Midinfrared Photoluminescence up to 290 K Reveals Radiative

Subscriber access provided by Fudan University

Communication

Mid-infrared photoluminescence up to 290 K reveals radiative mechanisms and substrate doping-type effects of InAs nanowires Xiren Chen, Qiandong Zhuang, Hayfaa Alradhi, ZhiMing Jin, Liangqing Zhu, Xin Chen, and Jun Shao Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.6b04629 • Publication Date (Web): 23 Feb 2017 Downloaded from http://pubs.acs.org on February 24, 2017

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 free 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 accessible to all readers and 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.

Nano Letters 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 21

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

Nano Letters

Mid-infrared photoluminescence up to 290 K reveals radiative mechanisms and substrate doping-type effects of InAs nanowires Xiren Chen,† Qiandong Zhuang,∗,‡ H. Alradhi,‡ Zh. M. Jin,‡ Liangqing Zhu,†,¶ Xin Chen,† and Jun Shao∗,† †National Laboratory for Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 200083 Shanghai, China ‡Physics Department, Lancaster University, LA14YB Lancaster, United Kingdom ¶Key Laboratory of Polar Materials and Devices, East China Normal University, 200062 Shanghai, China E-mail: [email protected]; [email protected] Phone: +86 (0)21 25051860. Fax: +86 (0)21 65830734

Abstract Photoluminescence (PL) as a conventional yet powerful optical spectroscopy may provide crucial insight into the mechanism of carrier recombination and bandedge structure in semiconductors. In this study, mid-infrared PL measurements on vertically aligned InAs nanowires (NWs) are realized for the first time in a wide temperature range of up to 290 K, by which the radiative recombinations are clarified in the NWs grown on n- and p-type Si substrates, respectively. A dominant PL feature is identified to be from the type-II optical transition across the interfaces between the zinc-blend (ZB) and the wurtzite (WZ) InAs, a lower-energy feature at low temperatures is ascribed

1

ACS Paragon Plus Environment

Nano Letters

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

to impurity-related transition, and a higher-energy feature at high temperatures originates in the interband transition of the WZ InAs being activated by thermal-induced electron transfer. The optical properties of the ZB-on-WZ and WZ-on-ZB interfaces are asymmetric, and stronger non-radiative recombination and weaker carrier-phonon interaction show up in the NWs on p-type substrate, in which built-in electric field forms and leads to carrier assembling around the WZ-on-ZB interface. The results indicate that wide temperature-range infrared PL analysis can serve as efficient vehicle for clarifying optical properties and bandedge processes of the crystal-phase interfaces in vertically aligned InAs NWs.

Keywords InAs nanowires, infrared photoluminescence, substrate, built-in electric field, non-radiative recombination InAs nanowires (NWs) grown on Si substrate have recently attracted intensive research interests because of the great potential for high performance Si-based infrared photodetector, 1 solar cell 2 and transistor, 3 of which the bandedge structure and carrier recombination play crucial roles in optoelectronic device applications, and remain a widely concerned yet to be further clarified issue in the recent years. 4–6 One of the distinct characteristics of InAs NWs is the axial alternation of zinc-blend (ZB) and wurtzite (WZ) phases, which is induced by the competition of surface free energy and is clearly different from InAs bulk. 7–9 Theoretical study of the band alignment of the ZBand WZ-phase InAs showed that the conduction-band edge of the ZB InAs is 86 meV lower than that of the WZ InAs while the valence-band edge of the WZ InAs is 46 meV higher than that of the ZB InAs. 10 As a consequence, the ZB and WZ InAs constitute a typical type-II interfacial structure and hence modulate the band structure along the axial direction of the InAs NWs: electrons distribute in ZB while holes in WZ InAs. For optoelectronic

2


Page 2 of 21

Page 3 of 21

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

Nano Letters

device applications, the formation of p-n heterojunction is necessary. As the InAs NWs are essentially of n-type, a possible configuration is to realize aligned InAs NWs arrays on a p-type Si substrate for the NW/substrate heterojunction. 1 The built-in electric field is in this case along the axial direction, and hence the questions arise if the field is affected by a particular ZB/WZ alternation and how the differences will be for the photo-induced carriers recombinations in the InAs NWs grown on p- and n-type Si substrates, respectively? Photoluminescence (PL) spectroscopy has been proved an efficient and accessible routine for clarifying bandedge structure and optical properties of semiconductor nanowires. 11–13 In the recent years, significant efforts have also been conducted on PL analyses of InAs NWs at variable temperatures for bandgap evolution and carrier recombination. 14–17 However, the PL measurements were conducted in temperature ranges far below room temperature 18 and with rather poor signal-to-noise ratio (SNR), 14 which may cast doubt on the identification of PL mechanisms. Previously, Möller et al analyzed PL transitions in InAs NWs grown on GaAs(100) substrate by chemical-beam epitaxy and ascribed PL features to the impuritiesrelated, band-to-band and type-II radiative processes, 14 though it was rather strange that the considered type-II transition-related PL component shifted to higher energies, in clear opposite to the general bandgap reduction of semiconductors at higher temperatures, and for which either the thermal-induced carrier redistribution in localization states 19 or the competition of unresolved adjacent multiple PL features 20 could not be simply excluded. This indicates that for reliable identification of band structures and PL mechanisms in InAs NWs, high quality samples and the ability in performing PL measurements at possibly high temperatures are both crucial factors 21 for any further investigations. In this study, mid-infrared PL spectra are acquired from InAs NWs with significantly improved SNR in a wide temperature range of 8−290 K, by using an optimized step-scan Fourier transform infrared (FTIR) spectrometer-based modulated-PL method. 22,23 The quality of the InAs NWs samples 8 and the SNR of the PL spectra are high enough for a reliable quantitative PL analysis, by which the origins of the PL transitions are revealed, the non-

3


Nano Letters

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

radiative recombination rate is estimated, and the behavior of photo-induced carriers and the electron-phonon interaction are clarified. The results show the asymmetric nature of the ZB-on-WZ and WZ-on-ZB InAs interfaces in optical properties that deserves attention in material and device fabrication. Two InAs NWs samples were grown directly on n- and p-type bare Si(111) substrates, respectively, by molecular-beam epitaxy (MBE) via an identical droplet-assisted method, 8 and are labeled hereafter as sample-N and sample-P for conciseness. The growth was initiated by a pre-calibrated In droplets acting as the nucleation site to prompt the NWs growth. It was then followed by opening In and As shutters simultaneously for InAs NWs growth, at a temperature range of 420−460 ◦ C and an As-rich condition (As4 equivalent beam pressure of 6−9 × 10−6 mbar). The local scanning electron microscopy (SEM) images of the sample-N and sample-P are depicted in Fig. 1(a) and (b), respectively, from which the average length and diameter of the NW are estimated to be about 1 µm and 50−100 nm, respectively. Such a diameter implies that the radial quantum effect is negligible and the properties of the NWs are bulk- rather than quantum-like. The NWs stand upright on the Si substrate as shown in Fig. 1(c) and (d). Figure 1(e) depicts a representative transmission electron microscopy (TEM) image of InAs NWs, which exhibits typical polytypic ZB/WZ crystal structure in the self-catalyzed NWs. The length of the ZB segment is about 5−8 nm. 8

(a)

(b) (c) (d) (e)

Figure 1: Local SEM images of sample-N (a) and sample-P (b) at a scale bar of 500 nm. Single NW manifests upright standing as in (c) and (d) for sample-N and sample-P. (e) TEM image of InAs NWs. The TEM scale bar is 10 nm.

4


Page 4 of 21

Page 5 of 21

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

Nano Letters

Temperature-dependent PL measurements were conducted with a modulated PL method based on a step-scan FTIR spectrometer. 22,24 A 639-nm continuous-wave laser was modulated by a mechanical chopper for optical pumping, and a liquid-nitrogen cooled InSb detector was used for PL signal detection. More details about the PL method and its advantages can be found elsewhere. 22 The sample-N and sample-P were mounted on the cold finger of a closed-cycle compressor for an adjustable temperature range of 8−290 K. For temperaturedependent PL measurements, the laser output power was set at 100 mW, while for the excitation power-dependent PL measurements the samples’ temperature was set at 8 K. The laser spot is about 200 µm in diameter, which is much larger than the vision field of SEM in Fig. 1(a) and (b), and may cover statistically more than ten thousands NWs. The PL property can hence be regarded as a statistical average of the unintentional random morphology and surface condition of the NWs. It is worthy to emphasize that although from the structure viewpoint residual clusters are clearly seen on the substrates, they do not contribute to the PL spectra as no PL signal was detectable when the NWs were removed, 8 due possibly to the low crystal quality of the clusters. This ensures the PL analysis to be solely related to the InAs NWs. Figure 2 depicts the infrared PL spectra of the sample-N and sample-P at temperatures of 8−290 K. Magnification is made for each of the spectra for intensity normalization. At 8 K, only one asymmetric PL peak shows up at about 0.415 eV for the two samples. This means the bandgap transition of the WZ InAs is not activated, as it should be at about 0.45 eV. 10,17,25 As temperature rises up, the PL peak redshifts monotonously, in agreement with the negative temperature coefficient of the bulk InAs bandgap. 26 The PL intensity gets weakened, and at 290 K it reduces to about 1/70 and 1/307, respectively, of that at 8 K for the sample-N and sample-P, evidenced by the magnification. For quantitative analysis of the temperature effects, lineshape fitting is performed on each of the PL spectra by a typical Lorentzian-Gaussian composite function, 27–29 by which the energies, linewidthes and intensities of the PL features are derived quantitatively. Rep-

5


Nano Letters

Wavelength (µm) 4.5 4.0

3.5

3.0

2.5 sample-P

Photoluminescence (arb. units)

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

Page 6 of 21

307 194 119 53.8 28.0 13.1 7.37 3.70 3.19 1.98 1.35 1.15 1.05 1

290 K 250 K 210 K 180 K 150 K 120 K 100 K 80 K 65 K 50 K 35 K 25 K 15 K 8K sample-N 290 K 250 K 210 K 180 K 150 K 120 K 100 K 80 K 65 K 50 K 35 K 25 K 15 K 8K

0.3

70.3 39.4 31.2 23.1 15.4 10.2 6.83 4.53 3.19 2.18 1.54 1.23 1.07 1

0.4

0.5

0.6

Energy (eV)

Figure 2: Temperature-dependent PL spectra of sample-N (a) and sample-P (b), magnified by particular factors for similar peak heights. resentative fittings are depicted in Fig. 3(a) for two temperatures of 8 and 290 K. At 8 K, a single Lorentzian-Gaussian composite function is insufficient, and a dominant feature (DF) at 0.415 eV and a low-energy feature (LEF) at around 0.401 eV are required for well reproducing the PL lineshape. At 290 K, the fitting reveals the DF at about 0.380 eV and a high-energy feature (HEF) around 0.425 eV. The HEF is revealed by plotting the PL spectra as in Fig. 3(b) in logarithmic intensity, which is obvious beyond the high-energy exponential tail introduced by carriers Fermi-edge distribution. 30 As the temperature of 290 K is high enough to depress the impurity-related transitions in InAs, the surviving of the DF and HEF at 290 K indicates the two features to be of band-to-band-like transitions. The energy evolution with temperature is depicted in Fig. 4(a) for each of the PL features of sample-N and sample-P. While the DF remains in the whole temperature range and redshifts monotonously with temperature, its energy versus temperature can be well accounted for as plotted in red lines in Fig. 4(a) by the bandgap shrinking behavior of semiconductor 6


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

Nano Letters

Photoluminescence (arb. units) Ln(I) (arb. units)

Page 7 of 21

(b)

290 K

sample-N sample-P

(a) sample-P 290 K

HEF

sample-N 290 K

sample-P 8 K DF LEF

0.30

0.35

0.40

sample-N 8 K

0.45

0.50

Energy (eV)

Figure 3: (a) Curve fittings of PL spectra for sample-N and sample-P at 8 and 290 K. (b) PL spectra of the two samples at 290 K plotted in logarithmic intensity. that is applicable to type-I and type-II transitions, 31,32

E(T ) = E0 − ShΘi[coth(hΘi/2T ) − 1],

(1)

where E0 is the band gap at 0 K, S a coupling coefficient, and hΘi an average phonon temperature. The derived E0 , S, and hΘi are 0.414 and 0.416 eV, 0.08 and 0.10 meV/K, and 178 and 220 K, respectively, for sample-N and sample-P. It is also clear that the DF energy changes linearly with temperature in the range above about 65 K, and the slope is −0.135 and −0.179 meV/K, respectively, for sample-N and sample-P. Meanwhile, the LEF energy is insensitive to temperature for the both sample-N and sample-P. As the LEF shows up at the low-energy side and quenches at high temperatures beyond 100 K, it is likely corresponding to donor-like impurity-related transition. For the HEF, the slope of the energy vs temperature plot is about −0.056 and −0.084 meV/K, respectively, for the sample-N and sample-P. The rapid redshifts of the DF and HEF reflect

7


Nano Letters

high sensitivity to temperature and hence indicate the nature of conduction band-related transition. 21,33 Furthermore, the difference in the redshift with temperature of the DF and HEF may imply that the DF and HEF correspond to different conduction bands. It is worthy to emphasize that neither the DF nor the HEF shows blueshift at higher temperatures, indicating the side effect of thermal-induced carrier redistribution in localization states 19 and/or competition of unresolved multiple PL features 20 be negligible.

0.38

sample-P

0.44

LEF DF HEF

0.42 0.40 0.38

sample-N 50

100

150

200

250

8

LEF DF HEF

6 4 2

sample-P LEF DF HEF

6 4 2

sample-N

300

Temperature (K)

50

FWHM (meV)

0.40

8

(c)

FWHM (meV)

0.42

LEF DF HEF

Ln(I) (arb. units)

Energy (eV)

0.44

(b)

ln(I) (arb. units)

(a)

Energy (eV)

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

Page 8 of 21

60 40 20

LEF DF HEF

sample-P

100

LEF DF HEF

80 60 40

sample-N

20

100

150

200

250

300

Temperature (K)

50

100

150

200

250

300

Temperature (K)

Figure 4: (a) Energy, (b) integral intensity, and (c) FWHM versus temperature for DF, LEF and HEF PL components of sample-N and sample-P. The PL intensities of the two samples are very similar at 8 K as shown in Fig. 2. Nevertheless, they are not taken as a criterion for sample’s optical quality as they may be affected by sample’s positioning and/or local NW density within the laser spot. Instead, the evolution of the PL integral intensity with temperature is more reliable, and can directly reflect the status of non-radiative recombination and carrier transferring. 34–36 The integral intensities are plotted against temperature in Fig. 4(b) for the PL features of the two samples. The DF and LEF get weakened monotonously as temperature rises, and can be interpreted with the quenching model containing multiple non-radiative processes, 35–37

I(T ) =

1+

P 8

I0 , Ci e−Ei /kB T


(2)

Page 9 of 21

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

Nano Letters

where Ci and Ei represent the non-radiative coefficient and quenching activation energy of the ith non-radiative channel, I0 the integral intensity extrapolated to 0 K. It is noteworthy that while the LEF is well reproduced with single non-radiative channel, the DF can be well fitted only when a second channel is invoked. The fitting parameters are listed in Table 1. The quenching activation energy of the first channel is about 9 meV for sample-N and 11 meV for sample-P, respectively, for either LEF or DF components. This channel is suggested to be correlated to the non-radiative dislocation and/or Auger recombination. 38,39 The activation energy of the second quenching channel for the DF is about 46 meV, which is very close to the energy deviation between the HEF and DF at 290 K, 0.045 eV. This hints the second channel to be thermal-induced carrier transfer from the DF- to HEF-related energy level. Such carrier transfer is sustained by the HEF integral intensity that first enhances and then weakens, manifesting competition of thermal-induced carrier injection and non-radiative recombination. 40 The thermal-induced carrier transfer is verified by the evolution of full-width at halfmaximum (FWHM) with temperature for sample-N and sample-P as shown in Fig. 4(c). In general, the FWHM (Γ) broadens as temperature rises up due to the carrier-phonon scattering, which can be described with 41,42

Γ(T ) = Γ0 + γT +

2b eεLO /kB T

−1

,

(3)

where εLO is the longitudinal optical (LO) phonon energy, kB the Boltzmann constant, and γ and b the carrier-acoustic and carrier-LO phonon interaction coefficients, respectively. This is in fact the real case for the FWHM of the LEF and DF at low temperatures as depicted in solid line and dash-dots in Fig. 4(c), and the derived values of Γ0 , εLO , γ and b are listed in Table 1. The εLO of the LEF is far smaller than the LO phonon energy of InAs, i.e. 29 meV 17,43 for the two samples, which may suggest that the LEF-related carriers locate in the region the elastic coefficients and/or effective atom mass significantly differing from

9


Nano Letters

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

Page 10 of 21

those of InAs NWs, e.g, around impurities. This supports the consideration of the LEF as impurity-related transition in the above PL energy analysis. The εLO of the DF at low temperatures, on the other hand, is very close to the LO phonon energy of InAs for the two samples, which indicates the DF to be of the band-to-band radiative recombination. As the HEF shows up at high temperatures beyond 100 K, the FWHM of DF deviates remarkably from the broadening described by Eq. (3), and meanwhile the broadening of the HEF can not be described with Eq. (3). Such failure of Eq. (3) implies the insufficiency of the twolevels assumption on which the equation is based, 44 and the carriers thermal-induced transfer between the DF and HEF has to be taken into account. Table 1: Ci , Ei , Γ0 , b, εLO and γ derived from the evolution of integral intensity and FWHM with temperature for the DF and LEF of sample-N and sample-P.

I

Γ

Sample-N LEF DF C1 16.6 5.8 E1 (meV) 8.9 8.5 C2 255 E2 (meV) 46.8 Γ0 (meV) 69.0 36.6 γ (meV/K) 0.10 0.07 b (meV) 61.3 84.3 εLO (meV) 13.6 27.6

Sample-P LEF 17.1 11.5 53.8 0.06 19.9 18.4

DF 9.3 11.0 377 45.3 29.4 0.04 24.2 27.8

To gain further insight into the transition mechanisms, excitation power-dependent PL spectra are collected as shown in Fig. 5(a) for the two samples at 8 K. Similar curve-fitting analysis is performed as for the temperature-dependent PL measurement, and the energy is plotted against excitation power in Fig. 5(b) for the LEF and DF PL components. While the DF blueshifts monotonously with excitation power, the LEF first moves to higher energy quickly at low excitations but slows down significantly at higher excitation, which is typically correlated to impurity-related transition. 35 The blueshift of the DF can be attributed to the band bending effect and type-II transition, as it well follows the equation

10


Page 11 of 21

(a)

(b) 0.42

320 mW sample-P

10 mW

Energy (eV)

8K

(c) 0.41

ZB

320 mW sample-N

10 mW

0.40

0.45

Energy (eV)

0.50

LEF DF

sample-P

confinement HEF DF

0.41 0.40 0.39

0.35

WZ

0.40 0.39 0.42

Energy (eV)

Photoluminescence (arb. units)

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

Nano Letters

LEF DF

sample-N 50

100

150

200

250

300

Excitation Power (mW)

Figure 5: (a) Excitation power-dependent PL spectra of sample-N and sample-P at 8 K. (b) Energy vs excitation power for DF and LEF PL components. (c) Schematic of the DF- and HEF- transitions along the NWs’ axial direction, red arrows for radiative recombination and blue arrow for thermal-induced electron transfer. for the band bending effect due to spatially separated carriers, 45

E(P ) = E0 + CP 1/3 ,

(4)

where E0 is the PL energy without band bending effect, P the excitation power and C a coefficient. The derived E0 and C are 0.406 eV and 1.98 meV/mW1/3 for sample-N, and 0.407 eV and 1.74 meV/mW1/3 for sample-P, respectively. With the above-mentioned experimental results and also, the coexistence of ZB/WZ phases in NWs as shown by TEM image in the previous report, 8 a type-II band structure is drawn and the mechanisms of the DF- and HEF-related PL transitions are clarified as in Fig. 5(c). The DF corresponds to the radiative recombination of the photo-induced electrons in the ZB- and holes in the WZ-phase InAs, while the HEF is due to the inter-band transition in the WZ-phase InAs. The HEF is activated at high temperatures due to the electrons thermally transferring from the ZB- to WZ-InAs conduction band. It is worthy to indicate at this point that a previous theoretical band structure of ZB/WZ

11


Nano Letters

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

InAs by Murayama et al 10 agrees quite well with this experimental result. If further the theoretical band offset is taken into account that the conduction- and valence-band edges of the ZB-phase InAs are 86 and 46 meV lower than those of the WZ-phase InAs, 10 quantitative description can be deduced with the experimental type-II model that (i) the electrons confinement energy is about 44 meV with the heavy-hole confinement being neglected thanks to the large effective mass, and (ii) the effective length of the ZB segment is about 6 nm, which is very close to the length by TEM image, 8 and in turn indicates the rationality of the type-II band structure. The WZ-InAs bandgap is about 0.425 eV at 290 K, 68 meV higher than the ZB-InAs value of 0.357 eV. 26 The energy difference between the WZ- and ZB-InAs is slightly larger than the theoretical value of 40−64 meV, 10,25,46 due possibly to the fact that the theoretical values were derived for 0 K. As illustrated in Fig. 4(a), the redshift of the DF is obviously quicker than that of the HEF as temperature rises. This originates in the relaxation of the band bending effect. 47 The photo-induced electrons in ZB InAs runs away significantly at higher temperatures because of non-radiative recombination and thermal transferring to the WZ conduction band, the band bending effect is hence diminished, and the type-II transition energy decreases much rapidly than that of the band gap. With the clarified PL features, it is interesting to probe how the doping type of the substrate will affect the behavior of the photo-induced carriers, and for which a comparative temperature-dependent PL analysis of the sample-N and sample-P is sure to be beneficial. Since InAs NWs are essentially of n-type, 1 a dominant difference between the two samples is that in sample-P a NW/substrate p-n heterojunction will form as schematically shown in Fig. 6(a). Such type of heterojunction in InAs NWs/p-substrate was previously demonstrated and utilized for opto-electronic devices. 1,48,49 The photo-induced carriers in the space charge region (SCR) is hence influenced by the built-in electric field. The SCR length x in the

12


Page 12 of 21

Page 13 of 21

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

Nano Letters

InAs-NW side is

x =

s

NA 2ǫ V · , e ND (ND + NA )

(5)

where ǫ is the dielectric coefficient of InAs, ND and NA are the effective donor and acceptor concentrations in InAs NWs and Si substrate, respectively, and V the built-in potential and V ≥0.8 V for InAs NWs on p-type Si substrate. 1 The electron concentration in InAs NW is as suggested nonuniform. 50 The concentration of the inner NW is close to that of bulk InAs, and at the level of 1015 − 1016 cm−3 , 51 while that at the NW surface is ∼ 1018 cm−3 . 52,53 The inset in Fig. 6(a) shows the estimated SCR length x in the NW side against effective donor concentration in NW, with the effective acceptor concentration in the substrate to be 5 × 1016 cm−3 . The x is 1000−400 nm for ND = 1015 − 1016 cm−3 whereas is negligible for ND = 1018 cm−3 , indicating that the built-in electric field hardly affects the carriers at the NW surface.

(b)

(a)

sample-N

Root

h

x

Si

EC

WZ

e

ZB DF

x

Top

DF

InAs

sample-P DF

Root

EF EV

h

Top

e

Vbi

p-Si substrate

Built-in Electric Field

InAs NW

Figure 6: (a) Band alignment of sample-P, inset for the length of SCR vs electron concentration in InAs NWs. (b) Schematic of carrier migration in sample-N and sample-P, with built-in electric field marked in sample-P. In sample-N, the photo-induced carriers distribute around the ZB-on-WZ and WZ-on-ZB interfaces and produce the DF PL component with equal probability. In contrast, a built-in electric field forms in sample-P with the direction pointing from the top to the root of the NW. The photo-induced electrons in the ZB- and holes in the WZ-InAs migrate under the 13


Nano Letters

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

electric field and assemble around the WZ-on-ZB interface, leading to the DF be mainly produced therein as schematically shown in Fig. 6(b). The temperature dependence of the DF can hence serve as a probe for the ZB-/WZ-InAs interfaces: while the DF of sampleN is due to the average of the ZB-on-WZ and WZ-on-ZB interfaces, that of sample-P is dominated by the WZ-on-ZB interface effects. As shown in Table 1 for the DF, the C1 of sample-P is larger than that of sample-N, suggesting the WZ-on-ZB interfaces be of lower optical quality. The slight difference in the E1 of the two samples implies the non-radiative recombination be asymmetric for the ZBon-WZ and WZ-on-ZB interfaces, respectively. Hetero-structural interfacial asymmetry was previously observed in no-common-atom structures like InAs/GaSb superlattices, 54,55 it is however rather different that here the interface in the NWs is identical-atom but different crystal phases, and is the reversal to the InAs/GaSb structure. The difference in the C2 of the DF, on the other hand, indicates different possibility of the thermal electron transfer between the ZB- and WZ-InAs regions. Such asymmetry in the optical properties of ZB/WZ-InAs interfaces may provide new vehicle for understanding nanoscale semiconductors. The optical asymmetry of the ZB-on-WZ and WZ-on-ZB interfaces is also sustained by the remarkably different carrier-phonon interaction between the two samples. The b is known p −1 56 to be positively correlated with the Fröhilich coupling constant CF ∝ εLO (ǫ−1 ∞ − ǫ0 ), significant b difference between sample-N and sample-P may indicate variation of the dielec-

tric coefficient ǫ around ZB-on-WZ and WZ-on-ZB interfaces as εLO is nearly the same for the two samples. The LEF also exhibits obvious difference in carriers-phonon interaction, and may hint for the LEF-related impurities to be at the interfaces. To summarize, vertically aligned InAs NWs grown by MBE on n- and p-type Si(111) substrates, respectively, were characterized by infrared PL analysis in a wide temperature range of up to 290 K, with a focus on the mechanisms of photo-induced carrier recombinations and the effects of the substrates on the NWs’ properties. Three PL features were resolved with dominant intensity (DF), lower energy (LEF) and higher energy (HEF), respectively,

14


Page 14 of 21

Page 15 of 21

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

Nano Letters

and the origins were clarified by the evolutions of the feature’s energy, integral intensity and width with temperature that: the DF is due to the type-II radiative recombination near the ZB-/WZ-InAs interfaces, the LEF correlates to impurity-related optical transition, and the HEF corresponds to the WZ-InAs inter-bandgap transition. The InAs NWs on p-type Si substrate (Sample-P) manifests significantly different non-radiative recombination rate and carrier-phonon interaction in comparison with those on n-type substrate (Sample-N), because built-in electric field forms in sample-P due to the doping type of the substrate and leads to carriers assembling near the WZ-on-ZB interfaces. The WZ-on-ZB interface is of lower optical quality. The results indicate that temperature-dependent infrared PL analysis can serve as a useful tool for clarifying the interfacial properties and mechanisms behind in the vertically alignment InAs nanowires.

Acknowledgement This work was supported by the MOST 973 Program (2014CB643901 and 2013CB632805), the NSFC (61604157, 61675224 and 11274329) and the STCSM (16ZR1441400 and 16JC1402400) of China.

References (1) Wei, W.; Bao, X.-Y.; Soci, C.; Ding, Y.; Wang, Z.-L.; Wang, D. Nano Lett. 2009, 9, 2926–2934. (2) Shin, J. C.; Kim, K. H.; Yu, K. J.; Hu, H.; Yin, L.; Ning, C.-Z.; Rogers, J. A.; Zuo, J.M.; Li, X. Nano Lett. 2011, 11, 4831–4838. (3) Persson, K.-M.; Malm, B. G.; Wernersson, L.-E. Appl. Phys. Lett. 2013, 103, 033508. (4) Thelander, C.; Caroff, P.; Plissard, S.; Dey, A. W.; Dick, K. A. Nano Lett. 2011, 11, 2424–2429. 15


Nano Letters

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

(5) Tchernycheva, M.; Travers, L.; Patriarche, G.; Glas, F.; Harmand, J.-C.; Cirlin, G. E.; Dubrovskii, V. G. J. Appl. Phys. 2007, 102, 094313. (6) Ford, A. C.; Ho, J. C.; Chueh, Y.-L.; Tseng, Y.-C.; Fan, Z.; Guo, J.; Bokor, J.; Javey, A. Nano Lett. 2008, 9, 360–365. (7) Caroff, P.; Dick, K. A.; Johansson, J.; Messing, M.; Deppert, K.; Samuelson, L. Nature Nanotech. 2009, 4, 50–55. (8) Zhuang, Q.; Anyebe, E. A.; Chen, R.; Liu, H.; Sanchez, A. M.; Rajpalke, M. K.; Veal, T. D.; Wang, Z.; Huang, Y.; Sun, H. Nano Lett. 2015, 15, 1109–1116. (9) Anyebe, E. A.; Sanchez, A. M.; Hindmarsh, S.; Chen, X.; Shao, J.; Rajpalke, M. K.; Veal, T. D.; Robinson, B. J.; Kolosoy, O.; Anderson, F.; Sundaram, R.; Wang, Z. M.; Falko, V.; Zhuang, Q. Nano Lett. 2015, 15, 4348–4355. (10) Murayama, M.; Nakayama, T. Phys. Rev. B 1994, 49, 4710. (11) Cui, J.; Beyler, A. P.; Coropceanu, I.; Cleary, L.; Avila, T. R.; Chen, Y.; Cordero, J.; Heathcote, S. L.; Harris, D. K.; Chen, O.; Cao, J.; Bawendi, M. G. Nano Lett. 2016, 16, 289. (12) Tedeschi, D.; Luca, M. D.; Fonseka, H. A.; Gao, Q.; Mura, F.; Tan, H. H.; Rubini, S.; Martelli, F.; Jagadish, C.; Capizzi, M.; Polimeni, A. Nano Lett. 2016, 16, 3085. (13) Chauvin, N.; Mavel, A.; Patriarche, G.; Masenelli, B.; Gendry, M.; Machon, D. Nano Lett. 2016, 16, 2926. (14) Möller, M.; de Lima Jr, M.; Cantarero, A.; Chiaramonte, T.; Cotta, M.; Iikawa, F. Nanotechnology 2012, 23, 375704. (15) Sun, M.; Leong, E.; Chin, A.; Ning, C.; Cirlin, G.; Samsonenko, Y. B.; Dubrovskii, V.; Chuang, L.; Chang-Hasnain, C. Nanotechnology 2010, 21, 335705. 16


Page 16 of 21

Page 17 of 21

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

Nano Letters

(16) Koblmüller, G.; Vizbaras, K.; Hertenberger, S.; Bolte, S.; Rudolph, D.; Becker, J.; Döblinger, M.; Amann, M.-C.; Finley, J.; Abstreiter, G. Appl. Phys. Lett. 2012, 101, 053103. (17) Morkötter, S. et al. Phys. Rev. B 2013, 87, 205303. (18) Treu, J.; Bormann, M.; Schmeiduch, H.; Doeblinger, M.; Morkoetter, S.; Matich, S.; Wiecha, P.; Saller, K.; Mayer, B.; Bichler, M.; Amann, M.-C.; Finley, J. J.; Abstreiter, G.; Koblmueller, G. Nano Lett. 2013, 13, 6070–6077. (19) Olsthoorn, S. M.; Driessen, F. A. J. M.; Eijkelenboom, A. P. A. M.; Giling, L. J. J. Appl. Phys. 1993, 73, 7798–7803. (20) Driessen, F. A. J. M.; Bauhuis, G. J.; Olsthoorn, S. M.; Giling, L. J. Phys. Rev. B 1993, 48, 7889–7896. (21) Zhang, X.; Shao, J.; Chen, L.; Lü, X.; Guo, S.; He, L.; Chu, J. J. Appl. Phys. 2011, 110, 043503. (22) Shao, J.; Lu, W.; Lü, X.; Yue, F.; Li, Z.; Guo, S.; Chu, J. Rev. Sci. Instrum. 2006, 77, 063104. (23) Shao, J.; Chen, L.; Lu, W.; Lü, X.; Zhu, L.; Guo, S.; He, L.; Chu, J. Appl. Phys. Lett. 2010, 96, 121915. (24) Shao, J.; Yue, F.; Lü, X.; Lu, W.; Huang, W.; Li, Z.; Guo, S.; Chu, J. Appl. Phys. Lett. 2006, 89, 182121. (25) De, A.; Pryor, C. E. Phys. Rev. B 2010, 81, 155210. (26) Vurgaftman, I.; Meyer, J.; Ram-Mohan, L. J. Appl. Phys. 2001, 89, 5815. (27) Shao, J.; Winterhoff, R.; Dörnen, A.; Baars, E.; Chu, J. Phys. Rev. B 2003, 68, 165327.

17


Nano Letters

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

(28) Chen, X.; Song, Y.; Zhu, L.; Wang, S.; Lu, W.; Guo, S.; Shao, J. J. Appl. Phys. 2013, 113, 153505. (29) Shao, J.; Qi, Z.; Zhao, H.; Zhu, L.; Song, Y.; Chen, X.; Zha, F.-X.; Guo, S.; Wang, S. J. Appl. Phys. 2015, 118, 165305. (30) Gogineni, C.; Riordan, N. A.; Johnson, S. R.; Lu, X.; Tiedje, T. Appl. Phys. Lett. 2013, 103, 041110. (31) Odonnell, K.; Chen, X. Appl. Phys. Lett. 1991, 58, 2924. (32) Guha, S.; Cai, Q.; Chandrasekhar, M.; Chandrasekhar, H. R.; Kim, H.; Alvarenga, A.; Vogelgesang, R.; Ramdas, A.; Melloch, M. Phys. Rev. B 1998, 58, 7222. (33) Chen, X.; Jung, J.; Qi, Z.; Zhu, L.; Park, S.; Zhu, L.; Yoon, E.; Shao, J. Opt. Lett. 2015, 40, 5295–5298. (34) Titova, L. V.; Hoang, T. B.; Jackson, H. E.; Smith, L. M.; Yarrison-Rice, J. M.; Kim, Y.; Joyce, H. J.; Tan, H. H.; Jagadish, C. Applied Physics Letters 2006, 89, 173126 – 173126–3. (35) Shao, J.; Lu, W.; Tsen, G.; Guo, S.; Dell, J. J. Appl. Phys. 2012, 112, 063512. (36) Chen, X.; Zhou, Y.; Zhu, L.; Qi, Z.; Xu, Q.; Xu, Z.; Guo, S.; Chen, J.; He, L.; Shao, J. Jpn. J. Appl. Phys. 2014, 53, 082201. (37) Sun, H.; Calvez, S.; Dawson, M.; Gupta, J.; Aers, G.; Sproule, G. Appl. Phys. Lett. 2006, 89, 101909. (38) Lee, H.; Choi, S.-H. J. Appl. Phys. 1999, 85, 1771–1774. (39) Seetoh, I.; Soh, C.; Fitzgerald, E.; Chua, S. Appl. Phys. Lett. 2013, 102, 101112. (40) Shao, J.; Ma, L.; LÃĳ, X.; Lu, W.; Wu, J.; Zha, F.-X.; Wei, Y.-F.; Li, Z.-F.; Guo, S.-L.; Yang, J.-R.; He, L.; Chu, J.-H. Appl. Phys. Lett. 2008, 93, 131914. 18


Page 18 of 21

Page 19 of 21

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

Nano Letters

(41) Bayer, M.; Forchel, A. Phys. Rev. B 2002, 65, 321–325. (42) Murotani, H.; Yamada, Y.; Taguchi, T.; Ishibashi, A.; Kawaguchi, Y.; Yokogawa, T. J. Appl. Phys. 2008, 104, 053514–053514–8. (43) Möller, M.; de Lima Jr, M.; Cantarero, A.; Dacal, L.; Madureira, J.; Iikawa, F.; Chiaramonte, T.; Cotta, M. Phys. Rev. B 2011, 84, 085318. (44) Lee, J.; Koteles, E. S.; Vassell, M. O. Phys. Rev. B 1986, 33, 5512–5516. (45) Alonso-Alvarez, D.; Alen, B.; Garcia, J. M.; Ripalda, J. M. Appl. Phys. Lett. 2007, 91, 263103. (46) Zanolli, Z.; Fuchs, F.; Furthmüller, J.; von Barth, U.; Bechstedt, F. Phys. Rev. B 2007, 75, 245121. (47) Pan, W.; Zhang, L.; Zhu, L.; Li, Y.; Chen, X.; Wu, X.; Zhang, F.; Shao, J.; Wang, S. J. Appl. Phys. 2016, 120, 094011. (48) Dalmau Mallorquí, A.; Alarcónlladó, E.; Russoaverchi, E.; Tütüncüoglu, G.; Matteini, F.; Rüffer, D.; Morral, A. F. I. J. Phys. D 2014, 47, 394017–394024(8). (49) Tomioka, K.; Fukui, T. Appl. Phys. Lett. 2011, 98, 083114–083114–3. (50) Katzenmeyer, A. M.; Leonard, F.; Talin, A. A.; Toimil-Molares, M. E.; Cederberg, J. G.; Huang, J. Y.; Lensch-Falk, J. L. IEEE T. Nanotechnol. 2011, 10, 92–95. (51) Chen, Y. Q.; Unuvar, T.; Moscicka, D.; Wang, W. I. J. Vac. Sci. Technol. B 2006, 24, 1599–1603. (52) Noguchi, M.; Hirakawa, K.; Ikoma, T. Phys. Rev. Lett. 1991, 66, 2243–2246. (53) Olsson, L. O.; Andersson, C. B. M.; Håkansson, M. C.; Kanski, J.; Ilver, L.; Karlsson, U. O. Phys. Rev. Lett. 1996, 76, 3626–3629.

19


Nano Letters

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

(54) Mohseni, H.; Litvinov, V.; Razeghi, M. Phys. Rev. B 1998, 58, 15378. (55) Wei, Y.; Razeghi, M. Phys. Rev. B 2004, 69, 085316. (56) Muñoz, M.; Pollak, F. H.; Zakia, M. B.; Patel, N. B.; Herrera-Pérez, J. L. Phys. Rev. B 2000, 62, 16600–16604.

20


Page 20 of 21

Page 21 of 21

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

Nano Letters

Graphical TOC Entry

(Left) Wide temperature-range infrared photoluminescence, and (Right) schematic of carrier migration and recombination in vertically aligned InAs nanowires on n- and p-type Si substrates (sample-N and sample-P).

21


Midinfrared Photoluminescence up to 290 K Reveals Radiative

Recommend Documents