Photoluminescence

The schematic setup of the photoluminescence system is shown in Fig. 2-2.

The samples were loaded on the cold finger of a closed cycle cryostat. The temperature can be controlled between 10 and 325 K. The sample was excited by a GaN pulsed laser (405 nm) with a pulse width of 50 ps and a repetition rate of 2.5 MHz. The incident laser beam is focused on the sample by a convex lens (L₁). The combination of a set of convex lenses (L₂ and L₃) guide the luminescence into the double-grating spectrometer. A SPEX 1403 double-grating spectrometer equipped with a thermoelectrically cooled photo-multiplier tube (PMT; R928, Hamamatsu) is used to detect the PL spectra.

The spectrometer is controlled by a computer which is used to store and plot the collected data.

2.3 Time-resolved photoluminescence (TRPL) system

The setup of TRPL system is shown in Fig. 2-3. The GaN pulsed laser (405 nm) with a pulse width of 50 ps and a repetition rate of 2.5 MHz was used as an excitation source. The laser light was focused on the sample by a convex lens (L₁). The combination (L₂ and L₃) lenses guides the luminescence into the double-grating spectrometer. The signal is dispersed by a 0.85 m double-grating spectrometer and detected by a high-speed photomultiplier tube. The singles were further analyzed by a computer plug-in time-correlated counting card. The overall temporal resolution of the TRPL measurement is about 300 ps.

Fig 2-1. Veeco Applied EPI 620 molecular beam epitaxy (MBE) system.

Closed-cycle refrigerator SPEX 1403 Double-Grating Spectrometer

GaN pulsed Laser: 405 nm Repetition rate: 2.5 MHz Pulse width: 0.05 ns

sample

L₁ L₂

L₃

PMT

Fig. 2-2. The experimental setup of the PL measurements.

GaN pulsed Laser: 405 nm Repetition rate: 2.5 MHz Pulse width: 0.05 ns Closed-cycle refrigerator SPEX 1403 Double-Grating Spectrometer sample

L₁ L₂ L₃

High speed PMT

Trigger

Fig. 2-3. The experimental setup of the TRPL measurements.

Chapter 3: Results and Discussion

In this chapter, we study the recombination kinetics of ZnSe_1−xTe_x ternary alloys with wide Te composition ranges from 0.005 to 0.420 by PL and TRPL measurements. The results are described below.

3.1 Photoluminescence study of ZnSe

_1-x

Te

_x

The normalized PL spectra of ZnSe1−xTex (0.005 ≦ x ≦ 0.42) at 13 K are shown in Fig. 3-1. It is obvious that the main peak of PL emission shifts to lower energies with the increasing of x, from 2.66 to 2.17 eV for x = 0.005 and 0.42, respectively. These PL emissions have generally been attributed to excitons bound to isoelectronic Te_Se (atoms and/or clusters) in ZnSe [6]. As shown in Fig 3-1, the PL peak at 2.78 eV for the x = 0.005 sample is attributed to excitons localized at a single Te atom (Te₁) [1]. A peak at around 2.66 eV is attributed to the Te₂ complexes [6], which dominates the PL spectrum for x = 0.005 sample. When the Te concentration is further increased, the PL emissions evolve from the Te2 complexes to the exction bound to Ten (n ≧ 3) clusters [6].

The origins of different PL emissions are indicated by dashed lines in Fig. 3-1.

measured PL peak energy is defined as the exciton binding energy (denoted as E_b) [10]. The band gap of ZnSe_1-xTe_x can be written as

2

E (x) = 1.57x - 1.953x+2.82g , (1)

which is fitted by Brasil et al.[11] from their optical results, as plotted in Fig.

3-3 with our PL peak energy data. From Fig. 3-2, it can be seen that binding energy increases as the Te concentration increases. The binding energy reaches a maximum of about 272 meV when Te is increased from 0 to 10 %. However, when Te composition is further increased, the binding energy decreases monotonically. This is because the exciton localization of Te isoelectronic centers is closely related to the size of the Te_n clusters. The larger Te_n clusters have higher binding energy. Nevertheless, when the Te_n clusters become to large enough, the difference between the isoelectronic potential and the host atomic potential will be reduced.

Fig. 3-4 shows the measured PL line-width as a function of Te composition.

The composition dependence of the line width has a similar tendency to the dependence of exciton binding energy on Te concentration. The full width at half maximum (FWHM) of the PL spectra increases from 120 to 160 meV as x increases from 0.005 to 0.060. As the Te concentration is further increased,

broadening cannot be totally ascribed to Te compositional fluctuation because the line-width broadening of non-isoelectronic ternary semiconductors is only 16 ~ 39 meV in Zn_1-xMg_xTe (0.10 ≦ x ≦ 0.52), 10 ~ 40 meV in Zn_1-xMn_xTe (0.10 ≦ x ≦ 0.56) [12] and 10 ~ 50 meV in Al_xGa_1-xN (0 ≦ x ≦ 0.50) [13].

Due to the fact that a strong localization is accompanied by isoelectronic centers and results in the increase of the line-width of the photon emission.

However, it is difficult to specify the line-width of one specific isoelectronic center, because more than one Te_n centers are involved in the PL spectrum in most samples. Nevertheless, for the samples with small Te composition (x = 0.005 – 0.060), the dominated PL peaks roughly correspond to the Te₂ complexes and Te_n clusters. Therefore, it is reasonable to assume that the line-width increases as the Te impurity evolves from Te₂ complexes to Te_n clusters.

The temperature dependence of the integrated PL intensity (I_PL) can be expressed as

where T is the temperature, k_B is the Boltzmann constant, I₀ is the integrated PL

E_a is responsible for the thermal quenching of PL intensity in the

temperature-dependent PL spectra. The PL intensity versus temperature of ZnSe_0.94Te_0.06 epilayer is shown in Fig. 3-5. The thermal activation energy of ZnSe_0.94Te_0.06 epilayer can be fitted using equation 2 as shown in Fig. 3-5.

Figure 3-6 presents the thermal activation energies E_a as a function of Te composition of all the ZnSe_xTe_1−x samples. As the Te concentration of the ZnSe_xTe_1−x alloys increases from 0.005 to 0.100, E_a increases initially from 88 to 130 meV. If the Te concentration is further increased, the thermal activation energy decreases monotonously. The temperature-induced quenching of luminescence in the epilayers mainly attributes to the thermal dissociation of excitons into free electrons and holes [16].

3.2 Time-resolved photoluminescence study of ZnSe

1-x

Te

x

In order to understand the recombination mechanism of ZnSe_1−xTe_x, the TRPL measurements were performed. The PL decay times were focused on the peak position of each PL spectrum. At low temperature, the PL is dominated by the radiative recombination and the nonradiative recombination could be neglected. Therefore, the measured decay time can be regarded as the radiative recombination lifetime of the bound excitons. The PL decay profiles were plotted in the logarithmic scale, shown in Fig. 3-7. As the Te content increases from 0.5 to 10 %, the PL decay lifetime increases. However, when Te composition is further increased, the PL decay lifetime decreases monotonically.

Additionally, the decay profiles are clearly non-single exponential decay. It could be due to different emission origins from various isoelectronic centers, which results in multiple exponential PL decay. Therefore, all the TRPL spectra can be fitted by using stretched exponential

I(t) = I exp

0

t

^β

where β is the stretching parameter, and I(t) is the PL intensity as a function of time [17 – 25]. Stretched exponentials have been used to describe the

disordered semiconductors. Figure 3-8 presents the PL decay curves in a double logarithmic scale and fitted by stretched exponential function.

The PL decay lifetimes as a function of Te concentration are shown in Fig.

3-9. The recombination lifetime varies from a few nanoseconds to tens of nanoseconds with different Te concentration. When x is small (x＜0.100), the lifetime increases with x. As the Te content increases, the recombination lifetime increases and reaches a maximum at about 78 ns for x = 0.10. This implies that when the Te isoelectronic centers gradually evolve from Te₂ complexes to Te_n clusters, the corresponding radiative recombination lifetime increases. The Te_n clusters have deeper localized short-range potentials and thus result in a larger binding energy. Therefore, the bound exciton states are more localized in the real space, leading to the reduced electronic - hole overlap for the self - trapped excitons and excitons localized at one center tunnel over barrier to another center. As a result, the radiative recombination lifetime increase [7, 26, 27].

When x＞0.100, however, the PL decay times are found to decrease with further increase of x. It is due to the fact that when the Te_n clusters become large enough, the bound exciton states decrease (such as FWHM). Thus, the difference between the different localize states will be diminishing resulting in

decrease of recombination lifetime with x reflects the decreasing percentage of excitons that tunnel between different localize states [7, 28]. For x close to 0.420, the delocalized nature of the host band states leads to the observed short recombination lifetime.

Figure 3-10 shows the PL decay time as a function of the monitored emission energies for the three samples. The general trend is that when the monitored energy decreases, the measured decay time increases. This result is consistent with the fact that the lower energy position corresponds to larger Te_n clusters [1].

For example, for the sample of x=0.005, the high energy side of the PL contain the contribution of Te₁, the central part of the PL is dominated by Te₂ complexes, and the lower energy side comprises the PL contribution of Te_n clusters.

FIG. 3-1. Normalized PL spectra of ZnSe1−xTex at 13 K.

FIG. 3-2. The dependence of exciton binding energy on Te content.

0.0 0.2 0.4 0.6 0.8 1.0 100

150 200 250

300 ZnSe

_1-x

Te

_x

T = 13 K

Te composition x

Bin d in g Energy (meV)

FIG. 3-3. The dependence of the PL peak energy on Te content. The solid

curve is band gap energy of ZnSe1−xTex at 13 K from Ref. [11].

0.0 0.2 0.4 0.6 0.8 1.0 2.2

2.4 2.6

2.8 ZnSe

_1-x

Te

_x

T = 13 K

PL Peak Energy( eV )

Te composition x

E

_g

E

_PL

FIG. 3-4. The dependence of FWHM on Te content.

0.0 0.2 0.4 0.6 0.8 1.0 60

90 120 150 180

ZnSe

_1-x

Te

_x

T = 13 K

FWHM (meV)

Te composition x

FIG. 3-5. The PL intensity versus temperature of ZnSe0.94Te0.06 epilayer.

0.00 0.02 0.04 0.06 0.08

0

FIG. 3-6. The dependence of thermal activation energy (Ea) on Te

content.

0.0 0.2 0.4 0.6 0.8 1.0 0

30 60 90 120 150

ZnSe

_1-x

Te

_x

Te composition x

Thermal Activation Energy (meV)

FIG. 3-7. TRPL spectra of ZnSe_1-xTe_x (0.005 ≦ x ≦ 0.42) at 13 K.

0 100 200 300

x=0.420 x=0.340 x=0.230 x=0.120 x=0.100 x=0.060 x=0.050 x=0.005 ZnSe _1-x Te _x

T = 13 K

PL Intensity (arb. units)

Time(ns)

FIG. 3-8. Stretched exponential of ZnSe_1-xTe_x (0.005 ≦ x ≦ 0.42) at 13 K.

1 2 3 4 5 6 7 8

-2 -1 0 1 2 3

x=0.420

x=0.120 x=0.100 x=0.005

ln(time(ns)) ZnSe

_1-x

Te

_x

T = 13 K

FIG. 3-9. The dependence of PL decay lifetime as a function of x.

0.0 0.2 0.4 0.6 0.8 1.0 0

10 20 30 40 50 60 70 80 90

T=13 K ZnSe

_1-x

Te

_x

Te composition x

Decay time(ns)

FIG. 3-10. The PL spectra and the decay lifetime as a function of the

monitored emission energies for samples with x=0.005, 0.060,

and 0.340.

2.0 2.2 2.4 2.6 2.8 0 30 60 90 120

Decay time (ns)

x=0.340 x=0.060 x=0.005 T = 13 K

ZnSe

_1-x

Te

_x

Photon Energy(eV)

PL Intensity(arb. units)

Chapter 4: Conclusions

In summary, the decay dynamics of ZnSe_1-xTe_x (0.005 ≦ x ≦ 0.42) epilayers were investigated by TRPL measurements. The PL lifetime was analyzed by using the stretched exponential function. The recombination lifetime varies from a few nanoseconds to tens of nanosecond with Te concentration. As the Te content increases, the recombination lifetime increases and reaches a maximum at about 78 ns for x = 0.10. It is because the more Te content causes larger localization potential. As the Te composition further increases (x > 0.10), the recombination lifetime decreases with the increasing of Te concentration. This is due to the fact that the hybridization between the different localize states will be diminishing resulting in the formation of the hybridized states and towards the band state of ZnTe. The decrease of recombination lifetime with x reflects the decreasing percentage of excitons that tunnel between different localize states.

This tendency was supported by the dependence of the PL line width, the exciton binding energy, and thermal activation energy on Te concentration.

Reference

[1] C. S. Yang, D. Y. Hong, C. Y. Lin, W. C. Chou, C. S. Ro, W. Y.

Uen, W. H. Lan and S. L. Tu, J. Appl. Phys, 83, 2555 (1998).

[2] P. J. Klar, H. Grüning, W. Heimbrodt, J. Koch, F. Ho¨hnsdorf, W. Stolz, P. M. A. Vicente and J. Camassel, Appl. Phys Lett. 76 3439 (2000).

[3] Marco Felici, Antonio Polimeni, Mario Capizzi, Y. Nabetani, T. Okuno, K.

Aoki, T. Kato, T. Matsumoto, and T. Hirai, Appl. Phys Lett. 88 101910 (2006).

[4] X. D. Yang, Z. Y. Xu, Z. Sun, B. Q. Sun, I. K. Sou and W. K. Ge, Appl.

Phys Lett. 86 162108 (2005).

[5] D. B. Eason, Z. Yu, W. C. Hughes, W. H. Roland, C. Boney, J. W. Cook, Jr.

G. Cantwell and W. C. Harsch, Appl. Phys Lett. 66 115 (1995).

[6] Igor L. Kuskovsky, C. Tian, G. F. Neumark, J. E. Spanier, Irving P.

Herman, W.-C. Lin, S. P. Guo, and M. C. Tamargo, Phys. Rev. B 63, 155205 (2001).

[7] Qiang Fu, D. Lee, A. V. Nurmikko, L. A. Kolodziejski and R. L. Gunshor, Phys. Rev. B 39, 3173 (1989).

[9] Y. Gu, Igor L. Kuskovsky, M. van der Voort, G. F. Neumark, X. Zhou, and M. C. Tamargo, Phys. Rev. B 71, 045340 (2005).

[10] Jingbo Li and Lin-Wang Wang, Phys. Rev. B 67, 205319 (2003).

[11] M. J. S. P. Brasil, R. E. Nahory, F. S. Turco-Sandroff, H. L. Gilchrist, and R. J. Martin, Appl. Phys. Lett. 58, 2509 (1991).

[12] X. Liu, U. Bindley, Y. Sasaki, and J. K. Furdyna, J. Appl. Phys. 91,

[17] Lorenzo Pavesi and Matteo Ceschini, Phys. Rev. B 48, 17625 (1993).

[18] X. Chen, B. Henderson, and K. P. O'Donnell, Appl. Phys Lett. 60, 2672 (1992).

[20] M. Pophristic, F. H. Long, C. Tran, I. T. Ferguson, and R. F. Karlicek, Jr., J. Appl. Phys. 86, 1114 (1999).

[21] M. Dovrat, Y. Goshen, J. Jedrzejewski, I. Balberg, and A. Sa’ar, Phys. Rev.

B 69, 155311 (2004).

[22] M. Cazzanelli and L. Pavesi*, Phys. Rev. B 56, 15264 (1997).

[23] Germar Schlegel, Jolanta Bohnenberger, Inga Potapova, and Alf Mews, Phys. Rev. Lett. 88, 137401 (2002).

[24] Satoshi Tanaka, Hideki Koyama, and Nobuyoshi Koshida, Appl. Phys.

Lett. 73, 2334 (1998).

[25] Olivier Labeau, Philippe Tamarat, and Brahim Lounis, Phys. Rev. Lett.

90, 257404 (2003).

[26] C. D. Lee, H. L. Park, C. H. Chung, and S. K. Chang, Phys. Rev. B. 45, 4491 (1992).

[27] D. Lee, A. Mysyrowicz, A. V. Nurmikko, and B. J. Fitzpatrick, Phys. Rev.

Lett. 58, 1475 (1987).

[28] J. Wu, W. Walukiewicz, K. M. Yu, J. W. Ager, III, E. E. Haller, I.

在文檔中 碲硒化鋅磊晶層中的碲等電性中心之光激螢光時間解析光譜 (頁 10-0)