reilly, E. P.; Lindsay, A.; Klar, P. J.; Polimeni, A.; Capizzi, M. Trends in the electronic structure of dilute nitride alloys. Semicond. Sci. Technol. 2009, 24, 033001.


Page 14 of 30

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[16] Chamings, J.; Ahmed, S.; Sweeney, S. J.; Odnoblyudov, V. A.; Tu, C. W. Physical properties and efficiency of GaNP light emitting diodes. Appl. Phys. Lett. 2008, 92, 021101. [17] Stehr, J. E.; Dobrovolsky, A.; Sukrittanon, S.; Kuang, Y.; Tu, C. W.; Chen, W. M.; Buyanova, I. A. Optimizing GaNP coaxial nanowires for efficient light emission by controlling formation of surface and interfacial defects. Nano Lett. 2015, 15, 242-247. [18] Dobrovolsky, A.; Sukrittanon, S.; Kuang, Y.; Tu, C. W.; Chen, W. M.; Buyanova, I. A. Energy upconversion in GaP/GaNP core/shell nanowires for enhanced near‐infrared light harvesting. Small 2014, 10, 4403-4408. [19] López, N.; Reichertz, L. A.; Yu, K. M.; Campman, K.; Walukiewicz, W. Engineering the electronic band structure for multiband solar cells. Phys. Rev. Lett. 2011, 106, 028701. [20] Zhang, Y.; Fluegel, B.; Mascarenhas, A.; Xin, H. P.; Tu, C. W. Optical transitions in the isoelectronically doped semiconductor GaP:N: an evolution from isolated centers, pairs, and clusters to an impurity band. Phys. Rev. B 2000, 62, 4493-4500. [21] Buyanova, I. A.; Chen, W. M.; Tu, C. W. Recombination processes in N-containing III– V ternary alloys. Solid state electron. 2003, 47, 467-475. [22] Kuang, Y. J.; Sukrittanon, S.; Li, H.; Tu, C. W. Growth and photoluminescence of selfcatalyzed GaP/GaNP core/shell nanowires on Si(111) by gas source molecular beam epitaxy. Appl. Phys. Lett. 2012, 100, 053108. [23] Sukrittanon, S.; Dobrovolsky, A.; Kang, W. M.; Jiang, J. S.; Kim, B. J.; Chen, W. M.; Buyanova, I. A.; Tu, C. W. Growth and characterization of dilute nitride GaNxP1-x nanowires and GaNxP1-x/GaNyP1-y core/shell nanowires on Si (111) by gas source molecular beam epitaxy. Appl. Phys. Lett. 2014, 105, 072107.



[24] Buyanova, I. A.; Izadifard, M.; Kasic, A.; Arwin, H.; Chen, W. M.; Xin, H. P.; Hong, Y. G.; Tu, C. W. Analysis of band anticrossing in GaNxP1-x alloys. Phys. Rev. B 2004, 70, 085209. [25] Dobrovolsky, A.; Persson, P. O. Å.; Sukrittanon, S.; Kuang, Y.; Tu, C. W.; Chen, W. M.; Buyanova, I. A. Effects of polytypism on optical properties and band structure of individual Ga(N)P nanowires from correlative spatially resolved structural and optical studies. Nano Lett. 2015, 15, 4052-4058. [26] Wood, D. L.; Tauc, J. Weak absorption tails in amorphous semiconductors. Phys. Rev. B 1972, 5, 3144-3151. [27] Klingshirn, C. Semiconductor optics; Springer: Heidelberg, 2005. [28] Oueslati, M.; Zouaghi, M.; Pistol, M. E.; Samuelson, L.; Grimmeiss, H. G.; Balkanski, M. Photoluminescence study of localization effects induced by the fluctuating random alloy potential in indirect band-gap GaAs1-xPx. Phys. Rev. B 1985, 32, 8220-8227. [29] Dobrovolsky, A.; Stehr, J. E.; Chen, S. L.; Kuang, Y. J.; Sukrittanon, S.; Tu, C. W.; Buyanova, I. A. Mechanism for radiative recombination and defect properties of GaP/GaNP core/shell nanowires. Appl. Phys. Lett. 2012, 101, 163106. [30] Dobrovolsky, A.; Chen, S. L.; Kuang, Y. J.; Sukrittanon, S.; Tu, C. W.; Buyanova, I. A. Optical properties of GaP/GaNP core/shell nanowires: a temperature-dependent study. Nanoscale Res. Lett. 2013, 8, 239. [31] Gourdon, C.; Lavallard, P. Exciton transfer between localized states in CdS1-xSex alloys. Phys. Stat. Sol. 1989, 153, 641-652. [32] Narukawa, Y.; Kawakami, Y.; Fujita, S.; Nakamura, S. Recombination dynamics of localized excitons in In0.2Ga0.8N-In0.05Ga0.95N multiple quantum wells. Phys. Rev. B 1997, 55, R1938.


Page 16 of 30

Page 17 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[33] Mair, R. A.; Lin, J. Y.; Jiang, H. X.; Jones, E. D.; Allerman, A. A.; Kurtz, S. R. Timeresolved photoluminescence studies of InxGa1-xAs1-yNy. Appl. Phys. Lett. 2000, 76, 188. [34] Kaschner, T.; Lüttgert, T.; Born, H.; Hoffmann, A.; Egorov, A. Y.; Riechert, H. Recombination mechanisms in GaInNAs/GaAs multiple quantum wells. Appl. Phys. Lett. 2001, 78, 1391-1393. [35] Buyanova, I. A.; Pozina, G.; Bergman, J. P.; Chen, W. M.; Xin, H. P.; Tu, C. W. Timeresolved studies of photoluminescence in GaNxP1-x alloys: Evidence for indirect-direct band gap crossover. Appl. Phys. Lett. 2002, 81, 52-54. [36] Chen, S. L.; Chen, W. M.; Ishikawa, F.; Buyanova, I. A. Suppression of non-radiative surface recombination by N incorporation in GaAs/GaNAs core/shell nanowires. Sci. Rep. 2015, 5, 11653. [37] Chen, S. L.; Chen, W. M.; Buyanova, I. A. Dynamics of donor bound excitons in ZnO. Appl. Phys. Lett. 2013, 102, 121103. [38] Yong, C. K.; Noori, K.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C.; Giustino, F.; Johnston, M. B.; Herz, L. M. Strong carrier lifetime enhancement in GaAs nanowires coated with semiconducting polymer. Nano Lett. 2012, 12, 6293-6301. [39] Chang, C. C.; Chi, C. Y.; Yao, M.; Huang, N.; Chen, C. C.; Theiss, J.; Bushmaker, A. W.; LaLumondiere, S.; Yeh, T. W.; Povinelli, M. L.; et al. Electrical and optical characterization of surface passivation in GaAs nanowires. Nano Lett. 2012, 12, 44844489. [40] Jiang, N.; Gao, Q.; Parkinson, P.; Wong-Leung, J.; Mokkapati, S.; Breuer, S.; Tan, H. H.; Zheng, C. L.; Etheridge, J.; Jagadish, C. Enhanced minority carrier lifetimes in GaAs/AlGaAs core-shell nanowires through shell growth optimization. Nano Lett. 2013, 13, 5135-5140.



[41] Arab, S.; Chi, C. Y.; Shi, T.; Wang, Y.; Dapkus, D. P.; Jackson, H. E.; Smith, L. M.; Cronin, S. B.; Etheridge, J.; Jagadish, C. Effects of surface passivation on twin-free GaAs nanosheets. ACS Nano 2015, 9, 1336-1340. [42] Higuera-Rodriguez, A.; Romeira, B.; Birindelli, S.; Black, L. E.; Smalbrugge, E.; van Veldhoven, P. J.; Kessels, W. M. M.; Smit, M. K.; Fiore, A. Ultralow surface recombination velocity in passivated InGaAs/InP nanopillars. Nano Lett. 2017, 17, 26272633. [43] Alekseev, P. A.; Dunaevskiv, M. S.; Ulin, V. P.; Lvova, T. V.; Filatov, D. O.; Nezhdanov, A. V.; Mashin, A. I.; Berkovits, V. L. Nitride surface passivation of GaAs nanowires: impact on surface state density. Nano Lett. 2015, 15, 63-68. [44] Woo, R. L.; Xiao, R.; Kobayashi, Y.; Gao, L.; Goel, N.; Hudait, M. K.; Mallouk, T. E.; Hicks, R. F. Effect of twinning on the photoluminescence and photoelectrochemical properties of indium phosphide nanowires grown on silicon (111). Nano Lett. 2008, 8, 4664-4669. [45] Parkinson, P.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Zhang, X.; Zou, J.; Jagadish, C.; Herz, L. M.; Johnston, M. B. Carrier lifetime and mobility enhancement in nearly defect-free coreshell nanowires measured using time-resolved terahertz spectroscopy. Nano Lett. 2009, 9, 3349-3353. [46] Szymura, M.; Klopotowski, L.; Mitioglu, A. A.; Wojnar, P.; Karczewski, G.; Wojtowicz, T.; Maude, D. K.; Plochocka, P.; Kossut, J. Exciton and carrier dynamics in ZnTe-Zn1xMgxTe core-shell nanowires. Phys. Rev. B 2016, 93, 155429. [47] Casey, H. C.; Jayson, J. S. Cathodoluminescent measurements in GaP (Zn, O). J. Appl. Phys. 1971, 42, 2774-2776. [48] Stringfellow, G. B. Effect of surface treatment on surface recombination velocity and diode leakage current in GaP. J. Vac. Sci. Technol. 1976, 13, 908.


Page 18 of 30

Page 19 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[49] Chen, S. L.; Filippov, S.; Ishikawa, F.; Chen, W. M.; Buyanova, I. A. Origin of radiative recombination and manifestations of localization effects in GaAs/GaNAs core/shell nanowires. Appl. Phys. Lett. 2014, 105, 253106. [50] Jansson, M.; Chen, S. L.; La, R.; Stehr, J.; Tu, C. W.; Chen, W. M.; Buyanova, I. A. Effects of nitrogen incorporation on structural and optical properties of GaNAsP nanowires. J. Phys. Chen. C. 2017, 121, 7047-7055.



Fig. 1 (Color online) (a) an SEM image of the GaP/GaNP core/shell NWs with the 90 nm shell (D1) and a sketch of the core/shell NW structure. The scale bar denotes 1 µm in length. (b) A room-temperature time-integrated PL spectrum from the D1 NWs (the open circles). The dashed-dotted line represents the best fit to the PL tail by the exponential function ~ exp(W/W ), where W is the localization energy of tail states. The band edge of the GaNP shell at 300 K is marked by the arrow at 2.06 eV. The time evolution of the PL spectrum after the excitation pulse is depicted by the solid curves. The PL peak position is traced by the dashed arrow. (c) Schematic diagram of LEs related to N-related band-tail states, which are subject to non-radiative recombination at surface (bulk) with the ( ) recombination lifetime of ! ( ). The dashed bars represent defects at the two

spatial regions. (d) PL decays at RT under the 400 nm excitation as a function of the

emission energy.


Page 20 of 30

Page 21 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 2 (Color online) (a) Spectral dependent 1/e lifetime, , of LEs measured at of 400 nm (the filled circles) and 266 (the open circles) at 300 K. Solid line represents fitting by Eq. (1) and horizontal dashed line is guide to the eye. (b) PL decays of LEs from the D1 NWs, as a function of at 400 nm, 360 nm, 300 nm, and 266 nm. The PL transients were spectrally integrated below 1.7 eV for the deep N states. The inset shows the magnified time window for fast transient process under the 266 nm excitation. The dashed lines are fittings to the experimental data.



Fig. 3 (color online) (a)-(d) Cross-sectional view of the laser profiles under different . (e) Normalized exciton generation profile, 2Z , as a function of depth from the NW surface under different . (f) Depth-dependent exciton recombination dynamics in the D1 NWs after the 400 nm pulsed laser excitation at 300 K. The inset displays the dynamics for = 266 nm.


Page 22 of 30

Page 23 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 4 (Color online) (a)-(d) Distributions of the excitation light inside the D1-D4 NWs under = 400 \] obtained from the FDTD simulations. (e) PL transients of LEs at 300 K from the D1-D4 NWs with different GaNP shell thicknesses. The dashed lines are fitting curves to the experimental data obtained within the proposed model of the LE

dynamics.



Page 24 of 30

Fig. 5 (Color online) (a) Temperature-dependent PL decays of LEs measured from the D1 NWs under of 400 nm (the solid curves) and 266 nm (the open circles). The dashed lines are fittings to the experimental data. They are vertically shidted for clarity. (b) Timeintegrated PL intensity versus inverse temperature under the 400 nm (the filled circles) and 266 nm (the open circles) laser excitation. The fittings by Eq.4 are shown by the solid and dashed lines. The inset summarizes the temperature dependence of surface (bulk) non radiative lifetime, ( ), which are represented by the open (filled) circles, ()

respectively. The data are fitted by an exponential funcion, (T) ~ exp(W() /_` a), where W() stands for the thermal activation energy of the non-radiative recombination via the surface (bulk) states. The fitting results is shown as the dashed (solid) lines. The intrinsic radiative lifetime is indicated as the dash-dot line.


Page 25 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC Graphic



Fig. 1 (Color online) (a) an SEM image of the GaP/GaNP core/shell NWs with the 90 nm shell (D1) and a sketch of the core/shell NW structure. The scale bar denotes 1 µm in length. (b) A room-temperature timeintegrated PL spectrum from the D1 NWs (the open circles). The dashed-dotted line represents the best fit to the PL tail by the exponential function ~ exp (E/E_0), where E_0 is the localization energy of tail states. The band edge of the GaNP shell at 300 K is marked by the arrow at 2.06 eV. The time evolution of the PL spectrum after the excitation pulse is depicted by the solid curves. The PL peak position is traced by the dashed arrow. (c) Schematic diagram of LEs related to N-related band-tail states, which are subject to nonradiative recombination at surface (bulk) with the recombination lifetime of τ_nr^s (x) (τ_nr^b). The dashed bars represent defects at the two spatial regions. (d) PL decays at RT under the 400 nm excitation as a function of the emission energy. 266x244mm (96 x 96 DPI)


Page 26 of 30

Page 27 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 2 (Color online) (a) Spectral dependent 1/e lifetime, τ, of LEs measured at λ_exc of 400 nm (the filled circles) and 266 (the open circles) at 300 K. Solid line represents fitting by Eq. (1) and horizontal dashed line is guide to the eye. (b) PL decays of LEs from the D1 NWs, as a function of λ_exc at 400 nm, 360 nm, 300 nm, and 266 nm. The PL transients were spectrally integrated below 1.7 eV for the deep N states. The inset shows the magnified time window for fast transient process under the 266 nm excitation. The dashed lines are fittings to the experimental data. 199x261mm (96 x 96 DPI)



Fig. 3 (color online) (a)-(d) Cross-sectional view of the laser profiles under different λ_exc. (e) Normalized exciton generation profile, I_g, as a function of depth from the NW surface under different λ_exc. (f) Depthdependent exciton recombination dynamics in the D1 NWs after the 400 nm pulsed laser excitation at 300 K. The inset displays the dynamics for λ_exc = 266 nm. 227x195mm (96 x 96 DPI)


Page 28 of 30

Page 29 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 4 (Color online) (a)-(d) Distributions of the excitation light inside the D1-D4 NWs under λ_exc= 400 ݊݉ obtained from the FDTD simulations. (e) PL transients of LEs at 300 K from the D1-D4 NWs with different GaNP shell thicknesses. The dashed lines are fitting curves to the experimental data obtained within the proposed model of the LE dynamics. 242x213mm (96 x 96 DPI)



Fig. 5 (Color online) (a) Temperature-dependent PL decays of LEs measured from the D1 NWs under λ_exc of 400 nm (the solid curves) and 266 nm (the open circles). The dashed lines are fittings to the experimental data. They are vertically shidted for clarity. (b) Time-integrated PL intensity versus inverse temperature under the 400 nm (the filled circles) and 266 nm (the open circles) laser excitation. The fittings by Eq.4 are shown by the solid and dashed lines. The inset summarizes the temperature dependence of surface (bulk) non-radiative lifetime, τ_nr^s (τ_nr^b), which are represented by the open (filled) circles, respectively. The data are fitted by an exponential funcion, τ_nr^s(b) (T) ~ exp(E_(s(b))/k_B T), where E_(s(b)) stands for the thermal activation energy of the non-radiative recombination via the surface (bulk) states. The fitting results is shown as the dashed (solid) lines. The intrinsic radiative lifetime is indicated as the dash-dot line. 285x199mm (96 x 96 DPI)


Page 30 of 30

Effects of Strong Band-Tail States on Exciton Recombination

Recommend Documents