Dynamics of time-resolved photoluminescence in GaInNAs and GaNAsSb solar cells

  • Alexander Gubanov1Email author,

    Affiliated with

    • Ville Polojärvi1,

      Affiliated with

      • Arto Aho1,

        Affiliated with

        • Antti Tukiainen1,

          Affiliated with

          • Nikolai V Tkachenko2 and

            Affiliated with

            • Mircea Guina1

              Affiliated with

              Nanoscale Research Letters20149:80

              DOI: 10.1186/1556-276X-9-80

              Received: 18 November 2013

              Accepted: 21 January 2014

              Published: 17 February 2014


              We report a time-resolved photoluminescence study for GaInNAs and GaNAsSb p-i-n bulk solar cells grown on GaAs(100). In particular, we studied the extent to which the carrier lifetime decreases with the increase of N content. Rapid thermal annealing proved to significantly increase the decay times by a factor of 10 to 12 times, for both GaInNAs and GaNAsSb heterostructures, while for the 1-eV bandgap GaNAsSb structure, grown at the same growth conditions as the GaInNAs, the photoluminescence decay time remained slightly below 100 ps after annealing; the approximately 1.15-eV GaInNAs p-i-n solar cell exhibited a lifetime as long as 900 ps.


              78.47.D; 78.55.Cr; 88.40.hj


              Solar cells Dilute nitrides GaInNAsSb Time-resolved photoluminescence Carrier lifetime


              In recent years, multijunction III-V semiconductor solar cells have experienced remarkable improvements, not only for space applications but also for terrestrial concentrated photovoltaic systems. The highest photovoltaic conversion efficiency reported so far is 44.7% and has been obtained with four junction solar cell [1]. A very promising way to further improve the performance of solar cells is to utilize dilute nitride and dilute antimonide materials, which can be grown lattice matched onto GaAs and Ge substrates [2]. These materials provide suitable absorption bands to harvest photons down to 1 eV and even below. Recently, a conversion efficiency of 44% was reported for a triple junction solar cell including a bottom junction based on GaInNAs(Sb) grown by molecular beam epitaxy (MBE) [3]. Adding antimony to ternary GaAsN to form GaAsNSb compounds can be also used to lower the bandgap beyond the 1-eV limit, serving as an alternative to quinary alloys, which are somewhat more difficult to grow due to the presence of three elements of group V [4, 5]. The drawback in using dilute nitrides/antimonides is related to challenges in material fabrication [6] and formation of defects [7, 8]. Careful growth parameter optimization and thermal annealing are known to increase the material quality and carrier lifetimes [9]. Carrier lifetime correlates with solar cell performance via the minimum diffusion length required for the carriers to travel without recombination, and it should be maximized in order to harvest efficiently the photogenerated carriers [10]. Time-resolved photoluminescence (TRPL) using up-conversion technique [11] is commonly used for estimating carrier lifetimes of optoelectronic heterostructures and has been extensively used in connection with optimization of GaInNAs heterostructures [2, 1214]. However, most of the studies have been concerned with analyses of quantum wells [15]. Studies on GaInAsN epilayers have reported a wide variety of lifetimes in the range of 70 to 740 ps [8, 16]. In this paper, we report TRPL values for bulk GaInAsN and GaNAsSb p-i-n solar cells. In particular, we focus on correlating the effects of thermal annealing and the nitrogen composition.


              The samples studied were grown on GaAs(100) substrate by MBE equipped with radio-frequency plasma source for atomic nitrogen incorporation. Their structures are presented in Figure 1. The thickness of the intrinsic region of the p-i-n solar cells grown was modified throughout the series, but other growth parameters were kept constant. The intrinsic regions of samples 1, 2, and 3 consist of lattice-matched GaInNAs with nitrogen compositions of 1%, 2%, and 3%, and were 320-, 600-, and 600-nm thick, respectively. In order to obtain lattice matching, the In composition was 2.7 times the nitrogen composition in each of the samples. Sample 4 comprised a lattice-matched GaN0.02As0.93Sb0.05 intrinsic region with a bandgap of approximately 1 eV and, unlike the other samples, had also an AlInP window layer. After growth, wafers were diced and thermally annealed. Rapid thermal annealing (RTA) treatments were done in N2 atmosphere. Sample temperature was monitored by optical pyrometer through the Si carrier wafer. In order to avoid desorption of As, the samples were protected with a GaAs proximity cap during RTA [17]. The annealing temperatures and the corresponding times for samples 1 to 3 were optimized to maximize the PL intensity [18].
              Figure 1

              Schematic sample structures for (a) samples 1, 2, 3, and (b) sample 4. The thickness of the lattice-matched N-based intrinsic regions is ranging from 300 to 1,300 nm.

              TRPL measurements were carried out with an up-conversion system [19]. For instrumentation details, see [20]. The excitation source was an 800-nm mode-locked Ti-sapphire pulsed laser, which delivered 50-fs pulses enabling a final time resolution of approximately 200 fs (FWHM). The excitation density was approximately 3 × 10-4 J/cm2, with a 20-μm diameter spot on the sample.

              The population dynamics of a single radiative level is given by a rate equation:
              dn t dt = - k × n t , http://static-content.springer.com/image/art%3A10.1186%2F1556-276X-9-80/MediaObjects/11671_2013_Article_1906_Equ1_HTML.gif
              which results in a monoexponential photoluminescence decay [21]:
              n t = A exp - t / τ decay . http://static-content.springer.com/image/art%3A10.1186%2F1556-276X-9-80/MediaObjects/11671_2013_Article_1906_Equ2_HTML.gif

              This model ignores thermalization of carriers after excitation, which is typically a very fast process and was not time-resolved in these measurements. To account for limited time resolution of the instrument, emission decays were fitted using deconvolution with the instrument response function. The monoexponential fits gave satisfactory results for all measured decays.

              Results and discussion

              Figure 2 shows the fit results for TRPL data for samples 1 to 3 measured at different wavelengths. Emission wavelength depends on the nitrogen and indium composition, as shown by lines and open points in Figure 2. The photoluminescence emission spectra appear to be rather broad, which is typical for bulk-like heterostructures. The decay time increased steadily with the wavelength, being within 400 to 600 ps for sample 1 and in 200 to 400 ps range for samples 2 and 3.
              Figure 2

              Wavelength dependences of decay time constants for samples 1-3 with GaInAsN i-region and PL intensities.

              The spectral dependence of carrier lifetime in GaInNAs can be explained in terms of interplay between the radiative recombination and hopping energy relaxation of localized excitons as described by Rubel et al. [22] and references therein. According to Takahashi et. al [23], the increased nitrogen concentration, as in samples 1 to 3, merely increases the band bowing and reduces the dipole interaction by involving more non-Γ states, which has a direct effect on the radiative lifetime when larger N concentrations are used. Moreover, nitrogen increases the density of nonradiative recombination centers in the bandgap which strongly contributes to the carrier lifetime.

              Annealing indeed increases the decay time of GaInNAs, and this is shown in Figure 3, where the as-grown sample decay time is also plotted. Lifetime increases by one order of magnitude following RTA, underlining the importance of thermal annealing for dilute nitride solar cells. Optimal annealing conditions for GaInNAs depend on the amount of nitrogen and growth parameters. Typically, good results for solar cells are obtained when annealing is performed at 750°C to 800°C for a few hundred seconds [24, 25]. This significant increase of decay time is related to reduction of nonradiative recombination and removal of defects due to thermal annealing [26, 27]. Furthermore, the decrease of decay times for the higher nitrogen content points out to the fact that that nitrogen-related defects are responsible for decreasing the carrier lifetime [13].
              Figure 3

              Decay time versus wavelength for as-grown and annealed sample 1.

              The effect of RTA was further investigated on the GaNAsSb structure. Figure 4 shows TRPL decays for sample 4 for as-grown wafer and annealing times of 300 and 1,800 s at a temperature (Tann) of 750°C. The dependences of decay time on detection wavelength are presented in Figure 5. An increase in decay time is observed when moving towards the band edge, which is similar to samples 1 to 3. The change in the τ(λ) slope upon RTA can be linked to carrier energy relaxation processes in the vicinity of the conduction band edge [22]. Although lifetime increases with annealing, it remained below 100 ps. Furthermore, sample 4 has AlInP window layer which suppresses effectively surface recombination rates. This lifetime is approximately one fourth of that for sample 3 and one half of the value obtained for the quinary GaInNAsSb [8]. Furthermore, as high as 900 ps, lifetime (not shown) was measured from an optimized GaInNAs p-i-n solar cell structure with an approximately 1.15-eV bandgap [9]. The fact that the lifetime after annealing is one order of magnitude less than for optimized GaInNAs and less than what has been published for GaInNAsSb indicates that there is still room for further optimization for GaNAsSb growth and annealing parameters.
              Figure 4

              Decay profiles for sample 4 comprising GaNAsSb measured at λ= 1,250 nm. Annealing time at Tann = 750°C was 0, 300, and 1,800 s.

              Figure 5

              Wavelength-dependent decay times τ for sample 4 with GaNAsSb i-region. Annealed at Tann = 750°C for 0, 30, and 1,800 s.


              We investigated the carrier lifetime dynamics in lattice-matched GaInNAs and GaNAsSb p-i-n solar cells using TRPL. The increase of nitrogen content decreases the carrier lifetime owing to increase of defect densities. An increase of the lifetime by at least tenfold was observed after thermal annealing of bulk GaInNAs layers. Thermal annealing was also found to affect the carrier energy relaxation process in GaNAsSb. Further growth and annealing parameter optimization is needed to improve the quality of GaNAsSb to make it an effective subjunction material in high-efficiency terrestrial and space solar cells.



              The authors acknowledge the Finnish Funding Agency for Technology and Innovation, Tekes, via projects “Solar III-V” (40120/09) and “Nextsolar” (40239/12). Alexander Gubanov and Ville Polojärvi acknowledge the National Doctoral Programme in Nanoscience (NGS-NANO). Joel Salmi and Wenxin Zhang are acknowledged for their support in sample processing.

              Authors’ Affiliations

              Optoelectronics Research Centre, Tampere University of Technology
              Department of Chemistry and Bioengineering, Tampere University of Technology


              1. World Record Solar Cell with 44.7% Efficiency. http://​www.​ise.​fraunhofer.​de/​en/​press-and-media/​press-releases/​presseinformatio​nen-2013/​world-record-solar-cell-with-44.​7-efficiency
              2. Harris JS, Kudrawiec R, Yuen HB, Bank SR, Bae HP, Wistey MA, Jackrel D, Pickett ER, Sarmiento T, Goddard LL, Lordi V, Gugov T: Development of GaInNAsSb alloys: growth, band structure, optical properties and applications. Phys Stat Sol (b) 2007, 244: 2707–2729. 10.1002/pssb.200675620View Article
              3. Green MA, Emery K, Hishikawa Y, Warta W, Dunlop ED: Solar cell efficiency tables (version 41). Prog Photovolt Res Appl 2013, 21: 1–11. 10.1002/pip.2352View Article
              4. Tan KH, Yoon SF, Loke WK, Wicaksono S, Ng TK, Lew KL, Stöhr A, Fedderwitz S, Weiβ M, Jäger D, Saadsaoud N, Dogheche E, Decoster D, Chazelas J: High responsivity GaNAsSb p-i-n photodetectors at 13μm grown by radio-frequency nitrogen plasma-assisted molecular beam epitaxy. Opt Express 2008, 16: 7720. 10.1364/OE.16.007720View Article
              5. Harmand J, Caliman A, Rao EVK, Largeau L, Ramos J, Teissier R, Travers L, Ungaro G, Theys B, Dias IFL: GaNAsSb: how does it compare with other dilute III V-nitride alloys? Semicond Sci Technol 2002, 17: 778–784. 10.1088/0268-1242/17/8/306View Article
              6. Zhang S, Wei S: Nitrogen solubility and induced defect complexes in epitaxial GaAs:N. Phys Rev Lett 2001, 86: 1789–1792. 10.1103/PhysRevLett.86.1789View Article
              7. Buyanova I: Physics and Applications of Dilute Nitrides. New York: Taylor & Francis; 2004.
              8. Jackrel DB, Bank SR, Yuen HB, Wistey MA, Harris JS, Ptak AJ, Johnston SW, Friedman DJ, Kurtz SR: Dilute nitride GaInNAs and GaInNAsSb solar cells by molecular beam epitaxy. J Appl Phys 2007, 101: 114916. 10.1063/1.2744490View Article
              9. Aho A, Tukiainen A, Polojärvi V, Korpijärvi VM, Gubanov A, Salmi J, Guina M: Lattice matched dilute nitride materials for III-V high-efficiency multi-junction solar cells: growth parameter optimization in molecular beam epitaxy. In 26th European Photovoltaic Solar Energy Conference, 5–9 September 2011; Hamburg. Edited by: Ossenbrink H. Munich: WIP; 2011:58–61.
              10. Friedman D, Geisz J, Kurtz S, Olson J: 1-eV solar cells with GaInNAs active layer. J Cryst Growth 1998, 195: 409–415. 10.1016/S0022-0248(98)00561-2View Article
              11. Shah J, Damen TC, Deveaud B, Block D: Subpicosecond luminescence spectroscopy using sum frequency generation. Appl Phys Lett 1987, 50: 1307. 10.1063/1.97890View Article
              12. Ng TK, Yoon SF, Wang SZ, Lin L, Ochiai Y, Matsusue T: Photoluminescence characterization of GaInNAs/GaAs quantum well carrier dynamics. J Appl Phys 2003, 94: 3110. 10.1063/1.1601297View Article
              13. Geisz JF, Friedman DJ: III N V semiconductors for solar photovoltaic applications. Semicond Sci Technol 2002, 17: 769–777. 10.1088/0268-1242/17/8/305View Article
              14. Kaschner A, Lüttgert T, Born H, Hoffmann A, Egorov AY, Riechert H: Recombination mechanisms in GaInNAs/GaAs multiple quantum wells. Appl Phys Lett 2001, 78: 1391. 10.1063/1.1355014View Article
              15. Baranowski M, Kudrawiec R, Syperek M, Misiewicz J, Zhao H, Sadeghi M, Wang SM: Contactless electroreflectance, photoluminescence and time-resolved photoluminescence of GaInNAs quantum wells obtained by the MBE method with N-irradiation. Semicond Sci Technol 2011, 26: 045012. 10.1088/0268-1242/26/4/045012View Article
              16. Mair RA, Lin JY, Jiang HX, Jones ED, Allerman AA, Kurtz SR: Time-resolved photoluminescence studies of In x Ga1-xAs1-yN y . Appl Phys Lett 2000, 76: 188. 10.1063/1.125698View Article
              17. Pakarinen J, Peng CS, Puustinen J, Laukkanen P, Korpijärvi V, Tukiainen A, Pessa M: Postgrowth annealing of GaInAs/GaAs and GaInAsN/GaAs quantum well samples placed in a proximity GaAs box: a simple method to improve the crystalline quality. Appl Phys Lett 2008, 92: 232105. 10.1063/1.2943157View Article
              18. Kudrawiec R, Syperek M, Latkowska M, Misiewicz J, Korpijärvi V, Laukkanen P, Pakarinen J, Dumitrescu M, Guina M, Pessa M: Influence of non-radiative recombination on photoluminescence decay time in GaInNAs quantum wells with Ga- and In-rich environments of nitrogen atoms. J Appl Phys 2012, 111: 063514. 10.1063/1.3695457View Article
              19. Shah J: Ultrafast luminescence spectroscopy using sum frequency generation. IEEE J Quantum Electron 1988, 24: 276–288. 10.1109/3.124View Article
              20. Tkachenko NV, Rantala L, Tauber AY, Helaja J, Hynninen PH, Lemmetyinen H: Photoinduced electron transfer in phytochlorin - [60]fullerene dyads. J Am Chem Soc 1999, 121: 9378–9387. 10.1021/ja9915605View Article
              21. Syperek M, Leszczyński P, Misiewicz J, Pavelescu EM, Gilfert C, Reithmaier JP: Time-resolved photoluminescence spectroscopy of an InGaAs/GaAs quantum well-quantum dots tunnel injection structure. Appl Phys Lett 2010, 96: 011901. 10.1063/1.3280384View Article
              22. Rubel O, Stolz W, Baranovskii SD: Spectral dependence of the photoluminescence decay in disordered semiconductors. Appl Phys Lett 2007, 91: 021903. 10.1063/1.2755927View Article
              23. Takahashi M, Moto A, Tanaka S, Tanabe T, Takagishi S, Karatani K, Nakayama M, Matsuda K, Saiki T: Observation of compositional fluctuations in GaNAs alloys grown by metalorganic vapor-phase epitaxy. J Cryst Growth 2000, 221: 461–466. 10.1016/S0022-0248(00)00741-7View Article
              24. Aho A, Tukiainen A, Polojärvi V, Salmi J, Guina M: High current generation in dilute nitride solar cells grown by molecular beam epitaxy. In Proc. SPIE8620, Physics, Simulation, and Photonic Engineering of Photovoltaic Devices II. Edited by: Freundlich A, Guillemoles J. San Francisco: SPIE; 2013. doi:10.1117/12.2002972 doi:10.1117/12.2002972
              25. Aho A, Tukiainen A, Polojärvi V, Salmi J, Guina M: MBE growth of high current dilute III-V-N single and triple junction solar cells. In 27th European Photovoltaic Solar Energy Conference, 24–28 September 2012; Frankfurt. Edited by: Novak S. Munich: WIP; 2012:290–292.
              26. Kurtz S, Webb J, Gedvilas L, Friedman D, Geisz J, Olson J, King R, Joslin D, Karam N: Structural changes during annealing of GaInAsN. Appl Phys Lett 2001, 78: 748. 10.1063/1.1345819View Article
              27. Chen W, Buyanova I, Tu C, Yonezu H: Point defects in dilute nitride III-N–As and III-N–P. Phys B Condens Matter 2006, 376–377: 545–551.View Article


              © Gubanov et al.; licensee Springer. 2014

              This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://​creativecommons.​org/​licenses/​by/​2.​0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.