Potential of semiconductor nanowires for single photon sources

(1)

Potential of semiconductor nanowires for single photon sources

Jean-Christophe Harmand

a*

, Linsheng Liu

a

, Gilles Patriarche

a

, Maria Tchernycheva

a

, Nikolai

Akopian

b

, Umberto Perinetti

b

, Valery Zwiller

b

a

_{Laboratoire de Photonique et de Nanostructures, CNRS, route de Nozay,}

91460 Marcoussis, France ;

b

_{Kavli Institute of Nanoscience, Delft University of Technology, P.O. Box 5046,}

2600 GA Delft, The Netherlands

ABSTRACT

The catalyst-assisted growth of semiconductor nanowires heterostructures offers a very flexible way to design and fabricate single photon emitters. The nanowires can be positioned by organizing the catalyst prior to growth. Single quantum dots can be formed in the core of single nanowires which can then be easily isolated and addressed to generate single photons. Diameter and height of the dots can be controlled and their emission wavelength can be tuned at the optical telecommunication wavelengths by the material composition. The final morphology of a wire can be shaped by the radial/axial growth ratio, offering the possibility to form single mode optical waveguides with a tapered end for efficient photon collection.

Keywords: Semiconductor nanowires, quantum dots, single photon source, photonic wire

1. INTRODUCTION

Nanowires (NWs) are defined as structures that have an unconstrained longitudinal size while lateral size is restricted to tens of nanometers or less. Typical NWs exhibit aspect ratios (length-to-width ratio) of 100 or more. As such, they are often referred to as one-dimensional (1D) materials. There are many applications where NWs may become important. They could be used to interconnect or to address components into extremely small circuits. They could become the future building blocks of electronics, opto-electronics, nanoelectromechanical devices and biomolecular nanosensors. They also have potential as field-emitters. Several demonstrations of devices based on semiconductor NWs have already been demonstrated [1]: resonant tunneling diodes, photodetectors as well as light emitting diodes and lasers.

Semiconductor quantum dots (QDs) are well known sources of single [2, 3] and entangled photons [4, 5] and are naturally integrated with modern semiconductor electronics. The most popular fabrication technique is to exploit the Stranski Kratanow growth where strain is the driving force leading to the self-assembling of QDs [6, 7]. Although very successful, this method presents several limitations. In lattice-matched material systems, QDs cannot be formed by the SK mechanism. The shape and the size of SK dots strongly depend on the value of the strain. Before QD formation, a wetting layer covers the whole substrate surface uniformly. This wetting layer, which subsists even after the QD capping, is a pathway for carriers to escape more easily from the QD when the temperature increases, drastically degrading the QD properties at room temperature [8]. The exact size and location of SK QDs are not predetermined due to the self-forming nature which makes their spatial distribution and size dispersion unavoidable. To solve these problems, substrate patterning or substrate masking have been attempted to localize the nucleation of the QDs [9]. However, the QDs must be formed directly on these processed surfaces where residual contamination may play a role in degrading the optical properties of the QDs. Finally, addressing a single SK QD electrically is a real technological challenge. All the attempts reported until now suffer from a poor carrier injection efficiency. These limitations certainly hinder the use of SK QDs for optical applications.

*jean-christophe.harmand@lpn.cnrs.fr; phone 33 1 69 63 60 81; fax 33 1 69 63 60 06 Invited Paper

Quantum Sensing and Nanophotonic Devices VI, edited by Manijeh Razeghi, Rengarajan Sudharsanan, Gail J. Brown, Proc. of SPIE Vol. 7222, 722219 · © 2009 SPIE · CCC code: 0277-786X/09/$18 · doi: 10.1117/12.810929

Proc. of SPIE Vol. 7222 722219-1

(2)

In comparison, the fabrication of QDs in semiconducting NWs does not need strained material systems and it brings additional unique features. It provides easiness in the manipulation of a single QD; this single dot can then be electrically addressed via an inherent one-dimensional channel for charge carriers: the NW itself. In addition, the natural alignment of vertically stacked QDs allows to design QD molecules. Furthermore, the unprecedented material and design freedom makes them very attractive for novel opto-electronic devices and quantum information processing. Recently, QDs embedded in NWs have been investigated and their atom-like behavior has been demonstrated by the observation of photon anti-bunching [10].

The bottom-up approach based on the catalytic growth is of particular interest for NW fabrication, because it avoids the top-down etching processes that can damage the crystalline structure, and whose control is not satisfactory at the nm scale. Conventional epitaxial techniques like chemical vapor deposition or molecular beam epitaxy (MBE) have demonstrated their ability to grow NWs with the pre-deposition of metallic particles onto the substrate, further acting as catalyst. This method has successfully been applied to synthesize NWs of almost all III-V semiconductors. Big effort has been devoted to the fabrication of NWs of In(As)P, which is a key material for high-speed electronics [11, 12] and for near infrared light emission/detection applications [13, 14, 15]. The band gap of InAsP can be tailored from 3.5 µm to 0.92 µm by adjusting the alloy composition and, in particular, can cover the 1.3-1.55 µm wavelength range of technological importance for optical telecommunications

In this paper, we briefly present the principles of NW growth focusing on Au-assisted MBE. We point out some mechanisms which are of particular importance for the fabrication of a single photon emitter in a photonic wire: heterostructure formation, axial and radial growth, control of the wire position, control of its morphology. More particularly, we present our recent progress in the fabrication of InAsP QDs inside InP NWs. We illustrate the advantage and the flexibility of this approach to obtain single photon sources.

2. NANOWIRE GROWTH

NWs can be fabricated by a bottom-up approach based on the Vapour-Liquid-Solid (VLS) synthesis method, firstly demonstrated at the micrometer scale in 1964 [16]. Since that time, this method has been investigated, analyzed and improved by several groups [12,17,18,19,20]. The VLS method can be implemented with several growth techniques such as vapor phase epitaxy (VPE) [16,17], laser ablation [19], metal-organic vapor phase epitaxy (MOVPE) [18], chemical beam epitaxy (CBE) [12], and molecular beam epitaxy (MBE) [20]. NWs of Si/Ge, III-V or II-VI compounds have been fabricated. The process starts with the deposition of metal particles, usually nanometer sized gold particles, on the substrate before growth. These nano-particles can act as chemical or physical catalysts. At growth temperatures, they generally form liquid droplets with the semiconductor constituents.

In this work we have used MBE to grow III-V NWs on (111) B oriented substrates using Au as a catalyst. The substrate surface was systematically deoxidised and buffer layers were grown prior to catalyst deposition. Au was evaporated directly in the growth chamber. An amount of Au equivalent to 1 nm layer or less was deposited on the clean surface of the buffer layer, the substrate being maintained at the growth temperature. This procedure resulted in the formation of droplets containing Au alloyed with the substrate constituents. The resulting morphology of NW ensembles was investigated by scanning electron microscopy (SEM). The crystallographic structure was analyzed by transmission electron microscopy (TEM) equipped with an energy dispersive X-ray (EDX) spectrometer for composition measurements. For TEM characterization, samples were thinned by mechanical polishing and Ar ion milling. Alternatively, to observe single NWs by TEM, they were separated from their substrate and picked up by touching the substrate with Cu TEM grid with a lacey carbon film. NWs were also dispersed on thin SiN membrane allowing the dual analysis of a single wire by TEM and micro-photoluminescence (micro-PL).

Figure 1 shows typical droplets obtained after Au deposition at 550°C on a GaAs surface. In this particular case, the droplet diameter was about 10 nm. Post-growth analysis by EDX revealed that, at room temperature, most of the droplets were composed of Au7Ga2 compound with no detectable As [21]. This observation is consistent with the high solubility of Ga and the low solubility of As in Au. More generally, when III-V compounds are exposed to Au at sufficiently high temperature, eutectic alloys are formed between Au and the group III elements, whilst the concentration of group V

(3)

500nm

5.0kV 10.2mm x70.Ok SE(U) 4/26/2007 6.0kv 10 4mm x6.00k SE(LJLAO) 611612006

elements in the liquid phase remains low. When the sample with catalyst droplets is exposed to a supersaturated vapor flow of source materials, the concentration of these constituents in the liquid phase increases until they condensate at the droplet/substrate interface: the NWs grow. The chemical catalyst effect consists in the preferential decomposition of precursor materials, such as organo-metallics used in MOVPE, at the surface of the metal droplets. The growth rate is then locally enhanced because free constituents are more abundant near the droplets [22]. This is the dominant mechanism for NW grown by MOVPE. However, this effect does not operate in MBE where elemental constituents are supplied to the sample. In that case, the faster nucleation rate at the droplet/substrate interface results from a physical catalytic effect: the periphery of the droplet, boundary between the 3 phases, vapor, liquid and solid, presents a lower energy barrier to the nucleation [23].

Fig. 1: Droplets of Au7Ga2 formed by Au deposition on a GaAs (111)B surface. The dashed line points out that few GaAs monolayers were etched from the GaAs surface. The corresponding Ga was supplied to the droplets. The observation was done at room temperature by transmission electron microscopy (TEM) .

This effect also leads to NW formation. In both cases, the final NW length can be adjusted by controlling the growth time, while the diameter is set by the catalyst particle size. GaAs NWs grown by MBE from self-assembled catalyst droplets are shown in figure 2a. In order to control the position of the NWs on their substrate, the catalyst particules must be organized. This can be done by processing the sample prior to growth. Figure 2b shows an example of an array of NWs obtained with the following process: a GaAs substrate was coated with a resist mask which was patterned with small apertures written by electron-beam lithography. A gold film was deposited and lifted off. The sample was cleaned under an atomic hydrogen flux before GaAs growth. This approach generates regular NWs with controlled positions. It should be implemented for the future developments of NW-based devices.

Fig. 2: GaAs nanowires grown by Au-assisted MBE: a) from self-assembled Au droplets; b) from an organized array of

Au particles defined by electron-beam lithography.

(4)

VA

Heterostructure of alternating materials can be created along the NW by switching sources during growth. In this respect, switching group V rather than group III sources is preferable in Au-assisted growth. As mentioned before, this is because Au droplets mainly alloy with group III elements whilst the solubility of group V elements remains very low. Therefore, forming abrupt interfaces between two different compounds is more straightforward if they contain the same group III constituent. As a matter of fact, the realisation of axial heterostructures by VLS has been most successful in systems like GaP/GaAsP [24] or InP/InPAs [12,25]. Another reason is found when comparing the range of temperature for which NWs of different III-V binary compounds can grow. In MBE, the suitable temperatures for growing InAs and InP are very similar, ranging from 330°C to 440°C typically; for GaAs and GaP, the optimal temperatures range from 420°C to 620°C. It is remarkable that the optimal temperature is fixed by the type of group III element (In or Ga), but it is not significantly affected by the group V element (P or As). Moreover there is only a small overlap between the ranges suitable for Ga-based or In-based compounds. This is not favorable to the fabrication of GaAs/InAs heterostructures for instance.

NW geometry offers further original possibilities to fabricate heterostructures: growth conditions can be varied in order to switch from axial growth, promoted by the catalytic effects explained above, to radial growth where material nucleates on the NW sidewalls. This process results in core-shell heterostructures [26] of particular interest to shield an active core region by a high bandgap shell material. The switch from axial growth to radial growth is illustrated in figure 3 in the case of MBE grown InP NWs.

Fig. 3: a) InP nanowires grown at 420°C. The growth is purely axial and corresponds to the VLS mechanism as

indicated by the perfect cylindrical shape of the nanowires. The nanowire diameter is close to the diameter of the catalyst droplet. b) InP nanowires grown at lower temperature (380°C). Radial growth occurs in parallel to the axial growth. The diameter of the lower part of the nanowire is larger than the diameter of the catalyst droplet. c) schematic of the axial and radial growth.

An additional advantage of NWs over conventional two-dimensional growth is the flexibility in structuring highly lattice-mismatched material systems: because of the very small diameter and the free sidewalls of the NW, the strain can relax elastically without introducing dislocations at the mismatched interfaces, increasing the choice of material combinations as compared to two dimensional growth [27]. This advantage may however be limited in radial heterostructures where the core can be strained by the shell. It is worth noting that the growth of any III-V NWs can be easily transferred to Si or other substrates: their crystalline quality will be preserved despite a large lattice mismatch with the substrate on which they are grown.

(5)

500 nm

3. InP/InAsP HETEROSTRUCTURE FORMATION

For InP/InAsP/InP heterostructure NW formation, we used InP (111)B substrates with self-assembled Au catalyst droplets prepared as described previously. The MBE system was equipped with solid sources supplying In atoms, and As and P cracking sources to produce dimers. The nominal growth rate, i.e. the growth rate on a clean and Au-free InP surface, was 0.2 nm/s. The first growth stage was realized at relatively high temperature (420°C) in order to favor the formation of an ideal axial heterostructure. During this stage the lower part of the NW was formed with 15 min of InP growth, then the As source was opened for several seconds in order to form an InAsP segment and the first stage was completed with 2 min deposition of InP. In the second growth stage, the temperature was lowered to 380°C in order to promote the radial growth of InP. As a result, InP shell was formed around the InAsP segment inserted during the axial growth stage. The shell thickness was adjusted with the growth time of this second stage. Typically, for 30 min growth, InP shells of 100 nm were obtained. Figure 4 shows TEM images of a NW grown with such conditions. In this wire, the size of the InAsP segment was intended to be small enough to produce three-dimensional carrier confinement.

Fig. 4: Transmission electron microscope images of a single InP wire containing a single cylindrical InPAs QD. a)

The wire is 4 µm long with a 150 nm diameter. The end of the wire is tapered. b) The InAsP QD can be seen in the core of the wire. The contrast is due to strain. Near the wire edges, fringes can be observed and correspond to the variable thickness that electrons have to cross in these areas. C) High resolution image showing the cylindrical InAsP QD. The crystalline structure is wurtzite. The dot has a diameter of about 9 nm and a length of 16 nm.

(6)

Wavelength (pm) 1.6 1.4 1.2 1 0.8 -- --As/P=1 As/P=1.5

/

\

0.8 1.0 1.2 1.4 Energy (eV)

The total length of the wire is about 4 µm and the diameter is 150 nm in the lower part. The last portion of the wire is tapered over 800 nm length, the diameter varying from 150 nm down to the catalyst diameter. The extension of this tapered end is related to the radial to axial growth ratio which can be adjusted by the temperature on the second growth stage. The short InAsP insertion is observed in the core of the NW. It is a quantum-dot-like cylinder of 16 nm length and 9 nm diameter. This diameter is close to the size of the catalyst particle, as expected for axial growth by VLS. On the contrary, the much larger diameter of the lower part confirms that the InP shell around the InAsP dot was formed by radial growth. At the vicinity of the dot, the crystalline structure is perfect. As commonly observed in III-V NWs, the dominant crystalline phase is wurtzite [28,29]. Some stacking faults were observed in the rest of the NW. Stacking faults and changes in crystal structure from sphalerite to wurtzite are widely observed in III-V NWs. They represent one of the major problem whose origin is being clarified [23]. The supersaturations of the III-V constituents in the catalyst phase during growth are a key factor. In the sample presented in Fig.4, the growth conditions of the InAsP dot were favorable to prevent the occurrence of stacking faults near the heterostructure.

4. OPTICAL PROPERTIES

Room temperature photoluminescence (PL) of NWs was investigated using the following setup. The sample was excited at normal incidence with 532 nm line of frequency doubled NdYag laser at low excitation density (~0.1 W×cm-2). The luminescence signal was dispersed with Jobin Yvon spectrometer and detected with nitrogen cooled Ge photodetector. Macro-PL measurements were performed on NW ensembles on their substrate. Spectra are presented in figure 5. The corresponding samples have InAsP insertions in InP wires. During the formation of the InAsP insertions, the As to P flux ratio was varied from sample to sample to investigate different alloy compositions. The high energy peak, situated around 1.35 eV corresponds to the luminescence of InP. The line-width is broadened by the presence of mixed sphalerite and wurzite InP crystal phases in the NWs. Indeed these two crystal phases have been reported to have a bandgap difference of 60 to 80 meV [30]. The low energy peak is ascribed to the luminescence of the embedded InAsP segments. This peak has an inhomogeneous broadening of 100 – 150 meV, which is related to the wire-to-wire composition fluctuations of the InAsP alloy. Indeed, the diameter of the catalyst particles has some impact on the growth kinetics. Different diameters lead to different lengths and different compositions of the InAsP insertions.

Fig. 5: Room-temperature PL of InAsP/InP heterostructure nanowires grown with different As to P ratio. The emission

from the InAsP insertions is easily tuned between 1.2 and 1.6 µm.

(7)

a)

c) d)

1 1.15

(Hill

The present samples were grown with a relatively large distribution of particle sizes which explains the PL broadening. The two samples of figure 5 were grown by changing the As to P flux ratio from 1 to 1.5. As a consequence, the peak emission energy was tuned from 1.025 eV (1.2 µm) to 0.8 eV (1.55 µm). This demonstrates the potential of these InP/InAsP heterostructure NWs for fabrication of emitters in the telecommunication wavelength range 1.3-1.55 µm. A crucial advantage of the NWs is that they can be easily manipulated, removed from their substrate, isolated and observed in detail. This flexibility is illustrated in figure 6 which shows that optical and structural studies can be performed on the same single nanostructure. To this end, the NWs were transferred to a thin SiN membrane in order to be observed later by TEM. The membrane was then loaded in a helium cryostat and optical measurements were performed at 5 K. The light from a semiconductor based laser (l=532 nm) was focused by a microscope objective (NA=0.75) to a spot size of ~1 µm allowing to excite a single NW. The signal was collected with the same objective, dispersed with a 0.75m spectrometer and detected with a linear array of InGaAs detectors. Fig. 4a shows a micro-photoluminescence image with a white light background where bright spots correspond to the luminescence of individual (or sometimes bundles of) NWs dispersed on the membrane. A spectrum taken on a single NW is shown in Fig. 4b. TEM studies were then performed on the same NW: Fig 4c shows the NW, Fig 4d shows the InAsP/InP heterostructure at the origin of the emission shown in fig. 4b. This unique ability to combine optical studies and structural studies on the same single nanostructure enables the correlation of the photoluminescence properties to the exact crystalline structure of the emitter. This kind of study is of particular relevance for NWs that may consist of mixed segments of different crystalline structure, wurtzite and sphalerite. In this particular sample, stacking faults can be seen at the proximity of the InAsP/InP heterointerfaces. These structural characteristics can have a direct impact on the photoluminescence properties of the QD. This method provides an efficient mean to study this impact and helps to the growth optimization.

Fig. 6: (a) Micro photoluminescence image of nanowires (bright spots) transferred on a thin SiN

membrane. (b) Photoluminescence spectrum from an InAsP section in a nanowire on the membrane measured at 5 K under 532 nm excitation. (c) TEM image of the nanowire studied optically in (b); scale bar is 200 nm. (d) High resolution image of the nanowire heterostructure shown in (c); scale bar is 20 nm. The dark contrast corresponds to the InAsP insertion.

In figure 7, we show a 4.2K PL spectrum from a NW including a single InAsP QD, similar to the sample presented in figure 4. In figure 7a, the 838 nm and 874 nm emissions are respectively attributed to wurtzite and sphalerite InP, both crystalline structures being present in the NW. The narrow peak at lower energy comes from the single InAsP dot. As shown in figure 7b, this emission presents an excitation power dependence revealing a typical exciton-biexciton behavior. The peak linewidths are below 100 µeV. Similar results have been reported by Borgström et al. [10] for the

(8)

x 1O PL wavelength (nm) 12000 10000-8000 -0 6000-xl. l2W A xlO6 4000 Xl. = a 2000 A xl,O.l6V. X2, 0.08 p x5. 0.04 0 -000 002 904 906 908 PL Wavelength (nm)

GaP/GaAsP material system, with slightly broader linewidths. They verified the exciton-biexciton behavior by time resolved PL measurements. In addition, they performed photon correlation measurements. They observed photon antibunching which demonstrates the emission of single photons. Few other demonstrations of single photon emission from semi-conductor QDs formed in NWs are being reported, in CdSe/ZnSe QDs for instance [31].

Fig. 7: 4.2K PL spectra from a single nanowire including a single InAsP quantum dot. a) PL peaks are attributed

to wurtzite InP (838 nm), sphalerite InP (874 nm), InAsP QD (904 nm).b) excitation power dependence of the InAsP quantum dot emission.

5. PHOTONIC WIRE DESIGN

A single photon source does not only require the suitable emitter. The emitted photons should also be collected efficiently. Various designs have been proposed and tested to improve the collection of single photons emitted by semiconductor QDs [32]. A novel strategy was proposed recently which does not require a high-finesse cavity [33]. This approach is based on photonic wires which are defined as one-dimensional dielectric systems that could confine a single optical mode. The authors consider a single radial dipole emitting at the wavelength λ, located on the axis of a photonic wire of diameter d. Their calculations show that the spontaneous emission can be efficiently coupled to a single optical mode for d/λ close to an optimal value of 0.22. At lower diameter, the mode is mainly localized outside the wire; at larger diameter the optical confinement weakens. According to this theory, if one considers the case of a 1.55µm QD emitter located in an InP wire, the optimal wire diameter should be about 110 nm. This size is in the range of final NW diameters that can easily be obtained by radial growth, as demonstrated in section 3. Thus, spontaneous emission coupling factor above 0.95 could be obtained [33].

An other requirement is the efficient collection of the optical power emitted in the guided mode of the wire by the outside optics. A recent study has shown that it is possible to design a low divergence output beam by the introduction of a conical taper at the end of the NW [34]. The transmission to a realistic lens of NA=0.5 was found to be significantly improved (more than 50% of the total free space transmission) for opening angles of the taper below 10°. The experimental NW shown in figure 4 has a taper which was naturally formed during the growth of the InP shell around the initially cylindrical NW. The taper angle is 8°, fitting satisfactorily the condition mentioned above. Sharper taper would be easily obtained by adjusting the ratio of axial/radial growth rates which depends on substrate temperature. The next step will be to find strategies to prevent the escape of photons from the other end of the wire, i.e toward the substrate for a self-standing wire. This could be implemented by growing the NWs on a substrate which integrates a reflecting structure at the surface.

(9)

6. CONCLUSION

The results presented in this paper highlight the potential of NWs for their application to quantum information processing. Several characteristics of the NWs realized by catalyst-assisted growth are indeed very well suited to the fabrication of efficient single photon sources. Heterostructures can be formed in the core of the wire to obtain optically activeQDs, precisely positioned on the NW axis. The InAsP/InP material system is well adapted to this goal and InAsP dots can emit in the telecommunication wavelength range with sharp emission linewidth. These QDs in NWs offer important new functionalities over the well known self-assembled QDs: a single emitter is easily isolated and can be addressed electrically by doping and contacting the NW. The NW can be shaped by lateral growth in order to form in fine a cylindrical optical waveguide with a single guided-mode. Spontaneous emission of the dot is expected to couple efficiently to this single mode. Such a photonic wire is naturally ended by a conical taper presenting a small opening angle. This should allow an efficient outside collection of photons. Finally, the position of the wires on their substrate can be predetermined by organizing the catalyst prior to growth.

ACKNOWLEDGEMENTS

This work was partly supported by the SANDIE Network of Excellence of the European Commission, by the Partenariat Hubert Curien between the Netherlands and France and by Agence Nationale de la Recherche (contract FILEMON35). The electron-beam lithography used to obtain the arrays of organized NWs was realized at LPN by Giancarlo Faini and Laurent Vila. The authors want to thank them.

REFERENCES

[1]_{Y. Li, F. Qian, J. Xiang, and C. M. Lieber, “Nanowire electronic and optoelectronic devices” Mater. Today, 9(10),} 18-27 (2006).

[2]_{P. Michler, A. Kiraz, C. Becher, W. V. Schoenfeld, P. M. Petroff, Lidong Zhang, E. Hu, A. Imamoglu, “A Quantum} Dot Single-Photon Turnstile Device”, Science 290(5500), 2282-2285 (2000)

[3]_{V. Zwiller, H. Blom, P. Jonsson, N. Panev, S. Jeppesen, T. Tsegaye, E. Goobar, M.-E. Pistol, L. Samuelson, G.} Bjork, “Single quantum dots emit single photons at a time: Antibunching experiments”, Appl. Phys. Lett 78, 2476 (2001)

[4]_{N. Akopian, N. H. Lindner, E. Poem, Y. Berlatzky, J. Avron, D. Gershoni, B. D. Gerardot and P. M. Petroff,} "Entangled Photon Pairs from Semiconductor Quantum Dots", Physical Review Letters 96(13), 130501 (2006) [5]_{R. Young, M. Stevenson, et al.. "Improved fidelity of triggered entangled photons from single quantum dots." New}

Journal of Physics 8(2), 29 (2006)

[6]_{L.Goldstein, F. Glas, J. Y. Marzin, M. N. Charasse, and G. Le Roux, “Growth by molecular beam epitaxy and} characterization of InAs/GaAs strained-layer superlattices”, Appl. Phys. Lett. 47, 1099 (1985)

[7]_{J.-Y. Marzin, J.-M. Gérard, A. Izraël, D. Barrier, G. Bastard, "Photoluminescence of Single InAs Quantum Dots} Obtained by Self-Organized Growth on GaAs", Phys. Rev. Lett. 73, pp. 716-719 (1994)

[8]_{S. Moehl, L. Maingault, K. Kheng, and H. Mariette, “Enhanced carrier confinement in quantum dots by raising} wetting layer state energy”, Appl. Phys. Lett. 87(3), 033111 (2005)

[9]_{D. Bimberg; M. Grundmann; and N. N. Ledentsov, Quantum dot heterostructures Chapter 1 Chichester: Wiley} (1999)

[10]_{M.T. Borgstrom, V. Zwiller, E. Muller, A. Imamoglu; “Optically Bright Quantum Dots in Single Nanowires”; Nano} Lett 5(7); 1439-1443 (2005)

(10)

[11]_{Erik Lind, Ann I. Persson, Lars Samuelson, and Lars-Erik Wernersson, “Improved Subthreshold Slope in an InAs} Nanowire Heterostructure Field-Effect Transistor”, Nano Lett. 6(9), 1842-1846 (2006)

[12]_{M. T. Bjork, B. J. Ohlsson, T. Sass, A. I. Persson, C. Thelander, M. H. Magnusson, K. Deppert, L. R. Wallenberg,} and L. Samuelson, “One-dimensional heterostructures in semiconductor nanowhiskers”, Appl. Phys. Lett. 80, 1058 (2002)

[13]_{X. Duan, Y. Huang, Y. Cui, J. Wang, and C.M. Lieber, „Indium phosphide nanowires as building blocks for} nanoscale electronic and optoelectronic devices”, Nature 409(6816), 66-69 (2001).

[14]_{H. Pettersson, J. Tragårdh, A. I. Persson, L. Landin, D. Hessman, and L. Samuelson, “Infrared Photodetectors in} Heterostructure Nanowires”, Nano Lett., 6 (2), 229-232 (2006)

[15]_{E. D. Minot, F. Kelkensberg, M. van Kouwen, J. A. van Dam, L. P. Kouwenhoven, V. Zwiller, M. T. Borgström, O.} Wunnicke, M. A. Verheijen, and E. P. A. M. Bakkers, “Single Quantum Dot Nanowire LEDs”, Nano Lett. 7, 2, pp 367 - 371 (2007)

[16]_{R. S. Wagner and W. C. Ellis, “Vapor-liquid-solid mechanism of single crystal growth”, Appl. Phys. Lett. 4, 89} (1964)

[17]_{E.I. Givargizov, “Fundamental aspects of VLS growth”, J. Cryst. Growth 31, pp. 20-30, (1975)}

[18]_{M.Yazawa, M. Koguchi, A. Muto, M. Ozawa, and K. Hiruma “Effect of one monolayer of surface gold atoms on the} epitaxial growth of InAs nanowhiskers”, Appl. Phys. Lett. 61, 2051 (1992)

[19]_{A. M. Morales and C.M. Lieber, “A laser ablation method for the synthesis of crystalline semiconductor nanowires} “, Science 279(5348), 208-211 (1998)

[20]_{Z. H. Wu, X. Y. Mei, D. Kim, M. Blumin, and H. E. Ruda, “Growth of Au-catalyzed ordered GaAs nanowire arrays} by molecular-beam epitaxy”, Appl. Phys. Lett. 81, 5177 (2002)

[21]_{J.C. Harmand, M. Tchernycheva, G. Patriarche, L. Travers, F. Glas, G. Cirlin; "GaAs nanowires formed by} Au-assisted molecular beam epitaxy: Effect of growth temperature", J. Cryst. Growth 301, pp.853-856 (2007)

[22]_{M. T. Borgström, G. Immink, B. Ketelaars, R. Algra, E.P.A.M. Bakkers, “Synergetic nanowire growth”,Nature} Nanotechnology 2, 541 - 544 (2007)

[23]_{F. Glas, J.-C. Harmand, G. Patriarche, “Why does wurtzite form in nanowires of III-V zinc blende}

semiconductors?”, Phys. Rev. Lett. 99, p 146101 (2007)

[24]_{M.T. Borgstrom, M.A. Verheijen,G. Immink, T. de Smet, E.P.A.M Bakkers, “Interface study on heterostructured} GaP-GaAs nanowires”; Nanotechnol. 17, 16, pp. 4010-4013, (2006)

[25]_{M. Tchernycheva, G.E. Cirlin, G. Patriarche, L. Travers, V. Zwiller, U. Perinetti, J.-C. Harmand,“Growth and} characterization of InP nanowires with InAsP insertions”; Nano Lett. 7, pp.1500-1504, (2007)

[26]_{J. Xiang, W. Lu, Y. Hu, Y. Wu, H. Yan and C. M. Lieber,“Ge/Si nanowire heterostructures as high-performance} field-effect transistors”, Nature 441, 489-493 (2006)

[27]_{F. Glas, “Critical dimensions for the plastic relaxation of strained axial heterostructures in free-standing nanowires”,} Phys. Rev. B 74, 121302 (2006)

[28]_{A. I. Persson, M.W. Larsson, S. Stengstrom, B. J. Ohlsson, L. Samuelson, L. R. Wallenberg, “Solid-phase diffusion} mechanism for GaAs nanowire growth”, Nature Mater. 3(10), 678-681 (2004)

[29]_{M. Tchernycheva, L. Travers, G. Patriarche, F. Glas, J.-C. Harmand, G.E. Cirlin, V.G. Dubrovskii;"Au-assisted} molecular beam epitaxy of InAs nanowires: Growth and theoretical analysis", J. Appl. Phys. 102, 094313 (2007) [30]_{A. Mishra, L. V. Titova, T. B. Hoang, H. E. Jackson, L. M. Smith, J. M. Yarrison-Rice, Y. Kim, H. J. Joyce, Q.}

Gao, H. H. Tan, and C. Jagadish, « Polarization and temperature dependence of photoluminescence from zincblende and wurtzite InP nanowires” Appl. Phys. Lett. 91, 263104 (2007)

[31]_{S. Tatarenko; “CdSe quantum dots in ZnSe nanowire as efficient source of single photon up to 220K”, 5}th_bilateral Russian-French Workshop on Nanoscience and Nanotechnologies, Moscow, 1-2 dec 2008.

[32]_{W.L. Barnes, G. Björk, J.M. Gérard, P. Jonsson, J.A.E. Wasey, P.T. Worthing and V. Zwiller, “Solid-state single} photon sources: light collection strategies”, Eur. Phys. J. D 18, 197-210 (2002)

[33]_{Y.-R. Nowicki-Bringuier, R. Hahner, J. Claudon, G. Lecamp, P. Lalanne and J.M. Gérard, "A novel high-efficiency} single-mode single photon source", Ann. Phys. Fr. 32(2-3), 151-154 (2007)

[34]_{N. Gregersen, T. R. Nielsen, J. Claudon, J.M. Gérard, J. Mørk,; "Controlling the emission profile of a nanowire with} a conical taper", Optics Lett. 33(15), 1693-1695 (2008)

Proc. of SPIE Vol. 7222 722219-10