event.itmo.infoevent.itmo.info/images/pages/45/nsp16june27.pdf · physics and technology of...

Physics and technology of nanostructured materials for photonic applicationshas become an important area of research that is poised to transform ongoingtechnologies and ultimately our everyday life in the future. Synthesis, modelingand characterization of such materials help to better understand the fundamentalphysics at the nanoscale.

The main goal of this summer school is to review recent achievements invarious fields related to advanced photonic nanostructures and provide highlevel courses and training for early stage researchers.

We sincerely hope that you will enjoy the scientific sessions as well as thebeautiful surroundings of Saint Petersburg — the “cultural capital” of Russia —and Peterhof, the former summer residence of the Russian emperors, in themysterious season of White Nights.

Vladimir G. Dubrovskii and Frank GlasConference co-chairs

ProgramTuesday, June 28 Holiday Inn Hotel, St Petersburg

08:45–09:00 Opening Remarks from Session Chairs V. Dubrovskii and F. Glas09:00–11:00 Oral Session • SPb1 Chair: F. Glas09:00–09:30 i-SPb1 Interface dynamics and crystal phase switching in GaAs nanowires F. Ross09:30–10:00 i-SPb2 Semiconductor nanostructures for lasers and optoelectronics applications C. Jagadish10:00–10:30 i-SPb3 Growth of organized III–V nano structures for quantum technology and energy applications

A. Fontcuberta i Morral10:30–10:45 o-SPb1 Self–catalyzed growth of GaAs nanowires and nanostructures on silicon by HVPE Z. Dong, Y. Andre,

V. Dubrovskii, C. Bougerol, G. Monier, R. Ramdani, A. Trassoudaine, C. Leroux, D. Castelluci, E. Gil10:45–11:00 o-SPb2 MBE growth and optical properties of GaN nanowires on SiC/Si(111) hybrid substrate R. Reznik, K.P. Kotlyar,

I.V. Ilkiv, S.A. Kukushkin, A.V. Osipov,I.P. Soshnikov, E.V. Nikitina, G.E. Cirlin11:00–11:30 Coffee break

11:30–13:30 Oral Session • SPb2 Chair: F. Ross11:30–12:00 i-SPb4 Photonic wires and trumpets: an attractive novel platform for quantum optoelectronic devices J.-M. Gerard12:00–12:30 i-SPb5 Recent progress on patterned Ga–assisted growth of GaAs nanowires for optoelectronic applications

R.R. LaPierre, J. Boulanger, A. Chia, M. Leyden, S. Yazdi, T. Kasama, M. Aagesen, H. Tavakoli Dastjerdi12:30–13:00 i-SPb6 Gate–controlled plasmonics in single nanostructures L. Sorba, F. Rossella, A. Arcangeli, J. Xu, D. Ercolani,

A. Tredicucci, F. Beltram, S. Roddaro13:00–13:15 o-SPb3 Metal mesoscopic contact as a source of plasmons for plasmonic nanocircuitries A.V. Uskov, I.V. Smetanin,

I.E. Protsenko, J.B. Khurgin, M. Buret, A. Bouhelier13:15–13:30 o-SPb4 Phase and amplitude modulations of THz waves in carbon-based derivatives M. Irfan, J.-H. Yim, Y.-D. Jho13:30–15:00 Lunch

15:00–16:30 Oral Session • SPb3 Chair: L. Sorba15:00–15:15 o-SPb5 High-sensitivity side-coupled symmetric-shaft-shape photonic crystal sensor arrays Z. Fu, J. Zhou, L. Huang,

F. Sun, H. Tian15:15–15:30 o-SPb6 Laser formation of the metal-carbon islands thin films for optical application A. Kucherik, A. Antipov,

S. Arakelian, S. Kutrovskaya, A. Osipov, T. Vartanyan, A. Povolotckaia, A. Povolotskiy, A. Manshina15:30–15:45 o-SPb7 Saturation parameters studies of carbon nanotube-based thin-film saturable absorbers for erbium fiber laser

mode-locking A.A. Krylov, S.G. Sazonkin, N.R. Arutyunyan, V.V. Grebenyukov, A.S. Pozharov,D.A. Dvoretskiy, A.B. Pnev, V.E. Karasik, E.D. Obraztsova, E.M. Dianov

15:45–16:00 o-SPb8 Ferrofluid as promising magnetically controlled material for optofluidics and microstrutured fiber-based sensingA.V. Prokofiev, A.V. Varlamov, P.M. Agruzov, I.V. Pleshakov, E.E. Bibik, S.I. Stepanov, A.V. Shamray

16:00–16:15 o-SPb9 Novel hybrid materials based on various oxyquinoline organic phosphour complexes and oxyfluoride glassM.O. Anurova, C.V. Ermolaeva, O.B. Petrova, A.V. Khomyakov, A.A. Akkuzina, R.I. Avetisov, I.Ch. Avetissov,D. Mendeleev

16:15–16:30 o-SPb10 Laser correlation spectroscopy and nonlinear magnetooptic response of structures formed by nanoparticlesin magnetic fluid E.K. Nepomniashchaia, A. V. Prokofiev, E.T. Aksenov, I.V. Pleshakov, E.E. Bibik,

E.N. Velichko, Yu.I. Kuzmin16:30–18:30 Poster Session • SPb17:00–17:30 Coffee break

18:30–20:30 Transfer from St Petersburg to New Peterhof Hotel Check-in20:30–23:30 Registration and Welcome party New Peterhof Hotel

Wednesday, June 29 New Peterhof Hotel, Peterhof

08:45–09:00 Opening remarks V. Dubrovskii and F. Glas09:00–10:30 Oral Session • Fundamental Aspects of Nanostructures and Photonics Chair: V. Dubrovskii09:00–09:30 i-Wed1 Optical antennas; spontaneous emission faster than stimulated emission E. Yablonovitch09:30–10:00 i-Wed2 Nucleation statistics in nanowires from single nano-objects to ensemble length distributions F. Glas09:30–10:00 i-Wed3 Nanostructured semiconductors for thermoelectric applications: a theoretical perspective P. Kratzer10:30–11:00 Coffee break

11:00–13:00 Oral Session • Growth of Nanostructures Chair: M. Tchernycheva11:00–11:30 i-Wed4 InP based nanoflags and core–shell nanowires D. Ritter11:30–12:00 i-Wed5 Growth of high aspect ratio semiconductor (nano)structures: back to basics of crystallogenesis E. Gil12:00–12:30 i-Wed6 Thermodynamics of VLS growth J. Johansson, M. Ghasemia12:30–13:00 i-Wed7 Nanostructured silicon carbide on silicon: a new material for photonics and optoelectronics A. Osipov,

S. Kukushkin13:00–14:30 Lunch14:30–17:30 Excursion Peterhof Grand Palace and Lower Park Refreshments after

17:30–19:30 Oral Session • Widegap Nanostructures Chair: E. Yablonovitch17:30–18:00 i-Wed8 III-nitride based photonic crystals: recent achievements and prospects R. Butte18:00–18:30 i-Wed9 Optical properties of III-nitride nanowires: early expectations, current knowledge B. Gayral18:30–19:00 i-Wed10 Nitride nanowires for light emission: from single wire properties to flexible LEDs M. Tchernycheva,

N. Guan, X. Dai, A. Messanvi, H. Zhang, F. Bayle, V. Neplokh, V. Piazza, F.H. Julien, L. Rigutti,A. Babichev, C. Bougerol, G. Jacopin, J. Eymery and C. Durand

19:00–19:30 i-Wed11 AlGaN nanoheterostructures for mid-UV photonics S. Ivanov

Thursday, June 30 New Peterhof Hotel, Peterhof

09:00–11:00 Oral Session • Characterization of Nanostructures Chair: R. LaPierre09:00–09:30 i-Thu1 In situ microscopy for growing nanostructures and measuring structure/property relationships F. Ross09:30–10:00 i-Thu Strategies for building nanostructures in nanowires F. Panciera10:00–10:30 i-Thu3 Making defective shell in III–V core-shell nanowires: how and why? B. Grandidier10:30–11:00 i-Thu4 Title to be announced later V. Zwiller11:00–11:30 Coffee break

11:30–13:30 Oral Session • Nanostructures for Optoelectronic Devices Chair: V. Zwiller11:30–12:00 i-Thu5 Semiconductor nanowires for optoelectronics and energy applications C. Jagadish12:00–12:30 i-Thu6 Photonic wires and trumpets: an attractive novel platform for quantum optoelectronic devices J.-M. Gerard12:30–13:00 i-Thu7 Growth of III–V nanowires for infrared sensors R. LaPierre13:00–13:30 i-Thu8 Gas sensors based on nanomaterials and VLSI technology Z. Tang, X. Chen13:30–15:00 Lunch

15:00–17:00 Oral Session • Nanomaterials for Terahertz Applications Chair: D. Zeze15:00–15:30 i-Thu9 Metamaterials for THz detection Y. Todorov15:30–16:00 i-Thu10 Nanostructured materials for terahertz systems A. Gallant16:00–16:30 i-Thu11 Carbon-based nanostructures for THz applications M. Portnoi16:30–17:00 i-Thu12 Title to be announced later P. Kuzel17:00–17:30 Coffee break

17:30–18:30 Poster Presentations Chair: F. Ross18:30–20:30 Poster Session (with refreshments)

Friday, July 1 New Peterhof Hotel, Peterhof

08:30–10:30 Oral Session • Nanostructure Science and Technology I Chair: A. Gallant09:00–09:30 i-Fri1 Nanoimprint lithography as enabler of large-scale fabrication of nanostructures for photonics I. Maximov,

M. Graczyk, M. Heurlin, R.J. Jam, N. Nilsson, G. Otnes, M.T. Borgstrom09:30–10:00 i-Fri2 Integration of GaAs nanowires in capacitive pressure sensing D. Zeze, A. Chandramohan, N. Sibirev,

G. Cirlin, V. Dubrovskii, B. Mendis, M. Petty, A. Gallant10:00–10:30 i-Fri3 HVPE of III-Nitride nanostructures and nanowires towards optoelectronic devices Y. Andre, E. Roche,

Z. Dong, G. Avit, V. Dubrovskii, C. Bougerol, D. Castelluci, E. Gil, J. Leymarie, F. Medard,G. Monier, F. Reveret, A. Trassoudaine

10:30–11:00 Coffee break11:00–12:30 Oral Session • Semiconductor nanowires Chair: J. Johansson11:00–11:30 i-Fri4 Advances in nanowire-based quantum dots for nanolasers and single photon emitters Y. Arakawa11:30–12:00 i-Fri5 Strategies for narrowing the size distributions of III–V nanowires V.G. Dubrovskii12:00–12:30 i-Fri6 Quantum dot inside nanowire: GaAs in AlGaAs case G. Cirlin, N. Akopian

iv Oral Sessions

12:30–13:30 Lunch

13:30–17:30 Bus trip to Kronstadt and Lomonosov

17:30–18:30 Poster Presentations (3 min each) Chair: Y. Arakawa18:30–20:30 Poster Session (with refreshments)21:00–24:00 Conference Dinner and Poster Award Ceremony

Saturday, July 2 New Peterhof Hotel, Peterhof

09:00–11:00 Hotel check-out and free time11:00–13:00 Oral Session • Nanostructure Science and Technology II Chair: E. Gil11:00–11:30 i-Sat1 Optical characteristics of oxide nanowire X. Chen11:30–12:00 i-Sat2 Independence of nanowire length distribution on the initial growth conditions N.V. Sibirev, Y. Berdnikov,

V.G. Dubrovskii12:00–12:30 i-Sat3 A compact photometer based on metal-waveguide-capillary: application to detecting glucose of nanomolar

concentration H. Huang, M. Bai, J. Hao, J. Zhang, H.Wu, B. Qu12:30–13:00 i-Sat4 Density functional theory calculations on layered materials S. Clark13:00–13:30 Closing remarks13:30–15:00 Lunch

List of Poster Presentations

Poster Session SPb,Tuesday, June 28 Holiday Inn Hotel, St Petersburg, 16:30–18:30

p-SPb1 Precision uv vacuum spectral reflectivity test system Y. Jiang, S. Xup-SPb2 THz-wave gain in asymmetric grapheme-SiC hyperbolic metamaterial O.N. Kozina, L.A. Melnikov, A.S. Zotkina,

I.S. Nefedovp-SPb3 Laser-assisted deposition of the bimetal thin films with pre-difined optical and electrical properties S. Kutrovskaya,

A. Antipov, S. Arakelian, A. Kucherik, A. Osipov, T. Vartanyan, A. Istratov, and T. Itinap-SPb4 Search of optimal conditions of Nd:Y2O3 nanopowder synthesis by using a powerful fiber ytterbium laser

G.S. Evtushenko, V.V. Lisenkov, V.V. Osipov, V.V. Platonov, A.V. Podkin, A.V. Spirina, E.V. Tikhonov, M.V. Trigub,K.V. Fedorov

p-SPb5 The influence of the dipole-dipole interaction on the radiative properties of point-like impurity centers in Fabry–Perotmicrocavity A.S. Kuraptsev, I.M. Sokolov

p-SPb6 Light-matter coupling in nonideal array of coupled microresonators with quantum dots V.V. Rumyantsev, S.A. Fedorovp-SPb7 Temperature dependent optical properties of the titanium nitride broadband perfect absorber J. Wang, M. Zhu and J. Shaop-SPb8 Matrix photoreceiver based on carbon nanotubes for control laser radiation E.V. Blagov,A.Yu. Gerasimenko,A.A. Dudin,

L.P. Ichkitidze, E.P. Kitsyuk, A.P. Orlov, A.A. Pavlov, A.A. Polokhin, Yu.P. Shamanp-SPb9 IR and raman spectroscopy of biocomposite with carbone nanotubes A.A. Polohin, L.P. Ichkitidze, A.A. Pavlov,

Yu.P. Shaman, A.Yu. Gerasimenkop-SPb10 The copper nanostructures produced by in situ laser synthesis reveal catalytic activity D.I. Gordeychuk, M.S. Panov,

I.I. Tumkin, A.G. Kuzmin, V.A. Kochemirovsky, I.A. Balovap-SPb11 The nanostructured membrane investigation by optical methods A.A. Mikhaylina, A.V. Prikhodko, O.I. Konkov,

N.N. Rozhkovap-SPb12 The electric-dipole transitions in an emitter K.K. Pukhovp-SPb13 Quantum dots luminescence in the photonic cristal fibers modified with polymer layers S.A. Pidenko, S.D. Bondarenko,

A.A.Chibrova, A.A. Shuvalov, N.A. Burmistrova, Y.S. Skibina, I.Y. Goryachevap-SPb14 NO2 gas sensor based on Au-tZnPc-OH Langmuir–Blodgett thin film D.M. Krichevsky, A.V. Zasedatelev, A.Yu. Tolbin,

T.V. Dubinina, V.I. Krasovskii, A.B. Karpop-SPb15 Eu3+-doped transparent lead fluoroborate glass-ceramics T.S. Sevostjanova, E.V. Zhukova, A.V. Khomyakov,

O.B. Petrovap-SPb16 Yb3+-doped glasses and glass ceramics based on Bi2O3 and GeO2 in different proportions I.V. Stepanova,

A.V. Khomyakovp-SPb17 Synthesis condition influence on stability of metal-organic phosphor based on 8-hydroxyquinoline A.A. Akkuzina,

A.V. Khomyakov, R.I. Avetisov, I.Ch. Avetissovp-SPb18 Synthesis and study of efficient up-conversion luminophores based M1−x−yYbxEryF2+x+y (M = Ca, Ba)

for biomedical applications M.N. Mayakova, E.O. Solovyeva, R.G. Vahrenev, S.V. Kuznetsov, D.V. Pominova,A.V. Ryabova, V.V. Voronov, P.P. Fedorov

Saturday, July 2 v

p-SPb19 New type of nanocomposite material for SERS N.V. Mitetelo, A.I. Maydykovskiy, S.E. Svyakhovskiy, A.A. Tepanov,A.D. Gartman, T.V. Murzina

p-SPb20 The obtaining and deposition of silicon nanoparticles: size control, luminescence in visible spectra A. Osipov,A. Kucherik, S. Kutrovskaya, A. Evlyukhin, B. Chichkov

p-SPb21 Optical properties of cyanine dyes in the nanoporous chrysotile asbestos A.A. Starovoytov, V.I. Belotitskii,Yu.A. Kumzerov, A.A. Sysoeva

p-SPb22 Novel transparent glass-ceramics based on Co:Li(Al,Ga)5O8 nanocrystals for passive Q-switching of Er lasersO.S. Dymshits, A.A. Zhilin, I.P. Alekseeva, M.Ya. Tsenter, A.M. Malyarevich, K.V. Yumashev, V.V. Vitkin, P.A. Loiko,N.A. Skoptsov, K.V. Bogdanov, I.V. Glazunov

p-SPb23 Photodesorbtion of Rb atoms from glass and sapphire surfaces P.A. Petrov, A.S. Pazgalev, T.A. Vartanyanp-SPb24 Glass-ceramics with Yb, Tm:YNbO4 nanocrystals: novel NIR-to-NIR upconversion phosphor E.V. Vilejshikova,

P.A. Loiko, O.S. Dymshits, A.A. Zhilin, I.P. Alekseeva, M.Ya. Tsenter and K.V. Yumashev

Poster Session,Thursday, June 30 New Peterhof Hotel, Peterhof, 17:30–20:30

p-Thu1 DNA–Based Semiconductor Nanowires A. Aldana, B. Horrocks, A. Houltonp-Thu2 Modelling the irreversible growth of nanostructures Y. Berdnikov, N.V. Sibirev, V.G. Dubrovskiip-Thu3 Capacitive pressure sensor with GaAs nanowire-PMMA dielectric layer A. Chandramohan, G. Cirlin, B. Mendis,

M. Petty, A. Gallant and D. Zezep-Thu4 Temperature dependent optical properties of single core-shell CdSe/ZnSe nanowire quantum dots grown along (111)B

T. Cremel, W. Lee, M. Jeannin, E. Bellet-Amalric, G. Nogues, K. Kyhm, and K. Khengp-Thu5 Terahertz emission from low-temperature grown GaAs nanowires A. Dıaz Alvarez, G. Tutuncuoglu,

T. Xu, M. Berthe, J.-P. Nys, S. Plissard, A. Fontcutberta i Morral, J.-F. Lampin, B. Grandidierp-Thu6 Chemical potentials and growth rates of gold-catalyzed ternary InGaAs nanowires J. Grecenkov, V.G. Dubrovskiip-Thu7 Flexible white light emitting diodes based on nitride nanowires and nanophosphors N. Guan, X. Dai, A. Messanvi,

H. Zhang, J. Yan, F. Julien, C. Durand, J. Eymery and M. Tchernychevap-Thu8 Influence of As flux on heterostructure formation in AlxGa1−xAs nanowires obtained via vapor-liquid-solid growth

A. Koryakin, N. V. Sibirevp-Thu9 Narrowing of diameter distribution during growth of Ga-catalyzed GaAs nanowires E. Leshchenko, M.A. Turchina,

V.G. Dubrovskii, T. Xu, A. Dıaz A lvarez, S.R. Plissard, P. Caroff, F. Glas, B. Grandidierp-Thu10 Electron beam induced current microscopy investigation of GaN nanowire arrays grown on Si substrates V. Neplokh,

A. Ali, F. H. Julien, M. Foldyna, I. Mukhin, G. Cirlin, J.-C. Harmand, N. Gogneau, M. Tchernychevap-Thu11 Investigation of GaN nanowires containing AlN/GaN multiple quantum discs V. Piazza, A. Babichev, N. Guan,

M. Morassi, V. Neplokh, P. Quach, F. Bayle, L. Largeau, F.H. Julien, J.–C. Harmand, N. Gogneau, M. Tchernychevap-Thu12Formation of (Al,Ga)As axial heterostructures in self-catalyzed nanowires G. Priante, F. Glas, G. Patriarche, K. Pantzas,

F. Oehler and J.-C. Harmandp-Thu13MBE growth and optical properties of GaN nanowires on SiC/Si(111) hybridsubstrate R.R. Reznik, K.P. Kotlyar, I.V. Ilkiv,

S.A. Kukushkin, A.V. Osipov, I.P. Soshnikov, E.V. Nikitina, G.E. Cirlinp-Thu14 SiC nanofilm on Si obtained by atoms substitution as a substrate for III–Nitride optoelectronics S.A. Kukushkin,

A.V. Osipov, R.S. Telyatnikp-Thu15 Study of initial stages of ordered GaAs NW growth in views of optimizing the yields J. Vukajlovic Plestina, W. Kim,

F. Matteini G. Tutuncuoglu, H. A. Potts and A. Fontcuberta i Morralp-Thu16 Tuning the growth of GaAs/InAs heterostructured nanowires by catalyst composition V. Zannier, D. Ercolani,

U.P. Gomes, J. David, M. Gemmi and L. Sorbap-Thu17 Nucleation and growth mechanism of self-catalyzed InAs nanowires on silicon D. Ercolani, U.P. Gomes, V. Zannier,

J. David, M. Gemmi and L. Sorbap-Thu18 Optical characterization of MBE-grown GaAs nanomembranes and related heterostructures G. Tutuncuoglu,

M. Friedl, M. de la Mata, D. Deiana, H. Potts, F. Matteini, J.-B. Leran, J. Arbiol, A. Fontcuberta i Morralp-Thu19 Ultrafast optical response in Au and Ag nanoparticles formed on silica nanowires arrays L. Tian, L. di Mario, D. Catone,

P. O’Keeffe, S. Turchini, F. Martelli

Poster Session, Friday, July 1 New Peterhof Hotel, Peterhof, 17:30–20:30

p-Fri1 Silver nanoisland films: self-assembly and patterning S. Chervinskii, I. Reduto, A. Kamenskii, A. A. Lipovskiip-Fri2 Local gate tuning of quantum rings T.P. Collier, V.A. Saroka and M.E. Portnoip-Fri3 Angular and positional dependence of Purcell effect for layered metal-dielectric structures A.R. Gubaydullin,

V.A. Mazlin, K.A. Ivanov, M.A. Kaliteevski, C. Balocco

vi Poster Sessions

p-Fri4 Artificial dielectric based antireflection layer for terahertz applications M. Hajji, D. Zeze, C. Balocco and A. J. Gallantp-Fri5 Characterization of 1D nanostructures using optical pump-terahertz probe time-domain spectroscopy P. Karlsen,

M.E. Portnoi and E. Hendryp-Fri6 Quality factor comparison of terahertz cavities formed by photonic crystal slabs A.K. Klein, D. Zeze, C. Balocco,

A.J. Gallantp-Fri7 Intense THz pulse emission from InAs-based epitaxial structures grown on InP substrates I. Nevinskas, R. Butkute,

S. Stanionyte, A. Biciunas, A. Geizutis and A. Krotkusp-Fri8 Antenna-coupled microcavity quantum infrared detectors D. Palaferri, Y. Todorov, C. Sirtorip-Fri9 Terahertz photoconductivity in silicon nanoparticles networks V. Pushkarev, H. Nemec, S. Gutsch, D. Hiller, J. Laube,

M. Zacharias, T. Ostatnicky and P. Kuzelp-Fri10 Ferroelectric epitaxial thin films for optoelectronics A. Razumnaya, Yu. Yuzyuk, A. Mikheykin, I. Lukyanchuk,

V. Mukhortovp-Fri11 Terahertz transitions in narrow-gap carbon nanotubes and graphene nanoribbons V.A. Saroka, R. Hartmann and

M.E. Portnoip-Fri12 Spin currents of exciton polaritons in microcavities with (110)-oriented quantum wells V. Shahnazaryan, S. Morina,

S. Tarasenko, I. Shelykhp-Fri13 Dynamic magnetoelectric coupling for THz devices S. Skiadopoulou, F. Borodavka, C. Kadlec, F. Kadlec, X. Bai,

B. Dkhil, M. Retuerto, Zh. Deng, M. Greenblatt and S. Kambap-Fri14 Second-harmonic generation for crystal structure characterization along GaAs nanowires M. Timofeeva, A. Bouravleuv,

G. Cirlin, M. Reig Escale, A. Sergeev, R. Grangep-Fri15 1180 nm GaInNAs quantum well based high power DBR laser diodes H. Virtanen, J. Viheriala, A.T. Aho,

V.-M. Korpijarvi, M. Koskinen, M. Dumitrescu, M. Guinap-Fri16 Modeling of topological nanostructures in ferroelectrics for THz radiation applications S. Kondovych, I. Lukyanchuk

Poster Sessions vii

Contents

Tuesday, June 28

i-SPb1 F.M. RossInterface dynamics and crystal phase switching in GaAs nanowires . . . . . . . . . . . . . . . . . . . . . . . . 1

i-SPb2 C. JagadishSemiconductor nanostructures for lasers and optoelectronics applications . . . . . . . . . . . . . . . . . . . . . 2

i-SPb3 A. Fontcuberta i MorralGrowth of organized III-V nano structures for quantum technology and energy applications . . . . . . . . . . . 3

o-SPb1 Z. Dong, Y. Andre, V. Dubrovskii, C. Bougerol, G. Monier, R. Ramdani,A. Trassoudaine, C. Leroux, D. Castelluci, E. GilSelf-catalyzed growth of GaAs nanowires and nanostructures on silicon by HVPE . . . . . . . . . . . . . . . . 4

o-SPb2 R.R. Reznik, K.P. Kotlyar, I.V. Ilkiv, S.A. Kukushkin, A.V. Osipov, I.P. Soshnikov, E.V. Nikitina, G.E. CirlinMBE growth and optical properties of GaN nanowires on SiC/Si(111) hybrid substrate . . . . . . . . . . . . . . 5

i-SPb4 J.-M. GerardPhotonic wires and trumpets: an attractive novel platform for quantum optoelectronic devices . . . . . . . . . . 6

i-SPb5 R.R. LaPierre, J. Boulanger, A. Chia, M. Leyden, S. Yazdi, T. Kasama, M. Aagesen,H. Tavakoli DastjerdiRecent progress on patterned Ga-assisted growth of GaAs nanowires for optoelectronic applications . . . . . . . 7

i-SPb6 F. Rossella, A. Arcangeli, J. Xu, D. Ercolani, A. Tredicucci, F. Beltram, S. Roddaro and L. SorbaGate-controlled plasmonics in nanostructures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

Wednesday, June 29

i-Wed2 F. GlasNucleation statistics in nanowires from single nano-objects to ensemble length distributions . . . . . . . . . . . 9

i-Wed3 P. KratzerNanostructured semiconductors for thermoelectric applications: a theoretical perspective . . . . . . . . . . . . . 10

i-Wed5 E. GilGrowth of high aspect ratio semiconductor (nano)structures: back to basics of crystallogenesis . . . . . . . . . . 11

i-Wed6 J. Johansson and M. GhasemiaThermodynamics of VLS growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

i-Wed7 S. Kukushkin, A. OsipovNanostructured silicon carbide on silicon: a new material for photonics and optoelectronics . . . . . . . . . . . 14

i-Wed8 R. ButteIII-nitride based photonic crystal: recent achievements and prospects . . . . . . . . . . . . . . . . . . . . . . . 16

i-Wed9 B. GayralOptical properties of III-nitride nanowires: early expectations, current knowledge . . . . . . . . . . . . . . . . 17

i-Wed10 M. Tchernycheva, N. Guan, X. Dai, A. Messanvi, H. Zhang, F. Bayle, V. Neplokh, V. Piazza, F.H. Julien,L. Rigutti, A. Babichev, C. Bougerol, G. Jacopin, J. Eymery and C. DurandNitride nanowires for light emission: from single wire properties to flexible LEDs . . . . . . . . . . . . . . . . 18

i-Wed11 S.V. Ivanov, V.N. Jmerik, A.A. Toropov, E.V. Lutsenko, V.I. Kozlovsky, X. Rong, X. WangAlGaN nanoheterostructures for mid-UV photonics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Thursday, June 30

i-Thu3 B. GrandidierMaking defective shell in III–V core-shell nanowires: how and why? . . . . . . . . . . . . . . . . . . . . . . . 20

i-Thu8 Z. TangGas sensor based on nanomaterials and VLSI technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

i-Thu9 Y. TodorovMetamaterials for THz detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

i-Thu11 M.E. PortnoiCarbon-based nanostructures for THz applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

i-Thu12 P. Kuzel, L. Nadvornık, P. Nemec, T. Janda, K. Olejnık, V. Novak, V. Skoromets, H. Nemec, F. Trojanek,T. Jungwirth, J. WunderlichLong-range and high-speed electron and spin transport at GaAs/AlGaAs interface . . . . . . . . . . . . . . . . 24

ix

x

Friday, July 1i-Fri1 M. Graczyk, M. Heurlin, R.J. Jam, N. Nilsson, G. Otnes, M.T. Borgstrom and I. Maximov

Nanoimprint lithography as enabler of a large-scale fabrication of nanostructures for photonics . . . . . . . . . . 25i-Fri2 A. Chandramohan, N. Sibirev, G. Cirlin, V. Dubrovskii, B. Mendis, M. Petty, A. Gallant and D. Zeze

Integration of GaAs nanowires in capacitive pressure sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . 26i-Fri3 Y. Andre, E. Roche, Z. Dong, G. Avit, V. Dubrovskii, C. Bougerol, D. Castelluci, E. Gil, J. Leymarie, F. Medard,

G. Monier, F. Reveret, A. TrassoudaineHVPE of III-Nitride nanostructures and nanowires towards optoelectronic devices . . . . . . . . . . . . . . . . 28

i-Fri5 V.G. DubrovskiiStrategies for narrowing the size distributions of III–V nanowires . . . . . . . . . . . . . . . . . . . . . . . . . 30

i-Fri6 G.E. Cirlin, N. AkopianQuantum dot inside nanowire: GaAs in AlGaAs case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Saturday, July 2i-Sat1 X. Chen, B. Geng and S. Li

Impact of side reservoir on electromigration of copper inter-connects . . . . . . . . . . . . . . . . . . . . . . . 32i-Sat2 N.V. Sibirev, Y. Berdnikov and V.G. Dubrovskii

Independence of nanowire length distribution on the initial growth conditions . . . . . . . . . . . . . . . . . . . 34i-Sat3 H. Huang, M. Bai, J. Hao, J. Zhang, H. Wu and B. Qu

A compact photometer based on metal-waveguide-capillary: application to detecting glucoseof nanomolar concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

i-Thu7 S.J. ClarkAdvances in non-local density functional theory with applications in 2D layered materials . . . . . . . . . . . . 37

Poster Session,Thursday, June 30p-Thu1 A. Aldana, B. Horrocks, A. Houlton

DNA-based semiconductor nanowires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38p-Thu2 Y. Berdnikov, N.V. Sibirev and V.G. Dubrovskii

Modelling the irreversible growth of nanostructures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39p-Thu3 A. Chandramohan, G. Cirlin, B. Mendis, M. Petty, A. Gallant and D. Zeze

Capacitive pressure sensor with GaAs nanowire-PMMA dielectric layer . . . . . . . . . . . . . . . . . . . . . 40p-Thu4 T. Cremel, W. Lee, M. Jeannin, E. Bellet-Amalric, G. Nogues, K. Kyhm and K. Kheng

Temperature dependent optical properties of single core-shell CdSe/ZnSe nanowire quantum dotsgrown along (111)B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

p-Thu5 A. Dıaz Alvarez, G. Tutuncuoglu, T. Xu, M. Berthe, J-P. Nys, S. Plissard,A. Fontcuberta-i-Morral, J-F. Lampin, B. GrandidierTerahertz emission from low-temperature grown GaAs nanowires . . . . . . . . . . . . . . . . . . . . . . . . . 42

p-Thu6 J. Grecenkov, V.G. DubrovskiiChemical potentials and growth rates of gold-catalyzed ternary InGaAs nanowires . . . . . . . . . . . . . . . . 43

p-Thu7 N. Guan, X. Dai, A. Messanvi, H. Zhang, J. Yan, F.H. Julien, C. Durand, J. Eymery and M. TchernychevaFlexible white light emitting diodes based on nitride nanowires and nanophosphors . . . . . . . . . . . . . . . . 45

p-Thu8 A. Koryakin, N. SibirevInfluence of As flux on heterostructure formation in AlxGa1−xAs nanowires obtainedvia vapor-liquid-solid growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

p-Thu9 E.D. Leshchenko, M.A. Turchina, V.G. Dubrovskii, T. Xu, A. Dıaz A lvarez, S.R. Plissard, P. Caroff, F. Glas,B. GrandidierNarrowing of diameter distribution during growth of Ga-catalyzed GaAs nanowires . . . . . . . . . . . . . . . 47

p-Thu10 V. Neplokh, A. Ali, F.H. Julien, M. Foldyna, I. Mukhin, G. Cirlin, J-C. Harmand, N. Gogneau, M. TchernychevaElectron beam induced current microscopy investigation of GaN nanowire arrays grown on Si substrates . . . . . 49

p-Thu11 V. Piazza, A. Babichev, N. Guan, M. Morassi, V. Neplokh, P. Quach, F. Bayle, L. Largeau, F.H. Julien,J.-C. Harmand, N. Gogneau, M. TchernychevaInvestigation of GaN nanowires containing AlN/GaN multiple quantum discs . . . . . . . . . . . . . . . . . . . 50

p-Thu12 G. Priante, F. Glas, G. Patriarche, K. Pantzas, F. Oehler and J-C. HarmandFormation of (Al,Ga)As axial heterostructures in self-catalyzed nanowires . . . . . . . . . . . . . . . . . . . . 52

p-Thu14 S.A. Kukushkin, A.V. Osipov, R.S. TelyatnikSiC nanofilm on Si obtained by atoms substitution as a substrate for III-Nitride optoelectronics . . . . . . . . . . 53

p-Thu15 J. Vukajlovic Plestina, W. Kim, F. Matteini, G. Tutuncuoglu, H.A. Potts and A. Fontcuberta i MorralStudy of initial stages of ordered GaAs NW growth in views of optimizing the yields . . . . . . . . . . . . . . . 55

p-Thu16 V. Zannier, D. Ercolani, U.P. Gomes, J. David, M. Gemmi and L. SorbaTuning the growth of GaAs/InAs heterostructured nanowires by catalyst composition . . . . . . . . . . . . . . . 56

xi

p-Thu17 D. Ercolani, U.P. Gomes, V. Zannier, J. David, M. Gemmi and L. SorbaNucleation and growth mechanism of self-catalyzed InAs nanowires on silicon . . . . . . . . . . . . . . . . . . 58

p-Thu18 G. Tutuncuoglu, M. Friedl, M. de la Mata, D. Deiana, H. Potts, F. Matteini, J.-B. Leran, J. Arbiol,A. Fontcuberta i MorralOptical characterization of MBE-grown GaAs nanomembranes and related heterostructures . . . . . . . . . . . 60

p-Thu19 L. Di Mario, L. Tian, D. Catone, P. O’Keeffe, S. Turchini, F. MartelliUltrafast optical response in Au and Ag nanoparticles formed on silica nanowires arrays . . . . . . . . . . . . . 62

p-Thu20 S. Skiadopoulou, F. Borodavka, C. Kadlec, F. Kadlec, X. Bai, B. Dkhil, M. Retuerto, Z. Deng, M. Greenblattand S. KambaDynamic magnetoelectric coupling for THz devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

Poster Session, Friday, July 1p-Fri1 S. Chervinskii, I. Reduto, A. Kamenskii, A.A. Lipovskii

Silver nanoisland films: self-assembly and patterning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66p-Fri2 T.P. Collier, V.A. Saroka and M.E. Portnoi

Local gate tuning of quantum rings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67p-Fri3 A.R. Gubaydullin, V.A. Mazlin, K.A. Ivanov, M.A. Kaliteevski, C. Balocco

Angular and positional dependence of Purcell effect for layered metal-dielectric structures . . . . . . . . . . . . 68p-Fri4 M. Hajji, D. Zeze, C. Balocco and A.J. Gallant

Artificial dielectric based antireflection layers for terahertz applications . . . . . . . . . . . . . . . . . . . . . . 69p-Fri5 P. Karlsen, M.E. Portnoi and E. Hendry

Characterization of 1D nanostructures using optical pump-terahertz probe time-domain spectroscopy . . . . . . 70p-Fri6 A.K. Klein, D. Zeze, C. Balocco, A.J. Gallant

Quality factor comparison of terahertz cavities formed by photonic crystal slabs . . . . . . . . . . . . . . . . . 71p-Fri7 I. Nevinskas, R. Butkute, S. Stanionyte, A. Biciunas, A. Geizutis and A. Krotkus

Intense THz pulse emission from InAs-based epitaxial structures grown on InP substrates . . . . . . . . . . . . 73p-Fri8 D. Palaferri, Y. Todorov, C. Sirtori

Antenna-coupled microcavity quantum infrared detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74p-Fri9 V. Pushkarev, H. Nemec, S. Gutsch, D. Hiller, J. Laube, M. Zacharias, T. Ostatnicky, and P. Kuzel

Terahertz photoconductivity in silicon nanoparticles networks . . . . . . . . . . . . . . . . . . . . . . . . . . . 75p-Fri10 A. Razumnaya, Y. Yuzyuk, A. Mikheykin, I. Lukyanchuk, V. Mukhortov

Ferroelectric epitaxial thin films for optoelectronics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77p-Fri11 V.A. Saroka, R.R. Hartmann and M.E. Portnoi

Terahertz transitions in narrow-gap carbon nanotubes and graphene nanoribbons . . . . . . . . . . . . . . . . . 78p-Fri12 V. Shahnazaryan, S. Morina, S. Tarasenko, I. Shelykh

Spin currents of exciton polaritons in microcavities with (110)-oriented quantum wells . . . . . . . . . . . . . . 79p-Fri13 S. Kondovych, I. Lukyanchuk

Modeling of topological nanostructures in ferroelectrics for THz radiation applications . . . . . . . . . . . . . . 80p-Fri14 M. Timofeeva, A. Bouravleuv, G. Cirlin, M. Reig Escale, A. Sergeev, R. Grange

Second-harmonic generation for crystal structure characterization along GaAs nanowires . . . . . . . . . . . . . 81p-Fri15 H. Virtanen, J. Viheriala, A.T. Aho, V.-M. Korpijarvi, M. Koskinen, M. Dumitrescu, M. Guina

1180 nm GaInNAs quantum well based high power DBR laser diodes . . . . . . . . . . . . . . . . . . . . . . . 83

International Summer School and Workshop • Nanostructures for Photonics i-SPb1St Petersburg, Russia, June 27–July 2, 2016

Interface dynamics and crystal phase switching in GaAs nanowiresF.M. RossIBM T. J. Watson Research Center, Yorktown Heights, NY, USA

Abstract. Controlled formation of non-equilibrium crystal structures is one of the most important challenges in nanowiregrowth. For III-V nanowires, the ability to switch between two polytypes, wurtzite and zincblende, offers the excitingpotential for fabricating novel optoelectronic devices based on crystal phase engineering. In order to understand themechanism that controls crystal phase, we use in situ electron microscopy to image catalytically-grown GaAs nanowiresduring growth as they are switched between polytypes by varying growth conditions. We find striking differences betweenthe growth dynamics of the two phases, including differences in interface morphology, step flow, and catalyst geometry. Weexplain the differences, and the phase selection, through a model that relates the catalyst volume, contact angle at thetrijunction, and nucleation site of each new layer. This allows us to predict the conditions under which each phase should bepreferred, and use these predictions to design GaAs heterostructured nanowires. We discuss the extent to which these resultsmay apply to crystal phase selection in other nanowire materials.

1


Semiconductor nanostructuresfor lasers and optoelectronics applicationsC. JagadishAC, FAA, FTSE, FTWAS, Research School of Physics and Engineering, The Australian National University,Canberra, ACT 2601, Australia

Abstract. Semiconductors have played an important role in the development of information and communicationstechnology, solar cells, solid state lighting. Nanostructures such as quantum wells, quantum dots and nanowires played animportant role in the development of semiconductor lasers, infrared photodetectors and solar cells. In this talk I will discussabout the synthesis of nanostructures and their characterization and device fabrication and testing. Role of plasmoniccavities in improving the quantum efficiency of nanostructures will be discussed. Strengths and weaknesses of each of thesenanostructures will be presented and future perspective will be provided.

2


Growth of organized III-V nano structuresfor quantum technology and energy applicationsA. Fontcuberta i MorralLaboratory of Semiconductor Materials, Institute of Materials, Ecole Polytechnique Federale de Lausanne, 1015Lausanne, Switzerland

Abstract. Nanowires are filamentary crystals with a tailored diameter ranging from few to ∼100 nm. The special geometryand reduced dimensions of these nanowires results in interesting optical and electrical properties and provides a greatpotential for many applications of the XXI century. In this talk we will first review the growth mechanisms of Ga-assistedgrowth of GaAs nanowires by molecular beam epitaxy. We will follow by elucidating the photonic properties of singlenanowires standing and lying on a substrate to show how they can be used for quantum science and technology and energyharvesting applications.

Semiconductor nanowires have attracted an increasing amountof attention thanks to their special properties. Especially in-teresting is the growth of III-V nanowires on silicon as theynaturally allow the integration of the functionalities of the III-V and Si platforms.

Ga-assisted growth of GaAs nanowires was initially demon-strated on GaAs substrates. Here we discuss some elements totake into account for the translation of this method to Si sub-strates. The most important aspect to consider is the surfaceenergy of the silicon oxide deposited on Si. This oxide favorsthe formation of Ga droplets that will result in GaAs nanowires.Depending on the wetting angle of the Ga droplets, nanowireswill grow oriented perpendicularly to the substrate or in a dif-ferent angle. The importance of the wetting angle of the Gadroplets at the beginning of the growth has been validated bothfor the growth of GaAs nanowires in an organized (Fig. 1) andin a self-assembly manner [1,2].

The dimensions and shape of the nanowires render themideal for a variety of applications in photonics. The needle-like shape results also in a high degree of polarization response:light absorption is very low for incoming light polarized acrossthe nanowire axis, while it is very high for light polarized alongthe nanowire axis. In this talk we will show it is possible to varythe intrinsic absorption and emission properties by coupling thenanowires to plasmonic nanostructures [3,4].

Finally we report on the use of GaAs nanowires standingon a Si substrate for solar cells. We show how nanowires areextremely good absorbers when they are standing on a substrateand how this property can be used to obtain high efficiency solarcells [5]. We also discuss how this property can be used formultiple junction solar cell applications [6] and provide someguidelines for the design of these devices.

1 µm1 µm

Fig. 1. (a) Scanning electron microscope image of a GaAs nanowirearray obtained on a (111)Si substrate [ref].

References

[1] E. Russo-Averchi, J. Vukajlovic Plestina, G. Tutuncuoglu, et al,Nano Lett. 15, 2869 (2015).

[2] F. Matteini, G. Tutuncuoglu, H. Potts, et al, Cryst. Growth &Des. 15, 3105 (2015).

[3] M. Ramezani, A. Casadei, G. Grzela, et al, Nano Lett. 15, 4889(2015).

[4] A. Casadei, E. Alarcon-Llado, F. Amaduzzi, et al, ScientificReports 5, 7651 (2015).

[5] P. Krogstrup, H.I. Jørgensen, M. Heiss, et al, Nature Photonics7, 306-310 (2013).

[6] A. Dorodnyy, E. Alarcon-Llado, V. Shklover, et al, ACS Pho-tonics 2, 1284 (2015).

3

International Summer School and Workshop • Nanostructures for Photonics o-SPb1St Petersburg, Russia, June 27–July 2, 2016

Self-catalyzed growth of GaAs nanowires and nanostructureson silicon by HVPEZ. Dong1,2, Y. Andre1,2,3, V. Dubrovskii4,5, C. Bougerol6,7, G. Monier1,2, R. Ramdani1,2,A. Trassoudaine1,2,8, C. Leroux9,10, D. Castelluci1,2, E. Gil1,2,31 Clermont Universite, Universite Blaise Pascal, Institut Pascal, 63000 Clermont-Ferrand, France2 CNRS, UMR 6602, 63178 Aubiere, France3 ITMO University, Kronverkskiy pr. 49, 197101 St Petersburg, Russia4 St Petersburg Academic University, St Petersburg, Russia5 Ioffe Institute, St Petersburg, Russia6 Univ. Grenoble Alpes, 38000 Grenoble, France7 CNRS, Institut Neel, 38042 Grenoble, France8 Institut Universitaire de Technologie, Dep. Mesures Physiques, Universute d’Auvergne, 63178 Aubiere, France9 Universite du Sud Toulon-Var, IM2NP, Bat.R, B.P.20132, 83957 La Garde Cedex, France10 CNRS, UMR 6242, 83957 La Garde Cedex, France

Hydride vapor phase epitaxy (HVPE) is the only III–V semi-conductor crystal growth process known as working close toequilibrium. GaAs nanowires (NWs) grown by gold catalyst-assisted HVPE with a high elongation rate (170 μm/h) showa constant cylinder shape over unusual length (100 μm) witha constant cubic phase regardless of the NW diameter [1–2].However, the use of gold harms the properties of semiconduc-tors by causing for example deep-level trap sites in GaAs.

We report here on the first in-situ self (Ga)-catalyzed growthof GaAs nanowires and nanostructures by HVPE on patternedand non-patterned silicon wafers. Nanowires exhibit cylindri-cal rod-like shape morphology with a mean diameter of 50 nmand are randomly distributed. The elongation rate of nanowireswas up to 30 μm/h. The nanowires grew along the [111]B di-rection. HRTEM analysis showed a zincblende structure witha low density of twin defects. A large number of nanostruc-tures competed with the nanowire growth on the silicon sub-strate as shown in Fig. 1. A discussion based on the crystalanisotropy of the facets of both nanowires and nanostructuresis proposed. A model of the first self-catalyzed growth usingGaCl III-precursor as a function of experimental parameters(temperature, III and V fluxes) is considered.

References

[1] E. Gil, et al, Nano Lett., 14, 3938–3944, (2014).[2] M.R. Ramdani, et al, Nano Lett. 10, 1836–1841 (2010).

(a) (b)

[111]

200 nm1 µm

Fig. 1. (a) Scanning electron microscope image showing a GaAsnanowire, a nano-polyhedron and a bulk GaAs structure grown si-multaneously; (b) Top view of two GaAs nano-polyhedrons grownalong the (111)B direction.

4

International Summer School and Workshop • Nanostructures for Photonics o-SPb2St Petersburg, Russia, June 27–July 2, 2016

MBE growth and optical properties of GaN nanowireson SiC/Si(111) hybrid substrateR.R. Reznik1,2,5, K.P. Kotlyar1, I.V. Ilkiv1,2, S.A. Kukushkin5,6, A.V. Osipov5,6, I.P. Soshnikov1,3,4,E.V. Nikitina1, G.E. Cirlin1,2,4,5

1 St Petersburg Academic University, St Petersburg, Russia2 Peter the Great St.Petersburg Polytechnic University, Polytechnicheskaya 29, 195251, St Petersburg, Russia3 Ioffe Institute, St Petersburg, Russia4 Institute for Analytical Instrumentation RAS, Rizhsky 26, 190103, St Petersburg, Russia5 ITMO University, Kronverkskiy pr. 49, 197101, St Petersburg, Russia6 Institute of Problems of Mechanical Engineering RAS, Bolshoj 6, 199178, St Petersburg, Russia

Abstract. The fundamental possibility of the growth of GaN nanowires by molecular-beam epitaxy on a silicon substratewith nanoscale buffer layer of silicon carbide has been demonstrated for the first time. Morphological and spectralproperties of the resulting system have been studied and compared properties of GaN nanowires on silicon substrate.

Introduction

The wide-gap nanoheterostructures based on GaN are of greatinterest for creating electronic [1] and optoelectronic [2] de-vices. Works in growing GaN layers on silicon [3] have beenvery promising recently. However, the lattice misfit of suchmaterials is 17%, which leads to the formation of defects ofdifferent nature It is known that the optoelectronic GaN baseddevices can operate for a long time without degrading despitethe high density linear defects. Nevertheless, to extend thelifetime of optoelectronic devices is necessary to increase theperfection of GaN structures.

In this work, in order to reduce the number of misfit dis-locations a nanometer (about 50 nm) buffer layer of SiC wasused. It is grown on Si by solid-phase epitaxy, which providesextremely low values of the density of misfit dislocations, sincethe difference in the lattice parameters is only 3%, and also,instead of a planar layer, growth GaN nanowires (NWs), whichcan radically reduce the density of structural defects [7]. Man-ageable synthesis of GaN NWs allow one to control their elec-tronic properties, including the degree of doping of the n- andp-type, and create ultraviolet lasers and LED.

1. Experiments

Forming of a buffer layer SiC on the substrate Si (111) wasfulfilled by the method described in [5]. Growth experimentsare carried out using Riber Compact 12 MBE setup equippedwith the effusion Ga cell and the nitrogen source. Firstly, thesubstrate was transferred into the growth chamber, and the sub-strate temperature was set at 950 ◦C for further purification.Then, the substrate temperature was lowered to 600 ◦C andwe opened gallium source for 20 seconds to form thin layerof gallium to fabricate small particles on surface of the sub-strate for growing NWs, and the substrate temperature wasraised to the growth point — 800 ◦C. After that, using nitrogensource plasma was ignited and nitrogen flow on the sample wasformed, and the gallium source was opened too, the tempera-ture of which was chosen 870 ◦C, which corresponds to growthrate 0.01 ML/s. Growth time of GaN NWs was 16 hours.

After the growth, the samples are studied by applying thescanning electron microscopy (SEM) and low-temperaturephotoluminescence (PL) techniques.

2. Results

Comparison of photoluminescence spectra of grown GaN/SiC/Si and the most successful GaN NWs structures on silicon.The figure shows that the intensity of radiation grown on SiCbuffer layer GaN NWs is more than two times higher than theintensity of the best grown on silicon structures of GaN. Thisfact leads to the conclusion that grown in this work structureshave fewer defects compared with GaN NWs on silicon sub-strate, which, in its turn, have few defects. This is caused bya smaller lattice constant mismatch between GaN and SiC ascompared with GaN and Si.

In summary, in this paper the possibility of MBE growth ofGaN nanowires on nanoscale silicon carbide buffer layer hasbeen demonstrated for the first time. Also, we have shown thatthe intensity of PL spectrum from such structures is more thantwo times brighter than PL spectrum from the best structuresof GaN NWs without a buffer layer of silicon carbide, whichindicates that these structures are more attractive for applica-tions.

References

[1] S.J. Pearton, F. Ren, Adv. Mater., 11, 1571-1580 (2000).[2] S. Nakamura, G. Fasol, New York: Springer-Verlag, Berlin

(1997).[3] I.G. Aksyanov, V.N. Bessolov, S.A. Kukushkin, Techn. Phys.

Lett., 34, 479–482 (2008).[4] V.G. Dubrovskii, G.E. Cirlin, V.M. Ustinov, Semociond., 43,

1539–1584 (2009).

5


Photonic wires and trumpets:an attractive novel platform for quantum optoelectronic devicesJ.-M. GerardCEA, Institute for nanoscience and cryogenics, Grenoble, France

Over the last 20 years, major efforts have been devoted to thetailoring of the optical properties of semiconductor emittersusing optical microcavities and photonic crystals. We have re-cently introduced photonic wires as a novel platform for quan-tum optics. I will review recent studies which demonstratean excellent control over the spontaneous emission of InAsquantum dots (QDs) embedded in vertical single- mode GaAsphotonic wires and first applications in the field of quantumoptoelectronic devices.

On the basic side, we have demonstrated a strong inhibition(x 1/16 [1]) of QD SpE in thin wires (d < λ/2n) and a nearlyperfect coupling of the SpE to the guided mode (β > 0.95 ford ∼ λ/n) in circular photonic wires [2]. The polarization ofQD SpE can also be tailored by playing with the shape of thecross section of the photonic wire. For elliptical cross sections,a strong (>90%) linear polarization oriented along the long axisof the ellipse is observed [3].

In view of practical applications, a proper engineering ofthe radiation pattern of the photonic wire is required. Wehave therefore developed novel hybrid (metal+dielectric) mir-rors displaying a high modal reflectivity, as well as integratedtip-shaped or trumpet-like adiabatic tapers, in order to reducethe divergence of the emitted beam. The recently developedphotonic trumpet (see Fig. 1) exhibits superior performancesin this context, since it ensures a perfectly Gaussian and lowNA far-field emission [4].

As a first application of SpE control in photonic wires, wehave developed single mode QD single-photon sources (SPS).Unlike microcavity-based devices, such SPS display an ex-cellent purity (g(2)(0) < 0.01) under non-resonant excitation,over the whole range of excitation powers. Furthermore, effi-ciencies exceeding 0.7 photon per pulse (within NA = 0.75)have been obtained for tip-shaped [5] as well as trumpet-like [4]SPS. Beyond these first results, photonic wires are also veryattractive for developing high efficiency sources of entangledphoton pairs or wavelength tuneable SPS, thanks to the broad-band SpE control they provide.

More generally, photonic trumpets appear as a very promis-ing template to explore and exploit in a solid-state system theunique optical properties of “one-dimensional atoms”. Possi-ble long term applications in the field of quantum informationprocessing will be discussed, including the optimal quantumcloning of single photons, using the amplification by stimulatedemission provided by a single 1D atom [6].

Finally, photonic trumpets containing a single QD consti-tute a hybrid optomechanical system, whose remarkably largecoupling between the two-level system and the mechanical de-gree of freedom, mediated by the strain, opens promising novelperspectives [7].

Au+ SiO2mirror

1 µm

adiabatictaper

anti-reflectioncoating

InAs QD

Fig. 1. Colorized scanning electron micrograph of a GaAs photonictrumpet (from [4]).

Acknowledgements

This work has been done in collaboration with J. Claudon,J. Bleuse, M. Munsch, P. Stepanov, N.S. Malik (CEA Greno-ble), N. Gregersen, J. Moerk (DTU Fotonik, Copenhagen),P. Lalanne (Institut d’Optique, Palaiseau),A.Auffeves, J.P. Poi-zat, M. Richard and coworkers at CNRS/Neel; it has been sup-ported by the IST FET European project “HANAS”.

References

[1] J. Bleuse et al, Phys. Lett. Lett. 106, 103601 (2011).[2] I. Friedler et al, Opt. Exp. 17, 2095–2110 (2009).[3] M. Munsch et al, Phys. Rev. Lett. 108, 077405 (2012).[4] M. Munsch et al, Phys. Rev. Lett. 110, 177402 (2013);

P. Stepanov et al, Appl. Phys. Lett. 107, 141106 (2015).[5] J. Claudon et al, Nature Photon. 4, 174 (2010).[6] D. Valente et al, New J. Phys. 14, 083029 (2012)

and Phys. Rev. A 86, 022333 (2012).[7] I. Yeo et al, Nature Nanotech. 9, 106 (2014).

6


Recent progress on patterned Ga-assisted growthof GaAs nanowires for optoelectronic applicationsR.R. LaPierre1, J. Boulanger1, A. Chia1, M. Leyden1, S.Yazdi2,3, T. Kasama2, M. Aagesen4, H. TavakoliDastjerdi11 Department of Engineering Physics, Centre for Emerging Device Technologies, McMaster University,Hamilton, ON, Canada, L8S 4L72 Center for Electron Nanoscopy, Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark3 Current address: Department of Materials Science and NanoEngineering, Rice University,6100 Main Street MS-325, Houston, TX 770054 Gasp Solar ApS, Gregersensvej 7, DK-2630 Taastrup, Denmark

Abstract. Semiconductor nanowires are being developed for the next generation of optoelectronic devices such as lightemitting diodes, lasers, photodetectors, photovoltaics, and transistors. The free lateral surfaces of nanowires allow elasticrelaxation of lattice misfit strain without the generation of dislocations, permitting the integration of III–V nanowires oninexpensive silicon substrates. Furthermore, nanowires permit high optical absorption due to an optical antenna effect. Theself-assisted vapor-liquid-solid method is now a well-established technique for the growth of III–V nanowires on siliconsubstrates. In this method, an array of holes in an SiO2 film is used for metal droplet collection, which seeds the growth ofnanowires. Some of the challenges associated with this technique will be illustrated, using Ga-assisted growth of GaAsnanowires on silicon as an example. A single junction core-shell GaAs nanowire solar cell on Si (111) substrates ispresented. A Ga-assisted vapor-liquid-solid growth mechanism was used for the formation of a patterned array of radialp-i-n GaAs nanowires encapsulated in AlInP passivation. Novel device fabrication utilizing facet-dependent properties tominimize passivation layer removal for electrical contacting is demonstrated. Electrical characterization and analysis of thecell is reported. The electrostatic potential distribution across a radial p-i-n junction GaAs nanowire is investigated byoff-axis electron holography. These characterization methods illustrate some of the challenges in nanowire growth anddevice fabrication, including dopant incorporation, consumption of the Ga droplet, crystal structure, passivation, andelectrical contacting.

7


Gate-controlled plasmonics in nanostructuresF. Rossella1, A. Arcangeli1, J. Xu1, D. Ercolani1, A. Tredicucci1,2, F. Beltram1, S. Roddaro1 and L. Sorba1

1 NEST, Istituto Nanoscienze and Scuola Normale Superiore, Pisa Italy2 Dipartimento di Fisica, Universita di Pisa, Italy

Nanoplasmonics is emerging as a powerful tool for moderninformation and communication technologies, as suggested bythe recent realization of gate-tunable plasmons in graphenenanostructures [1]. Here we demonstrate that a similar ap-proach can be also very useful for the spatially resolved inves-tigation of fundamental properties of nanostructures. In thepresent implementation we achieve field-effect control of theplasma resonance in InAs nanowire (NW) devices, detectedby scattering-type scanning near-field optical microscopy (s-SNOM) in the mid-infrared region (λ ∼ 10.5 μm) [2].

The NWs are grown by chemical beam epitaxy with a lin-early graded n-doping which modulates the free carrier axialdensity, and are deposited on a substrate consisting of 300 nmthick thermal SiO2 grown on top of bulk n-doped silicon. Op-tical nano-imaging of individual NWs shows a typical resonantfeature that mirrors the axial doping profile. Electrical contactsare fabricated on the NWs while the doped silicon substrate actsas a back gate.

Keeping the NW grounded by the contact electrode, weapply a dc voltage to the substrate and simultaneously probethe plasma resonance along the NW. Using a back-gate volt-

x

Seiϕ

(a)

0x

S E M

500

n(1

0cm

)16

–3

(b)

Fig. 1. Local plasmon resonance in a steep doping profile. (a): wecreate a linear modulation of the carrier density profilen(x) as a func-tion of the axial position x along an InAs nanowire. The induceddensity ranges from a nominally undoped value n ∼ 1×1016 cm−3

(in the position labelled by the letter S) to a maximum dopingn ∼ 5×1018 cm−3 (in the position labelled by the letterE). An inter-mediate doping segment was introduced in the growth sequence andused as marker (M) for the s-SNOM maps. The resulting graded-doping NW is sketched at the bottom of the diagram: the grey colourmodulation reflects the amount of doping in the NW body (withdark grey meaning high carrier concentration), while the yellowhalf sphere represents the metallic (Au) tip of the NW. (b): NWswere deposited on a SiO2/n-Si substrates and a (λ = 10.5];μm)laser beam (red arrows) is focused on an s-SNOM tip oscillating at250 kHz. The laser impinging on the tip was vertically polarized,with the wave vector forming an angle of 30◦ with the surface of theNW. The amplitude s and phase φ of the reflected beam (blue ar-row)are detected using an interferometric pseudo-heterodyne technique,demodulated at the fourth harmonic of the tip tapping frequency andused to reveal the local dielectric response of the NW.

70 nm6050403020100

(a)

500 nm

500 nmS

S

EM

(c)+1.0

0.0

–1.0(au)

+0.5

–0.5–1.0

–1.5

–2.0–2.5(au)

0.0

500 nmS

EM

ϕ(b)

Fig. 2. (a) AFM topography map of an isolated NW, acquired inparallel to the SNOM signal. (b) Fourth harmonic s-SNOM phase(φ4) map: a marked phase dip in the scattered light is observed incorrespondence with the plasma frequency. (c) Fourth harmonic s-SNOM amplitude (s4) map: the plasmon mode is highlighted by astrong modulation of the scattered field ampli-tude in proximity ofthe region where the laser matches the local plasma frequency. Thedoping marker is visible as a region of modified scattering (both inamplitude and phase) and provides a reference to identify the gradeddoping region.

age in the range (−100, +100) V we detect a 0.75 nm/V shiftof the axial position of the plasma resonance, induced by themodification of the free carrier density in the NW due to thefield effect. Moreover, we implement a setup for low-noiselocal measurements of the field effect on the carrier density inthe NW. From the analysis of our results we are able to ex-tract quantitative information on the electron mobility aroundthe plasma resonance. Our approach leads to the experimen-tal evaluation of key electrical parameters with 20 nm spatialresolution using a non-invasive and all-optical method.

References

[1] J. Chen, et al, Nature 487, 77 (2012).[2] J.M. Stiegler, et al, Nano Lett. 10, 1387 (2010).

8

International Summer School and Workshop • Nanostructures for Photonics i-Wed2St Petersburg, Russia, June 27–July 2, 2016

Nucleation statistics in nanowires from single nano-objectsto ensemble length distributionsF. GlasCentre for Nanoscience and Nanotechnology CNRS, Universite Paris Sud, Universite Paris-Saclay C2N – Sitede Marcoussis, Route de Nozay, 91460 Marcoussis, France

Our current understanding of vapor-liquid-solid (VLS) growthof nanowires (NWs) is largely based on the 2D nucleationof monolayers (MLs) at the solid-liquid interface. This phe-nomenon is intrinsically stochastic so that, even in the station-ary regime where a NW grows on average at a constant rate,the nucleation events do not occur periodically.

We will first describe an effect which consists in a temporalanti-correlation of the formation of successive MLs in VLSgrowth [1]. In short, because the nucleation events occur ina nanodroplet which is depleted by the ensuing rapid growthof a ML, these events, and hence the formation of successiveMLs, tend to be anticorrelated. This nucleation “antibunching”induces sub-Poissonian nucleation statistics In the case of III–V NWs, we modelled these statistics quantitatively allowingfor the depletion of the catalyst droplet in the dilute group-Velement, following the formation of each ML [2]. We willdescribe both numerical and generic analytical calculations.

We shall then investigate the consequences of nucleationantibunching for ensemble of NWs growing at the same rate.Namely, we calculate the length distributions (LDs) of such en-sembles as a function of time. This is done either by modellingnumerically the growth of a large number of NWs, or elseanalytically using the results obtained for single NWs. Themain result is that, similarly to what occurs in single NWs forthe length distribution of successive segments grown in equaltimes, the standard deviation of the LD saturates with time.However, the saturation value depends on the initial conditionsthat dictate the times at which the various NWs start growingand, in case of VLS growth, the state of the droplet at this initialinstant.

In the recent years, self-catalyzed growth has emerged asa powerful technique for growing NWs of III–V semiconduc-tors. In this variant of vapor-liquid-solid (VLS) growth, whichhas been mainly applied to Ga-based materials elaborated bymolecular beam epitaxy, the foreign metal that forms the basisof the catalyst droplet is replaced by the group-III constituent(s)of the NW. At variance with standard MBE of planar III–Vstructures, the stationary growth velocity is then determinedby the group-V flux and ultimately by the concentration ofgroup-V element in the droplet, which is only of the order of1.3%. If time permits, we will discuss how a joint modellingof the kinetics and statistics of self-catalyzed growth of GaAsNWs allows one to set narrow bounds for the key parameter ofboth models, namely the energy of the edge of the 2D nucleithat form at the liquid-solid interface.

References

[1] 1 F. Glas, J.-C. Harmand, G. Patriarche, Phys. Rev. Lett., 104,135501 (2010).

[2] F. Glas, Phys. Rev.B, 90, 125406 (2014).[3] F. Glas, M. R. Ramdani, G. Patriarche, J.-C. Harmand, Phys.

Rev. B., 88, 195304 (2013).

9


Nanostructured semiconductors for thermoelectricapplications: a theoretical perspectiveP. KratzerFaculty of Physics and Center for Nanointegration (CENIDE), University Duisburg-Essen, 47048 Duisburg,Germany

Nanostructuring offers a promising route to modified propertiesof semiconductor materials. In my presentation, I will demon-strate how the capabilities of nanostructuring may be exploitedto enhance the thermoelectric performance of semiconductors,following the strategy of reducing their thermal conductivitywhile retaining large thermopower and electrical conductivity.Atomistic calculations are presented for the high-temperaturethermoelectric material Si/Ge, as well as for recently developedternary semiconductors in the C1b “half-Heusler” structure.

From a theoretical point of view, the simplest nanostruc-tured material is a stack of layers. The interface between ma-terials of different acoustic impedance hinders the transportof heat perpendicular to the layers. I will show that the har-monic phonon spectra, which are computationally accessiblevia Density Functional Perturbation Theory, form the basis fora simple estimate of the interface thermal resistance. Fromthe class of half-Heusler compounds, I will consider the elec-tronic properties of the closely lattice-matched heterostructureZrNiSn/ZrCoBi, and show that the high thermoelectric powerfactor of these materials persists, while the cross-plane thermalconductivity may be reduced by a factor of three compared tothe bulk materials [1].

200

0.02

0.04

0.06

0.08

400T (K)

600 800 1000

(/

)m

Wcm

KS

22

σ

Lx

Lz w

Hot

Cold

Fig. 1. Regular arrays of Ge quantum dots embedded in Si matrixto be used as a compound material for a thermoelectric generator.The theoretically predicted power factors for n-doped samples (1017,1018 and 1019 cm−3 dopant concentration) clearly exceeds that ofbulk Si (dash-dotted line) and shows a sharp maximum below roomtemperature.

In materials systems with a larger lattice mismatch, suchas Si/Ge, the internal strain may be further exploited for bandstructure engineering. Epitaxial growth techniques allow forfabrication of a regular array of nanoscale Ge inclusions in aSi host crystal. Due to the strain in these type-II quantum dotcrystals, conducting channels for electrons open up right be-low the conduction band edge in the Si host. These channelsallow for a large thermoelectric power factor of the compoundmaterial already at room temperature [2], while the thermalconductivity drops even below the limit of a random SiGe al-loy [3].

References

[1] G. Fiedler and P. Kratzer, J. Elec. Mater. 45, 1762 (2015).[2] G. Fiedler and P. Kratzer, New J. Phys. 15, 125010 (2013).[3] G. Pernot et al, Nature Materials 9, 491 (2010).

10


Growth of high aspect ratio semiconductor (nano)structures:back to basics of crystallogenesisE. Gil 1−3

1 Universite Clermont Auvergne, Universite Blaise Pascal, Institut Pascal, BP 10448, F-63000 Clermont-Ferrand,France2 CNRS, UMR 6602, IP, F-63178 Aubiere, France3 ITMO University, Kronverkskiy pr. 49, 197101 St Petersburg, Russia

Metal-organic vapor phase epitaxy (MOVPE) and molecularbeam epitaxy (MBE) are the most commonly-used growthtechniques for the synthesis of nanostructures for photonics,especially nanowires. Hydride vapor phase epitaxy (HVPE)involving high mass inputs of chloride III-Cl and hydride V-H3gaseous growth precursors, is a third less-well known growthprocess. The use of chloride III-Cl molecules induces par-ticular growth properties as compared to MOVPE and MBE.The talk will revisit the basics of crystallogenesis that rule theshaping of crystal materials: how can we control morpholo-gies and aspect ratio of (nano)structures as a function of ther-modynamics and surface kinetics? HVPE provides the largestrange of growth rates (1 to 300 μm/h) for the planar growth ofIII–V semiconductors. HVPE also demonstrates perfect selec-tive growth capability and ability to reach large crystal growthanisotropy [1]. It is a perfect tool for giving educational modelsof crystallogenesis. Illustrative examples, including reversiblecrystallogenesis, will be commented so as to introduce epitax-ial issues to students and physicists and designers of photonicdevices: morphology-controlled selective area growth (SAG)for gratings and nanowires arrays; vapor-liquid-solid (VLS)growth of ultra-long nanowires (NW) [2,3] (Fig. 1).

References

[1] E. Gil, et al, Handbook of Crystal Growth (Elsevier) 3, 51(2015).

[2] E. Gil, et al, Nano Letters 14, 3938 (2014).[3] M.R. Ramdani, et al, Nano Letters 10, 1836 (2010).

GaAs nanowire: 50 µm-long, constsntdiameter of 120 nm1 µm

LP

= 500 nm= 3 µm

20 µm

780 nm(a)

(b)

(c)Pure cubic GaAs NW,diameter of 10 nm(c)

Fig. 1. GaAs grating element grown by SAG-HVPE in 15 minutes(left, after SEM). GaAs nanowires grown by VLS-HVPE with elon-gation rate of 170 μm/h: constant diameter-high aspect ratio NW(top right, after TEM, by C. Leroux, IM2NP UMR CNRS 7334);pure cubic phase regardless of the diameter of the NW (bottom right,after HRTEM, by C. Leroux, IM2NP UMR CNRS 7334).

11


Thermodynamics of VLS growthJ. Johansson and M. GhasemiaSolid State Physics and NanoLund, Lund University, Box 118, 221 00 Lund, Sweden

Thermodynamics is a very powerful discipline, which is widelyapplied in materials science. Using thermodynamics, it is pos-sible to explain which state of a given material that is stableunder given conditions. This thermodynamic understanding isbased on equilibrium between phases of matter, and we willshow how the necessary equilibrium conditions, based on en-tropy maximization in a multicomponent heterogeneous sys-tem, form the basis for phase diagram calculations. Moreover,these equilibrium conditions lead to Gibbs phase rule, whichdictates the dimensional properties of phase diagrams [1–2].

The modern way to assess thermodynamic systems isknown as CALPHAD, which is an abbreviation for calculationof phase diagrams [3]. In this context assess means to fit exper-imental or calculated thermodynamic data to specified Gibbsenergy models. This procedure results in a thermodynamicdatabase from which the phase diagram and other propertiesof interest can be calculated.

We will demonstrate how the CALPHAD approach can beused to deepen the thermodynamic understanding of vapor-liquid-solid (VLS) growth. VLS growth is a technologicallyhighly important growth mode used to grow semiconductornanowires and the two examples we will consider concerngrowth of gold alloy seeded III–V semiconductor nanowires.Even if several properties of nanowire growth, most promi-nently the growth rate, rely on kinetic rather than thermody-namic properties, the thermodynamics of the constituting ma-terials systems still sets the boundaries of what is possible toachieve.

The first example is gold seeded GaAs nanowire growth.Here two materials systems are of interest, the ternary Au-Ga-As and the sub-binary Ga-As. What we are interested in here ishow a small amount of arsenic influences the gold and galliumdominated liquid seed particle and at which temperatures solidGaAs can form from such an alloy. In Fig. 1, we show somerelevant phase diagrams. In Fig. 1a, the Au-Ga binary phasediagram is displayed. This is a complicated phase diagramcontaining a number of intermediate phases between pure Auand pure Ga. At low Ga contents, Au becomes liquid at around620 K (347 ◦C) by dissolving 30 or 33 at% Ga through oneof the eutectic reactions at 623 K or 619 K, respectively. Thiscorresponds well to the lower growth temperature of 320 ◦Creported for MBE growth of gold seeded GaAs nanowires [4].In Fig. 1b, the Au-GaAs vertical section is shown. Along thiscompositional range, there is a ternary eutectic reaction at about880 K, below which the liquid phase is not stable. In Fig. 1c,we show a vertical section corresponding to the one in Fig. 1a,but with 2 at% As, which is probably a relevant As compositionduring gold seeded VLS growth of GaAs.

The two-phase field labelled “L + GaAs” in Fig. 1c repre-sents the compositional and temperature range where growthof such nanowires is thermodynamically possible.

Specifically we conclude that VLS growth of GaAs nano-wires from a ternary Au-Ga-As liquid is possible in the tem-

(b)

Mole fraction Ga

400

600

800

1000

1200

1400

1600

Tem

pera

ture

(K

)

Liquid Au Ga

GaAs

GaAs+Liquid(Au)+Liquid

GaAs+(Au)+Liquid

GaAs+(Au)+(As)

As

(Au)+(As)

Au GaAsMole fraction Ga0.0 0.1 0.2 0.3 0.4 0.5

Mole fraction Ga GaAu0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

L+(A

u)

Liquid (L)

Au Ga

As

Au

Ga

-LT

72

AuG

a 2

AuG

a

Au

Ga

73

400

600

800

1000

1200

1400

Tem

pera

ture

(K

)

(a)

623 K: Au Ga -LT + Au Ga→ 7 2 7 3

619 K: Au Ga + AuGa→ 7 2

Mole fraction Ga Ga As0.98 0.02Au As0.98 0.02

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

400

600

800

1000

1200

1400

Tem

pera

ture

(K

)

GaAu

As

2 at.% AsLiquid (L)

(c)

L+GaAs

L+(Au)+GaAs

(Au)

+G

aAs

Au

Ga

-LT

72

Au

Ga

73

L+(A

u)

Fig. 1. (a) TheAu-Ga binary phase diagram calculated using the datain [5]. (b) The Au-GaAs vertical section of the As-Au-Ga ternaryphase diagram [6]. (c) The vertical section of the As-Au-Ga phasediagram at 2.0 at% As [6]. The insets show the respective verticalcuts in the ternary phase diagram.perature interval 619–1041 K (346–768 ◦C) if the wires growfrom a liquid with a composition close to the eutectic reactions

12

13

in Fig. 1a, that is, at about 30 at% Ga. This compares wellto the upper temperature of 620 ◦C reported for MBE growthof gold seeded GaAs nanowires [4]. However, if this reportedupper temperature is thermodynamically determined, it wouldcorrespond to a lower Ga concentration in the liquid of about23 at% Ga.

Our second example is slightly more complicated. It con-cerns growth of ternary InxGa1−xAs nanowires from a qua-ternary In-Ga-As-Au liquid alloy. Understanding the thermo-dynamics of this system is important for composition controlas well as for heterostructure control in these nanowires. Re-cently, the chemical potentials in these systems, the ternarysolid and the quaternary liquid, and their difference which con-stitutes the driving force for InxGa1−xAs nanowire growth,have been characterized [7] and we will briefly address this.

We will further discuss the thermodynamics of this systemand for instance answer questions like: what is the compositionof the solid InxGa1−xAs phase that is in equilibrium with a liq-uid of a given state (composition and temperature)? The answerto this will be purely thermodynamically limited compositionalrelations between x and y, where x is the InAs fraction in thesolid nanowire and y is the In fraction of the total amount ofgroup III in the liquid. Beyond this, we will also show somepreliminary results on nucleation limited composition control.

References

[1] M. Hillert, International Metals Reviews, 30, 45-67 (1985).[2] R.T. DeHoff, Thermodynamics in materials science, McGraw-

Hill, 1993.[3] H.L. Lukas, S.G. Fries, B. Sundman, Computational thermo-

dynamics: the Calphad method, Cambridge University press,2007.

[4] M. Tchernycheva, J.C. Harmand, G. Patriarche, et al, Nanotech-nology, 17, 4025-4030 (2006).

[5] J. Wang, Y.J. Liu, L.B. Liu, et al, Calphad, 35, 242-248 (2011).[6] M. Ghasemi, Thermodynamic modeling of materials systems

for nanowires: CALPHAD, DFT and experiments, PhD thesis,Lund University, Lund, 2016.

[7] J. Grecenkov, V.G. Dubrovskii, M. Ghasemi, et al, Quaternarychemical potentials for gold-catalyzed growth of ternary In-GaAs nanowires, (submitted manuscript) (2016).


Nanostructured silicon carbide on silicon:a new material for photonics and optoelectronicsS. Kukushkin, A. Osipov1 Institute of Mechanical Engineering RAS, St Petersburg, Russia2 St Petersburg University of Information Technologies, Mechanics and Optics, Russia

The tremendous elastic stress appearing during the growth ofsingle crystal films with the lattice para-meter strongly differ-ing from the lattice parameter of the substrate does not allowus to obtain high quality layers of new wide-bandgap semicon-ductors without misfit dislocations. In this work, we suggestnew relaxation mechanism of the elastic energy for growingdislocation-free heteroepitaxial films. The essence of this ap-proach, which differs from all the existing methods of filmgrowth, is based on the idea of preliminary incorporation ofpoint defects into the crystal lattice of the silicon host. Whengrowing the SiC film on the Si substrate, such defects are thecarbon atom C placed in the Si interstitial position and the va-cancy formed as a result of removal of one of the Si atoms.If these defects are attracted to each other by the elastic in-teraction in the non-isotropic Si matrix, the resulting elasticenergy caused by their incorporation into the substrate host isconsiderably lower than the energy of non-interacting defects.

In order to provide the effective usage of this relaxationmechanism due to the interaction of point defects, we suggestthe deposition process of SiC not from the vapor phase but im-mediately from the matrix of the single crystalline Si substratedue to the topochemical reaction between the crystalline Si andgaseous carbon monoxide CO [1–3]

2Si(solid)+ CO(gas) = SiC(solid)+ SiO ↑ (gas). (1)

We selected this reaction because of the fact that the form-ing gaseous silicon monoxide SiO partially carries the atomsfrom the Si matrix inducing vacancies in it, while gaseous car-bon monoxide CO is the source of carbon atoms C arranged inatomic voids of the silicon lattice. Both Si vacancies and incor-porated C atoms are the dilatation centers in the cubic Si latticeand interact with each other [1]. Figure 1 represents the differ-ence between the classical mechanism of thin film growth andthe mechanism of epitaxy suggested here. In the first case thefilm grows on the substrate surface which results in enormousstress energy, whereas in the last case the film grows inside thesubstrate, and the attraction between point defects provide thecomplete relaxation of the stress energy. The merging processof vacancies after film nucleation results in the pore formationunder the film [2].

The interaction of point defects in cubic and hexagonal crys-tals has been recently investigated in [4]. In particular, it wasproved that in a cubic crystal like Si the attraction energy be-tween C atom in the interstitial position and Si vacancy equalsto

Eint = −E0

(cos4 ϕx + cos4 ϕy + cos4 ϕz − 3

5

), (2)

where cosϕi = xi/r are the cosines between the axes x, y, zand a direction of the line connecting the centers of interacting

➡ ➡

(a)

Si Si

SiC

SiH +C H4 3 8 CO

SiC

Si Si

(b)

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

●

Fig. 1. Schematic pattern of regular thin film growth (a) and the sug-gested one‘(b). In the second case dumbbells represent the attractionbetween different point defects (dilatation dipoles).

SiC

Si with pores200 nm

Fig. 2. Microscopic image of the cut of SiC sample grown inside Siby the topochemical reaction with CO. An average thickness of thesample is 180 nm. Si under the layer of SiC contains some amountof pores and voids.

defects. The expression cos4 ϕx + cos4 ϕy + cos4 ϕz − 3/5reaches its minimum, equaled to −0.27, in the direction 〈111〉.Therefore, the direction 〈111〉 is the most energetically favor-able for pairs of point defects like C atoms and Si vacancies.

To confirm experimentally the suggested relaxation mech-anism of elastic energy, Si (111) substrates 35 mm in diameterwere held in a vacuum furnace at T = 1200−1350 ◦C in theatmosphere of CO at p = 10−300 Pa for 5–60 min [1–3]. ASiC film 50–200 nm thick grew inside the Si substrate duringthis time (Fig. 2). The average value of tensile elastic stressesin SiC films measured by an FLX-2320-S thin film stress mea-surement system is 0.5 GPa in the absence of lattice misfit andcracks [1–3]. Microscopic analysis revealed an almost idealconjugation of the lattices of silicon and silicon carbide [3].Such a low measured elastic energy with the ideal conjugationof lattices in the absence of misfit dislocations and cracks canbe interpreted by the relaxation of elastic stresses due to theensemble of dilatation dipoles. Monocrystalline nanoscaledlayers of SiC of hexagonal polytypes 4H and 6H are obtainedon Si for the first time. Best of them have FWHM (ω − θ ) inthe range 20–40 arcmin at the SiC thickness of 100 nm, andFWHM (ω − 2θ ) in the range 5–10 arcmin. At the presenttime SiC layers on Si substrates of 6 inch are available [8]. Tofind out if it is possible to use SiC/Si samples obtained by the

14

15

⟨ ⟩e1 73.00°⟨ ⟩e2 73.00°

0.00

0

–2

–42

2

4

4

6

6

8

10

12

14

1.0 2.0 3.0Energy (eV)

4.0 5.0 6.0 7.0

⟨⟩

e1 ⟨⟩

e2

Variable angle spectroscopic ellipsometric (VASE) data

Fig. 3. Ellipsometric spectrum of GaN/SiC/Si (ε1 and ε2 are realand parts of image dielectric function correspondingly). The directbandgap 3.3 eV of GaN is clear observed. 1 μm epitaxial GaN filmis completely transparent up to the bandgap.

topochemical synthesis as the templates for the further growthof wide-gap semiconductors we deposited such materials asAlN, GaN, ZnO, CdS, CdSe, Ga2O3 by different techniques,mainly, HVPE, CVD, ALE, and vacuum evaporation. Andall of them are epitaxial without polycrystalline phase. Wehave performed a comprehensive analysis of structural, crystal-lographic, physicochemical, electro-optical and spectroscopicstudies of AlN, GaN and ZnO films grown on the SiC/Si sub-strates (Fig. 3). A lot of experimental results are presented inreviews [2,3].

Thus, a new relaxation mechanism of elastic energy due tothe formation of dilatation dipoles is put forward. Usage of thismechanism allowed us to originally grow epitaxial SiC on thesilicon substrate by the topochemical reaction of the substratewith CO. The employment of this relaxation mechanism alsopermits us to obtain heteroepitaxial films of wide-gap semi-conductors (such as SiC, AlN, GaN, AlGaN, ZnO, CdS, CdSe)on silicon nearly without misfit dislocations and cracks havingsufficient quality to fabricate many devices of microelectronicsand optoelectronics.

References

[1] S.A. Kukushkin and A.V. Osipov, J. Appl. Phys., 113, 024909(2013).

[2] S.A. Kukushkin, A.V. Osipov and N.A. Feoktistov, Phys. SolidState, 56, 1507 (2014).

[3] S.A. Kukushkin and A.V. Osipov, J. Phys. D: Appl. Phys, 47,313001 (2014).

[4] S.A. Kukushkin, A.V. Osipov and R.S. Telyatnik, Phys. SolidState, 58, 971 (2016).

[5] S.A. Kukushkin, A.V. Lukyanov, A.V. Osipov and N.A. Feok-tistov, Tech. Phys. Lett., 40 1, 36 (2014).


III-nitride based photonic crystal:recent achievements and prospectsR. ButteInstitute of Physics, Ecole Polytechnique Federale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland

III-nitrides (III-N) are considered as a very promising materialsystem for the realization of novel integrated optoelectronic de-vices on silicon offering different functionalities and a reducedfootprint. This is inherited from their wide direct bandgaptunability, ranging from the UV to the IR spectral range, butalso from features such as their enhanced light-matter inter-action, chemical inertness, thermal stability, mechanical resis-tance and biocompatibility.

In this talk, I will report on recent results from the literatureobtained on high quality factor (Q) III-N based photonic crys-tal (PhC) cavities including results from our group. I will focuson the optical characterization of both passive L3 GaN-on-Sicavities relying on a genetic algorithm optimization of the Q-factor operating in the near IR wavelength range [1] and GaNPhC nanobeam cavities containing a single InGaN quantumwell gain medium exhibiting low threshold room temperaturecw blue lasing [2]. For this latter geometry, a twofold increasein theQ-factor using a single-step e-beam lithography processand improvement in the integrated far-field intensity by nearlyone order of magnitude by introducing a sidewall Bragg cross-grating coupler will be reported [3]. A critical assessment ofthose photonic structures will be discussed and potential appli-cations will be outlined.

(a)

1 µm

(b)

1322.1 1322.2 1322.3 1322.4

Q = 37000 Fano fit

RS

sig

nal (

a.u.

)

Wavelength (nm)

0

5

10

15

Fig. 1. (a) Scanning electron microscopy top view of a fabricated L3cavity with a close-up view of one of the holes illustrating a trend to ahexagonal shape at the bottom left. (b) Resonant scattering spectrumand Fano fit of the cavity with the highest measured quality factor.

200 nm

1 µm

(b)

(a)

(c)

1

β = 1ββ

= 0.9= 0.8

ββ

= 0.7= 0.2

10P P/ thr

Inte

grat

ed P

L in

tens

ity (

a.u.

)

Fig. 2. (a) Scanning electron micrograph at 75◦ tilt of a III-N PhCsuspended in air over silicon substrate. (b) Overhead enlarged viewof a nanobeam cavity incorporating a cross-grating coupler. (c) Log-arithmic plot of theLin −Lout curve of a nanobeam cavity laser emit-ting at 460 nm acquired at 300 K under cw excitation together withtheoretical curves deduced from rate equations for various sponta-neous emission coupling factor (β) values supporting β > 0.8.

References

[1] N.V. Trivino, et al, Appl. Phys. Lett., 105, 231119 (2014).[2] N.V. Trivino, et al, Nano Lett., 15, 1259 (2015).[3] I. Rousseau, et al, Appl. Phys. Lett., 108, in press (2016).

16


Optical properties of III-nitride nanowires:early expectations, current knowledgeB. Gayral1,21 Univ. Grenoble Alpes, F-38000 Grenoble, France2 CEA, INAC-PHELIQS, Nanophysique et semiconducteurs group, F-38000 Grenoble, France

The first report of GaN nanowire growth by plasma-assistedmolecular beam epitaxy (PA-MBE) was done 20 years agowhen Yoshizawa and co-workers presented their work ongrowth of self-organized GaN nanostructures on saphire at the23rd symposium on compound semiconductor held in St Pe-tersburg in september 1996 [1,2]. Shortly after that, Calleja andco-workers published an article showing the remarkable prop-erties of GaN nanowires grown on Si (111) [3]: few 100 nmlong nanowires show optical properties typical of relaxed highquality GaN, usually observed for carefully grown thick 2Dlayers. The group of Kishino then pioneered nanowire basedLEDs [4].

While it has been early foreseen that such nanowires presentextraordinary optical properties, much has been learned sincethese pioneer articles.

I will discuss in what sense such nanowires have allowedto explore some fundamental physics of GaN and III-N het-erostructures, I will also discuss to what extent such nanowiresshow specific properties due to their large surface/volume ratio.

References

[1] M. Yoshizawa, et al, 23rd Int. Symp. Compound Semicond(1996).

[2] M. Yoshizawa, et al, Jpn. J. Appl. Phys 36, L459 (1997).[3] E. Calleja, et al, Phys. Rev. B 62 16826 (2000).[4] A. Kikuchi, et al, Jpn. J. Appl. Phys. 43, L1524 (2004).

17


Nitride nanowires for light emission:from single wire properties to flexible LEDsM. Tchernycheva1, N. Guan1, X. Dai1, A. Messanvi1,2, H. Zhang1, F. Bayle1, V. Neplokh1, V. Piazza1,F.H. Julien1, L. Rigutti3, A. Babichev1,4, C. Bougerol2, G. Jacopin5, J. Eymery2 and C. Durand2

1 Institut d’Electronique Fondamentale, UMR 8622 CNRS, Universite Paris Sud XI, 91405 Orsay, France2 Equipe mixte “Nanophysique et semiconducteurs,” CEA/CNRS/Universite Joseph Fourier, CEA, INAC, SP2M,17 rue des Martyrs, 38054 Grenoble Cedex 9, France3 Groupe de Physique des Materiaux, UMR CNRS 6634, Normandie University, University of Rouen and INSARouen, 76801 St. Etienne du Rouvray, France4 ITMO University, 197101 St Petersburg, Russia5 ICMP LOEQ Ecole Polytechnique Federale de Lausanne, 1015 Lausanne, Switzerland

Flexible devices are today a topic of intense research, motivatedby their numerous applications (e.g. flexible light emittingdiodes for rollable displays). The flexible technology is domi-nated by organic semiconductors integrated on lightweight andflexible plastic substrates. Taking the example of light emit-ters, organic devices possess many advantages, but present se-vere limitations in the short wavelength range. Blue organiclight emitting diodes suffer from a rather low (below 10 %)external quantum efficiency and a limited lifetime, while bluelight emitting diodes based on nitride semiconductors revealhigh brightness and efficiency. However, difficulties exist inthe fabrication of flexible devices with conventional thin filmstructure. In this context, nitride nanowires offer an elegantsolution to the problem. Single nanowires and nanowire arrayLEDs based on InGaN/GaN core/shell heterostructures havebeen successfully demonstrated on rigid substrates, showingexcellent performances in the blue spectral range, thanks totheir high crystalline quality and non-polar active region.

In this presentation I will present our recent work on ni-tride nanowire based light emitters and photodetectors. Thesenanomaterials have the potential to boost the device perfor-mance, to improve the energy efficiency, to reduce the cost andto bring new functionalities. In particular, I will discuss our re-cent progress towards flexible nitride nanowire devices [1]. Wepropose a method to combine high flexibility of polymer filmswith high quantum efficiency provided by nitride nanowires toachieve flexible inorganic light emitting diodes and sensors.

Silver NW mesh

Peel offPDMS

Deposit Ti/Au Disperse Ag NWs

FlipFlip

Fig. 1. Top: flexible nanowire-based LED fabrication; Bottom: flex-ible nanowire blue LED (SEM image and photos under operation).

References

[1] X. Dai, et al, Nano letters 15 10, 6958-6964 (2015).[2] N. Guan et al, ACS Photonics in press.

18


AlGaN nanoheterostructures for mid-UV photonicsS.V. Ivanov1, V.N. Jmerik1, A.A. Toropov1, E.V. Lutsenko2, V.I. Kozlovsky3, X. Rong4, X. Wang4

1 Ioffe Institute, St Petersburg, Russia2 Stepanov Institute of Physics of NAS Belarus, 68 Independence ave., Minsk 220072, Belarus3 Lebedev Physical Institute, Leninsky pr. 53, Moscow 119991, Russia4 School of Physics, Peking University, Beijing, 100871, China

The paper describes the unique features and current problems ofplasma-assisted molecular beam epitaxy (PA MBE) of AlGaN-based semiconductor heterostructures for basic research andmanufacturing the high-efficiency emitters and photodetectorsfor a middle ultraviolet (mid-UV) spectral range (λ<300 nm).This is the essential segment of photonics where low-efficient,bulky and Hg-toxic vacuum bulbs are still mainly used forthe numerous ecological, medical, and military/security ap-plications. The overview of the state-of-art of semiconductormid-UV photonics will be given in the introduction.

First, different approaches used in PA MBE of (Al,Ga)Nheterostructures grown on c-Al2O3 substrates will be consid-ered, aimed at the reduction of initially high threading dislo-cation (TD) density originating at the AlN/c-Al2O3 interface.They involve (i) the high-temperature growth of the AlN nu-cleation layer in a migration enhanced epitaxy mode with anenhanced grains size, (ii) the metal-modulated epitaxy of athick (up to ∼2 μm) AlN buffer layer with atomically smoothand droplet-free surface, and (iii) the incorporation of multipleultrathin 3D GaN interlayers into the 2DAlN buffer layer, facil-itating annihilation of TD’s and their reorientation in the basalplane (0001). As a result, the lowered density of the screw andedge TD’s down to ∼108 and ∼109 cm−2, respectively, hasbeen achieved.

Second, specific features of the PA MBE of AlxGa1−xNlayers (x>0.4) at the metal-rich stoichiometric conditions withdifferent modes of growth fluxes and substrate temperaturevariation will be discussed with an impact on obtaining the ho-mogeneous alloys with an accurate composition control. Theunrivaled possibilities of PA MBE for controllable fabricationof (1–2)-monolayer(ML)-thick and fractional MLAlxGa1−xN/AlyGa1−yN (y = 0.6–1, x = 0.1–0.4) quantum wells (QW)by a sub-monolayer digital alloying (SDA) technique will bedemonstrated by using high-resolution transmission electronmicroscopy (HAADF STEM) [1]. The optimum design of theAlGaN QWs and technological parameters of their growth bySDA technique will be discussed to provide the maximum ex-citon binding energy and strong carrier localization in the QWswhich would result in significant increase in the internal quan-tum efficiency (IQE) of the structures. Special attention willbe paid to increase in the output UV-radiation (external quan-tum efficiency) from AlGaN-based heterostructures throughsuppression of the TE/TM switch of the light polarization dueto the valence band crossover by introducing the compressivestress into the AlGaN QWs. The problems of p-type dopingof AlxGa1−xN layers with high Al-content (x > 0.4) will beconsidered and the effectiveness of the use for this purpose ofthe so-called polarization doping in graded AlxGa1−xN layer(∇x ≥ 0.005 nm−1) will be demonstrated.

Finally, different AlGaN-based spontaneous and laser mid-UV emitters grown by PA MBE will be demonstrated. Amongthem the e-beam-pumped spontaneous mid-UV (λ = 285 nm)emitters with the record output pulse-scanning (cw-) opticalpower up to ∼160(39) mW, as well as optically pumped laser(stimulated) emission sources (λ = 255−290 nm) with theminimum threshold excitation power density of 150 kW/cm2willbe presented [2,3]. Moreover, different UV-photodetectors(p-i-n photodiodes and photocathodes with negative electronaffinity) working within the solar-blind range (λ <290 nm)with responsivity of 10–35 mA/cm2 will be demonstrated.

References

[1] V.N. Jmerik et al, J. Mater. Res. 30, 2871 (2015).[2] S.V. Ivanov et al, Semicond. Sci. Technol. 29, 084008 (2014).[3] X. Rong, S.V. Ivanov et al, Adv. Mater., 2016 (in print).

19

International Summer School and Workshop • Nanostructures for Photonics i-Thu3St Petersburg, Russia, June 27–July 2, 2016

Making defective shell in III–V core-shell nanowires:how and why?B. GrandidierInstitut d’Electronique, de Microelectronique et de Nanotechnologie (IEMN), CNRS - UMR 8520, DepartementISEN, 41 bd Vauban, 59046 Lille Cedex, France

Abstract. Central to the increasingly use of semiconductor nanowires (NWs) into miniaturized devices has been thecontinuous improvement of the material crystal quality. For instance, gaining control over the phase purity in III–Vsemiconductor nanowires has led to narrow photoluminescence emission in GaAs/GaAsSb heterostructure nanowires and toballistic transport of electrons in InAs NW transistors. Similarly, the synthesis of defect-free interfaces in core-shellnanowires was a key achievement in getting higher collection efficiency of the photoexcited charge carriers in solar cells.While defects in semiconductor NWs are generally seen as deleterious, a few applications might however rely on theintentional incorporation of point defects during or after the growth. Here, I will describe a few examples of III–Vsemiconductor nanowires, with defective shells obtained in a control way.

First, we will discuss the direct incorporation of point defects through the fabrication of nonstoichiometric GaAs shellgrown at low temperature. Based on atomic-scale characterizations, we will show that all these defects introduce midgapstates in the shell leading to a short carrier lifetime and a low electrical conductivity. Such properties are essential toproduce THz emission from III–V nanowires as it will be demonstrated. In a second part, we will show how the morphologyof the sidewalls of III–V nanowires can be modified by post-growth treatments and we will highlight the physical reasons atthe origin of the sidewall roughness. Finally, we will investigate the chemical randomness in the shell of III–V ternarysemiconductor nanowires and will demonstrate a new type of ordering, that is directly caused by the nanowire growthscheme.

20


Gas sensor based on nanomaterials and VLSI technologyZ. TangDalian University of Techlogy, Dalian, China

Abstract. This talk presents a gas sensor based on nanomaterials and VLSI technology. The gas sensor is fabricated withstandard CMOS and post-CMOS micromachining process. The standard CMOS technology is used to finish the wholeprocess of integrated circuits that includes power driver, amplifier, and signal sampling and output circuits The post-CMOSmicromachining is used to form a microhotplate (MHP) to heat operation temperature of the gas sensor up to 300 ◦C.Typically, the MHP is a free standing membrane supported by some bridges. A snake resistor is used to heat it up to operateat temperature in a range of room ambinet to several hundreds degree C, and another resistor is used to test the operationaltemperature. For gas sensor application, gas sensing thin-film is deposited on the top surface of the MHP. Due to the MHP isa platform for some other microsensors whose operation are dependent on heat transfer, vacuum sensor and infrared sensorare also described in this talk.

21


Metamaterials for THz detectionY. TodorovLaboratoire Materiaux et Phenomenes Quantiques Universite Paris Diderot – CNRS UMR 7162, 75205 ParisCedex 13, France

The THz spectral domain (1–20 THz) has numerous applica-tions in spectroscopy gas sensing security screening and imag-ing and is even seen as the next frontier for wireless communi-cations [1,2]. Compact and powerful sources of THz radiationssuch as quantum cascade lasers are now available and they de-liver more than 10 mW in continuous wave, even if they areconstrained to operate at cryogenic temperatures (<50 K). Onthe other hand the detection in the THz domain is a notoriouslydifficult problem owe to the large photon wavelengths involved.Indeed, neither of the existing commercial THz detectors suchas bolometers or Golay cells are altogether sensitive, fast androom temperature [3]. These issues can be tackled by adoptingcompletely novel approaches for the electromagnetic confinement in the detector, inspired from the recent progress of elec-tromagnetic metamaterials [4]. In this approach engineeredmetamaterial resonators are used to provide highly subwave-length confinement of the electromagnetic field, and direct THzphotons into detector absorbers with high efficiency

(a)

(b)

Fig. 1. (a)Array of THz patch antenna detectors loaded with a QWIPabsorbing region (b) THz optomechanical splitring detector with ananosized cantilever.

We will present two classes of devices where the use of suchresonators can be highly beneficial In the first case (Fig. 1(a))the absorbing region is a quantum well infrared photodetector(QWIP) [5]. We will detail how the increased electromagneticfield confinement leads to a drastic suppression of the darkcurrent and higher operating temperatures In the second typeof device, a THz metamaterial resonator is upgraded with amechanical element (Fig. 1(b)), enabling a nanoscale optome-chanical coupling This approach allows detection at room tem-perature with high speed, with sensitivities that can potentiallyreach those of commercial semiconductor bolometers operat-ing at cryogenic temperatures [6].

References

[1] M. Tonouchi, Nature Phot. 1, 97–15 (2007).[2] I.F. Akyildiz J.M. Jornet, C. Han, Phys. Comm. 12, 16 (2014).[3] A. Rogalski and F. Sizov, Opto-Electron Rev. 19, 346404 (2011).[4] Cai and Shalaev, Optical Metamaterials: Fundamentals and

Applications (Springer 2009).[5] D. Palaferri et al., Appl. Phys. Lett. 16, 161102 (2015).[6] Optomechanical transducer for terahertz electromagnetic

waves, EP16305288.

22


Carbon-based nanostructures for THz applicationsM.E. PortnoiSchool of Physics, University of Exeter, Exeter, United Kingdom

Abstract. In this review lecture I outline several original ideas on using carbon nanotubes and graphene for various THzapplications [1].

References

[1] R.R. Hartmann, J. Kono and M.E. Portnoi, Nanotechnology 25,322001 (2014).

23


Long-range and high-speed electron and spin transportat GaAs/AlGaAs interfaceP. Kuzel1, L. Nadvornık1,2, P. Nemec2, T. Janda1,2, K. Olejnık1, V. Novak1, V. Skoromets1, H. Nemec1,F. Trojanek2, T. Jungwirth1, J. Wunderlich1

1 Institute of Physics, Czech Academy of Sciences, Prague, Czech Republic2 Faculty of Mathematics and Physics, Charles University in Prague, Czech Republic

Abstract. We combine optical pump-probe (time and spatially resolved Kerr effect measurements) and opticalpump-terahertz probe (ultrafast THz photoconductivity measurements) techniques to determine fundamental spin andcharge transport parameters in a semiconductor heterostructure [1]. Unprecedented combination of long-range andhigh-speed spin transport is achieved by suppression of processes limiting the carrier lifetime and mobility. Our undopedMBE-grown GaAs/AlGaAs heterostructure enables an efficient spatial separation of photoexcited electrons and holes due tothe built-in electric field in the GaAs layer. Electrons accumulated close to the GaAs/AlGaAs interface exhibit a highmobility (>105 cm2V−1s−1) and long lifetime at 10 K. Highly mobile electrons are detected by THz probing even at times>200 μs after photoexcitation. Since the spin decay channel due to the electron-hole recombination is suppressed, spinlifetimes of 20 ns and spin diffusion length of 13 μm are observed.

References

[1] L. Nadvornık et al, Sci. Rep., 6, 22901 (2016).

24

International Summer School and Workshop • Nanostructures for Photonics i-Fri1St Petersburg, Russia, June 27–July 2, 2016

Nanoimprint lithography as enablerof a large-scale fabrication of nanostructures for photonicsM. Graczyk, M. Heurlin, R.J. Jam, N. Nilsson, G. Otnes, M.T. Borgstrom and I. MaximovLund Nano Lab, NanoLund, Division of Solid State Physics, Lund University, Box 118, SE-211 00, Lund, Sweden

Nanoimprint lithography (NIL) is a high resolution, highthroughput lithographic technology, suitable for mass fabrica-tion of a variety of nanostructures, including nanowire-basedsolar cells, light-emitting diodes and for other photonic andelectronic applications The NIL technique is based on patterntransfer from a hard stamp (mold) into a soft polymer resistvia a mechanical contact between them. The process can bepurely thermal (thermal-NIL), where the polymer is displacedby application of pressure (10–50 bar) and temperature abovethe Tg of the polymer (typically 150–180 ◦C) or a UV-lightbased, where the resist is in a liquid form and can be hardenedby UV-light irradiation. There are also a number of combinedmethods, including simultaneous thermal and UV-imprint tech-nology. The nanoimprint lithography can be used to reach sin-gle nanometer resolution and throughput of several wafers perhour, i.e. it has a big industrial potential.

Here we present the optimized NIL-based technology ofa large-scale fabrication of arrays of Au seed particles on 2′′wafers to be used for growth of III–V nanowires (NWs) forphotonic applications. This method utilizes the IntermediatePolymer Stamp (IPS®) technique developed by Obducat AB,Lund, Sweden to perform a simultaneous thermal and UV-imprint (STU) process. The IPS stamps are produced as copiesfrom the 2.5′′ electroplated Ni “mother” stamps and includedhexagonal or rectangular arrays of 260 nm high pillars, withdiameter of 240 nm. For the imprint experiments we used 2′′ Sior InP wafers (with and without SiNx mask) covered by spin-coated double resist layers of LOR and TU7 polymers. The topTU7 layer was used as the pattern transfer layer during NIL,while the LOR layer served to make a negative slope in the re-sist for a lift-off process. After evaporation of 20–30 nm thickAu layer and lift-off in remover, the samples were transferredinto MOVPE reactor to grow InP or GaAs NWs. Alternatevely,theAu particles can be deposited by electroplating. As the goldseeds define the position and size of the grown nanowires weperformed investigation of the optimal process parameters tominimize the seed particle movement. In this work we demon-strate a defect-free NIL process to produce arrays of NWs overthe 2′′ wafers and report on the optimization of the nanoimprinttechnology and the related process steps.

1 µm

1 µm

1 µm

1 µm

(a)

(c)

(b)

(d)

Fig. 1. SEM images of (a) InP NWs grown from particles definedby resist mask, (b) InP NWs grown on a substrate with SiNx mask,(c) GaAs NWs grown on a substrate with SiNx mask, and (d) largearea image demonstrating a high growth yield of GaAs NWs [5].

Acknowledgements

The present work was performed within NanoLund and sup-ported by the Swedish Research Council, the Swedish Foun-dation for Strategic Research (grant RIF14-0090), by the Knutand Alice Wallenberg Foundation and by the Swedish EnergyAgency. The authors are thankful for the dedicated fundingfrom NanoLund to develop the NIL-technology.

References[1] S. Chou, P.R. Krauss and P.J. Renstrom, Appl. Phys. Lett., 67,

3114–6 (1995).[2] B. Heidari, I. Maximov, E.-L. Sarwe and L. Montelius, Technol-

ogy B: Microelectronics and Nanometer Structures 17, 2961–4(1999).

[3] J. Wallentin, N. Anttu, D. Asoli, et al, Science (2013).[4] H. Schift J. Vac. Sci. Technol. B 26 458–80 (2008).[5] R.J. Jam, M. Heurlin, V. Jain, et al, III–V Nanowire Synthesis

by Use of Electrodeposited Gold Particles, NanoLetters, 15,134–138 (2015).

[6] G. Otnes, M. Heurlin, J. Eklof, et al, Optimisation of patternfidelity for NW growth from large area nanoimprint lithographypatterned surfaces, submitted to Nanotechnology (2016).

25


Integration of GaAs nanowires in capacitive pressure sensingA. Chandramohan1, N. Sibirev2, G. Cirlin2, V. Dubrovskii2,3, B. Mendis1, M. Petty1, A. Gallant1

and D. Zeze1,3

1 Durham University, United Kingdom2 St Petersburg Academic University, St Petersburg, Russia3 ITMO University, Saint Petersburg, Russia

Interest in self-assembly processes to produce periodic microand nanostructures has grown significantly over the past twodecades because they are relatively inexpensive, fast and do notrequire any sophisticated lithography. This paper discusseslarge area patterning of silicon substrates using nanosphere-lithography (NSL). The bespoke technique enables the fab-rication of nanoholes or nanodots onto Si(111)/SiO2 surfacewhere semiconductor nanowires (NW) are subsequently grownby self-induced or catalysed nucleation. The principle con-sists of depositing, onto the substrate, a monolayer of closelypacked polystyrene nanospheres which is exploited as a col-loidal mask for the subsequent post processing, i.e. the creationof nanodots or nanoholes using controlled dry etching. Unlikeexisting literature, this work focuses on the fabrication of a uni-form monolayer over large areas (∼2 inch wafer), essential tothe industrial viability of the process. A model which evaluatesthe relative contribution of gravity, liquid friction, centrifugalforce and surface tension, was developed to describe the for-mation of a dense and uniform monolayer of 300 nm diameterpolystyrene nanospheres and then related to various processingstages. The monolayers produced were etched controllably toexploit the interstitial gaps between spheres to expose the sub-strate beneath where nanoholes (∼200 nm) or gold nanoseeds(30–70 nm) are created for the nucleation and growth of NWs(Fig. 1).

The nanostructure arrays produced were exploited to de-velop low cost nanoimprint polydimethylsiloxane (PDMS)master templates to reproduce the nanopatterns on large areasubstrates. GaAs NWs were subsequently grown on the nanos-tructures produced. The paper will discuss the fabricationof the nanopatterns, the particular features of the GaAs NWsgrown and their integration in a highly sensitive capacitive sen-sor.

High resolution microscopy (TEM and SEM) characterisa-tion revealed that the GaAs grown exhibit an alternating crys-tallography (periodic along their length), with wurtzite andzinc blend regions of ∼10 nm and less than 5 nm), respectively

(a) (b)

Fig. 1. Scanning electron micrograph of (a) nanholes exposing Si(111) in SiO2 (b) gold nanodots on Si (111).

(b)

Cap

acita

nce

(nF

)

Time (s)

0.5 N/m1.0 N/m

2

2

0 20 40 60 80 100 120

23.2

23.3

23.4

23.5

23.6

(a)

Fig. 2. (a) HR-TEM of GaAs nanowire with ZB and WZ alternatingphases (b) measured capacitive response under dynamic loading andunloading.

(Fig. 2a). The GaAs NWs were embedded in a poly (methylmethacrylate) (PMMA) matrix, under a metal-insulator-metalconfiguration to investigate the potential for semiconductorNWs to be exploited a highly sensitive capacitive pressure.This is particularly appealing for touch screens and wearableelectronics applications where measuring accurately the sensi-tivity to low pressure, below 1 N/m2, still remains challenging.The incorporation of GaAs NWs into the PMMA layer (∼2–5%wt.) led to a significant leap of the relative permittivity(∼2 orders of magnitude), consistent with the literature report-ing that adding a small number of conductive fillers near thepercolation threshold to polymer matrix increases its permit-tivity [1]. This, in turn, provided a large capacitance-voltagemeasurement range to sense capacitive pressure loads.

For illustration, sensitivities lower than 0.5 N/m2 success-fully demonstrated (Fig. 2b). Although PMMA is not particu-larly suitable for pressure sensing applications, the devices in-corporating GaAs NWs performed 10 times better than thosewith pristine PMMA. These preliminary results demonstratethat the integration of GaAs, recently shown to exhibit piezo-electric properties [2] and semiconductor NWs in general, of-fers a real potential to overcome the performance limits of exist-ing piezoelectric, resistive and capacitive pressure sensors [3].

26

27

Acknowledgements

This project received financial support from the EuropeanCommission FP7 grants NanoEmbrace (GA-316751) andFUNPROB (GA-269169) and from the UK Royal Academyof Engineering/Leverhulme Trust.

References

[1] D. Tan, et al, Materials Sciences and Applications 4, 6-15(2013).

[2] V. lysak et al, Physica Status Solidi – Rapid Research Letters10 2, 172-175 (2015).

[3] G. Schwartz et al, Nature Communications 4, 1859 (2013).


HVPE of III-Nitride nanostructures and nanowirestowards optoelectronic devicesY. Andre1,2,3, E. Roche1,2, Z. Dong1,2, G. Avit1,2, V. Dubrovskii4,5, C. Bougerol6,7, D. Castelluci1,2,E. Gil1,2,3, J. Leymarie1,2, F. Medard1,2, G. Monier1,2, F. Reveret1,2, A. Trassoudaine1,2

1 Universite Clermont Auvergne, Universite Blaise Pascal, Institut Pascal, BP 10448, F-63000 Clermont-Ferrand,France2 CNRS, UMR6602, Institut Pascal,F-63178 Aubiere, France3 ITMO University, Kronverkskiy pr. 49, 197101 St Petersburg, Russia4 St Petersburg Academic University, St Petersburg, Russia5 Ioffe Institute, St Petersburg, Russia6 Univ. Grenoble Alpes, F-38000 Grenoble, France7 CNRS, Institut Neel, F-38042 Grenoble, France8 Institut Universitaire de Technologie, Dept. Mesures Physiques, Universite d’Auvergne, 63172 Aubiere cedex,France

Gallium nitride and related semiconductor materials enable awide range of novel devices. Nanometer-scale nitride semicon-ductor structures are often at the hearth of such applications.The performances of such devices are strongly dependent onthe crystallographic, electronic and optical properties of thesemiconductor material and thus on the growth processes ap-plied to synthesize the device.

Hydride Vapor Phase Epitaxy (HVPE) process exhibits un-expected properties when growing III–V and III-Nitride semi-conductor micro- and nanostructures. With respect to the clas-sical well-known methods such as Metal Organic Phase Epi-taxy (MOVPE) and Molecular Beam Epitaxy (MBE), this near-equilibrium process based on hot wall reactor technology, en-ables for example, the synthesis of nanowires with a constantcylinder shape over unusual length and free of crystal defectswith great optical properties (Fig. 1) [1]. The aim of this pre-sentation is to investigate the potential of the versatile HVPEprocess implementing III chloride precursors and to demon-strate why it has been developed in the recent decade to growseveral III-Nitride semiconductors with controlled morpholo-gies.

(a)

0.1 µm

0.1 µm

0.1 µm

0.1 µm

2 nm

(b)

Fig. 1. HRTEM images of GaN nanowires grown by Vapor-Liquid-Solid (VLS)-HVPE: (a) 40 µm long GaN nanowire; (b) HRTEMimage of the sidewalls of the nanowire.

Indeed, recently, 1D c-axis InGaN core/ GaN shell rodshave emerged as good candidates for LEDs. Heteroepitaxyon highly mismatched substrates allows strain relaxation, en-hanced by the wire morphology. Crack and dislocation den-sities can be drastically reduced, inducing less non-radiativerecombinations and exalting light emission. Particularly, highaspect ratio columns offer a larger active zone per substrate

unit area than planar layers. In order to improve the GaN rodsmorphology and distribution, selective area growth (SAG) isdeveloped. In SAG, the growth selectively occurs on patternedsubstrates in the precisely defined apertures of an inert dielec-tric mask, ideally without addressing the filling factor (open-ing/mask pattern ratio) or the pattern periodicity. Chlorideprecursors, applied in HVPE are so volatile that they providethe most suitable environment for implementing selective andlocalized growth without any adsorption on the dielectric sur-face. This growth process is ruled by surface kinetics thatis, by the intrinsic growth anisotropy of crystals. The facetgrowth rate can be set by varying the experimental parame-ters of temperature and vapor phase composition. SAG-HVPEand MOVPE processes were coupled for the synthesis of highquality c-axis InGaN/GaN core/shell hybrid structures. Thecore consists in high aspect ratio and high optical quality GaNrods grown by SAG-HVPE on patterned AlN-Npolar/Si(100)substrates (Fig. 2). The shell is then grown by MOVPE. Thesilane free HVPE process ensured the whole lateral claddingof the core. The hybrid HVPE core/MOVPE shell structuresexhibit high optical quality without yellow luminescence [2].

(a) (b)

10 µm 4 µm

Fig. 2. SEM image of dense array of GaN rods grown by SAG-HVPEon Si(100)/AlN /SiO2 substrate on a zone with 0.7 µm diameterholes. As an insert a single GaN rod.

The growth of GaN nanowires (NWs) on silicon has drawnmuch attention because it provides a unique way for the mono-lithic integration of high-quality GaN nanostructures on a Siplatform for producing low-cost devices. Today, growers usean intermediate AlN to bridge the mismatch between the GaNmaterial and the substrate. Here again, the HVPE process ap-pears to be powerful: hexagonal individual, tripod and hyper-bunched GaN nanorods (NRs) were grown on silicon withoutperforming any intentional pretreatment in a short growth pro-

28

29

cess (< 1 h). The obvious effect of the temperature and theelement V gas flow on the NRs size distribution was demon-strated. The crystal symmetry of the nuclei imposes the finalNRs structure. The high potential of SAG-HVPE process forsynthesis of dense GaN rod arrays and the direct growth onsilicon is still harnessed.

More recently, significant advancements have been madewith the HVPE growth of single-crystalline higher indiumInxGa1−xN composition using the strain-relieving propertiesof nanowire geometries [3]. The capability of the HVPE pro-cess to provide InxGa1−xN nanowires using tri-chloride pre-cursors with varying x between 0 and 1 will be discussed sup-ported by thermodynamic calculations.

References

[1] G. Avit et al, Nano Letters 14, 559 (2014).[2] G. Avit,Y. Andre et al, Submitted to Crystal Growth and Design

(2016).[3] C. Hahn et al, ACSNANO 5, 3970 (2011).


Strategies for narrowing the size distributions of III–V nanowiresV.G. Dubrovskii 1−3

1 St Petersburg Academic University, St Petersburg, Russia2 Ioffe Institute, St Petersburg, Russia3 ITMO University, Kronverkskiy pr. 49, 197101 St Petersburg, Russia

Semiconductor nanowires (NWs) and in particular III–V NWsare promising building blocks for novel electronic, photonicand sensing devices. It is well documented that the size unifor-mity (i.e., sharp length and diameter distributions over the NWensemble) greatly enhances fundamental properties and deviceperformance of these structures. Most III–V NWs are obtainedby the vapour-liquid-solid (VLS) method with either gold orgroup III metal (e.g., gallium) droplets as the growth catalysts.In this talk, I will discuss several strategies to narrow up thelength and diameter distributions of different III–V NWs.

Nor

mal

ized

dis

trib

utio

n

00.0

0.2

0.4

0.6

0.8

1.0

1.2

10 20 30 40Diameter (nm)

50 60 70 80

Ga dropletsGaAs NWs, pitch 250 nmGaAs NWs, pitch 500 nmGaAs NWs, pitch 1000 nm

Fig. 1. Histograms showing the distribution of the gallium dropletsize and NW diameter for gallium-catalyzed GaAs NWs grown onpatterned Si(111) substrates with a hole size of 60 nm and differentpitches (left: 250 nm, middle: 500 nm, right: 1000 nm). Inset:30tilted SEM images of the corresponding GaAs NW arrays. Linesshow the theoretical fits [1].

First, I will discuss the effect of diameter self-equilibrationwhich is specific for self-catalyzed III–V NWs [1,2] and hasbeen recently observed experimentally in gallium-catalyzedVLS GaAs NWs grown by molecular beam epitaxy on pat-terned SiOx /Si(111) substrates [1]. In this case, the catalystdroplet serves as a non-stationary reservoir of gallium and mayself-regulate its size to a certain stable one under a balancedV/III flux ratio. The diameter distribution becomes almostdelta-like in the asymptotic stage and thus one can completelysuppress the fluctuation-induced spreading of the size distri-bution [3]. Second, I will consider Poissonian versus super-Poissonian length distributions of VLS InAs NWs grown bydifferent techniques with the gold and indium catalysts andshow why surface diffusion of the group III adatoms yields verybroad length distributions compared to the Poissonian case [4].Finally, I will briefly discuss an interesting effect of nucleationantibunching in individual NWs [5-8] and its possible narrow-ing effect on the length distributions within the NW ensembles(Fig. 2).

0.00

0.02

0.04

0.06

0.08

F(s

,)τ

100 200 300 400Number of NW monolayers s

τ = 50

τ = 100

τ = 200τ = 400

Poisson, = 0

= 0.02 invariant

= 0.02

εεε

Fig. 2. Narrowing the length distributions of NWs by nucleation an-tibunching (with the parameter ε > 0), compared to the Poissoniangrowth with the time-independent nucleation events (ε = 0) underotherwise identical conditions.

References[1] V.G. Dubrovskii, et al, Nano Lett. 15, 5580 (2015).[2] J. Tersoff, Nano Lett. 15, 6609 (2015).[3] V.G. Dubrovskii, Phys. Rev. B, accepted (2016).[4] V.G. Dubrovskii, et al, Cryst. Growth Des. 16, 2167 (2016).[5] F. Glas, et al, Phys. Rev. Lett., 104, 135501 (2010).[6] C.Y. Wen, et al, Phys. Rev. Lett. 105, 195502 (2010).[7] V.G. Dubrovskii, Phys. Rev. B, 87, 195426 (2013).[8] F. Glas, Phys. Rev. B 90, 125406 (2014).

30


Quantum dot inside nanowire: GaAs in AlGaAs caseG.E. Cirlin1,2,3, N. Akopian4

1 St Petersburg Academic University, St Petersburg, Russia2 ITMO University, Kronverkskiy pr. 49, 197101 St Petersburg, Russia3 Institute for Analytical Instrumentation RAS, Rizhsky 26, 190103, St Petersburg, Russia4 DTU Photonics, Technical University of Denmark, Kgs. Lyngby, Denmark 2800

Abstract. A combination of nanowires (NWs) with quantum dots (QDs), are promising building blocks for futureoptoelectronic devices, in particular, single-photon emitters. The most studied epitaxially grown QDs are self assembled,i.e., grown by island nucleation in the Stranski-Krastanow growth mode. The size, shape, and density of self-assembledQDs can be controlled by growth parameters such as temperature and growth time, but in the end it is a self organized straininduced process and controlling the properties of the array independently is a challenging task. QDs in nanowires have incontrast shown great potential as a highly controllable system. The diameter, height, and density of the QDs are defined bythe NW diameter, the growth time, and the NW density, respectively, and can be chosen more predictable.

Experimentally, all the samples in the present work were grown by molecular beam epitaxy (MBE). For AlGaAs/GaAsmaterial systems, different growth conditions were applied, but the strategy was the same: we have used Au-assisted growthof the NWs on Si(111) substrate, firstly we grew the AlGaAs base of the NW, secondarily, the GaAs nanoinsertion withlower bandgap, or QD, was formed and we end the structure with the core with the same material as the base. It was foundthat during the growth spontaneous, independently on the Al fraction, core-shell structures with lower aluminum content inthe cores are formed. Another important conclusion is that aluminum should enter the droplet at a much lower rate but leavethe droplet at a much higher rate than gallium. Optically, our growth method results in the formation of GaAs QD in aAlGaAs NW having very narrow spectral linewidth (< 10 µeV), single-photon emission in the wavelength range740–800 nm in dependence on the QD growth time.

31

International Summer School and Workshop • Nanostructures for Photonics i-Sat1St Petersburg, Russia, June 27–July 2, 2016

Impact of side reservoir on electromigration of copper inter-connectsX. Chen1, Geng1 and S. Li21 Dept. of Electrical and Information Engineering, Dalian University of Technology Dalian, China2 Dept. of Information Engineering, Dalian Ocean University, Dalian, China

Reservoir effect has been widely known and used both for Aland Cu interconnects [1,2]. For Cu interconnects, voids mostlikely occur at the interface of Cu/cap layer, which can causeelectromigration (EM) failure [3,4]. Reservoir appearing as anadditional metal segment closed to via can reduce EM failure.A reservoir at end of metal line can be called as end-of-linereservoir, and also can be called as side reservoir when it is atside of metal line. Most of current researches have focused onthe length of end-of-line reservoir. For current technology nodewith low-K dielectric, Tan et al. found that the effectiveness ofreservoir length on EM lifetime is still valid.

Recent experiments have found that side reservoir can fur-ther improve EM lifetime. Hau-Riege et al. studied the in-terconnect structure with both end-of-line and side reservoirand found that this double reservoirs structure at cathode cansignificantly improve EM lifetime than single end-of-line reser-voir structure. Through later experiment, Mario et al. foundthat there also exists critical length for side reservoir. In theabove works only simple 1-D stress distribution simulationswere conducted. Further accurate 3-D FEM simulation needto be conducted to figure out the mechanism of side reservoireffect.

In this work, driving force based FEM simulation methodusing ANSYS and Matlab is proposed. FEM simulations forCu interconnect structures with cathode end-of-line reservoirand side reservoir respectively are conducted to study the sidereservoir effects on EM lifetime.

With the continual decrease in interconnect line width, EMis no longer dominated by electron wind force. It has becomea combinational result of electron-wind force induced migra-tion (EWM), temperature gradient induced migration (TM)and thermomechanical stress gradient induced migration (SM).Dalleau et al. first proposed a simulation method of void for-mation, taking into account of EWM, TM and SM. But as linewidth shrinks, Tan et al. [6] found this method was no longeraccurate because some assumptions made in this method wasno longer valid. So Tan et al. proposed an accurate drivingforce formulation:

div(JA) =( EA

kBT 2 − 1

T+ α

ρ0

ρ

) N

kBTeZ∗ρD0

× exp(− EA

kBT

)j · ∇T (1)

div(Jth) =( EA

kBT 2 − 2

T

)NQ∗D0

kBT 2 exp(− EA

kBT

)∇T · (∇T )

−NQ∗D0

kBT 2 exp( EA

kBT

)∇ · (∇T ) (2)

div(Js) =( EA

kBT 2 − 1

T

)N�D0

kBT 2 exp(− EA

kBT

)∇σH · ∇T

+N�D0

kBT 2 exp(− EA

kBT

)∇ · (∇σH) (3)

Abowe vialocation

Line endlocation

Side reservoirend location

MNMN

(b)

(a)

Fig. 1. (a) AFD distribution of end-of-line reservoir structure andside reservoir structure; (b) Current density distribution of end-of-line reservoir structure and side reservoir.

Where div(JA), div(Jth), div(Js) are the atomic flux diver-gences due to EWM, TM and SM respectively. D the prefactorof the self-diffusion coefficient,EA the activation energy, T thelocal temperature,Z∗ the effective valence charge,Q∗ the heatof transport, N the atomic density,� the atomic volume, j thecurrent density, ρ the resistivity given asρ = ρ0[1+a(T−T0)],σH the hydrostatic stress.

Fig. 1 (a) illustrates the AFD distribution of end-of-linereservoir structure and side reservoir structure.

The sum of these three divergence components is the totalAFD, which can be used to predict void nucleation site. Inthe driving force formulation, current density and temperaturegradient distribution can be got through electro-thermal FEMsimulation. Stress distribution can be got through thermo-mechanical FEM simulation. However temperature divergence,stress gradient and stress divergence can’t be calculated just us-ing ANSYS postprocessor. So we use the driving force methodproposed by Tan et al. to do FEM simulation and use Matlabto calculate gradient and divergence. Finally total AFD is cal-culated and displayed in ANSYS.

32

33

The AFD distribution demonstrates that there are two max-imum atomic flux divergence (AFD) sites in end-of-line reser-voir structure. One is at above via location and the other isat line end location. The final failure void location for cath-ode reservoir has been proven to be at above via location. Butfor end-of-line reservoir structure with reservoir length largerthan critical length, EM lifetime is still limited by above viamaximum AFD. This limitation can be solved by side reservoirstructure. So to find the root cause of the EM lifetime improve-ment in side reservoir structure, this work analyze the impactof side reservoir on current density, temperature gradient andthermomechanical stress distributions. Considering the contri-bution of different driving forces to total AFD, SM is identifiedas the dominant driving force. As illustrated in Fig. 2, sidereservoir can impede stress built-up at above via location sothat the AFD of above via location can be decreased and EMlifetime improvement can be achieved.

So, side reservoir can be used to improve EM lifetime. Sidelocation is more prone to have void failure than end-of-linelocation in a side reservoir structure. The physical mecha-nism of side reservoir effect is that side reservoir can movethe maximum AFD sites from above via location to side reser-voir location. The root cause for EM lifetime improvement ofside reservoir structure is that side reservoir can impede stressbuilt-up at above via location.

Fig. 2. Stress distribution of end-of-line reservoir structure and sidereservoir structure.

References

[1] J. Lienig, Proc. of the 2013 ACM Int. Symp. on Physical Design,33–40.

[2] J.J. Clement, IEEE Transactions on Device and Materials Re-liability, 1(1), 33–42, (2001).

[3] C.M. Tan. Electromigration in ULSI interconnections. WorldScientific, 2010.

[4] E. Zschech, V. Sukharev, Microelectronic engineering, 82(3)629–638, (2005).


Independence of nanowire length distributionon the initial growth conditionsN.V. Sibirev 1,2, Y. Berdnikov1,2 and V.G. Dubrovskii1,21 St Petersburg Academic University, St Petersburg, Russia2 ITMO University, Kronverkskiy pr. 49, 197101 St Petersburg, Russia

This paper is devoted to the study of nucleation during thenanowire growth via the so-called vapor-liquid-solid (VLS)mechanism. The dependence of nanowire length distributionon growth conditions has been studied. Our studies provide anexplanation why nanowire length distribution is usually nar-rower than Poissonian.

Nanowire growth is an interesting process, as its behav-ior is determined by supersaturation dynamics in nano-sizeddroplet. The idea underlying the VLS mechanism is in lower-ing of nucleation barrier inside the catalyst droplet with respectto bare substrate. Nucleation process at the liquid-solid inter-face determines the crystal structure, growth rate and thereforenanowires physical properties. Uniformity of their propertiesis of paramount importance for different applications.

Here we study the length distribution of nanowires obtainedfrom the uniform size droplets. Recent studies have revealedthe following features of VLS growth nanowires:

(i) The temporal anticorrelation of individual nucleationevents during nanowire growth is explained by smallsize of catalyst droplet. Concentration inside the dropletfalls down when a new layer emerges in turn immediatelysuppresses the nucleation probability.

(ii) Nucleation antibunching and constant material transportto individual nanowire leads to saturation of the variancenanowire length.

(iii) Asyncrony in nanowire growth start does not affectlength distribution after the end of growth, see Fig. 1.

(iv) The dependence of incoming flux on diameter or contactangle could change the length distribution. In particu-lar the diffusion-induced growth of nanowires results inthe length distributions that are much broader than Pois-sonian. Shadowing and evaporation also increase thevariance of nanowire length.

25 50 75 100 125 150 175 200

1

10

100

α = 10α = 1α = 0.1α = 0.01α = 0.001

Mean nanowire length (Monolayer)

Var

ianc

e

Fig. 1. The dependence of length variance with growth time at dif-ferent initial conditions of nanowire formation. Here the parameterα describes the simultaneity of nanowires nucleation: at high α allnanowires emerge simultaneously while low α corresponds to ex-tended nucleation.

34


A compact photometer based on metal-waveguide-capillary:application to detecting glucose of nanomolar concentrationH. Huang1, M. Bai1, J. Hao1, J. Zhang1, H. Wu1 and B. Qu2

1 Department of Electronic Science and Technology, Faculty of Electronic Information and Electrical EngineeringDalian University of Technology, Dalian 116024, China2 Jordan Valley Semiconductors UK Ltd, Shanghai 201206, China

Trace analysis of liquid samples has wide applications in lifescience and environmental monitor. In this paper, a compactand low-cost photometer based on metal-waveguide-capillary(MWC) was developed for ultra-sensitive absorbance detec-tion. The optical-path can be greatly enhanced and muchlonger than the physical length of MWC, because the lightscattered by the rippled and smooth metal sidewall can be con-fined inside the capillary regardless of the incident-angle. Forthe photometer with a 7 cm long MWC, the detection limit isimproved ∼3000 fold compared with that of commercial spec-trophotometer with 1 cm-cuvette, owing to the novel nonlinearoptical-path enhancement as well as fast sample switching, anddetecting glucose of a concentration as low as 5.12 nM was re-alized with conventional chromogenic reagent.

Glucose detection is important for diagnosing diabetes mel-litus, hepatocirrhosis and psychopathy, etc., and many detec-tion methods, such as photometry (including spectrophotome-try [1–5] and paper based colorimetry [6–8], amperometry [9–11], fluoremetry [12–15], polarimetry [16], surface plasmonresonance [17], Fabry–Perot cavity [18], electrochemistry [19],and capillary electrophoresis [20–21] etc., have been reported.However, most of these methods require expensive equipmentsand detecting glucose of a few nanomolar concentration is still achallenge (e.g., for photometry measurement [2-1-28], till nowthe lowest detection limit of glucose is only 30 nM by usingPrussian blue nanoparticles as peroxidase mimetics [1]. For thecell researches on molecular level, nanomolar glucose detec-tion is normally required, such as inhibitive growth of humanprostate cancer [22] and CO2 fixation behaviors of Prochloro-coccus in the Ocean [23].

In this paper, a compact and low-cost photometer based onmetal-waveguide-capillary (MWC), which is a SUS316L stain-less steel capillary with electropolishing internal surface, wasdeveloped for ultra-sensitive absorbance detection. Becauselight can be confined inside the metal capillary regardless ofthe incident-angle, the optical-path can be greatly enhancedand much longer than the physical length of MWC, via lightscattering at the rippled and smooth metal surface. Moreover,for optical coupling and fluid inlet/outlet, a simple T-connectorwas developed to minimize the dead volume and avoid gas bub-ble trapping. For the photometer with a 7 cm long MWC, thedetection limit is improved ∼3000 fold compared with thatof commercial spectrophotometer with 1cm-cuvette, owing tothe novel nonlinear optical-path enhancement as well as fastsample switching, and detecting glucose of a concentration aslow as 5.12 nM was realized with conventional chromogenicreagent.

(a)

(b)

LED T-connector MWCphotodiode

lens

7 cmPeakpipe

Inletvalve

T-connector

MWCLED Photodiode

Outlet

PMMAtube

Quartzplate

Fig. 1. Schematic diagram of the MWC-based photometer.

Abs

orba

nce

Relative concentration10–9

0.001

0.01

0.1

1curvetteMWC

AEF

10–8 10–7 10–6 10–5 10–4 10–3 10–2

■

■

■

■

■

●

●●

●

●

●

●

●

●

●

●

■

■

Fig. 2. The relationship between the absorbance and concentration.

References

[1] W. Zhang, et al, Talanta 120 362–367 (2014).[2] T. Wang,et al, Chemistry – A European Journal 20, 2623–2630

(2014).[3] S. Chen,et al, Anal. Chem. 86, 6689–6694 (2014).[4] A.K. Dutta, et al, Sensors Actuators B: Chem, 177 676–683

(2013).[5] X. Chen, et al, Analytical Methods 4, 2183–2187 (2012).[6] W.-J. Zhu, et al, Sensors Actuators B: Chem,190 414–418

(2014).[7] M. Ornatska, et al, Anal. Chem. 83, 4273–4280 (2011).[8] A. Maattaen, et al, Sensors Actuators B: Chem,160, 1404–1412

(2011).[9] A.K. Dutta, et al, Sensors Actuators B: Chem, 173, 724-731

(2012).[10] C. Chen, et al, Biosensors Bioelectron, 26, 2311–2316 (2011).[11] C. Chen, et al, Biosensors Bioelectron, 24, 2726–2729 (2009).

35

36

[12] Y. Li, et al, Biosensors Bioelectron, 24, 3706–3710 (2009).[13] E.D. Paprocki, et al, Sensors Actuators B: Chem, 135, 145–

151 (2008).[14] Y.-S. Yuh, et al, J. Pharm. Biomed. Anal. 16, 1059–1066

(1998).[15] L.J. Blum, Enzyme Microb. Technol. 15, 407–411 (1993).[16] Y.-L. Lo, et al, Optics Communications 259, 40–48 (2006).[17] H. Nakamura, et al, Biosensors Bioelectron, 24, 455–460

(2008).[18] P. Liu, et al, Appl. Phys. Lett., 102, 163701 (2013).[19] K.-Y. Hwa, et al, Biosensors Bioelectron 62 127–133 (2014).[20] H. Yang, et al, Biosensors Bioelectron 26 295–298 (2010).[21] H.-L. Lee, et al, Talanta 64, 750–757 (2004).[22] Y. Mizushina, et al, Biochem. Pharmacol. 80, 1125–1132

(2010).[23] M. d. C. Munz-Marına, et al, PNAS 110, 8597–8602 (2013).


Advances in non-local density functional theory withapplications in 2D layered materialsS.J. ClarkDepartment of Physics, University of Durham, Science Labs, South Road, Durham DH1 3LE, United Kingdom

Abstract. The success of density functional theory is illustrated by the exponential growth in publications describing itsapplication to a wide range of areas in material science. Despite this success it has some well-known failings in describingsome properties of materials, most notably the electronic band gap and excitation energies. To compensate for this,empirical “hybrid functionals” have become commonplace where the underestimated DFT band gap can be opened byintroducing some Hartree–Fock mixing. However the “ab initio-ness” of the method is lost since fitting parameters arerequired; this is unsatisfactory. In this talk I will discuss some of my investigations into non-local density functional theoryand my various attempts at reintroducing the “ab initio” back into the method with applications in nanostructures such as the2D layered materials MX2 (M = Mo/W, X = S/Se/Te).

37

International Summer School and Workshop • Nanostructures for Photonics p-Thu1St Petersburg, Russia, June 27–July 2, 2016

DNA-based semiconductor nanowiresA. Aldana, B. Horrocks, A. HoultonChemical Nanoscience Labs, Newcastle University – NanoEmbrace ITN

One-dimensional nanostructures (1D-NS), including nanowi-res, nanotubes and quantum wires, have been regarded as themost promising building blocks for nanoscale electronic andoptoelectronic devices. Worldwide efforts in both the theoryand the experimental investigation of growth, characterizationand applications of 1D-NS have resulted in a mature, multidis-ciplinary field. As part of the NanoEmbrace ITN, the project in-volves the synthesis of nanowires (NWs) using DNA moleculesas a template, aiming to synthesize binary nanostructures andternary systems.

By varying the stoichiometric ratio of a system, the opticalproperties can be modified. Specifically, ZnxCd1−xS could beused in optoelectronic applications within the visible to UVspectral range, and also in solar energy driven devices, due totheir nonlinear optical and luminescence, quantum-size effect,and their tunable optical properties.

Recently, attention has been focused on the layer-structuredsemiconductor tin sulfide (SnS, p-type), due to its interestingproperties and potential application in photoconductors, near-infrared detector, holographic recording systems, and photo-voltaic materials with high conversion efficiency.

Aligned Zn Cd S DNA based nanowirex x1–

AFM TEM Fluorescence

1 µm2 µm

50 nm

2 µm

Fig. 1. Aligned ZnxCd1−xS DNA Based Nanowire.

These nanostructures are synthesized and then chemicaland physical characterized using a combination of techniquessuch as AFM, TEM, SEM, XRD, Raman spectroscopy andphotoluminescence (PL). Conductivity measurements are car-ried out using techniques such as scanned conductance Mi-croscopy (SCM), conductive AFM (cAFM) and I–V charac-terization with lithographically-defined microelectrodes.

38


Modelling the irreversible growth of nanostructuresY. Berdnikov1, N.V. Sibirev1,3 and V.G. Dubrovskii1,2,31 St Petersburg Academic University, St Petersburg, Russia2 Ioffe Institute, St Petersburg, Russia3 ITMO University, Kronverkskiy pr. 49, 197101 St Petersburg, Russia

Analysis of size distributions (SDs) of different nanosized“clusters” (droplets, surface islands, nanowires, linear rows)is paramount for controlling physical and mechanical proper-ties of nanomaterials and interesting from fundamental point ofview withal. Size uniformity is crucial for optic and optoelec-tronic applications of different types III–V of nanostructuredensembles. In this work we develop a theoretical model of ir-reversible growth (the growth with negligible decay of formingnanostructure) which can be adopted for different structures.

We discuss the pre-coalescence stage of epitaxial processwhich can be well understood within the rate equation (RE)approach to the concentrations ns of clusters of size s:

⎧⎨⎩

dnAdt = −DnA

∞∑s≥0

σsns + F

dnsdt = DnA(σs−1ns−1 − σsns), s ≥ 1

(1)

Here nA is the concentration of free monomers homoge-neously spread on the substrate and F is monomer influx.Particular form of the SDs depends on the capture rates σs ,which describe the strength of clusters of a given size to cap-ture monomers. Using the method of generating function theexact analytical solutions for ns can be found in cases of size-independent σs = const and σs = a+ s−1 size-linear capturecoefficients.

We distinguish the cases of heterogeneuos nucleation, whenclusters start from limited number of “seeds” fixed on the sub-strate, and homogeneous nucleation, when the cluster forms inrandom point of substrate as soon as two monomers collide.

We exploit the heterogeneous nucleation case to model thelength distributions of InAs nanowires (NWs) during the Au-catalized MOCVD [1] and CBE [2] growth with use of het-erogeneous approach. For size-linear and size-independentcapture coefficients (which correspond to collection of mate-rial from the the whole NW length and from part limited by

■

■

■

■

■

■

experimentsize-linear ( < )δ λs Lsize independent ( > )δ λs L

00

200

400

600

800

1000

1200

2000 4000 6000Mean length in MLs

9000 10000 12000

Leng

th d

ispe

rsio

n

Fig. 1. Dependences of length dispersion on mean length ofgold-seeded InAs NWs grown by MOCVD in a diffusion inducedregime [1]

▲

▲

▲

▲

▲

▲

▲

▲

▲ ▲▲

▲

▲▲

▲

▲

▲▲ ▲ ▲ ▲

▲▲

▲

▲

▲

▲ ▲

▲

▲

▲ ▲

■

■■

■

■

■

0.10 ML0.14 ML0.28 MLFit = 12 = 1a p

On CaF2

109 K139 K162 KFit = 9 = 0.5a p

On (Au,In)/Si(111)

0 1 2 3 4Scaled size

Sca

led

SD

0.0

2.0

4.0

6.0

8.0

1.0

Fig. 2. Scaled size distributions of fullerenes clusters on [3] and [4]substrates

constant diffusion length respectively) we obtain different de-pendences of length dispersion on mean length shown in Fig. 1.

Homogeneous approach was used to model size distribu-tions of fullerenes clusters on [3] and [4] substrates. Figure2 illustrates one of the most important features of irreversiblegrowth — the Family–Vicsek scaling property. It suggests thatSD as a function of the scaled size and the coverage in thelimiting regime is expected to scale as. The SDs scaling wasobserved for different coverages and growth temperatures.

References

[1] V.G. Dubrovskii et al., Cryst. Growth & Des., 16, 2167 (2016).[2] V.G. Dubrovskii et al., submitted to Nanotechnology.[3] N.V. Sibirev et al., Appl Surf Sci, 307, 46 (2014).[4] F. Loske et al., Phys. Rev. B, 82, 155428 (2010).

39


Capacitive pressure sensor with GaAs nanowire-PMMAdielectric layerA. Chandramohan1, G. Cirlin2, B. Mendis3, M. Petty1, A. Gallant1 and D. Zeze1

1 School of Engineering and Computing Sciences, Durham University, UK2 St Petersburg Academic University, St Petersburg, Russia3 Department of Physics, Durham University, UK

One of the most remarkable ways of perceiving information,the sense of touch, has led to a ubiquitous need for severalapplications like touch screens, and wearable electronics. Var-ious types of pressure sensors including piezoelectric [1], resis-tive [2] and capacitive [3] have been reported earlier. Despitethe success of the aforementioned approaches, the develop-ment of reliable pressure sensors, highly sensitive to low pres-sure (< 1 N/m2) values, still remains a challenge. It is knownthat the relative permittivity εr of polymer dielectric materi-als can be significantly increased by adding a small numberof conductive fillers near the percolation threshold [4]. Dueto the relationship between the percolation probability and thesize ratio of the filler, nanomaterials are ideal fillers for im-proving εr of thin layer dielectric composites. It is commonlyknown that II-VI and III–V compounds with wurtzite (WZ)structures are inherently piezoelectric. Recent published re-ports suggest that Gallium Arsenide (GaAs) nanowires (NWs)exhibit piezoelectric effect [5–6], although it immensely de-pends on the crystallographic phase but nevertheless makes itan interesting choice of material. However, GaAs alongsideInAs with zinc blende (ZB) lattices have rather low piezoelec-tric coefficients [7].

Nanosphere lithography [8] was used to produce nanodotsand nanoholes on 2 inch Si (111) wafers which were exploitedas master templates to create flexible stamps for nanoimprintlithography using polydimethylsiloxane (PDMS). This allowedprinting nanodots and nanoholes on 2 inch wafers, makingnanofabrication cost effective to pattern large-area substrates.GaAs NWs were grown on the bespoke samples by molecularbeam expitaxy (MBE) and exhibited a polytypism which mayplay a critical role in the generation of piezoelectric effect.

TEM and SEM investigation (Fig. 1) shows that the GaAsNWs have an aspect ratio (AR) over 150 and exhibit periodicregions of WZ (≥10 nm) and ZB (< 5 nm) along the lengthof the wire. However, the WZ phase was dominant. As grownnanowires (∼109/cm2) extracted from each of the 2 inch waferswere utilised to prepare polymer nanocomposites with varyingconcentrations. A suspension was produced by adding 2% (wt.)

5 nm

(a) (b)

Fig. 1. (a) SEM micrograph of a GaAs NW (b) TEM showing thealternating crystal phase.

0 50 100 150 200 250 300 3500.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07 GaAs in PMMAPMMAPolynomial Fit

Pressure (N/m )2

dC/C

0

Fig. 2. Change in capacitance with respect to applied pressure.

NWs to a solution of polymethylmethacrylate (PMMA) andanisole. Linear Au electrodes were fabricated on glass usingphotolithography. The solution was spin coated on the pat-terned substrate and baked to evaporate the solvent to create aNW-loaded dielectric film. Additional Au electrodes were de-posited on top of the film, perpendicular to bottom electrodes toform an addressable matrix of capacitive sensors. Capacitive-voltage (C-V) measurements were carried out to determine thecapacitance across the composite dielectric film. Known loadswere sequentially applied uniformly onto the device while thechange in the capacitance was measured. The devices withNW composite performed 10 times better than those contain-ing PMMA only (Fig. 2).

In summary, we have successfully demonstrated thatchanges in the capacitance for pressures as small as 0.5 N/m2

can be measured using the NW device produced, showing thathighly sensitive pressure sensors can be fabricated by integra-tion of GaAs nanowires in a PMMA matrix.

Acknowledgements

This project received financial support from the EuropeanCommission FP7 grants NanoEmbrace (GA-316751) andFUNPROB (GA-269169) and from the UK Royal Academyof Engineering/Leverhulme Trust.

References[1] J. Briscoe et al, Energy Environ. Sci. 6, 3035–3045 (2013).[2] Y. Noguchi et al, Applied Physics Letters 89, 253507 (2006).[3] G. Schwartz et al, Nature Communications 4, 1859 (2013).[4] D. Tan, et al, Materials Sciences and Applications 4, 6–15

(2013).[5] I.P. Soshnikov et al, Semiconductors 45, 9, 1082 (2011).[6] V. lysak et al, Physica Status Solidi – Rapid Research Letters,

10, 2, 172-175 (2015).[7] Y.M. Niquet, Physical Review B, 74, 155304 (2006).[8] A. Chandramohan, et al, SPIE, 9556 (2015).

40


Temperature dependent optical properties of single core-shellCdSe/ZnSe nanowire quantum dots grown along (111)BT. Cremel1,2, W. Lee3, M. Jeannin1,4, E. Bellet-Amalric1,2, G. Nogues1,4, K. Kyhm3,5 and K. Kheng1,2

1 Univ. Grenoble Alpes, F-38000 Grenoble, France2 CEA, INAC-SP2M, F-38000 Grenoble, France3 Dept. Cogno-mechatronics Eng., Pusan Nat’l University, Busan 609-735, Korea4 CNRS, Institut Neel, F-38000 Grenoble, France5 Physics and Physics Education, RCDAMP, Pusan Nat’l University, Busan, 609-735, Korea

Epitaxial CdSe quantum dots (QDs) are promising structuresfor applications as single photon sources since they can showsingle photon emission up to room temperature [1]. For thisreason, we have grown CdSe/ZnSe nanowire quantum dots(NW-QDs) along (111)B by molecular beam epitaxy passivatedwith a ZnMgSe shell. We report the temperature dependentoptical study of single QDs up to room temperature.

Exciton (X), biexciton (XX) and charged exciton (CX) linesare clearly identified by power dependent measurements witha cw laser excitation. The CX disappears around 75 K due tothe detrapping of the charge while the X and XX lines can beobserved up to room temperature.

Time-resolved photoluminescence (TRPL) at 5 K revealsthat the decay time of X (∼3 ns) is twice longer than the XXdecay time. This is in contrast with measurements in singleCdSe/ZnSe Stranski-Krastanov (SK) QDs [2] where the decay-times are similar for X and XX and shorter (∼300 ps). Thiseffect could be explained by the difference in the aspect ratioand the electromagnetic environment of the QDs.

The decay-times of our single QDs remain constant up to200 K and start to drop above this temperature due to non-radiative recombinations. This threshold temperature is 50Khigher compared to CdSe/ZnSe SK QDs [3] and indicates abetter charges confinement within our structure.

References

[1] S. Bounouar et al, Nano Lett. 12, 2977 (2012).[2] G. Bacher et al, Phys. Rev. Lett. 83, 4417 (1999).[3] I-C. Robin et al, Phys. Rev. B 74, 155318 (2006).

0 50 100 150 200 250 300

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0 XXX

Temperature (K)D

ecay

tim

e (n

s)

Fig. 1. Decay-time of the exciton and the biexciton in a CdSe/ZnSeNW-QD as a function of the temperature.

510 515 520 525 530 535 540 545 5500

5000

10000

15000

20000

25000

30000 CX

Wavelength (nm)

X

XX 7.53 µW15.1 µW19.8 µW38 µW101.7 µW

PL in

tens

ity (

coun

ts)

Fig. 2. Microphotoluminescence spectrum at 5 K of a single CdSe/ZnSe NW-QD as a function of a 488 nm cw laser power.

41


Terahertz emission from low-temperature grownGaAs nanowiresA. Dıaz Alvarez1, G. Tutuncuoglu2, T. Xu1,3, M. Berthe1, J-P. Nys1, S. Plissard4, A. Fontcuberta i Morral2,J-F. Lampin1, B. Grandidier1

1 Institut d’Electronique, de Microelectronique et de Nanotechnologies (IEMN), CNRS, UMR 8520, DepartementISEN, 41 bd Vauban, 59046, Lille Cedex, France2 Laboratoire des Materiaux Semiconducteurs, Institut des Materiaux, Ecole Polytechnique Federale deLausanne, CH-1015 Lausanne, Switzerland3 Sino-European School of Technology, Shanghai University, 99 Shangda Road, Shanghai, 200444, Poeple’sRepublic of China4 CNRS-Laboratoire d’Analyse et d’Architecture des Systemes (LAAS), Universite de Toulouse, 7 avenue ducolonel Roche, 31400 Toulouse, France

III–V Semiconductor Nanowires have been shown to be able tooperate as active components of optoelectronic devices such asLEDs, Lasers or Waveguides [1–4], sometimes outperformingthe equivalent thin-film version of the device. Nanowires in-heritely have cylindrical symmetry and nanometric sized diam-eters, which produces a quasi-1D system in which the surface-to-volume ratio is very large. As a consequence; first, the roleof the surface is determinant for the onset of light emissionin some nanowires such as GaAs [5], then, the tube-like ge-ometry of the nanowires combines micrometric axial lengths,which is the scale for the wavelenghts of photons in the visi-ble range, with nanometric scale diameters, in which quantumconfinement effects for electrons are important, allowing theappearance of novel light confining modes that can enhance theabsorption or emission of light in the nanowire [6–7]. Addi-tionaly, semiconductor nanowires —- normally grown follow-ing the VLS-growth scheme- offer great flexibility to produceaxially or radially extended heterostructures.

In this context, semiconductor nanowires for which is pos-sible to detect and emit pulses in the range of THz frequencieshave been succesfully grown during the last decade [8–10].The onset of THz sensing in semiconductor materials rely onthe ability of the material to follow the fast optical transients ofexciting light sources such as pulsed femtosecond lasers, forwhich is necessary to have materials with very high rates ofnon-radiative recombination such as low-temperature grownGaAs [11], making them very suitable for THz radiation de-tection. To emit THz radiation, the photoexcited electron-holepairs are split apart in different directions along the material,which, in conjunction with the fast recombination, forms aradiating dipole which emits a pulse in the range of THz fre-quencies, such electron-hole separation can be produced by anexternal potential, as it is done with LT-GaAs photoconductiveantennae [12], by built-in electric field formed at accumulationof depletion layers at the surface of some semiconductors suchas n or p-type doped GaAs or by other physical effects likephoto-dember, as in InAs materials [13].

Recently, it has been shown that it is possible to synthe-size LT-GaAs nanowires by growing the low-temperature GaAsshell conformally to self-catalyzed GaAs nanowires. The LT-GaAs nanowire shell presents structural and electrical proper-ties equivalent to the low-temperature grown film GaAs. Fur-thermore, these nanowires show a very short carrier lifetime of

few picoseconds [14].In the present work, the THz emission profile of core-shell

GaAs/LT-GaAs have been studied in the time domain. To dothis, a standard THz-TDS emission setup has been used inwhich a Ti:Sa femtosecond laser pumps the target substrate,the emitted THz radiation is collected and detected by a dipo-lar LT-GaAs antenna as a function of the probe beam delay.Complementary, self-catalyzed GaAs nanowires and core-shellGaAs-LT-GaAs nanowires with a AlGaAs outer shell on tophave been also characterized with THz-TDS in order to inves-tigate the impact of the LT-GaAs shell as well as the nanowiresurfaces on the emission profile of the nanowires.

References

[1] Y. Huang, X. Duan and C.M. Lieber, Small, 1, 142–147 (2005).[2] X. Duan, Y. Huang, R. Agarwal, C.M. Lieber and C.G. Fast,

Nature, 421, 241–245 (2003).[3] M.S. Gudiksen, L.J. Lauhon, J. Wang, D.C. Smith and

C.M. Lieber, Nature, 415, 617–620 (2002).[4] M. Law, Science (80-. ) 305, 1269–1273 (2004).[5] O. Demichel, M. Heiss, J. Bleuse, H. Mariette, and I.A. Fontcu-

berta Morral, Appl. Phys. Lett. 97, 2010–2012 (2010).[6] L. Cao, et al, Nat. Mater., 8, 643–647 (2009).[7] P. Krogstrup, et al, Nat. Photonics, 7, 1–5 (2013).[8] H. Ahn, et al, Appl. Phys. Lett., 91, 1-4 (2007).[9] D.V. Seletskiy, et al, Phys. Rev. B - Condens. Matter Mater.

Phys., 84, 1–7 (2011).[10] A. Arlauskas, et al, Nano Lett., 14, 1508–1514 (2014).[11] K.a. McIntosh, K.B. Nichols, S. Verghese, and E.R. Brown,

Appl. Phys. Lett., 70, 354 (1997).[12] Y.C. Shen, P.C. Upadhya, E.H. Linfield, H.E. Beere, and

A.G. Davies, Appl. Phys. Lett., 83 3117–3119 (2003).[13] V.L. Malevich, R.Adomavicius, andA. Krotkus, Comptes Ren-

dus Phys., 9, 130–141 (2008).[14] A. Dıaz Alvarez, et al, Nano Lett., 15, 6440–6445 (2015).

42


Chemical potentials and growth ratesof gold-catalyzed ternary InGaAs nanowiresJ. Grecenkov1, V.G. Dubrovskii1,2,31 St Petersburg Academic University, St Petersburg, Russia2 Ioffe Institute, St Petersburg, Russia3 ITMO University, Kronverkskiy pr. 49, 197101 St Petersburg, Russia

Abstract. By applying the Redlich–Kister polynomial model, we calculate chemical potentials of quaternary liquid alloysduring the gold-catalyzed vapor-liquid-solid growth of ternary InxGa1−xAs nanowires. We then use these chemicalpotentials in the Zeldovich nucleation rate to obtain the nanowire axial growth rates versus the deposition conditions andindium compositions.

Introduction

Due to their ability to relax elastic strain on free sidewalls,nanowires are promising for monolithic integration of III–Voptoelectronic materials with silicon electronic platform [1,2].III–V nanowires are usually grown by the gold-assisted vapor-liquid-solid (VLS) method wherein the III–V growth speciesare first transferred from vapor to liquid and then crystallize atthe liquid-solid interface under the catalyst droplet [3,4].

Chemical potentials of Au-III–V liquid alloys play a cru-cial value in understanding the VLS growth and crystal struc-tures of III–V nanowires. Previous works have shown that theanalysis of chemical potentials of gold-assisted binary III–Vnanowires allows for a delicate control over the VLS growthrates and wurtzite-zincblende polytypism within nanowires [5-12]. In this work, we consider a more complex case of ternarygold-catalyzed InGaAs nanowires and study how chemical po-tentials affect the VLS growth rates under variable depositionconditions and indium compositions [13].

1. Chemical potential

In the mononuclear growth mode described by classical nucle-ation theory, the axial growth rate of a nanowire is given by theZeldovich nucleation rate [10]

dL

dt= J0 exp

(− 2

�μkBT

)(1)

whereJ0 is a prefactor, is the effective surface, or edge energyof a two-dimensional island of a bilayer height, �μ is the dif-ference of chemical potentials of a III–V pair (InxGa1−xAs) inthe liquid and solid phases, T is the surface temperature and kBis the Boltzmann constant. No vapor-solid contributions [14]are considered in Eq. (1). Since the prefactor only weaklydepends on the chemical potential and surface energy, the nu-cleation probability and consequently the nanowire growth rateis mainly determined by the Zeldovich exponent of the nucle-ation barrier in Eq. (1). The latter is extremely sensitive tothe chemical potential value and hence the �μ needs to bedetermined as precisely as possible [11].

In the case of gold-catalyzed ternary III–V nanowires, weare dealing with quaternary Au-III–V liquid alloys such as Au-In-Ga-As. These chemical potentials can be calculated by ap-plying the Redlich–Kister polynomial corrections to the Gibbsthat take into account the interactions introduced by mixing the

components. The Redlich–Kister polynomials have the form

GR−K = 1

2

4∑i,j=1i �=j

cicj

n∑m=0

Lmij(ci − cj

)m + 1

3

4∑i,j=1i �=j �=k

cicj ck L0ijk

(2)with Ci denoting the concentrations of different elements inthe liquid alloy and Lmij are the interaction parameters betweenthe atoms of different elements [15–21]. Once the Gibbs en-ergy is calculated, the chemical potential of a given element isobtained as a derivative of the Gibbs energy with respect to thecorresponding concentration

μli = ∂G

∂ci(3)

The liquid-solid chemical potential difference can be writ-ten down as

�μ = x(μlIn+μlAs−μsInAs

)+(1−x)(μlGa+μlAs−μsGaAs

)(4)

under the central simplifying assumption on the identical in-dium content x in liquid and solid phases. The quantities μlIn,μlGa andμlAs denote chemical potentials of the indium, galliumand arsenic atoms in the liquid phase, while μsInAs and μsGaAsstand for chemical potentials of InAs and GaAs pairs in solid.

2. VLS growth rate

We can now exploit Exp. (1) to estimate the growth rates ofgold-catalyzed InGaAs nanowires under different conditions.Assuming, for simplicity, that the nucleation process occursin the centre of the the liquid-solid interface rather than at thetriple phase line, we can use the expression for the surfaceenergy of a newly formed island in the form [10]

= x In + (1 − x) Ga (5)

with In being the surface energy of an InAs island in an Au-In droplet (the influence of arsenic on the surface energy canbe neglected due to its low concentration inside the catalystdroplet) and respectively Ga in the same value for a GaAsisland in an Au-Ga droplet. We take Ga = Ga(cGa) and In = In(cIn) as the linear interpolations between pure goldand group III metal droplets [11].

The prefactor in the Zeldovich nucleation rate can be cal-culated using the expression of Ref. [22]

J0 = 33/4

√πD5

(h

�35

)2

eμS35cAs�μ

1/2 (6)

43

44

Here,D5 is the diffusion coefficient of the arsenic adatomsthrough the liquid, �35 is the elementary volume of a III–Vpair, h is the height of a bilayer andμS35 is the averaged value ofchemical potential of a III–V pair in the solid phase. Equations(1) to (6) constitute the complete system for computing theVLS growth rates of gold-catalyzed ternary nanowires versustemperature and droplet composition.

3. Results and discussion

Figure 1 shows the VLs growth rate of gold-catalyzedInxGa1−xAs nanowires versus the indium composition at750 K, for a fixed arsenic concentration c5 = 0.05 and threedifferent total group III concentrations c3 = 0.3, 0.5 and 0.7.Quite interestingly, we observe a non-monotonic dependenceof dL/dt on x for c3 = 0.3, while for larger c3 the growth rategradually increases with x. Figure 2 shows the chemical po-tentials in quaternary Au-In-Ga-As alloy with respect to solidInxGa1−xAs, computed for the same parameters as in Fig. 1.Clearly, these chemical potentials increase monotonically withx and hence the non-monotonic behavior of dL/dt on x atc3 = 0.3 is due to a combined effect of chemical potential andsurface energy.

c3 = 0.3cc

3

3

= 0.5= 0.7

0.00.01

0.1

1

10

0.2 0.4 x

d/d

(nm

/s)

Lt

0.6 0.8 1.0

Fig. 1. Growth rates of gold-catalyzed InxGa1−xAs nanowires ver-sus x at a fixed temperature of 750 K, arsenic concentration of0.05 and three different total concentrations of group III atoms in tedroplet c3 = c In + cGa.

Since both chemical potential and surface energy enter theZeldovich exponent, even the slightest changes in them resultin a dramatic effect on the VLS growth rate. In particular,our results show that InxGa1−xAs nanowires of the same com-position x grow at very different rates depending on the totalgroup III concentration (determined by the V/III ratio). Forexample, low total group III flux may stop completely the VLSgrowth for small enough x, while at higher group III fluxessame nanowires would continue growing normally.

4. Conclusion

The model presented is capable of describing the VLS growthrates of gold-catalyzed ternary InGaAs nanowire based on clas-sical nucleation theory involving the macroscopic Zeldovichnucleation rate and quaternary chemical potentials. The lat-ter has been calculated using the Redlich–Kister polynamials.The results show that changing the In/Ga as well as the totalV/III flux ratio under otherwise fixed experimental conditionsmay cause an abrupt growth stop within a plausible range ofparameters. We now plan to extend our approach for predict-ing the crystal structures of InGaAs nanowires and to apply

c3 = 0.3cc

3

3

= 0.5= 0.7

0.0 0.2 0.4 x 0.6 0.8 1.0

600

500

400

300

200

100

Δµ (

meV

)

Fig. 2. Chemical potentials versus x for the same parameters as inFig. 1.

this approach to other ternary nanowires based on both groupIII and group V intermixing.

Acknowledgements

JG acknowledges the FP7 project NanoEmbrace (Grant Agree-ment 316751). VGD thanks the Russian Foundation for BasicResearch for the financial support under different grants.

References

[1] V.G. Dubrovskii, N.V. Sibirev, et al, Cryst. Growth Des. 10,3949 (2010).

[2] X. Zhang, V.G. Dubrovskii, N.V. Sibirev, et al, Cryst. GrowthDes. 11, 5441 (2011).

[3] N. Wang, Y. Cai and R. Zhang, Mater. Sci. Eng. Res. 60, 1(2008).

[4] K.A. Dick, Prog. Cryst. Growth Charact. Mater. 54, 138(2008).

[5] V.G. Dubrovskii and J. Grecenkov, J. Phys.: Conference Series541, 012001 (2014).

[6] F. Glas, M.R. Ramdani, G. Patriarche et al, Phys. Rev. B 88,195304 (2013).

[7] F. Glas, J. Appl. Phys. 108, 073506 (2010).[8] C. Colombo, D. Spirkoska, M. Frimmer, et al, Phys. Rev. B

77, 155326 (2008).[9] V.G. Dubrovskii, Appl. Phys. Lett. 104, 053110 (2014)

[10] V.G. Dubrovskii, Theory of VLS Growth of Compound Semi-conductors. In: A.F. i Morral, S.A. Dayeh and C. Jagadish,editors, Semiconductors and Semimetals, 93, Burlington: Aca-demic Press, 2015, p. 1–78.

[11] V.G. Dubrovskii, J. Chem. Phys. 142, 204702 (2015).[12] E. Gil, V.G. Dubrovskii, G. Avit et al, Nano Lett. 14, 3938

(2014).[13] A.S. Ameruddin, P. Caroff, H.H. Tan et al, Nanoscale 7, 16266

(2015).[14] V.G. Dubrovskii, V. Consonni, L. Geelhaar et al, Appl. Phys.

Lett. 100, 153101 (2012).[15] Calphad 18, 2, 177 (1994).[16] J. Wang, Y. Liu, L. Liu et al, Calphad 35, 242 (2011).[17] H. Liu, Y. Cui, K. Ishida et al, Calphad 27, 27 (2003).[18] T. Anderson, I. Ansara, J. Phase Equilib. 12, 64, (1991).[19] M. Ghasemi, B. Sundman, S.G. Fries et al, J. of Alloys and

Compounds 600, 178 (2014).[20] A.T. Dinsdale, Calphad 15, 317 (1991).[21] I. Ansara, C. Chatillon, H.L. Lukas et al, Calphad 18, 177

(1994).[22] V.G. Dubrovskii and J. Grecenkov, Cryst. Growth Des. 15, 340

(2015).


Flexible white light emitting diodes based on nitride nanowiresand nanophosphorsN. Guan1, X. Dai1, A. Messanvi1,2, H. Zhang1, J.Yan1,3, F.H. Julien1, C. Durand2, J. Eymery2 andM. Tchernycheva1

1 Institut d’Electronique Fondamentale, UMR 8622 CNRS, 91, 91405 Orsay, France2 CEA/CNRS/Universite Grenoble Alpes, CEA, INAC, SP2M, 17 rue des Martyrs, 38054 Grenoble, France3 Institute of Semiconductors, Chinese Academy of Sciences, 100083 Beijing, China

Today white light emitting diodes (LEDs) are playing an impor-tant role in the light revolution all over the world, motivatedby their significant role in reducing global energy consump-tion and practical solid-state lighting applications. Moreover,flexible light sources attract huge attention for a number of ap-plications (e.g. curved surface displays). Nowadays, the whiteorganic LEDs (WOLEDs) generating white light by mixingof different colored emitters [1] and the phosphor-convertedWOLEDs [2–3] are the key technologies to realize the flexi-ble white emitters. After a decades-long effort, WOLEDs arenow commercialized thanks to their low cost, compatibilitywith various flexible substrates and relative ease of process-ing. However, compared to the nitride semiconductors, theorganic LEDs have a short lifetime and a lower luminance espe-cially for the blue component. Recently, we have demonstratedblue flexible LEDs based on vertical nitride nanowires (NWs)encapsulated in a flexible polymer [4]. Here we report theflexible white phosphor-converted LEDs based on core/shellInGaN/GaN NW blue LEDs grown by MOCVD [5], whichcombine the high flexibility of polymers with the high effi-ciency of the nitride NWs and nanophosphors [6]. Figure 1

(b)

2 µm

n-G

aN+

MQ

Ws

7x InGa/GaNMQWs

p-GaN

n-GaN

20 µm

(a)

200 nm

7x InGa/GaNMQWs

p-GaN

n-GaN

(c)

Fig. 1. (a) 30◦ tilted SEM image of core/shell InGaN/GaN NWgrown on c-sapphire. (b) Detail of a single wire showing thecore/shell QW region and its schematics. (c) Transversal cross-sectional STEM-HAADF image taken along the c-zone axis show-ing the shell structure with the MQWs and p-GaN.

➡ ➡

Sapphire

Peel off PDMS

YAG:Ce phosphor

Ni/Au Ti/Al/Ti/Au

Flip Flip

SilverNWs

Arbitary host substrate Metal foil

(a) (b) (c)

Fig. 2. Fabrication process flow of the flexible white LEDs basedon free-standing polymer-embedded NWs.

●

0.0 0.2

0.2

0.4

Location oflight emission

0.4

0.6

0.6

0.8

0.8

xy

(a) (b)

(c) (d)

Fig. 3. Photographs of the operating flexible white LED under bend-ing radii of (a) infinity (c) 5 mm (d) –5 mm. (b) CIE 1931 chromatic-ity diagram of flexible white LEDs under injection current densityof 3.9 A/cm2 (chromaticity coordinates x = 0.3011, y = 0.4688;CCT = 6306 K; CRI = 54).

shows the SEM image and the internal structure of core/shellInGaN/GaN NWs. Figure 2 illustrates the fabrication processof the flexible LED, which consists in NW embedding witha phosphor-doped PDMS layer, peel-off and contacting. Asshown in Figure 3, the white source has a good flexibility. Nodegradation of the current and the electroluminescence emis-sion was observed during bending or after 50 days storage inambient conditions without any external encapsulation.

References

[1] J. Kido, et al, Science, 267, 1332-1334 (1995).[2] M.A. Baldo, et al, Nature, 395, 151–154 (1998).[3] S. Reineke, et al, Nature, 459, 234–238 (2009).[4] X. Dai, et al, Nano Letters, 10, 6958-6964 (2015).[5] R. Koester, et al, Nano Letters, 11, 4839-4845 (2011)[6] N. Guan et al, ACS Photonics, in press.

45


Influence of As flux on heterostructure formationin AlxGa1−xAs nanowires obtained via vapor-liquid-solid growthA. Koryakin, N. Sibirev1 St Petersburg Academic University, St Petersburg, Russia2 ITMO University, Kronverkskiy pr. 49, St Petersburg, 197101, Russia

Semiconductor III–V nanowires (NWs) with axial heterostruc-tures are used as building blocks for novel electronic and opto-electronic devices. Vapor-liquid-solid growth (VLS) is themethod most common to synthesize NWs via molecular-beamepitaxy or chemical vapor deposition. Most of the publishedtheoretical works usually consider the group III influence onthe NW growth only. The importance of the group V influ-ence was shown recently [1,2]. A composition change in III–VNWs is usually formed by an interchange of group III or V ma-terial flux, i.e. the group III or V material flux is interchangedto create two neighbouring segment with different group III orV composition correspondingly. For instance, the NWs withthe AlAs/GaAs, GaInAs/InAs, InGaAs/GaAs and InP/InAs,InAsP/InAs, GaAsP/GaP heterojunctions can be grown by thismethod. In present work, we consider the influence of groupV flux on axial heterostructure formation in III–V NWs ob-tained via VLS growth. As an example of material system,we study Au-catalyzed AlxGa1−xAs NWs. We theoreticallyinvestigate a new method to form heterostructures in NWs viaan alteration of As flux. The idea of this method is based on thefact of different concentration changes of group III elementsin the catalyst droplet with concentration change of group Velement. The group III composition change can be performedby the group V flux change provided that the group III fluxesare constant. The method is only valid when two or more groupIII fluxes are used to grow multicomponent NWs. We modelheterostructure formation using the method developed in thework [3]. In the model [3], the chemical potential differenceof species dissolved in the droplet and in solid state that de-termine the NW growth rate and composition are calculated.The catalyst droplet at the NW top are considered as a four-component regular solution (three NW growth materials andone catalyst) and the interaction coefficients are found usingthe Stringfellow formula [4]. The NW composition changewith the alteration of As flux is explained by the difference inthe chemical potential variation of AlAs and GaAs pair.

Finally, we theoretically predict the opportunity to form ax-ial heterostructures in AlxGa1−xAs nanowires obtained via analteration ofAs flux only. Considering the typicalAlxGa1−xAsNW growth conditions, we show that the composition profileshape depends very strongly on the As flux.

Acknowledgement

This work was supported by the grant of the Ministry of Educa-tion and Science of RF (agreement No. 14.613.21.0044, uniqueproject identifier RFMEFI61315X0044).

References

[1] F. Glas et al, Phys. Rev. B 88, 195304 (2013).[2] V.G. Dubrovskii, J. Grecenkov, Cryst. Growth Des. 15, 340

(2015).[3] A.A. Koryakin, N.V. Sibirev, D.A. Zeze, V.G. Dubrovskii J.

Phys.: Conf. Ser. 643, 012007 (2015).[4] J.C. Stringfellow, J. Phys. Chem. Solids 33, 665 (1972).

46


Narrowing of diameter distribution during growthof Ga-catalyzed GaAs nanowiresE.D. Leshchenko1,2, M.A. Turchina1, V.G. Dubrovskii1,3,4, T. Xu5,6, A. Dıaz Alvarez5, S.R. Plissard5,7,P. Caroff5,8, F. Glas9, B. Grandidier5

1 St Petersburg Academic University, St Petersburg, Russia2 St Petersburg State University, St Petersburg 198504, Russia3 Ioffe Institute, St Petersburg, Russia4 ITMO University, St Petersburg 197101, Russia5 Institut d’Electronique, de Microelectronique et de Nanotechnologies (IEMN), CNRS, UMR 8520,Departement ISEN, Lille Cedex 59046, France6 Sino-European School of Technology, Shanghai University, Shanghai 200444,China7 CNRS-Laboratoire d’Analyse et d’Architecture des Systemes (LAAS), Universitede Toulouse,Toulouse 31400, France8 Department of Electronic Materials Engineering, Research School of Physics and Engineering,The Australian National University, Canberra, ACT 0200, Australia9 CNRS-Laboratoire de Photonique et de Nanostructures (LPN), Marcoussis 91460, France

Self-catalyzed nanowires based on group III–V materials areattractive for use in photovoltaic, optoelectronic and other ap-plications, due to their high crystalline quality, the possibilityof integration of NW-based photonic devices on Si and the ab-sence of unwanted foreign element contamination. Anotheradvantage of these structures is the recently discovered self-equilibration of the nanowire diameter [1–3]. This effect in-volves the nanowire diameters tending to a certain stable onein the course of vapor-liquid-solid growth, irrespective of theinitial Ga droplet diameter distribution. This requires that thearsenic vapor flux should be larger than that of gallium andthat the V/III influx imbalance should be compensated by adiffusion flux of gallium adatoms. In the opposite case, whenthe gallium vapor flux is larger than the arsenic vapor flux,the radial growth is infinite. Radius dependencies in self-equilibration regime at different initial radii of individual NWsare presented in Fig. 1.

But model [3] does not take into account the second deriva-tive in the Fokker–Planck equation which describes kineticfluctuations in analytical analysis. This results in the simpli-fied form of diameter distribution function, in particular theabsence of the Poissonian broadening. In this contribution wepresent the evolution of radius distribution function during Ga-

30 35 40 45 500

0.2

0.4

0.6

Dia

met

er d

istr

ibut

ion

(,

)f

D t (b)

t = 0

t = 51 s

t = 90 s

t = 201 s

Nanowire diameter , (nm)DNanowire diameter , (nm)D30 40 50 60 70

0.00

0.04

0.08

0.12

0.16(c)

t = 0

t = 51 s

t = 201 s

Diameter (nm)

Ga dropletGaAs NWs

0 10 20 30 40 50 60 70 80 90 1000

20

40

60

80

100

Sta

tistic

al c

ount

(%

) (a)

Fig 2. Histograms of the Ga droplet diameter distribution and the final nanowire diameter distribution (a). Inset: 30◦ tilted SEM imageof the corresponding GaAs nanowire array. Diameter distributions in self-equilibration regime corresponding to approximation of driftand diffusion coefficient equality model (b) and the full model (c). The solid curves are the analytical solution, the dotted curves are thenumerical calculations, the dash dotted curves are initial Gaussian distributions of Ga droplets.

Nanowire length , (nm)L

Nan

owire

rad

ius

, (nm

)R

0 500 1000 1500 200005

1015202530354045505560

Fig 1. The length-dependent NW radius at different initial radiiwith (solid curves) and without (dashed curves) accounting for thehyperbolic tangent dependence of the diffusion on the NW length [4]in the self-equilibration regime.

catalyzed vapor-liquid-solid growth of GaAs nanowires usingboth the full model and the approximation of drift and diffu-sion coefficient equality model. For this purpose, we solvethe Fokker–Planck equation by applying the Green’s functionmethod and convoluting Green’s function with the initial con-ditions. Different growth modes are obtained depending on theV/III flux ratio including the modes of infinite growth and self-

47

48

stabilized radius growth. For comparison, numerical solutionof the Fokker–Planck equation using by the implicit differencescheme is presented. Fig. 2(b,c) shows the analytical solutionof diameter distribution evolution in self-stabilized growth.

The obtained results can be used to choose the optimalgrowth conditions of self-catalyzed GaAs nanowire arrays withhigh degree of uniformity.

References

[1] G. Priante, S. Ambrosini, V.G. Dubrovskii,etal, Cryst. Growth Des. 13, 3976 (2013).

[2] J. Tersoff, Nano Lett. 15, 6609 (2015).[3] V.G. Dubrovskii, T. Xu, A. Dıaz Alvarez, et al, Nano Lett. 15,

5580 (2015).[4] V. Consonni, V.G. Dubrovskii, A. Trampert, et al, Phys. Rev. B

85, 155313 (2012).


Electron beam induced current microscopy investigationof GaN nanowire arrays grown on Si substratesV. Neplokh1, A. Ali1, F.H. Julien1, M. Foldyna2, I. Mukhin3,4, G. Cirlin3−5, J-C. Harmand6, N. Gogneau6,M. Tchernycheva1

1 Institut d’Electronique Fondamentale, UMR CNRS 8622, University Paris Sud 11 Orsay, 91405 France2 Laboratoire de Physique des Interfaces et Couches Minces (LPICM), CNRS, Ecole polytechnique, UniversiteParis Saclay, 91128 Palaiseau, France3 ITMO University, St. Petersburg 197101 Russia4 St Petersburg Academic University, St Petersburg, Russia5 Ioffe Institute, St Petersburg, Russia6 Laboratoire de Photonique et de Nanostructures Route de Nozay, 9146 Marcoussis

Abstract. We report on the electron beam induced current (EBIC) investigation of GaN nanowires grown on n-type Si (111)substrates. The objective of this study is to acquire information about the modifications of the substrate properties inducedby the wire growth in view of the future development of nitride- on-silicon tandem photovoltaic devices We show that thegrowth procedure using a deposition of an ultra-thin AlN film prior to the nanowire growth step leads to the formation of ap-n junction in the Si substrate with a high surface conductivity The induced p-n junction exhibits a photoresponse over thespectral range from 36 to 1100 nm The properties of the induced p-n junction are investigated in cross-section and topviewconfigurations with EBIC microscopy For a localized contact of the GaN nanowires the induced current collection range inSi extends over a few millimeters. The treatment of the surface using reactive ion etching with a CHF3 plasma leads to theinhibition of the surface conductivity and to the appearance of an S-shape in the currentvoltage characteristics underillumination Nevertheless, the conversion efficiency of the plasma-treated sample under AM1.5G solar spectrum isestimated to be in the 21–27% range.

49


Investigation of GaN nanowirescontaining AlN/GaN multiple quantum discsV. Piazza1, A. Babichev2, N. Guan1, M. Morassi1, V. Neplokh1, P. Quach1, F. Bayle1, L. Largeau3,F.H. Julien1, J.-C. Harmand3, N. Gogneau3, M. Tchernycheva1

1 Institut d’Electronique Fondamentale, Universite Paris Sud XI, 91405 Orsay Cedex, France2 ITMO University, St. Petersburg, 197101, Russia3 Laboratoire de Photonique et Nanostructure-CNRS, Route de Nozay, 91460 Marcoussis, France

In this work we investigate single GaN nanowires withGaN/AlN multiple quantum discs (MQDs) insertion using elec-tron beam induced current (EBIC) microscopy. The struc-tures were grown by plasma-assisted molecular beam epitaxy(PAMBE) on Si (111) substrate with a thin AlN buffer layer.The wires consist of a n-GaN base with a thin AlN externalshell, of an active region with 20 periods of 0.6 nm GaN discs/ 6 nm Al barriers and of an n-GaN cap, as confirmed by SEMand TEM (Fig. 1a). Single NWs were dispersed on Si/SiO2substrates with alignment marks and contacted using electronbeam lithography and Ti/Au metallization. Structural char-acterization results were used to build a band gap-vs-positionsimulation with NextNano© software which showed the pres-ence of a triangular confining potential well at the AlN/n-GaNinterface (similar to the 2D electron gas in HEMT transistorsinduced by the polarization discontinuity between AlN andGaN) (Fig. 1b). The position of this potential well is differentfrom the thin film case since the nanowires exhibit N polarityinstead of Ga-polarity [1].

EBIC mapping was used to confirm experimentally the sim-ulated band profile. Being a charge collection microscopy tech-

CB VB–400

–5–4–3–2–101234567

–300 –200 –100 0Position [nm] (x)

100 200 300 400 500

(b)n-GaNGaN + AlN shell MQDs

20 nm

0.6–0.8 nm

AlN barriers n-GaN

GaN QDs (a)

Fig. 1. (a) TEM image of a single NWs showing the internal struc-ture; (b) Band profile calculated with NextNano© software, showingthe MQD structure and the confining potential well at theAlN/n-GaNinterface.

200 nm

Schottkycontact

GaN/AlNMQW

(a)

(b)

SEM

EBIC(positive)

EBIC(negative)

EBICExponential fitting SchottkyExponential fitting MQDsSEM

– –4.00E 010

– –3.50E 010

– –3.00E 010

– –2.50E 010

– –2.00E 010

– –1.50E 010

– –1.00E 010

– –5.00E 011

0.00E+000

5.00E 011–

1.00E–010

EB

IC (

A)

0 200 400 600 800 1000 1200 1400Position (nm)

SE

M (

a.u.

)

Fig. 2. (a) RGB image superposing SE image and EBIC mapping ofa single NW at 0 V bias. It is possible to observe the different signalsdue to Schottky contact and to the MQDs region (legend aside); (b)EBIC and SE profiles correspondent to panel (a). An exponentialfitting was used for the estimation of the diffusion length.

nique, EBIC technique is able to trace a map of electrical sig-nals related to the structural features, which, together with afine tuning of operational parameters, allows to obtain quanti-tative results related to doping level and diffusion length [2].Evaluating the results, it was possible to identify the electronicfeatures within the wire and their behaviour with an applied bias(ranging from +4 V to -4 V). In these conditions, EBIC map-ping showed that MQDs region produces electrical activity,meaning that it corresponds to a depleted region with internalfield which is due to band structure bending. In addition, also aSchottky contact is revealed at the metal/nanowire interface atone of the two ends, due to the presence of a thin AlN layer on

50

51

the n-GaN base side, as coherently shown by TEM (Fig. 2a).The EBIC profile fitting (Fig. 2b) allows to estimate the minor-ity carrier diffusion length in GaN (around 110 nm and 220 nm,depending on experimental conditions). The investigation hasbeen enriched also by the I–V measurement.

These results allow to assess the electrical features (e.g.Schottky barrier, band bending in the heterostructures etc.) atthe nanoscale within a single nanowire, which is a fundamentalstep in the design of nanostructured devices. In accordancewith previous works [3,4], this investigation shows how thecombination of charge collection microscopy techniques, suchas EBIC, and SEM imaging can be a useful tool for a deepunderstanding of nanoscale objects and their features.

References

[1] L. Largeau, E. Galopin, N. Gogneau, et al, Cryst. Growth Des.,12, 2724 (2012).

[2] A. Togonal, M. Foldyna, W. Chen, et al, J. Phys. Chem. C, 120,2962–297 (2016).

[3] Y.-J. Lu, M.-Y. Lu, Y.-C.Yang, et al, ACSNano, 7 9, 7640–7647(2013).

[4] M. Tchernycheva, V. Neplokh, H. Zhang, et al, Nanoscale, 7,11692-11701, (2015).


Formation of (Al,Ga)As axial heterostructuresin self-catalyzed nanowiresG. Priante, F. Glas, G. Patriarche, K. Pantzas, F. Oehler and J-C. HarmandLaboratoire de Photonique et de Nanostructures, CNRS, Universite Paris-Sac/ay Route de Nozay, 91460Marcoussis, France

Semiconductor nanowires are presently fabricated from a widevariety of materials and methods. The less stringent require-ments of lattice matching, as compared to thin films, open theway for new material combinations and allows for the mono-lithic integration III–V materials on silicon.

III–V nanowires are most commonly synthesized via thevapor-liquid-solid method, where a “catalyst” droplet promotestheir one-dimensional growth. A very interesting case is pro-vided by the selfcatalyzed GaAs nanowires [l], for several rea-sons: the foreign catalyst is replaced by liquid gallium; thedroplet can be changed in size, consumed or reformed at anytime [2]; a pure zinc-blende crystal phase is easily obtained [3].

To go further with this model system and to evaluate its po-tential, we investigate the formation by MBE of self-catalyzed,axial heterostructures, using Al as a second group III element.The quality and composition of the AlGaAs insertions is ana-lyzed with monolayer resolution using high-angle annular darkfield scanning transmission microscopy.

We show that the reservoir effect, which tends to producebroad composition gradients at the heterointerfaces, has lim-ited impact. The interface widths (−10 monolayers) are in-deed some ten times better than those obtained using a goldcatalyst [4]. By performing a growth interruption and prefill-ing of the droplet with Al atoms, we reduce the GaAs/AlGaAsinterface width to only two monolayers.

Finally, starting from the thermodynamic data available inthe literature, we develop numerical and analytical models forthe interfaces composition profiles, showing very good agree-ment with experiments. The models provide towards the for-mation of ideal interfaces: atomically sharp interfaces shouldbe attainable for small nanowire radii and/or low growth tem-peratures.

References

[1] F. Jabeen et al, Nanotechnology, 19, 275711 (2008).[2] G. Priante et al, Cryst. Growth Des., 13, 3976 (2013).[3] G.E. Cirlin, et al, Phys. Rev. B, 82, 035302 (2010).[4] L. Ouattara et al, Nano Lett., 7, 2859 (2007).[5] G. Priante et al, Nano Lett., 16, 1917 (2016).

52


SiC nanofilm on Si obtained by atoms substitutionas a substrate for III-Nitride optoelectronicsS.A. Kukushkin1, A.V. Osipov1, R.S. Telyatnik1

Institute of Problems in Mechanical Engineering RAS, St Petersburg, Bolshoy pr. VO 61, 199178, Russia

Optoelectronic devices based on III–V semiconductors havewell known advantages, such as direct bandgap for diodes emit-ting light without phonon energy loss, p- and n-type conduc-tivity for transistors, high mobility of charge carriers for fastswitchers. III–N semiconductors [1] have additional advan-tages, such as high melting temperature and hardness, radia-tion resistance and low chemical reactivity allowing usage inaggressive media, e.g. for ionizing radiation detection. Theyalso have high thermal conductivity providing fast cooling, andIII–N are not toxic compared to arsenides III–As, that allowsbiological applications. The main purpose of high bandgapGaN, AlN, BN are short-wave emitters (blue and ultra-violetLEDs and laser diodes) and detectors, while low bandgap InNprovides an ability to obtain any visible spectrum by ternarysolid solutions like Inx Ga1−xN with bandgap fitting nonlinearVegard’s law with bowing parameter b ∼ 3 eV for InGaN [2].Full spectrum provided with blue LEDs enabled white lightsources, this is a reason I.Akasaki, H.Amano and S. Nakamuragot Nobel Prize in physics in 2014 for developing high-qualityGaN films on Al2O3 with buffer AlN layer [3].

However, for epitaxy of AlN and GaN single crystals, di-electric sapphireAl2O3 is not a good substrate considering a bigincoherence of crystal lattices, but silicon carbide SiC has al-most ideal coherence withAlN lattice, meanwhile recent worksconcerns the cheapest but very incoherent Si substrates for III–N deposition [4]. The ability to grow III–N on Si or SiC is pro-vided by the similar tetrahedral structure between correspond-ingly hexagonal wurtzite-like and cubic diamond/sphalerite-like crystals (Fig. 1). Despite the similar geometries, there isstill a misfit between different scales of lattice parameters offilm and substrate crystals, which is the major problem of het-eroepitaxy, while the minor one is a mismatch between thermalexpansions of crystals taking effect upon cooling a heterostruc-

90°

[0001]

[11-2]

[111]

SiC

AlN

[0-1-10]

[2-1-10]

Fig. 1. One of coherent junctions between wutzite 2H-AlN(0001)and sphalerite 3C-SiC(111) surfaces.

Si

CO

Si

Cvacancies

SiO

Si

SiCPoresSi

Fig. 2. SiC nanofilm production by surface chemical reaction viaintermediate state.

ture from the epitaxy temperature. These differences causestrains, which give rise to stresses causing appearance of dis-locations, cracks, delaminations, and other defects at certaincritical thicknesses of films [5,6]. The new method of solidphase epitaxy [7,8] produces almost unstrained SiC nanofilms(∼ 100 nm thickness, <10 nm roughness) of either cubic orhexagonal polytypes on large Si substrates with 99% relax-ation of lattice misfit and 50% of thermal mismatch. This is aperfect and cheap industrial solution for buffer layer betweenSi templates and III–N or SiC thick films grown further bystandard deposition methods like CVD, HVPE, or MBE.

SiC nanofilms are not grown in usual sense, they are orig-inated from the material of Si substrate via atoms substitutionduring surface chemical reaction 2Si+CO→SiC+SiO (Fig. 2)at temperature range 1100–1300 ◦C and CO gas pressure 70–700 Pa. This reaction passes an intermediate state (“activatedcomplex”) saturated with Si vacancies and C interstitial atomsuntil they collapse into SiC structure driven by elastic interac-tion of the point defects [9]. This collapse forms in Si underthe SiC film contraction pores (∼ 1 μm size) fully releasingthe lattice parameters mismatch (5aSiC = 4aSi) and allowingelastic damping of thermal strains. To prevent the evaporationof surface silicon atoms out of the activated complex, an ad-ditional source of Si atoms like silane SiH4 (decomposes at900 ◦C) with partial pressure ∼15 Pa is usually used. (111)-oriented Si surface and vicinal 4◦ — declined surface are thepreferred ones for reaction. Kinetics and thermodynamics ofsuch phase transition through the intermediate state have beendeveloped [8,10], while the molecular dynamics of reactionand diffusion of point defects will be soon clarified by com-putational quantum chemistry using DFT GGA approach inAbinit software [11].

Our laboratory owns vacuum furnace Nabertherm VHT8/18(22)-GR 1800 allowing mass production of 50 SiC/Sitemplates of 6 inches diameter for one cycle of 6–8 hours.We have comprehensive research facilities including WollamM-2000 and ultraviolet (wavelengths between 140 and1700 nm)VUV-VASE ellipsometers, optic profilometer ZYGO

Fig. 3. Working LED on nano-SiC/Si template.

53

54

pGaN + InGaN QW

AlGaN

AlNSiC ~150 nm

pores in Si

nGaN

Fig. 4. SEM photography of the LED.

NewView-6000, confocal Raman microscope WITec Alpha300 S, Nanosurf easyScan 2 atomic force microscope, MicroMaterials Nanotest 600 for hardness determination by nanoin-dentation, Toho FLX-2320-S measurement system of plate cur-vature and film stresses, electron difractometer ER-100, etc.Our collegues in the Institute of Technology help us with chem-ical milling, while our collegues in Ioffe institute support uswith III–N deposition and scanning electron microscopy.

First experimental cyan LED structure on the nano-SiC/Sitemplate was grown in 2012 by MOCVD [12]. The heterostruc-ture had total thickness 2.5 µm, dislocation density 8×108cm−2,and no cracks; it is presented on Fig. 3 and 4.

References

[1] H. Morkoc . Handbook of Nitride Semiconductors and De-vices. Wiley-VCH, Weinheim (2008).

[2] I. Vurgaftman, J.R. Meyer, L.R. Ram-Mohan, J. Appl. Phys.89, 5815 (2001).

[3] H. Amano, et al, Appl. Phys. Lett. 48, 353 (1986).[4] S.A. Kukushkin, et al, Rev. Adv. Mater. Sci. 17 (2008).[5] L.B. Freund, S. Suresh. Thin Film Materials. Stress, Defect

Formation and Surface Evolution. Cambridge University Press(2003).

[6] R.S. Telyatnik, A.V. Osipov, S.A. Kukushkin, Phys. SolidState, 57, 162 (2015).

[7] S.A. Kukushkin, A.V. Osipov, Phys. Solid State 50, 1238(2008).

[8] S.A. Kukushkin, A.V. Osipov, J. Phys. D: Appl. Phys. 47,313001 (2014).

[9] S.A. Kukushkin,A.V. Osipov, R.S. Telyatnik, Phys. Solid State58, 971 (2016).

[10] S.A. Kukushkin,A.V. Osipov, Phys. Solid State 56, 792 (2014).[11] X. Gonze, et al, Comp. Phys. Comm. 180, 2582 (2009).[12] S.A. Kukushkin, et al, Tech. Phys. Lett. 38, 297 (2012).


Study of initial stages of ordered GaAs NW growthin views of optimizing the yieldsJ. Vukajlovic Plestina, W. Kim, F. Matteini, G. Tutuncuoglu, H.A. Potts and A. Fontcuberta i MorralLaboratory of semiconductor materials, Ecole Polytechnique Federale de Lausanne, Switzerland

Nanowires (NWs) are single-crystalline, filamentary nanos-tructures with diameters between 20 and 200 nm that resultfrom the rapid growth along one crystalline direction. Thefinite lateral size gives rise to many interesting physical prop-erties which are not seen in bulk materials, such as electronquantum confinement and optical resonances. Moreover, smalldiameters of NWs allow their growth on lattice-mismatchedsubstrates what enables the integration of high-performanceIII–V semiconductors monolithically into mature silicon tech-nology, overcoming fundamental issues such as lattice, polarityand thermal expansion mismatch.

The physical and chemical properties of NWs determinetheir optical properties. The length and diameter of NWs, aswell as their spacing can affect the emission and absorptionproperties [1]. The doping level, defect concentration, crystalstructure, growth direction, and nature of the facets are alsocritical to the emission and absorption. All these properties areclosely related to the growth of NWs. Therefore, to exploitthe potential of NWs in optoelectronics, the growth mecha-nism needs to be fully understood and simple and reproduciblegrowth processes need to be developed.

In this work we focus on self-assisted vapor-liquid-solid(VLS) growth of GaAs NW arrays on patterned silicon sub-strates. The need of position controlled growth arises from thefact that if NWs are randomly positioned, each NW has a differ-ent surrounding during growth, resulting in a variation in theirmorphological as well as structural and functional properties.All that can strongly affect device performances, so our goalis to achieve uniform NW arrays that could enable fabricationof high performance devices. Up to date fabrication of GaAsNWs on a patterned Si surface has shown to very challengingand hard to reproduce. It is known that elements such as gal-lium pre-deposition, thickness and composition of the growthmask play important role for a successful growth [2-5]. In thatspirit we are trying to look for further answers by directing ourinvestigation to the initial stages of the growth, more preciseto the Ga pre-deposition step. Ga – droplet shape and posi-tion within the nanoscale opening can be directly correlatedwith NW morphology [6]. The droplet wetting angle needs tobe close to 90 ◦ in order to promote vertical NW growth [7].Changing the hole size and modifying the surfaces within theopenings we are aiming to optimize patterned silicon substratefabrication in order to obtain high yield GaAs NW arrays. InFigure 1 two types of patterned substrate used in our workare presented – standard planar (a) and oxide nanotube basedsubstrate (b). By changing the holes and tubes aspect ratio(diameter/depth) the droplet morphology inside of nucleationsite can be tuned to the desired one, for promoting verticalgrowth [6], as shown in Figure 1(c). Finally in Figure 1(d) wepresent single GaAs NW growing vertically from the nanotubetemplate.

(c) Desirable configurationof Ga droplet with the opening

SiO2 Ga

Si

(a)

SiO2

Silicon

Planar substrate

SiO2

Si

(b) Nanotubessubstrate

200 nm

(d)

Fig. 1. (a) Illustration of the standard planar patterned substrate.(b) Illustration of oxide nanotube substrate, (c) Ga droplets with equi-librium contact angle around 90◦ can promote vertical NW growth.(d) Vertical GaAs NW growing from the oxide nanotube.

References

[1] M. Heiss, et al, Nanotechnology, 25, 014015 (2014).[2] S. Plissard, et al, Nanotechnology, 21, 385602-1-8 (2010).[3] S. Plissard, et al, Nanotechnology, 22, 275602-1-7 (2011).[4] S. Gibbson et al, Semiconductor science and technology, 28,

105025-1-9 (2013).[5] E. Russo et al, Nano Lett., 15, 2869-2874 (2015).[6] F. Matteini et al, to be submitted to Cryst. growth and des.[7] F. Matteini et al, Cryst. Growth Des.. 15 7, 3105-3109 (2015).

55


Tuning the growth of GaAs/InAsheterostructured nanowires by catalyst compositionV. Zannier1, D. Ercolani1, U.P. Gomes1, J. David2, M. Gemmi2 and L. Sorba1

1 Istituto Nanoscienze-CNR and Laboratorio NEST Scuola Normale Superiore, Plazza S. Silvestro 12, 56127Pisa, Italy2 Istituto Italiano di tecnologia IIT, Plazza S. Silvestro 12, 56127 Pisa, Italy

Semiconductor nanowires (NWs) are widely considered one ofthe most promising technologies for future device applications.One of the greatest advantages of NWs is the ability to relaxelastic stress in two dimensions, which enables the growth ofdefect-free heterostructures in lattice-mismached systems andallows to obtain structures not achievable in conventional 2Dand 3D growth. Indeed, a number of devices based on ax-ial heterostructured nanowires, including single-electron tran-sistors [1], optically active quantum dots [2] and field-effecttransistors [3] have already been demonstrated.

Although the epitaxial growth of axial nanowire hetero-structures has been reported since 1996 [4], the growth is notstraightforward and far from fully understood. The main is-sue when growing axial heterostructures is to have straightnanowires with sharp interfaces between the two materials.Concerning Au-assisted III-V semiconductor NWs, straightwires with atomically sharp heterointerfaces are easier to ob-tain when group V interchange occurs, due to the lower sol-ubility of the group V species than that of the group III onesinto the Au seed nanoparticles [5,6]. In the specific case ofAu-assisted InAs/GaAs heterostructured nanowires, straightnanowire growth has been reported for the growth of GaAs ontop of InAs, whereas in the other direction kinked nanowiresare generally obtained [5,7]. Moreover, it is observed that theGa-to-In switch is much sharper than the In-to-Ga one. Allthese results have been explained using a simple model whichtakes into account the energy balance resulting from the inter-face energies between the Au particle and each material andthe interface energy between the two materials [7]. However,the model refers to the system at equilibrium and it treats theseed particle as a gold cluster, rather than an alloy particle.Nevertheless, it is known that both In and Ga have a high (butdifferent) affinity for the gold particle. Therefore, it is likelythat the chemical composition of the alloy nanoparticle dur-ing the NW growth, and the changes which it undergoes whenswitching from one material to the other one play an importantrole in the growth mode, from both points of view of thermo-dynamics and kinetics.

In this contribution we report on the growth of bothInAs/GaAs and GaAs/InAs heterostructured NWs, as well asthe growth of double heterostructured NWs (InAs/GaAs/InAsand GaAs/InAs/GaAs) by means of Chemical Beam Epitaxy.A careful investigation of the NWs morphology as a functionof growth parameters like growth temperature, III/V flux ra-tio and chemical composition of the catalyst nanoparticle hasbeen performed by means of SEM, TEM and EDX. The re-sults suggest that the chemical composition of the nanoparti-cle, rather than other growth parameters, strongly affects thegrowth. We found that GaAs on InAs easily grows straight,

(a) (b)

100 nm 100 nm

200 nm200 nm

(c)

InAs

GaAs

InAs

(d)

GaAs

InAs

GaAs

Fig. 1. (a,b) 45◦-tilted SEM images of InAs/GaAs (a) andGaAs/InAs (b) heterostructured NWs grown on InAs(111)B andGaAs(111)B substrates respectively. The insets at the bottom areSTEM images of representative NWs reported in each panel. (c,d)False-colour SEM images of the double heterostructured NWs:InAs/GaAs/InAs (c) and GaAs/InAs/GaAs (d). The colours high-light the various segments: InAs (green), GaAs (blue) and catalystnanoparticle (yellow).

while InAs on GaAs grows straight only if the chemical com-position of the nanoparticle is properly tuned. In particular,we found a strong correlation between the amount of group-III into the NPs before InAs growth and its following growthmode, straight or downwards. As a consequence, we obtainedstraight heterostructured NWs in both sequences by controllingthe chemical composition of the seed particles. We discuss ourresults in terms of group-III/Au ratio and nanoparticle stabilityupon the change of the growing material, and this scenario canbe extended also to other III-V heterostructured systems witha group-III interchange.

56

57

References

[1] H.A. Nilson et al, Nano Letters, 8 3, 872-875 (2008).[2] M.T. Borgstrom et al, Nano Letters, 5 7, 1439-1443 (2005).[3] E. Lind et al, Nano Letters, 6 9, 1842-1846 (2006).[4] K. Hiruma et al, Journal of Crystal Growth, 163, 226-231

(1996).[5] K. Dick et al, Nano Letters, 7, 1817-1822 (2007).[6] K. Dick et al, Nano Letters, 12, 3200-3206 (2012).[7] M. Paladugu et al, Small, 3, 1873-1877 (2007).


Nucleation and growth mechanismof self-catalyzed InAs nanowires on siliconD. Ercolani1, U.P. Gomes1, V. Zannier1, J. David2, M. Gemmi2 and L. Sorba1

1 NEST, Scuola Normale Superiore and Istituto Nanoscienze-CNR, Piazza S. Silvestro 12, I–56127 Pisa, Italy2 Center for Nanotechnology Innovation @ NEST, Istituto Italiano di Tecnologia, Piazza S. Silvestro 12, I–56127Pisa, Italy

Au-free growth of III–V semiconductor nanowires (NWs) onSi is a fundamental step for the integration of the III–V NWswith the present Si technology, to eliminate the problems stem-ming from Au-induced defects and traps in the Si. Two growthtechniques have been used to obtain Au-free NWs: vapor-solid (VS) growth mechanism for catalyst-free (CF) and vapor-liquid-solid (VLS) for self-catalyzed (SC) NWs. With bothtechniques, either CF or SC, growth can be achieved by tun-ing the growth conditions (growth temperature and fluxes) andsubstrates (III–V or Si with or without oxide mask) [1]. Sincethe growth mechanisms determine the morphology and crys-tal structure of the grown NWs, an in-depth understanding ofthe mechanisms is necessary to obtain the degree of controlenvisaged for device development.

We show that InAs NWs can be grown by chemical beamepitaxy (CBE) via VS CF or VLS SC growth by tuning themetal-organic (MO) line pressures (PTBAs and PTMIn) andgrowth temperature (Tgrowth). In CBE, the In and As fluxesare proportional to the MO line pressures. In particular CFNW growth was achieved for F(PTBAs/PTMIn) ≥ 1 while SCNW growth occurs at F ratio< 1. Within the parameter spaceof F < 1, it was possible to grow SC InAs NWs with an Indroplet throughout NW growth (see Fig. 1). At the optimizedgrowth parameter window (F = 0.75 and Tgrowth = 370 ◦C),the axial growth rate of SC InAs NWs was found to be radiusindependent and proportional only to the arsenic flux (φAs).On the other hand, the radial growth rate was determined bythe kinetics of both In (φIn) and As (φAs) fluxes. We also foundthat not all NWs nucleate at the beginning of the growth whenthe growth precursors were switched on; rather each individualNW began growth after a distinct nucleation time after the onsetof the precursor fluxes. This was evident from the NW lengthand radius (LNW − RNW) distribution of the SC InAs NWsgrown for various growth times (tgrowth) as shown in Fig. 2.

The spread and overlap of experimental points with growthtime indicates that SC InAs NWs do not nucleate simultane-ously. NWs that nucleated in the first seconds of growth con-tinue to increase in length and radius finally attaining maximumvaluesLmax andRmax (marked by colored pins in Fig. 2(a) and(b)) at the end of the growth, while NWs with shorter lengthand smaller radius nucleated at later nucleation times. Thelength of an individual NW is therefore a measurement of itsnucleation time. The observed experimental results of axialand radial growth rate of SC InAs NWs are in good agreementwith the theoretical growth model of SC III–V NWs [2], i.e.LNW ∝ φAs and RNW ∝ φIn − φAs. In accordance with themodel, the the NW length and diameter dependence on growthtime can be combined to obtain the functional relation between

(a)

●●

●

●

● ●

●

●

●

●

●

■

■ ■ ■

■

■ ■

■■

■■

■

(Tor

r)P

TM

In

(Tor

r)P

TM

In

(Torr)PTBAs

(Torr)PTBAs

F = 1

0.000.00

0.20

0.40

0.60

0.80

0.30 0.60 0.90 1.20 1.50 1.80 2.10

0.25

0.25

0.30

0.30

0.35

0.40

0.35

SC growth

CF growth

(b)F = 0.75PTMIn = 0.40

F = 0.86PTMIn = 0.35

F = 0.81PTMIn = 0.37

F = 1.00PTMIn = 0.30(d)

(c)

(e)

Fig. 1. (a) Map showing the line pressures PTBAs and PTMIn usedfor InAs NW growth. Each symbol corresponds to a grown sample:squares are CF InAs and circles correspond to SC InAs NWs. Theblack solid line corresponds to F = 1 line. The inset is a magnifiedview of the shaded region of the main panel. (b–e) Selected 45◦

tilted SEM micrographs of NWs (with a top-view inset of a singleNW) grown with different F (PTBAs/PTMIn) by changing PTMIn andfixing PTBAs at 0.30 Torr as shown in the inset of panel (a).

LNW and RNW as follows:

LNW = C/A · RNW + C/A · B/A · ln

(B/A

B/A+ RNW

)(1)

where B is a positive diffusion-induced term, A is the fluximbalance term and C is the axial growth rate. Eq. 1 is usedto fit the LNW −RNW distribution and the results of the fit areused to obtain important parameters like the effective V/III fluxratio, φAs/φIn = 0.250 ±0.095, and the (negligible) adatomdiffusion length, λ = 1.7 ±1.7 nm.

58

59

(nm)LNW

(nm)RIn(nm

)R

NW

(nm

)R

NW

0

0 20

20

40

40

60

60

80

100 200 300 400 5000

0

15

30

45

60

15

5

′30′4 ′70′

(a)

5′

15′

30′

4 ′5

70′

(b)

100 nm

Fig. 2. (a) RNW − LNW distributions for SC InAs NWs fortgrowth = 15, 30, 45, and 70 minutes. Colored points mark ex-perimental data points that correspond to measurements done on asingle NW. The different colors of the points distinguish sampleswith different growth times. The solid line is a theoretical fit of theexperimental data. The pins markLmax andRmax from samples withdifferent growth times. The inset is a plot of RNW as a function ofRIn; (b) 45◦ tilted SEM micrographs of longest NWs for differentsample-growth times (marked by the same colored pins as in panela).

References

[1] B. Mandl, et al, Nano Lett. 10, 4443 (2010).[2] F. Glas, et al, Phys. Rev. B 88, 195304 (2013).


Optical characterization of MBE-grown GaAs nanomembranesand related heterostructuresG. Tutuncuoglu1, M. Friedl1, M. de la Mata2, D. Deiana3, H. Potts1, F. Matteini1, J.-B. Leran1, J. Arbiol2,4,A. Fontcuberta i Morral11 Laboratoire des Materiaux Semiconducteurs, Ecole Polytechnique Federale de Lausanne, 1015 Lausanne,Switzerland2 Institut Catala de Nanociencia i Nanotecnologia (ICN2), CSIC and The Barcelona Institute of Science andTechnology (BIST), Campus UAB, Bellaterra, 08193 Barcelona, Catalonia, Spain3 Centre Interdisciplinaire de Microscopie Electronique, Ecole Polytechnique Federale de Lausanne, EPFL, 1015Lausanne, Switzerland4 Institucio Catalana de Recerca i Estudis Avanc ats (ICREA), 08010 Barcelona, Catalonia, Spain

The precise control of crystal quality is a long sought-aftergoal for the growth of semiconductor nanomaterials. In thiswork, defect-free GaAs nanomembranes were grown within amolecular beam epitaxy (MBE) reactor and characterized byoptical techniques.

III–V semiconductor materials are attractive for many elec-tronic and optoelectronic applications in part due to their directbandgap, making them good candidates for use as optical de-tectors. Additionally, their high carrier mobilities enable thepossibility to build high-speed electronics with them if they canbe integrated cheaply into current silicon CMOS processes.One possible way to cut production costs and achieve thisintegration is to sidestep the traditional top-down fabricationtechniques and instead use so-called bottom-up approaches ofcrystal growth to fabricate III–V nanostructures.

With these motivations in mind, bottom-up GaAs nano-membranes were grown on a GaAs (111)B substrates whichwere covered with a 30 nm layer of SiO2, patterned by elec-tron beam lithography and then etched with reactive ion etch-ing (RIE) to selectively expose the substrate below. The patternconsisted of trenches varying from 50 to 100 nm in width andlengths between 5 and 10 μm. The alignment of the patternwas critical since, due to the zincblende crystal structure ofthe GaAs substrate, defect-free membrane growth only occursin slits along (11–2), or equivalent, directions on the substrate.The result was epitaxial and vertical growth of defect-free GaAsfrom these trenches, shown in Fig. 1. As determined by trans-mission electron microscopy (TEM) analysis, the facets of theGaAs are terminated mostly by {110} family planes, exceptfor the rear of the membranes which is terminated by a singlehigh-index {221} family plane [1]. A diagram of the mem-brane faceting can be seen in Fig. 2. This is quite interestingas previous studies of metalorganic chemical vapor deposition(MOCVD) grown nanomembranes do not exhibit these high-index {221} facets [2]. Equally interesting, these TEM studiesalso revealed that the crystal structure of these membranes ismostly twin-free.

Low temperature cathodoluminescence (CL) spectroscopystudies were then performed to provide another assessment ofthe crystal quality and to measure the optical emission fromvarious points in the membranes. Shown in Fig. 3, the spec-trum of the GaAs free exciton was measured in the bulk of thenanomembranes while higher energy emission was seen at bothends of the membranes, though it was also observed at local-

500 nm

Fig. 1. SEM image of grown nanomembranes.

(1-10)

(0-11)

(-2-2-1) ( 01)-1

Fig. 2. Diagram and labelling of membrane facets as determined byTEM.

ized points along the whole length of the membranes in othersamples. This localized, high-energy emission is thought to besimilar to the emission observed in core-shell GaAs/AlGaAsnanowires where it has been reported that Al aggregation atvertices can cause quantum dot-like emission to be observedin these structures [3].

GaAs/AlGaAs quantum well (QW) heterostructures werethen also fabricated and analyzed using multiple methods.First, the crystallinity and interface of the GaAs/AlGaAs quan-tum well was analyzed by cross-sectional atomic-resolutionscanning-TEM (STEM), an image of which is shown in Fig. 4.Quantum well emission along with localized high-energy emis-sion has also been observed in these nanostructures with CLspectroscopy.

In conclusion, III–V bottom-up defect-free GaAs nano-membranes were grown in MBE using selective area growthtechniques. This kind of bottom-up approach has the advanta-ge of being able to produce self-assembled nanostructures notlimited by the resolution of traditional lithography techniques.In the future, we would also like to look into the possibilityof coupling of light into these nanostructures and use them aswaveguides or other elements in photonic circuits. Further-more, motivated by electronic and optoelectronic applications,

60

61

(a)

600 8000

500

1000

1500

2000

2500

3000

3500

4000

4500

830 nm724 nm

815 nm

BulkTip

Wavelength (nm)

Inte

nsity

(a.

u.)

1000

(b)

Fig. 3. (a) CL map showing sharp, high-energy emission from thetips of the nanomembrane and (b) the spectrum from two indicatedpixels in the CL map, one at the tip of the membranes (green), theother in the bulk of the membrane (blue).

5 nm

AlAggregation

AlGaAsBarrierGaAs QWAlGaAsBarrier

Fig. 4. High resolution cross-sectional HAADF-STEM image of aGaAs/AlGaAs quantum well showing Al aggregation at edges of thefacets.

combining other III–V materials and creating more complexheterostructures is being explored along with the fabrication ofintersecting membrane networks.

References

[1] G. Tutuncuoglu, M. de la Mata, D. Deiana, et al, Nanoscale, 7,19453–19460 (2015).

[2] C.-Y. Chi, C.-C. Chang, S. Hu, et al, Nano Lett., 13 6, 2506–2515, Jun. (2013).

[3] M. Heiss, Y. Fontana, A. Gustafsson, et al, Nat. Mater., 12 5,439–44 (2013).


Ultrafast optical response in Au and Ag nanoparticlesformed on silica nanowires arraysL. Di Mario1, L. Tian1, D. Catone2, P. O’Keeffe3, S. Turchini2, F. Martelli11 IMM-CNR Via del Fosso del Cavaliere 100, 00133 Roma, Italy2 ISM-CNR Via del Fosso del Cavaliere 100, 00133 Roma, Italy3 ISM-CNR Area della Ricerca Roma 1, 00016 Monterotondo Scalo, Italy

Metal nanostructures are characterized by unique optical prop-erties which derive from the localized surface plasmon reso-nance (LSPR), a collective oscillation of the conduction elec-trons. In particular, these nanostructures exhibit a sharp spec-tral absorption for incident photon frequencies resonant withthe LSPR. The plasmon resonance has high intensity and itis sensitive to the environment of the nanostructures and tothe coupling between them. This makes metal nanostructuresof great interest for molecular sensing and several biomedicalapplications [1]. Understanding the dynamics that occur fol-lowing absorption of photons in metal nanoparticles (NPs) ishence fundamental for many applications. The ways the differ-ent dynamical process depend on size, shape and compositionof the particles are reasonably well-known. However, the in-teraction of the particles with their environment still requiresfurther study to be fully understood [2].

In this work we investigated the ultrafast dynamics in Auand Ag NPs formed on silica nanowires (NWs) arrays (Fig. 1)using transient absorption spectroscopy. The silica NWs ar-rays are transparent in the visible to the near-UV region ofthe spectrum and offer a large surface area for attaching NPs,providing at the same time a macroporous support frameworkfor an efficient interaction between the particles and the envi-ronment. All these features make the metal decorated silicaNWs an optimal system for the study of the interaction withthe environment and coupling between NPs.

The silica NWs arrays have been fabricated via thermal ox-idation of Si NWs grown using the VLS method on a quartzsubstrate. They were subsequently decorated with Au (Fig. 1a)and Ag (Fig. 1b) NPs by dewetting thin metallic films evapo-rated on the NWs [3].

Transient absorption spectroscopy has been performed us-ing a pump-probe configuration in a femtosecond transient ab-sorption spectrometer (FemtoFrame II, IB Photonics). Theamplified second harmonic of a Ti:Sapphire laser at 400 nmwith a pulse length of about 50 fs and a repetition rate of 1 kHzhas been used as source of excitation for the measurement ontheAu-decorated silica NWs. For theAg-decorated silica NWsto avoid overlap between the pump and the plasmon resonance,we used a pump at 275 nm from an optical parameter ampli-

(a)

100 nm 100 nm

(b)

Fig. 1. SEM images of silica NWs decorated with: (a)Au and (b)Ag.

0.5 ps1 ps2 ps7 ps10 ps

psps

50480

0.5 ps1 ps2 ps7 ps10 ps

psps

50480

300 400

500

500

600

600

700

Wavelength (nm)

ΔAΔA

–0.05

–0.04

–0.03

–0.02

–0.02

–0.01

–0.01

0.00

0.00

0.01

0.01

0.02

0.03

0.04(b)

(a)

Fig. 2. Transient absorbance spectra at different time from the pumpexcitation for metal NPs on silica. NWs: (a) Au, (b) Ag NPs.

fier (OPA), with similar pulse length and the same repetitionrate. A white light supercontinuum generated in the FTAS hasbeen used as probe for both metals. Probe wavelength rangedbetween 320 and 750 nm.

In Fig. 2 the variation of the absorbance as a function of thewavelength at different pump-probe delay times is reportedfor both Au (Fig. 2a) and Ag-decorated silica NWs (Fig. 2b).The spectra show the expected decrease of the absorbance atthe plasmon resonance immediately after the pump excitationwhen the plasmon resonance is deformed/bleached due to anincrease in the electronic temperature of the nanoparticles. Forthe Ag NPs both the dipole and quadrupole contribute to theplasmon resonance are visible.

Figure 3a shows the dynamic of the transient absorbance forthe Agdecorated silica NWs at the wavelength of the minimumdue to the quadrupole (black curve) and the dipole (red curve).Figure 3b and 3c show the energy position of the two minimumas a function of the time delay between probe and pump. Asimilar shift has been observed also for the Au NPs plasmonresonance. A possible explanation for the shifts could be the

62

63

0

408411414417420423

–352

–354

–356

–360

1 2 3 4 5Time (ps)

Min

imum

pos

ition

(nm

)

6 7 8 9 10 11

n■■■

■

■■

■

■

■■

■

■ ■

■■ ■

■

■■

■

■■

■■■■■

■■ ■

■■

■■

■

■■

■

■

■■

■ ■■

0 1 2 3 4

■ Dipole

■ Quadrupole

5 6 7 8 9 10

0 1 2 3 4 5 6 7 8 9 10

11

–358

(c)

(b)

(a)

ΔA

359.5 nm422.3 nm

Fig. 3. (a) Transient absorbance of Ag-decorated silica NWs probedat 359.5 (black) and 412.3 nm (red). (b) and (c) energy position ofthe quadrupole and dipole minimum, respectively, as a function ofthe delay time.

non-spherical shape of the NPs formed on the solid surface ofthe NWs [4] or the formation of extended plasmons [2].

Acknowledgements

This work has received funding from the European Union’s7th Framework Programme for research, technological devel-opment and demonstration under grant agreement No. 316751(NanoEmbrace). We thank Silvia Rubini (IOM-CNR) for theSEM images.

References

[1] K.A. Willets et al, Annu Rev. Phys. Chem. 58, 267 (207)[2] G.V. Hartland Chem Rev. 111, 3858 (2011)[3] A. Convertino et al, J. Phys. Chem. C 118, 685 (2014)[4] Y. Guillet, et al, Phys. Rev. B 79, 195432 (2009)


Dynamic magnetoelectric coupling for THz devicesS. Skiadopoulou1, F. Borodavka1, C. Kadlec1, F. Kadlec1, X. Bai2, B. Dkhil2, M. Retuerto3, Z. Deng3,M. Greenblatt3 and S. Kamba1

1 Institute of Physics, Czech Academy of Sciences, Prague, Czech Republic2 Laboratoire SPMS, UMR 8580, Centrale Supelec CNRS, Universit’e ParisSaclay, Grande Voie des Vignes,Chatenay-Malabry, France3 Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, Piscataway, USA

Among the novel materials studied, an intense effort is concen-trated on the research in multiferroics materials that present atleast two of the ferroic properties: ferroelectricity, ferromag-netism and ferroelasticity [1,2]. A wide range of applicationssuch as information storage, sensing actuation and spintronicsawait pioneering materials and/or strategies that would pro-duce robust magnetoelectric coupling. The magnetoelectricmultiferroics’ ability of magnetization manipulation via elec-tric fields can be extremely promising for such applications,due to the simplicity and costefficiency of the use of the elec-tric field [3].

In addition to the static magnetoelectric coupling, a dy-namic magnetoelectric coupling can occur at the presence ofelementary magnetoelectric excitations. Such excitations arealso known as electromagons: electro-active magnons whichcan be tuned by electric or magnetic fields [4]. The possibilityof modulation of the index of refraction could promote the de-sign of novel optoelectronic devices. Due to the fact that theelectromagnons very often lie at the THz range of the electro-magnetic spectrum THz spectroscopy is an essential tool forthe detection of such excitations. However, a combination ofspectroscopic techniques is required to be able to account forthe nature of the detected excitations since those can be puremagnons (i.e. contribute only to the magnetic permeability μ)or electromagnons (i.e. influence at least partially the permittiv-ity ε). Thus Raman spectroscopy obeying to different selection

THz, Skiadopoulou .et al 6

IR, Talbayev l. Nagel et al.IR active modes only in H, Fishman

Raman, Rovillain

et a

et al.et al.

9 7

8

10

11Raman, Cazayous

0 50 100 150 200 250 300 350 400 450Temperature (K)

15

20

25

30

35

40

45

50

55

Spi

n-w

ave

freq

uenc

ies

(cm

)–1

Fig. 1. Temperature dependence of the THz spin excitation frequen-cies of BiFeO3 [1] compared with frequencies obtained from far IRspectra [7–9] and Raman scattering [10,11].

rules can be employed for the assignment of such modes. Thesimultaneous detection of spin excitations by both THz and Ra-man spectroscopies manifests the presence of electromagnons.

In the current work two examples of materials with dynam-ical magnetoelectric coupling detected by THz spectroscopyare presented: bismuth ferrite BiFeO3 and nickel tellurideNi3TeO6.

BiFeO3 as one of the few singlephase RT magnetoelectricmultiferroics is the center of attention, as it presents a ferro-electric phase transition at approximately 1100 K and an anti-ferromagnetic one at 643 K [5]. The knowledge of lattice andspin excitations in BiFeO3 is essential for the understanding ofthe underlying mechanisms that induce its multiferroic behav-ior. A series of Raman and Infrared (IR) spectroscopy studieshave presented controversial results concerning the assignmentof the magnon and phonon modes as well as of the highly ac-claimed electromagnons. Our THz spectroscopy studies re-vealed five low frequency spin modes for a temperature rangefrom 10 K up to RT the highest two appearing at 53 and 56 cm1

(Fig. 1 [6]. This corresponds to the frequency range where suchexcitations were theoretically predicted [7,8] but not experi-mentally confirmed up to now. In Fig. 1, one can also see thesimultaneously Raman and IR active modes in BiFeO3, sug-gesting the presence of electromagnons [9–11]. In addition, at

10 20 30 40 50 60 70 800.00

0.01

0.02

60 K

50 K

40 K

30 K

20 K

10 K

Ram

an In

tens

ity (

a.u.

)

4 K

(b)7 T @ 10 K0 T @ 10 K0 T @ 25 K0 T @ 48 K

Ext

inct

ion

coef

ficie

nt

Wavenumber (cm )–1

(a)

Fig. 2. (a) Raman and (b) THz spectra of Ni3TeO6 revealing twosimultaneously detected spin excitations.

64

65

5 K, the low-energy spin dynamics in the THz range were alsostudied in a varying magnetic field of up to 7 T. Softening ofthe (electro)magnon frequencies upon increasing the magneticfield was observed.

Ni3TeO6 presents a collinear antiferromagnetic order be-low 52 K giving rise to spin-induced-ferroelectricity. Amongthe spin-order driven multiferroics only Ni3TeO6 exhibits non-hysteretic colossal magnetoelectric effect near 8.5 and 52 Twhere spinflop and metamagnetic phase transitions occur, re-spectively [11,12]. The lack of hysteretic behavior in the mag-netic field dependence of magnetization and dielectric constantprecludes losses for a series of magnetoelectric applications.In the current work, we investigated the spin and lattice ex-citations of Ni3TeO6 ceramics and single crystals. Infrared,time-domain THz and Raman spectroscopy experiments wereconducted for a temperature range of 5 to 300 K. Time-domainTHz spectroscopy at external magnetic field was carried out atselected temperatures below and close to the antiferromagneticphase transition. The THz spectra revealed dynamic magneto-electric coupling i.e. tuning of THz spectra with magnetic field.Simultaneous Raman and IR active spin excitations correspondto electromagnons, highly sensitive on magnetic field (Fig 2(a)and (b)).

References

[1] M. Fiebig, J. Phys. D: Appl. Phys. 38, R123 (205)[2] Y. Tokura et al, Rep Prog Phys. 77, 76501 (2014)[3] W. Eerenstein et al, Nature (London) 442, 759 (206)[4] A. Pimenov et al, Nat Phys. 2, 97 (2006)[5] G. Catalan and J.F. Scott, Adv. Mater 21, 2463 (209)[6] S. Skiadopoulou et al, Phys. Rev. B 91, 174108 (2015)[7] U. Nagel et al, Phys. Rev. Lett 110, 25721 (2013)[8] R.S. Fishman Phys. Rev. B 87, 224419 (2013)[9] D. Talbayev, et al, Phys. Rev. B 83, 094403 (2011)

[10] M. Cazayous et al, Phys. Rev. Lett 101, 037601 (208)[11] P. Rovillain et al, Phys. Rev. B 79, 180411 (209)[12] Y.S. Oh, et al, Nature Communications 5, 321 (2014)[13] JW Kim, et al, Phys. Rev. Lett 115, 137201 (2015)

International Summer School and Workshop • Nanostructures for Photonics p-Fri1St Petersburg, Russia, June 27–July 2, 2016

Silver nanoisland films: self-assembly and patterningS. Chervinskii1,2, I. Reduto1,2,3, A. Kamenskii1,3, A.A. Lipovskii1,31 University of Eastern Finland, P. O. Box 111, Joensuu, 80101 Finland.2 Peter the Great St Petersburg Polytechnic University, 29 Polytechnicheskaya, St Petersburg, 195251 Russia3 St Petersburg Academic University, St Petersburg, Russia

Abstract. We developed a technique allowing self-organization of silver nanoparticles following the shape of appliedtemplate [1]. The technique is based on silver out-diffusion from ion-exchanged glass in the course of annealing in reducingatmosphere. To obtain patterned silver nanostructures we modified silver ions distribution in the exchanged soda-lime glassby thermal poling of this glass with a profiled electrode. The technique includes three steps: (i) during the ion exchange inthe AgxNa1−xNO3 (x = 0.01–0.15) melt a sodium-containing glass is enriched with silver ions in the subsurface layer; (ii) inthe thermal poling, the electric field under the 2D profiled anodic electrode moves these ions deeper into the glass, this shiftis smaller under the hollows in the electrode where the intensity of the field is minimal; (iii) annealing the glass in a reducingatmosphere of hydrogen results in silver out-diffusion only in the regions corresponding to the electrode hollows, as a resultsilver forms nanoislands following the shape of the electrode. Varying the electrode and mode of processing allowsgoverning the nanoisland size distribution and self-arrangement of the isolated single nanoislands, pairs, triples or groups ofseveral nanoislands—so-called plasmonic molecules [2].

References

[1] S. Chervinskii, et al, J. Appl. Phys., 114, 224301 (2013).[2] S. Chervinskii, et al, Faraday Disc., 186, 107–121 (2016).

66


Local gate tuning of quantum ringsT.P. Collier1,2, V.A. Saroka1,3 and M.E. Portnoi11 Department of Physics and Astronomy, University of Exeter, Stocker Road, Exeter EX4 4QL2 EPSRC Centre for Doctoral Training in Metamaterials (XM2), Department of Physics and Astronomy, Universityof Exeter, Stocker Road, Exeter EX4 4QL3 Institute for Nuclear Problems, Belarussian State University, Bobruiskaya 11, 220030 Minsk, Belarus

One of the most formidable, and therefore interesting, prob-lems of modern applied physics lies in bridging the aptly namedterahertz gap. The literature provides a variety of proposals forpractical THz emitters and detectors. The most heavily re-searched aspirants remain based on semiconductor nanostruc-tures hosting multiple quantum well [1], such examples includesuperlattice quantum cascade lasers [2] and optically induceddouble quantum wells [1]. However, promising candidates ex-ist in the form of non-simply connected nanostructures, suchas the use of carbon nanotubes [3] or quantum rings [4]. Theappeal in using these latter structures, in lieu of those former,lies in their practical tunability.

Quantum rings subject to external electric and magneticfields leads to an array of interesting phenomena [4-6]. Herewe theoretically investigate the system of an infinitely-narrow,single-electron quantum ring under the influence of two gatevoltages. We discuss the implications of this system as aprospective, tunable, THz emitter or detector; see Fig. 1 fora diagram of the systems energy levels.

Our single-electron quantum ring communicates with thetwo gate voltages only. The resulting double quantum wellsystem demonstrates voltage-tunable energy splitting betweenthe ground and first excited state, which permits analysis viathe WKB approximation or an exact treatment. We introducetwo system variations; one of an idealized set-up with exactlyequivalent voltages on each gate, and one of a more realisticsituation with differing gate voltages.

We discuss implications of the polarisability of such a quan-tum ring. Finally, we consider the intriguing case of re-intro-ducing magnetic flux through the ring which, at half integervalues in units of flux quanta, returns the ground-state to com-plete double degeneracy.

(a) V En( ),ϕ

−π 0 ππ− 2—π2—

01

2

3

4Vmax

ϕ

(b) V En( ),ϕ

Vmax

−π 0 ππ− 2—π2—

ϕ

01

2

3

Fig. 1. Energy levels and potentials for the two gate system; (a) withsymmetrically and (b) asymmetrically charged gate voltages.

References

[1] H.G. Roskos, M.C. Nuss, J. Shah, et al, Phys. Rev. Lett, 68,2216 (1992).

[2] J. Faist, F. Capasso, D.D.L. Sivco, et al, Science, 264, 553(1994).

[3] M.E. Portnoi, M. Rosenau da Costa, et al, Int. J. Mod. Phys. B,23, 2846 (2009).

[4] A.M. Alexeev and M.E. Portnoi, Phys. Rev. B, 85, 245419(2012).

[5] A.M. Fischer, V.L. Campo, M.E. Portnoi et al, Phys. Rev. Lett,102, 2 (2009).

[6] K.L. Koshelev, V.Y. Kachorovskii and M. Titov, Phys. Rev. B,92, 235426 (2015).

67


Angular and positional dependence of Purcell effectfor layered metal-dielectric structuresA.R. Gubaydullin1,2, V.A. Mazlin2, K.A. Ivanov2, M.A. Kaliteevski1,2, C. Balocco3

1 St Petersburg Academic University, St Petersburg, Russia2 Saint Petersburg National Research University of Information Technologies, Mechanics and Optics, ITMOUniversity, Kronverkskiy pr. 49, 197101, St. Petersburg, Russia3 School of Engineering and Computing Sciences, Durham University, Durham DH1 3LE, UK

We study the enhancement of spontaneous emission of thedipole embedded in the layered metal-dielectric structure.Such systems have attracted great scientific interest over thepast decade due to their special optical properties emanat-ing primarily from hyperbolic dispersion of isofrequency sur-face, and therefore are generally called hyperbolic metamate-rial (HMM) systems.

Hyperbolic metamaterials have attracted great scientific at-tention as an application for engineering the spontaneous emis-sion. The concept of the radiative decay enhancement in a cav-ity first proposed by Purcell in 1946 [1], this effect is describedby the Purcell factor, which in the broad sense is defined bythe ratio between lifetimes of spontaneous emission of a pointlight source inserted into a resonant cavity and in the infinitehomogeneous medium. According to the Fermi golden ruleand the definition of the density of photon states, the radiativedecay rate in a homogeneous lossless medium is proportionalto the density of states. The density of photonic states in HMMdiverges affording an enhancement of the spontaneous emis-sion [2].

Recently appeared estimations of the ultra-high values ofthe spontaneous emission rate enhancement of the dipole emit-ter centred in the stratified metal-dielectric metamaterial [3] arevery interesting since this effect offer a way to engineer manyfascinating applications. Nevertheless, generally only integra-tional characteristics were investigated, these results deservemore careful consideration.

In [4] we study the angular dependence of the spontaneousemission enhancement of a dipole source inserted into a lay-ered metal-dielectric metamaterial, figure 1(a). We analyse thedependence of Purcell effect from the position of the dipole inthe layered hyperbolic media. We analyse the impact of thecomplex structure of eigenmodes of the system operating inhyperbolic regime. We have shown that the spontaneous emis-sion rate of the dipole emitter depends on its position, whichmainly affect the interaction with Langmuir modes, figure 1(b).

003.0

3.5

4.0

4.5

5.0 11.00.7

0.0

8.1E-6

2.8E-8

1.0E-1015 30 45Angle (degree)

Freq

uenc

y,(e

V)

ω

60 75 90

Volume modes

Langmuir modes

(b)

(a)

x

y z

TEHx

HzEy

TMEx

Ez

Hy

z =0

k– k+

dm

D

εm

dd

εd

→μ

Fig. 1. (a) Structure, ⊥ and || orientations of dipole μ and polar-izations of propagating waves. (b) The angular dependence of thePurcell factor over the frequency calculated for the layered metal-dielectric metamaterial.

References

[1] E.M. Purcell, Phys. Rev. 69, 681 (1946).[2] Z. Jacob, et al, Applied Phys. B: Lasers and Optics 100, 215

(2010).[3] I. Iorsh, et al, Physics Letters A 376, 185187 (2012).[4] A.R. Gubaydullin, et al, Appl. Phys. A 122, 425 (2016).

68


Artificial dielectric based antireflection layersfor terahertz applicationsM. Hajji, D. Zeze, C. Balocco and A.J. GallantSchool of Engineering and Computing Sciences, Durham University, South Rd, Durham, DH1 3LE, UK

Abstract. We report on experimental and simulation results for low-loss micromachined silicon artificial dielectricsoperating as antireflection coating at terahertz frequencies. The artificial layers consist of subwavelength microgrooves withdifferent profiles. Finite-difference time-domain (FDTD) simulations are used to optimize the grating pitch, and tomaximize transmission over the frequency range of interest.

Introduction

Terahertz systems (THz) have received significant attention dueto their applications in spectroscopy and imaging. However, intime domain spectroscopy (TDS) undesired reflections, causedby the materialair interface introduce unwanted spectral fea-tures and losses [1]. In the optical regime, such reflectioncan easily be minimized through the use antireflection coat-ings. This is less well suited to the terahertz region wherecomparatively thick (in the order of tens to hundreds of mi-crons) dielectric coatings would be required. The fabricationof an artificial dielectric layer (ADL) with micromachined sub-wavelength features overcomes this problem. A key benefit ofthis approach is the ability to tune the dielectric properties ofthe artificial dielectric structure spatially for various focusingand Fourier optics applications: a long-term objective of thiswork. Here we report on a silicon ADL; which functions as ananti-reflection layer. High-resistivity silicon was used as it pro-vides high transmission and low dispersion in THz region [2].

1. Simulation and results

A commercial 3D FDTD solver (Lumerical) has been usedto simulate the transmission of the terahertz radiation throughsubwavelength grating structures of varying pitch in the fre-quency range 0.1–6 THz. By tuning the period of the gratings,the transmission could be maximized over a particular targetfrequency range.

Figure 1 shows a comparison of the three profiles with op-timized transmission between 0.75 and 1.1 THz (the region wecould access experimentally). The calculated peak transmis-sion was in excess of 90%. The groove pitch and depth havebeen chosen to be compatible with bulk micromachining tech-niques whilst maintaining an effective medium approximationin the region of interest. The trapezoidal shape can be obtainedthrough KOH based anisotropic etching of 〈100〉 oriented sili-

Frequency (THz)

Tran

smis

sion

00

0.2

0.4

0.6

0.8

1

1 2 3 4 5 6 7

RectangularTriangularTrapezoidal

Fig. 1. Simulated transmittance using gratings with different pro-files (rectangular, triangular, and trapezoidal) and an equal pitch of75 μm.

Fig. 2. SEM image of a subwavelength grating using KOH etchingwith depth of 40 μm.

SiliconADL

0.75 0.8 0.85 0.9 0.95Frequency (THz)

Tran

smis

sion

(dB

)

0.1–10

–5

0

5

1.05 1.10 1.15

Fig. 3. The reflectivity as a function of frequency for ADL andsilicon wafer with no grooves with pitch of 75 μm.

con. Figure 2 shows an example of an etched silicon surface.The etch depth is self-limited due to the formation of V groovesby the (111) planes; the depth of which is determined by thetop width of the groove. This leads to the possibility of defin-ing a range of groove depths and hence the ability to produce aspatially varying dielectric properties from a single mask andetching step.

2. Experimental sectionTo validate our simulations, the transmission through the grat-ing was experimentally measured using a THz vector networkanalyzer (VNA). Figure 3 shows a reduced reflectivity for asilicon wafer with the ADL compared to a control sample onewithout. The measurement was performed with the groovesperpendicular to the plane of incidence. The results showedthat the minimum transmission increased from −5 dB for baresilicon to −2 dB for silicon with an ADL, thus doubling thetransmitted power. We have demonstrated that the ADL canbe optimized to operate in the terahertz frequency range andwe also have determined these subwavelength structures workeffectively as an antireflection layer.

References[1] M. Naftaly, R.E. Miles, Opt. Comm. 280 291–295 (2007).[2] Y.W. Chen, Xi-Ch. Zhang, Optoelectron, 7 243–262 (2013).

69


Characterization of 1D nanostructuresusing optical pump-terahertz probe time-domain spectroscopyP. Karlsen, M.E. Portnoi and E. HendryUniversity of Exeter, School of Physics, Exeter, EX4 4QL, United Kingdom

One-dimensional nanostructures such as semiconductor nano-wires and Single-Walled Carbon-Nanotubes (SWCT’s) showenormous potential as active components in devices such as so-lar cells, gas sensors and photocatalytic applications [1]. SWC-NTs are particularly appealing, since their electronic structurecan be tuned via their “chirality” [2]. These materials can be ei-ther metallic or semiconducting, and are promising candidatesfor the development of novel THz devices with many proposedapplications such as THz emitters, detectors, polarisers andantennas [3]. Both nanowires and SWCNT’s are of special in-terest due to the exotic and novel properties that can manifestthemselves due to the increased surface-to-volume ratio andquantum confinement effects.

However characterizing the electronic properties of thesedevices can prove to be difficult due to the inherent smallscales involved with these materials. Many of the typical char-acterization techniques for bulk materials are unreliable fornanostructures, since they require addition of contacts, wiresand such, which can disturb or even destroy the fragile nanos-tructure arrangement in a sample or device and mask the trueresponse of the material. Here we report our results character-ising the electronic and optical properties of various nanowiresand SWCTs using Optical Pump-THz Probe Time-DomainSpectroscopy. Terahertz Time-Domain Spectroscopy (THz-TDS) is an all-optical and non-destructive technique that hasproven very effective at characterizing the high frequency elec-tronic properties of both bulk and nanomaterials. In particular,studying the charge-carrier dynamics in these materials can re-veal a number of interesting properties, since charge carrierscan have very different properties in semi-conductors and semi-conductor nanostructures, depending on morphology, temper-ature, and material properties such as the crystal structure,band gap, dielectric function, and electron-phonon couplingstrength [4]. Combining THz measurements with an optical“pump” stimulation, enables evaluation of the ultrafast photo-response due to photoexcitation. This versatile technique al-lows measurement of important parameters for device appli-cations, including carrier lifetimes, surface recombination ve-locities, carrier mobilities and donor doping levels, which areimportant for many applications such as photovoltaic devices,light-emitting diodes and transistors.

Acknowledgement

This work has been funded by the EU FP7 Initial TrainingNetwork NOTEDEV.

References

[1] H. Zheng, J.Z. Ou, M.S. Strano, et al, K. Kalantar-zadeh, Adv.Funct. Mater., 21, 12, 2175–2196, Jun. (2011).

[2] E.H. Haroz, J.G. Duque, X. Tu, et al, Fundamental OpticalProcesses in Armchair Carbon Nanotubes, Oct. 2012.

[3] R.R. Hartmann, J. Kono and M.E. Portnoi, Nanotechnology, 25,32, 322001, Aug. (2014).

[4] R. Ulbricht, E. Hendry, J. Shan, et al, Rev. Mod. Phys., 83, 2,543-586, Jun. (2011).

70


Quality factor comparison of terahertz cavitiesformed by photonic crystal slabsA.K. Klein, D. Zeze, C. Balocco, A.J. GallantSchool of Engineering and Computing Sciences, Durham University, United Kingdom

Introduction

Cavities are some of the most fundamental and important com-ponents for many optical applications. They are a basic require-ment for key technologies such as lasers or current scientificfrontier topics such as exciton-polaritons. While a variety ofcavity designs are available for most parts of the electromag-netic spectrum, quality factors (Q factor) for resonators in theTHz region have been fairly low with values below 30 for meta-material based resonators [1]. Microcavities fabricated on sil-icon show high Q factors of up to 1000 [2], but an integrationof an additional component is challenging, if not often techni-cally impossible. Achieving highQ factors for THz cavities ischallenging due to two major issues of THz technologies:

1. Many of the traditional concepts which have been per-fected for visible and near infrared optics rely on semi-conductor growth and, since the structure size scaleswith the wavelength, the concept cannot be adapted forthe hundred of times larger wavelengths associated withTHz radiation;

2. Most materials show a lower performance in the THz re-gion with absorption being a main challenge in the fab-rication of THz components. The structures discussedhere are one dimensional Bragg stacks and two dimen-sional rod or hole crystals consisting of either high resis-tivity Silicon (Si), which is one of the preferred materialsfor THz applications due to its low losses and high re-fractive index, or low-loss polymers, such as HDPE orPTFE. The polymers have a considerably higher absorp-tion and lower refractive index than Si, but offer moreflexibility in fabrication.

1. Structures

All photonic crystals consist of two stacks of layers flankingeach side of a defect. The defect is designed to have a cavitymode at ∼1 THz. All of the proposed structures are designedwithin the limits of realistic sample fabrication and take intoaccount considerations of finite thicknesses and number of lay-ers to guarantee a close match with experiments. The lengthin the x-direction, as indicated in Fig. 1, is many wavelengthsand can therefore be considered infinite, whilst the height, i.e.the length in the z-direction, is limited. The maximum possi-ble height for the different photonic crystals varies due to theirdifferent fabrication methods. The drilling of holes in poly-mers is limited by the length of the drill, which is as much as∼1.78 mm for a 100μm drill, and therefore many wavelengths.

The milling of the polymer rods has a stronger limitation,since the milling heads have an aspect ratio of 1:6 which sig-nificantly limits the maximum height, resulting in a maximumheight of a several 100μm, depending on the unit sell size. Allsilicon photonic crystals are ultimately limited by the thickness

(d) MaterialType

Rod HDPE 270 60 200 600

Rod 60020060270PTFE

Bragg 50032024104Si

Holes 500602064Si

Holes 17008050120HDPE

aµm

dmµ

zmµ

r t/mµ

a

d

t

(c)

y

z x

(a)

⎫⎬⎭

a

a

d

r

N

(b)

Fig. 1. (a) Schematic of a rod photonic crystal with the geometricparameters unit cell a, rod diameter r , defect length d and number ofperiods on each side of the defectN ; (b) Schematic of a hole photoniccrystal which has the same geometric properties as the rod structureof (a) respectively; (c) Photonic crystal based on Bragg mirrors with alayer thickness t ; (d) table with overview of the geometric parametersof all discussed photonic crystals.

of a wafer, the most likely source of material. Photonic crystalswith band gaps for different polarisations were chosen to inves-tigate the feasibility of the concepts within the constraint of thelimited height when the electric field oscillates in the directionof the finite thickness (Transverse Magnetic (TM) polarisationwith the magnetic field in the xy plane, rods, Fig. 1a) or whenthe electric field oscillates in the direction of the quasi-infiniteaxis (Transverse Electric (TE) polarisation with the elextricfield in the xy plane, hole structures, Fig. 1b). Additionally, astructure which in theory possesses no polarisation dependencewas investigated (Bragg, Fig. 1c).

2. Results

The presented data is simulated with Lumerical FDTD and willbe validated by spot set sample measurements. The differenttypes of photonic crystals need different approaches in theirdesign. It is known that the thickness of a photonic crystal slabcan affect the band gap size severely [3]. The effect on the bandgap is due to shifting of higher order modes in a non-infiniteextended crystal. When these modes shift into the region ofthe band gap, they offer energy states which allow propagationwithin the gap which results not only in a narrowing of the bandgap but also in a weakening, or even complete suppression,of the cavity modes. This observation is particularly relevant

71

72

Number of layers (N)

Qua

lity

fact

or (

log)

0 2 4 6 8 10 1210

100

1,000

Rod HDPERod PTFEBragg Si TMBragg Si TEHoles SiHoles HDPE

●

■

▲

▲

▼

◆

●■

▲

◆

◆

▲

▼▼

▼

▼

▼

▼▼

▼

▲ ▲

▲

▲

●

Fig. 2. Quality factors of the different photonic crystal types independence of the number of layers.

for the rod type photonic crystals with their TM polarisation,since the electric field oscillates along length of the rods andtheir height limitation is in the same order of magnitude asthe wavelength. In addition, the finite extension of only afew wavelengths in y and z directions allows the formationof diagonal modes for a certain number of layers which alsosuppress the cavity mode. Due to the large difference in thedielectric constant between Si and air compared to polymer andair, the splitting between the energy bands is larger resulting ina wider band gap for Si PCs. Yet the high dielectric constantalso results in a stronger influence on the critical slab thicknesswhich leads to a narrow range of slab thicknesses where a bandgap can be maintained. The same problem is less apparentwith the TE polarised hole structures. The largest band gapfor TE structures could be obtained by choosing the maximumpossible thickness which results in close agreement with theinfinite extended model. The more common design would beto choose a thickness of around half the effective wavelengthfor a TE hole slab, but the very thin layer poses an experimentaldifficulty in both manufacturing and handling of the samples.

TheQ factor of the rod structures reaches maximum valuesof up to 360 for three layers of HDPE rods. The hole basedphotonic crystals work over a wide range of different layernumbers, but for the HDPE photonic crystals theQ factor doesnot exceed 200, whilst the Si photonic crystals show, as ex-pected, a superior performance with Q factors above 2000.The Bragg structures show generally a low performance withQ factors below 200 and, despite their hypothetical indepen-dence of polarisation, theQ factors for TM polarisation are 10times lower.

References

[1] C. Jansen, et al, Applied Physics Letters 98, 051109 (2011).[2] C. Yee, et al, Applied Physics Letters 94, 154104 (2009).[3] J.D. Joannopoulos, et al, Photonic Crystals, 2nd Edition,

(2008).


Intense THz pulse emissionfrom InAs-based epitaxial structures grown on InP substratesI. Nevinskas1, R. Butkute1, S. Stanionyte1, A. Biciunas1, A. Geizutis1,2 and A. Krotkus1

1 Centre for Physical Sciences and Technology, Sauletekio al. 3, LT-10222, Vilnius, Lithuania2 Vilnius Gediminas Technical University, Naugarduko 41, LT-03227, Vilnius, Lithuania

When semiconductors such as GaAs, InAs or GaInAs are il-luminated with femtosecond laser pulses, they emit terahertz(THz) pulses. It has been known that the bulk p-InAs is themost efficient surface THz emitter to date [1] because of highdensity of donor surface states which form a surface inversionlayer and accelerate the photoexcited carriers.

●

●

● ● ● ● ●● ● ●

● ●● ● ● ●

●

●

●● ●

● ●●

●

●

●

●

●

●

●

●

sample Asample B

Photon energy (eV)

TH

z el

ectr

ic fi

eld

(arb

. uni

ts)

0.60.0

0.2

0.4

0.6

0.8

1.0

0.8 1.0 1.2

Fig. 1. THz pulse amplitudes registered in quasi-reflection directionas functions of the optical pulse photon energies for epitaxial InAssamples with (sample B) and without the p-n junction (sample A).

Undoped InAs and InAs p-n junction epitaxial layers weregrown on (100)-cut InP substrates with Molecular Beam Epi-taxy (MBE). The lattice difference between the substrate andthe InAs layers were matched with a graded AlInAs bufferlayer. The experimental Terahertz Time-Domain Spectroscopy(THz-TDS) setup contained an optical parametric amplifier(OPA). This allowed to examine our samples under a widerange of excitation wavelengths. It has been found that thebuilt-in electric field within the p-n junction enhances the THzemission. Registering THz signals in quasi-reflection directionat excitation wavelengths longer than ∼1 μm the p-n junctionemits more intense THz radiation in comparison to an undopedbulk InAs as shown in Fig. 1. At long excitation wavelengths(> 1.6 μm) the InAs p-n junction provides stronger THz pulsesthan those from (111)-cut p-InAs, Fig. 2.

In addition, the InAs-based epitaxial layers were exposedto a constant magnetic field from Neodymium permanent mag-nets. Exposure to a magnetic field enhances the THz radiationand allows to register THz pulses in the line-of-sight THz-TDSgeometry due to Lorentz force changing the travelling directionof photoexcited carriers.

p-InAssample B

TH

z el

ectr

ic fi

eld

(arb

. uni

ts)

–0.50

–0.25

0.00

0.25

0.50

0.75

1.00

Delay time (ps)0 2–2 4 6 8 10

Fig. 2. THz pulses radiated in quasi-reflection direction when ex-cited with optical pulses of 2.3 μm. A comparison of (111)-cutp-InAs and epitaxial InAs p-n junction.

Delay time (ps)0 2–2 4 6 8 10

sample B

TH

z el

ectr

ic fi

eld

(arb

. uni

ts)

–0.5

0.0

0.5

1.0

Fig. 3. THz pulse and its Fourier spectrum radiated in the line-of-sight direction from the sample B photoexcited by Yb:KGW laserpulses with an average optical power of 120 mW when placed in anexternal magnetic field of 433 mT.

References

[1] R. Adomavicius, A. Urbanowicz, G. Molis, et al, Appl. Phys.Lett., 85, 2463 (2004).

73


Antenna-coupled microcavity quantum infrared detectorsD. Palaferri, Y. Todorov, C. SirtoriLaboratoire Materiaux et Phenomenes Quantiques, Universite Paris Diderot, Sorbonne Paris Cite, CNRS-UMS7162, 75013 Paris, France

Quantum Well Infrared Photodetectors (QWIPs) have been in-vestigated in the past 20 years as valid solution to the demand offast and high sensitive detection in the infrared and far-infraredspectral region (5 μm < λ < 200 μm) [1,2]. These devicesuse intersubband (ISB) transitions in a semiconductor quantumwell (QW) superlattice (mainly n-type doped GaAs/AlGaAs) togenerate photocurrent. Recently, we demonstrated an antenna-coupled microcavity geometry for QWIPs operating at mid-infrared [3] and terahertz [4] frequencies, which enables animproved light coupling, a reduced dark current and a highertemperature performance. The benefit of our plasmonic archi-tecture on the detector performance is assessed by comparingit with detectors made using the same quantum well absorbingregion, but processed into a standard 45◦ polished facet mesa.The patch antenna concept can be further explored with theconcept of collection-area and the field enhancement factor inthe active region of the antennae. It will be shown that it ispossible to optimize the photon collection by using either pho-tonic resonators either lumped-circuit resonators [5] in orderto achieve the highest thermal response for this double-metalconcept.

The Focusing factor expresses quantitatively how the den-sity energy of free space photons is compressed into the cavityvolumeV of a specific resonator: this quantity is is an importantfigure of merit for any detector architecture, has a straightfor-ward interpretation in terms of the local electric field enhance-ment and corresponds to the quantity F = |Ein|2/|Eout|2. Byincreasing the unit-cell surface of an array of photonic resonatorit is possible to maximize the collection of the single detec-tor element, with the superior limit indicated by (see squaresin Figure 1a). In a circuit-coupled detector the external fieldin squeezed in a much smaller active region and it is possi-ble increase F of several orders of magnitude. Such a quan-tity has a direct impact of detector performances. Looking atthe background-limited temperature (BLIP) it can be seen thatQWIP operating at 5THz, normally limited around 10 K [2],can be optimized by the single patch antenna geometry giv-ing a thermal response which corresponds to T∞

BLIP = 35 K(Fig. 1b). A LC nano-resonator [5] coupled to same detectorcould give a TBLIP = 65 K close to liquid nitrogen operation.In the same way it can be predicted for a 9μm QWIP, normallylimited to a TBLIP = 70 K [1], could be enhanced with a LCarchitecture so to operate at the highest TBLIP around 135 K.

References

[1] H.C. Liu, Intersubband Transitions in Quantum Wells, ed. byH.C. Liu and F. Capasso, Academic Press, San Diego (2000).

[2] J.C. Cao and H.C. Liu Semiconductors and Semimetals, 84(2011).

[3] Y.N. Chen et al, Appl. Phys. Lett. 104 (2014).[4] D. Palaferri et al, Appl. Phys. Lett. 106, 161102 (2015).[5] Y. Todorov et al, Opt. Express 23, 16838-16845 (2015).

●

●●

●

●

●

●●

patchantennae

LC nano-resonators

liquid N2

★★★

Ref. [4]

TBLIP = 35 K∞

101 102 103 104

Focusing factor F105 106

10

20

30

40

50

60

70

80

90

100

TB

LIP

(K)

(b)

(a)F

ocus

ing

fact

orF

102 103 104

103

104

105

Array unit cell (µm )Σ 2

patch antennalimit F ∞

0.35 µm0.5 µm

1 µm

■

■

■

■ ■

●▲

▼ ▲

Fig. 1. (a) Focusing factors F from arrays of nano-resonators withdifferent geometries: the squares are the rectangular patches and theother symbols indicate the circuit-coupled resonators described inRef. [5], with the corresponding size of the square capacitance. Thisfigure combines data with structures that operate from 3 to 5 THz,assuming that the dependence of F on the frequency is weak in thespectral range. The dashed line is the limit for patch antennas fromEq. (13). (b) TBLIP as a function of the focusing factor F (con-tinuous line). The line+symbols curves correspond to the differentresonators discussed in this work, and the dashed line to the patchantenna limit. The stars corresponds to experimental values reportedin Ref. [4].

74


Terahertz photoconductivity in silicon nanoparticles networksV. Pushkarev1, H. Nemec1, S. Gutsch2, D. Hiller2, J. Laube2, M. Zacharias2, T. Ostatnicky3, and P. Kuzel11 Institute of Physics ASCR, Na Slovance 2, 18221 Prague 8, Czech Republic2 Laboratory for Nanotechnology, Department of Microsystems Engineering (IMTEK), University of Freiburg,Georges-Koehler-Allee 103,79110 Freiburg im Vreisgau, Germany3 Faculty of Mathematics and Physics, Charles University in Prague, Ke Karlovu 3, 12116 Prague 2,Czech Republic

Abstract. Silicon based nanocrystal superlattices prepared by thermal decomposition of silicon rich oxide layers wereinvestigated using time-resolved terahertz spectroscopy at 300 and 20 K. Analysis of the terahertz photoconductivity spectraprovided information about charge transport in the nanocrystal networks.

Introduction

Silicon based nanostructures play a significant role in optoelec-tronics, namely due to their enhanced light emission in the vis-ible region. Many applications rely on charge transport amongthe silicon nanocrystals; however, its investigation is difficultdue to the nanometer dimensions, complex morphology of thestructures and a typically broad size distribution of nanocrys-tals. These difficulties can be circumvented in time-resolvedterahertz (THz) spectroscopy which allows non-contact prob-ing of charge transport on nanometer distances [1]. The size ofnanocrystals and its distribution can be greatly controlled byusing the super-lattice approach in the sample preparation [2].

Samples

Here we investigated superlattices of silicon nanocrystals withvarious sizes [2]. A superstructure of silicon-rich silicon oxidelayers SiOx (0 ≤ x ≤ 1) with thicknesses 4.5 nm separatedby 4 nm thick SiO2 layers was deposited using nitrogen free

70 nm

(a)(a) (b)(b)

(c)(c)

70 nm

70 nm

Fig. 1. Plane-view energy-filtered transmission electron microscopyimages of samples [2] with 3 different SiOx stoichiometries:(a) SiO1.0, (b) SiO0.5, (c) SiO0.0. The white areas represent thesilicon phase.

SiH4+O2 plasma enhanced chemical vapor deposition. Sub-sequently, the samples were annealed in order to induce pre-cipitation and crystallization of the silicon phase. The size,density and connectivity of the nanocrystals are controlled bythe stoichiometry x of the SiOx layers.

Experiment

We carried out measurements at 20 and 300 K in a commonsetup for time-resolved THz spectroscopy based on a Ti:sap-phire amplifier. Silicon nanocrystals were photoexcited usingthe second harmonic (400 nm). Variable attenuator was em-ployed to control the excitation intensity by almost two ordersof magnitude.

Results and discussion

The local fields, which acts directly on charges in inhomoge-neous nanomaterials are differ from the “applied” probing THzfield due to the depolarization phenomena. In order to revealthe microscopic carrier response, one should apply an appro-priate effective medium theory. For this reason, in Fig. 2, weplot the normalized transient transmission function, which isproportional to the measured raw transient transmission spectra�T/T and normalized by the incident photon fluenceF [3]. Atlow fluences�Tnorm directly reflects the microscopic mobilityof carriers [3].

Measurements at room temperature with similar samples [4]were described by classical Monte Carlo calculations and theinterpretation was based on clustering of individual nanocrys-tals during the crystallization process upon approaching thepercolation threshold. The results at 20 K are quite surpris-ingly very similar to the room temperature ones (Fig. 2). Theshapes of the mobility spectra at 20 and 300 K are almost identi-cal. Such behaviour cannot be described classically and withinMaxwell–Boltzmann statistics, which would predict a dramaticincrease of the mobility at low temperatures [5]. Quantum na-ture of the carrier confinement must play a significant. Figure2 shows �Tnorm based on quantum mechanical calculations(reflecting the discrete nature of the energy levels). Here wealso assume presence of nanoclusters together with isolated Sinanoparticles.

Conclusions

Time-resolved THz spectroscopy was employed to investigatenetworks of silicon nanocrystals prepared by thermal decom-position of SiOx layers. Analysis of THz conductivity spectra

75

76

15

10

5

0

-5

–10

Photon fluence/cm2

experiment

simulation

0.4 0.8 1.2 1.6 2 2.4Frequency, THz

0 0.4 0.8 1.2 1.6 2 2.4Frequency, THz

15

10

5

0

–5

–10

T = 300 K T = 20 K

(a) (b)

(c) (d)

ΔTno

rm2

–1–1

(cm

Vs

)

2.8 101.3 102.8 108.5 107.3 10

×××××

14

14

13

12

11

2.8 101.1 102.8 108.5 105.3 10

×××××

14

14

13

12

11

ΔTno

rm2

–1–1

(cm

Vs

)

Fig 2. Normalized transient transmission spectra�Tnorm of Si nanocrystals made by decomposition of SiO0.5 for various excitation fluenes:(a) measured at 300 K; (b) measured at 20 K; (c) QM calculated for 300 K; (d) QM calculated for 20 K. In calculation spectra presence ofsome number of big nanoclasters is taken into account.

provided information about the morphology and charge trans-port in the networks of silicon nanocrystals. The semi-classicalpicture is sufficient to describe the response at room tempera-ture, whereas quantum-mechanical approach appropriately de-scribes both cases — 300 and 20 K.

References

[1] H. Nemec et al, J. Photochem. Photobiol. A, 215, 123 (2010).[2] J. Laube et al, Appl. Phys. Lett., 108, 043106 (2016).[3] P. Kuzel and H. Nemec, J. Phys. D: Appl. Phys., 47, 374005

(2014).[4] H. Nemec et al, Phys. Rev. B, 91, 195443 (2015).


Ferroelectric epitaxial thin films for optoelectronicsA. Razumnaya1, Y.Yuzyuk2, A. Mikheykin2, I. Lukyanchuk1, V. Mukhortov3

1 University of Picardy Jules Verne, LPMC, Amiens, France2 Southern Federal University, Faculty of Physics Rostov-on-Don, Russia2 Southern Scientific Center of RAS, Rostov-on-Don, Russia

Ferroelectric barium-strontium titanate Ba1−xSrxTiO3 thinfilms are promising materials for applications in microopto-electronics devices, both in the optical range and at microwa-ves, due to the low dielectric losses, high permittivity, refractiveindex, and electro-optic coefficients. Thereby, BST films canbe used to create of active photonic crystal devices such aselectro-optic modulators and switchers for fiber-optic commu-nication lines in the gigahertz frequency range.At room temperature bulk perovskite ceramics Ba0.4Sr0.6TiO3(BST04) is in paraelectric phase and usually used for applica-tions in phase, frequency and amplitude agile microwave sys-tems [1]. On cooling this solid solution exhibits the phase tran-sitions sequence similar to that observed in pure bulk BaTiO3(BT) but at reduced critical temperatures: from cubic m3mto tetragonal 4mm phase at ∼200 K, from tetragonal 4mm toorthorhombic mm2 phase at ∼160 K, and from orthorhombicmm2 to rhombohedral (trigonal) 3m phase at ∼130 K [2]. Thesequence of observed phase transitions in thin films with thesame composition is usually different from the one in BT crys-tal.

Heteroepitaxial BST-0.4 thin film (thickness 600∼nm) wasdeposited on a cubic (001)MgO substrate by rf sputtering. Weperformed Raman and synchrotron x-ray diffraction studiesof the film with the aim to determine the crystal structureand the soft mode behavior at ferroelectric phase transitions.Synchrotron XRD data confirm the (001)BST ‖ (001)MgOand [100]/[010]BST ‖ [100]/[010]MgO relative orientations.It was found that The BST film with the out-of-plane, c =0.3967 nm and in-plane, a = 0.3937 nm lattice parametersrevealed significant tetragonal distortion (c/a > 1) at roomtemperature. Room-temperature polarized Raman spectrumof the BST04 film in the crossed scattering geometry, whichcorresponds to the E-symmetry modes, contains the E(TO)soft mode at about 105 cm−1. Raman spectrum of BST-04film in the parallel scattering geometry corresponding to A1symmetry modes has a clear interference dip at 180 cm−1 dueto mode coupling and broad lines at about 250 and 535 cm−1.It should be noted that theA1(TO) component of the soft modein the BST04 film was observed at 250 cm−1.

We have studied polarized Raman spectra of BST04 film ina broad temperature range of 80–600 K to define the temper-atures of paraelectric-to-ferroelectric phase transitions. Thecubic-tetragonal ferroelectric phase transitions in BST04 bulksolid solutions occur at ∼200 K. We found that the ferroelec-tric phase transition in BST04 film occurs at about 500 K. Noevidence of the low-temperature phase transitions in the BSTfilm was found. Thus, our Raman study revealed a giant shiftof the phase transition temperature to the paraelectric state inthin film due to the misfit strain imposed by substrate. TheCurie temperature was found to be upshifted by about 300 K inBST04 with respect to the bulk solid solutions. The relatively

high strains in this film stabilize the ferroelectric phase overa wide temperature range that is very important for practicalapplications.

Acknowledgements

We are grateful to the European Research Network FP7-ITNfor the possibility of scientific exchange within the frameworkof the program ITN-NOTEDEV. This study was supported bythe Russian Science Foundation (grant No. 14-12-00258).

References

[1] S. Gevorgian, Ferroelectrics in Microwave Devices, Circuitsand Systems. Springer, London (2009) 394 ð.

[2] V.B. Shirokov, et al., Phys. Rev. B, 73, 104116 (2006).

77


Terahertz transitions in narrow-gap carbon nanotubesand graphene nanoribbonsV.A. Saroka1,2, R.R. Hartmann3 and M.E. Portnoi11 School of Physics, University of Exeter, Exeter, United Kingdom2 Institute for Nuclear Problems, Belarusian State University, Belarus3 Physics Department, De La Salle University, Manila, Philippines

Creating reliable portable tunable sources and detectors of ter-ahertz (THz) radiation is one of the most challenging tasksof contemporary applied physics. One of the recent trends inbridging the so-called THz gap is to use carbon-based nanos-tructures [1]. Several original schemes utilizing unique elec-tronic properties of carbon nanotubes (CNTs) and graphene forTHz application has been proposed [2–5]. These schemes in-clude THz generation by hot electrons in quasi-metallic CNTs,frequency multiplication in chiral-nanotube-based superlatti-ces controlled by a transverse electric field and tunable THzradiation detection and optically-pumped emission in metallicCNTs in a strong magnetic field. In this presentation we focuson the direct interband dipole transitions in narrow-gap CNTsand graphene nanoribbons.

In this work interband dipole transitions are calculated inmetallic single-walled carbon nanotubes and armchair graphe-ne nanoribbons. The main result is that the curvature effectsfor tubes and the edge effects for ribbons result not only in asmall band gap opening, corresponding to THz frequencies, butalso in a significant enhancement of the transition probabilityrate across the band gap. The absolute value of the velocityoperator matrix element for these transitions has a universalvalue equal to the graphene’s Fermi velocity when the photonenergy coincides with the band gap energy. For higher ener-gies of photon the absolute value of the matrix element rapidlydrops to zero and then increases. A similar dependence, exceptfor the increasing part, has been obtained for an armchair CNTwith a band gap opened and controlled by a magnetic fieldapplied along the nanotube axis [4–5]. The described sharpphoton-energy dependence of the transition matrix element to-gether with the van Hove singularity at the band gap edge of theconsidered quasi-one-dimensional systems make them promis-ing candidates for active elements of coherent THz radiationemitters. The effect of Pauli blocking of low-energy interbandtransitions caused by residual doping can be suppressed by cre-ating the population inversion using high-frequency (optical)excitation. Excitonic effects, which are known to dominate op-tical properties of semiconductor CNTs, are of less importancein narrow-gap CNTs andAGNRs where the exciton binding en-ergy is proportional to the bandgap [6] and dark excitonic statesbecome irrelevant.

Acknowledgements

This work was supported by the EU FP7 ITN NOTEDEV; FP7IRSES projects QOCaN, INTERNOM, CANTOR and H2020RISE project CoExAN.

References

[1] R.R. Hartmann, J. Kono and M.E. Portnoi, Nanotechnology, 25,322001 (2014).

[2] O.V. Kibis, D.G.W. Parfitt and M.E. Portnoi, Phys. Rev. B, 71,035411 (2005).

[3] O.V. Kibis, M. Rosenau da Costa and M.E. Portnoi, Nano Lett.,7, 3414 (2007).

[4] M.E. Portnoi, O.V. Kibis, M. Rosenau da Costa, SuperlatticesMicrostruct., 43, 399 (2008).

[5] M.E. Portnoi, M. Rosenau da Costa, O.V. Kibis, I.A. Shelykh,Intern. J. Modern Phys. B., 23, 2846 (2009).

[6] R.R. Hartmann, I.A. Shelykh and M.E. Portnoi, Phys. Rev. B,84, 035437 (2011).

78


Spin currents of exciton polaritons in microcavitieswith (110)-oriented quantum wellsV. Shahnazaryan1,2,3,S. Morina1,4, S. Tarasenko5,6, I. Shelykh1,3,4

1 University of Iceland, Reykjavik, Iceland2 Russian-Armenian (Slavonic) University, Yerevan, Armenia3 ITMO University, St Petersburg, Russia4 Nanyang Technological University, Singapore5 Ioffe Institute, St Petersburg, Russia6 Peter the Great St Petersburg Polytechnic University, St Petersburg, Russia

We study the polarization optical properties of microcavitieswith embedded (110)-oriented quantum wells. The spin dy-namics of exciton polaritons in such structures is governedby the interplay of the spin-orbit splitting of exciton states,which is odd in the in-plane momentum, and the longitudinal-transverse (LT) splitting of cavity modes, which is even in themomentum. The corresponding Hamiltonian for lower branchhas a form

H =(E0 + h2k2

2m + γ kx δ(kx − iky

)2δ(kx + iky

)2E0 + h2k2

2m − γ kx

), (1)

whereE0is polariton energy at k = 0 point,m is effective mass,γ denotes the spin-orbit coupling constant, and δ denotes theLT splitting of photonic mode.

To describe the dynamics of exciton polaritons we solvethe Schrodinger-type equation for macroscopic wavefunctionof polaritons in real space accounting for the external pumpand decay

ih∂�

∂t= H� − ih

2τ� + P (�r, t) , (2)

for the two-component wave function

� =(ψ+ (�r, t)ψ− (�r, t)

). (3)

We demonstrate the generation of polariton spin currents by alinearly polarized optical pump and analyze the arising polari-ton spin textures in the cavity plane. Namely, the Fig. 1 showsspatial distribution of stokes parameters obtained for a largepump-spot radius r0 = 20 μm. The upper and lower panelscontain plots for two different values of the spin-orbit couplingparameter which can be tuned in a wide range by changingthe QW width. One can see that the pumping with linearlypolarized light results in the emergence of a partial circular po-larization of polaritons at the right and left sides of the pumpspot, Figs. 1(a) and 1(d).

Further, by tuning of the excitation spot size, which con-trols the polariton distribution in the momentum space, oneobtains symmetric or asymmetric spin textures. A decrease inthe pump-spot size leads to the population of polariton stateswith higher wave vectors. As a consequence, the k-quadraticLT splitting takes over the k-linear SO splitting and the spatialdistribution of polarization changes. Such a transformationof the pattern of the polariton circular polarization is shownin Fig. 2. At large radii of the pump spot, the spatial distribu-tion of ρcirc is asymmetric and originates from the spin currentof polaritons. At smaller radii, the asymmetric distribution is

200

200

200 200 200

100

100

0

0

0

(a) (b) (c)

0 0

y(µ

m)

y(µ

m)

x (µm) x (µm) x (µm)

−100

−100

−200

−200−200 −200 −200

1

1

0.5

0.5

0

0

−0.5

−0.5

−1

−1

ρcirc ρlin ρdiag

(d) (e) (f)

Fig. 1. Spatial distribution of the Stokes parameters determining thepolarization of polaritons for linearly polarized pump and the pump-spot radius r0 = 20 μm. (a)–(c) and (d)–(f) Calculated for the spin-orbit coupling parameters γx = 50 and 100 meVÅ, respectively.The central area has a strong linear polarization stemming from thepump polarization, overshadowing the polarization conversion in(a), (c), (d), and (f).

1

1

0.5

0.5

0

0

−0.5

−0.5

−1

−1

(a) (b) (c)

(d) (e) (f)

ρcirc

r0 = 30 µm

r0 = 15 µmr0 = 15 µm

r0 = 25 µm

r0 = 10 µmr0 = 10 µm

r0 = 20 µm

r0 = 5 µmr0 = 5 µm

ρcirc ρcirc200

200

200 200 200

100

100

0

0

0 0 0

y(µ

m)

y(µ

m)

x (µm) x (µm) x (µm)

−100

−100

−200

−200−200 −200 −200

Fig. 2. Spatial distribution of the Stoke parameter ρcirc determiningthe circular polarization of polaritons for different pump-spot radii.The pump is linearly polarized along the x axis, γx = 100 meVÅ.The central area has a strong linear polarization stemming from thepump polarization, overshadowing the polarization conversion. Thefurther details of investigation can be found in Ref. [1].

superimposed with the symmetric cross-shape pattern of theoptical spin Hall effect stemming from the LT spslitting.

The further details of investigation can be found in Ref. [1].

References

[1] V. Shahnazaryan, et al., Phys. Rev. B, 92, 155305 (2015).

79


Modeling of topological nanostructures in ferroelectricsfor THz radiation applicationsS. Kondovych, I. LukyanchukLaboratory of Condensed Matter Physics, University of Picardie Jules Verne, 33 rue St Leu, 80039 Amiens,France

Since the size of physical systems goes nano, the majority ofthe material studies concentrates around the novel unique prop-erties of ultrathin films, nanowires, nanoparticles etc. Ferroic(ferroelectric, ferromagnetic, ferroelastic, multiferroic) nanos-tructures are of special interest, since their intrinsic propertiessignificantly differ from those in the bulk. Driven by the sur-face and edge effects, a large variety of the order parameterpatterns appear in ferroic nanostructures: domains, vortices,bubbles, skyrmions [1,2].

Polarization textures in ferroelectic nanostructures are in-tensively investigated both theoretically and experimentally, aswell as through ab-initio simulations, showing the high impactof the polarization distribution on static and dynamic proper-ties of nanosized ferroelectrics. Being easily tuned by externalelectric field, these patterned nanostructures find their nichein the development of modern nanoelectronic and informationstorage devices. They can be used as nanocapacitors, memoryelements, can be combined with magnetic or semiconductormaterials in hybrid devices [3]. Dynamical properties of tex-tured ferroelecrics can allow for their implementation as THz-radiation sources and detectors [4].

In the present work, we study the polarization ordering inthin ferroelectric films and nanodots, adopting the phenomeno-logical Ginzburg–Landau approach. We take into account thedetermining role of the depolarization field in the topologicaltexture confinement, and model the order parameter distribu-tion accordingly for two different systems: thin ferroelectricfilm with in-plane stripe domains (3D system) and cylindricalferroelectric nanoparticle (2D system). We show the depen-dence of the equilibrium state on the system size and geometryin both cases, and calculate the resulting permittivity.

References

[1] J. Seidel (Ed.), Topological structures in ferroic materials: do-main walls, vortices and skyrmions, Springer (2016).

[2] S. Prosandeev, et al., Ferroelectric vortices and related con-figurations, in nanoscale ferroelectrics and multiferroics: keyprocessing and characterization issues, and nanoscale effects,Volume I & II (eds. M. Alguero, J.M. Gregg and L. Mitoseriu),John Wiley & Sons, Ltd, Chichester, UK (2016).

[3] J.F. Scott, Applications of modern ferroelectrics, Science 315,954 (2007).

[4] I.A. Lukyanchuk, et al., Terahertz electrodynamics of 180◦ do-main walls in thin ferroelectric films, arXiv:1410.3124v3

80


Second-harmonic generation for crystal structure characterizationalong GaAs nanowiresM. Timofeeva1,2, A. Bouravleuv3,4, G. Cirlin2,3,4, M. Reig Escale1, A. Sergeev1, R. Grange1

1 Swiss Federal Institute of Technology Zurich, Phys. Dept, IQE, Auguste-Piccard-Hof 18093 Zurich, Switzerland2 ITMO University, Birzhevaja line 14, 199034 St Petersburg, Russia3 Ioffe Institute, St Petersburg, Russia4 St Petersburg Academic University, St Petersburg, Russia

Abstract. In our work we demonstrate an optical method for crystal structure characterization in semiconductor nanowires,based on the polarization-dependent second harmonic imaging. This method allows us to resolve areas with WZ and ZBcrystal structures along GaAs nanowires.

Introduction

Semiconductor III–V nanowires (NWs) are one of the keynanostructures for the new generation of optoelectronic de-vices. One of the most unique feature of the III–V NWs is thepossibility to fabricate structures with crystallographic phasesthat cannot be obtained at normal conditions in bulk III–V ma-terials, for example heterostructures based on zinc blende (ZB)and wurtzite (WZ) transitions in InAs NWs or GaAs NWs. Themost common technique to characterize the crystal structure ofNWs is the transmission electron microscopy (TEM), but thistechnique requires to deposit NWs on the special TEM sub-strates and cannot be used for a non-destructive analysis ofNWs, implemented, for example in lab-on-chip devices.

In our work, we present a method of NWs crystal structurecharacterization, based on the polarization-dependent second-harmonic generation (SHG) imaging with per-pixel resolution.This method was applied to resolve the areas with ZB and WZcrystal structure along the GaAs NWs.

The SHG is the second-order nonlinear optical process,where two photons with fundamental angular frequency (ω)convert into one photon with angular frequency (2ω). Thepolarization at the doubled frequency �P(2ω) is [1]:

�P(2ω) ∝ χ(2) · �E(ω) · �E(ω) (1)

where �E(ω) — is the applied field and χ(2) — the nonlinearsecond-order susceptibility tensor, that characterize nonlinearoptical properties of the material. The χ(2) tensor is dependenton the type of crystallographic point group of the studied ma-terial, which are different for WZ and ZB crystal structure [1].

1. Experiment

The studied GaAs NWs were grown with molecular-beam epi-taxy on Si(111) substrate with Au, as a catalyst material. Thenthese NWs were mechanically transfer to the TEM substratein order to perform together the TEM and the polarization-dependent SHG analysis. The scanning-transmission electronmicroscopy (STEM) studies of these NWs demonstrate, thatthe NWs have long areas with pure WZ and pure ZB, mixedWZ/ZB. Figure 1 demonstrate the example of STEM image ofthese GaAs NWs. The dark zones are WZ and the light zonesare ZB according to corresponding HRTEM measurements.

For experimental studies of polarization-dependent SHG,NWs are excited with a tuneable Ti:sapphire laser operating

ZB

WZ

500 nm

WZ

ZB

Fig. 1. STEM image of GaAs NWs with ZB and WZ zones, scale500 nm.

Half-waveplates Objective

10×

Beamsplitter Sample

IR la

ser

Filters

LensFlipmirrow

EMCDD

Spectrometer

Objective100×

Fig. 2. Schematic image of optical setup for nonlinear optical char-acterization of GaAs.

at 820 nm. Figure 2 presents a schematic image of the ex-perimental setup for polar SHG imaging [2]. The NWs wereplaced in the focal plane of the incident laser beam, SHG wasrecorded with an electron magnified CCD camera and anal-ysed with a spectrometer. All measurements were carried outat room temperature and pressure.

2. Results and discussion

The polarization-dependent SHG intensity images were re-corded at different polarizations from 0 to 360◦. Figure 3demonstrates the SHG images at different polarizations �E par-allel and rotated by 45◦ to NW longitudinal axis. Per-pixelanalysis of the recorded polar SHG responses allows to distin-guish the areas with different crystal structure along the singleGaAs NW. According to the polar SHG responses, the studiedNW can be split into zones with similar polar responses. Thevariations of the shapes of polar SHG responses (Fig. 3) allowto resolve the areas with WZ and ZB crystal structures alongthe NW.

The SHG intensity depends on the �E(ω) direction, χ(2) andthe orientation of the crystal structure within the NW (eq. 1).In our work, we developed a general theoretical model, that

81

82

WZ (1.5 µm)

WZ (1.5 µm)

ZB (300 nm)

ZB (300 nm)

WZ (700 nm)

WZ (700 nm)

E→

E→

Fig. 3. Image of SHG in GaAs NW at different polarizations of theincoming laser beam.

allows us to calculate the 3D SHG responses for each possiblecrystal structure and direction of the �E(ω). The shapes of these3D surfaces are defined by the components of the χ(2). TheZB GaAs belongs to 43 mm [1] class of the crystallographicpoint group and WZ GaAs — to 6 mm [3]. The correspond-ing χ(2)ZB and χ(2)WZ tensors have different nonzero components.Thus, study the variations of the shapes of SHG responses wecan estimate the areas with WZ and ZB crystal structures. Fig-ures 4 and 5 present the comparison of the measured (dots) andmodelled (lines) polar SHG responses for areas with WZ andZB crystal structures in NW.

0 20°

40°

60°

80°

100°

120°

140°

160°180°200°

220°

240°

260°

280°

300°

320°

340°

WZ crystal phase

Fig. 4. Measured (triangles) and modelled (lines) polar SHG re-sponses for areas with WZ crystal structure.

0 20°

40°

60°

80°

100°

120°

140°

160°180°200°

220°

240°

260°

280°

300°

320°

340°

ZB crystal phase

Fig. 5. Measured (dots) and modelled (lines) polar SHG responsesfor areas with ZB crystal structure.

The theoretical SHG responses on the Figs. 4 and 5 showthe cross-sections of the 3D shapes. Thus, by analysing the

variation of the SHG polar shapes along the NW, we can dis-tinguish the areas with WZ and ZB crystal structure. Studyingthe rotation of the corresponding shapes we can resolve therotation of crystal phases within the single NW.

3. Conclusions

In summary, in our work we demonstrated the method of per-pixel analysis of the polar SHG responses allows to resolve dif-ferent crystal phases along a single NW. The proposed theoret-ical model demonstrates the way of evaluation crystal structureby χ(2) of the material. This method is sensitive to structuraltransitions in NWs and does not require any special environ-mental conditions, like vacuum or low temperature environ-ment.

References

[1] Y. Zhang, et al, Nano Letters, 9, 2109–2112 (2009).[2] R. Grange, et al, Nano Letters, 12, 5412–5417 (2012).[3] R. Sanatinia, et al, Nano Letters, 12, 820–826 (2012).


1180 nm GaInNAs quantum wellbased high power DBR laser diodesH. Virtanen, J. Viheriala, A.T. Aho, V.-M. Korpijarvi, M. Koskinen, M. Dumitrescu, M. GuinaOptoelectronics Research Centre, Tampere University of Technology, Tampere, Finland

A single-mode 1180 nm distributed Bragg reflector (DBR) laserdiode with a state-of-the-art output power of ∼500 mW is re-ported in this paper. A high temperature stability was achievedby using GaInNAs/GaAs quantum wells (QWs), which ex-hibit improved carrier confinement compared to standard In-GaAs/GaAs QWs. The single-mode laser emission could betuned by changing the mount temperature (0.1 nm/◦C) or thedrive current (1.4 pm/mA). The laser showed no degradationin a room-temperature lifetime test at 900 mA drive-current.These compact and efficient 1180 nm laser diodes are instru-mental for the development of compact frequency doubledyellow–orange lasers, which have important applications inspectroscopy and medicine.

Many important applications in spectroscopy and medicine,such as the treatment of vascular lesions, would benefit fromthe availability of compact semiconductor lasers emitting atyellow–orange wavelengths. However, this wavelength rangecannot be reached with compact and efficient directly emittingsemiconductor lasers typically employed for red wavelengths(GaAs-based compounds) or for blue-green wavelengths(GaN-based materials). Moreover, the frequency doubling ap-proaches, traditionally used for reaching the green wavelengthrange, suffer from the lack of high-power narrow-linewidthfrequency-stable laser diodes emitting at 1170–1200 nm. Thisis due to the fact that GaInAs/GaAs QWs for this wavelengthrange require a high In content, leading to high strain that gen-erate a high amount of defects, which affect the laser efficiencyand life-time.

The 1180 nm wavelength range can be reached using dilutenitrides, i.e. GaInNAs/GaAs QWs. The use of a small amountof N, in the range of 1%, has been recognized for its benefitsrelated to reduced strain and good carrier confinement, en-abling power levels beyond 10 W in optically pumped vertical-external-cavity surface-emitting lasers (VECSELs) [1]. Theimproved carrier confinement translates to improved tempera-ture stability of the laser characteristics, a feature that has beenrecognized since the proposal of GaInNAs/GaAs QWs for un-cooled telecom lasers at 1.3μm [2]. The improved temperaturestability is expected to benefit especially the miniaturization offrequency-doubled lasers and in general the development ofphotonic integration approaches, which are currently limitedby thermal management issues. For example, the ability oflasers to operate at elevated temperatures will reduce the con-straints of mounting them close to frequency doubling crys-tals, which often require elevated operation temperatures [3].The developed 1180 nm DBR LDs achieved state-of-the-art490 mW single-mode output power in continuous-wave oper-ation. The emission spectrum variation with the bias current isshown in Fig. 1.

250 500 750 1000 1250 1500 1750 2000

1179

1178

1177

1176

1175

1174

Current (mA)

Wav

elen

gth

(nm

)

–90

–80

–70

–60

–50

–40

–30

–20

–10

00 250 500 750 1000 1250 1500 1750 2000

0

100

200

300

400

500

600

0.50

0.75

1.00

1.25

1.50

1.75

2.00

2.250 250 500 750 1000 1250 1500 1750 2000

Pow

er (

mW

)

Vol

tage

(V

)

(a)

(b)

Fig. 1. (a) LIV curves for 1180 nm DBR-LD device at room temper-ature. (b) the emission spectrum variation with bias current at 20 ◦Cmeasured at CW operation indicates that the side-mode suppressionratio varies from 38 to 48 dB, while the emission wavelength is tuned1.93 nm as the bias current changes from 100 to 1500 mA. The colorindicates the relative optical power in decibels.

References

[1] V. Korpijarvi, et al, Optics express 18, 25633-25641 (2010).[2] M. Kondow, et al, Selected Topics in Quantum Electronics,

IEEE Journal 3, 719-730 (1997).[3] A. Jechow, et al, Laser & Photonics Reviews 4, 633-655 (2010).

83

Author Index

Aagesen M., 7Aho A. T., 83Akopian A., 31Aldana A., 38Ali A., 49Andre Y., 4, 28Arbiol J., 60Arcangeli A., 8Avit G., 28Babichev A., 18, 50Bai M., 35Bai X., 64Balocco C., 68, 69, 71Bayle F., 18, 50Bellet-Amalric E., 41Beltram F., 8Berdnikov Y., 34, 39Berthe M., 42Biciunas A., 73Borgstrom M. T., 25Borodavka F., 64Bougerol C., 4, 18, 28Boulanger J., 7Bouravleuv A., 81Butkute R., 73Butte R., 16Caroff P., 47Castelluci D., 4, 28Catone D., 62Chandramohan A., 26, 40Chen X., 32Chervinskii S., 66Chia A., 7Cirlin G., 26, 40, 49, 81Cirlin G. E., 5, 31Clark S. J., 37Collier T. P., 67Cremel T., 41Dıaz A lvarez A., 42, 47Dai X., 18, 45David J., 56DavidJ., 58de la Mata M., 60Deiana D., 60Deng Z., 64Di Mario L., 62Dkhil B., 64Dong Z., 4, 28Dubrovskii V., 4, 26, 28Dubrovskii V. G., 30, 43, 47Dubrovskii V.G., 34, 39Dumitrescu M., 83Durand C., 18, 45Ercolani D., 8, 56, 58Eymery J., 18, 45Foldyna M., 49

Fontcuberta i Morral A., 3, 42, 60Fontcuberta i Morral A., 55Friedl M., 60Gerard J.-M., 6Gallant A., 26, 40Gallant A. J., 69, 71Gayral B., 17Geizutis A., 73Gemmi M., 56, 58Geng B., 32Ghasemi M., 12Gil E., 4, 11, 28Glas F., 47, 52Glas F. , 9Gogneau N., 49, 50Gomes U. P., 56, 58Graczyk M., 25Grandidier B., 20, 42, 47Grange R., 81Grecenkov J., 43Greenblatt M., 64Guan N., 18, 45, 50Gubaydullin A. R., 68Guina M., 83Gutsch S., 75Hajji M., 69Hao J., 35Harmand J-C., 49Harmand J.-C., 50, 52Hartmann R. R., 78Hendry E., 70Heurlin M., 25Hiller D., 75Horrocks B., 38Houlton A., 38Huang H., 35Ilkiv I. V., 5Ivanov K. A., 68Ivanov S. V., 19Jacopin G., 18Jagadish C., 2Jam R. J., 25Janda T., 24Jeannin M., 41Jmerik V. N., 19Johansson J., 12Julien F. H., 18, 45, 49, 50Jungwirth T., 24Kadlec C., 64Kadlec F., 64Kaliteevski M. A., 68Kamba S., 64Kamenskii A., 66Karlsen P., 70Kasama T., 7Kheng K., 41

Kim W., 55Klein A. K., 71Kondovych S., 80Korpijarvi V.-M., 83Koryakin A., 46Koskinen M., 83Kotlyar K. P., 5Kozlovsky V. I., 19Kratzer P., 10Krotkus A., 73Kuz el P., 75Kuzel P., 24Kukushkin S., 14Kukushkin S. A., 5, 53Kyhm K., 41Lampin J.-F., 42LaPierre R. R., 7Largeau L., 50Laube J., 75Lee W., 41Leran J.-B., 60Leroux C., 4Leshchenko E. D., 47Leyden M., 7Leymarie J., 28Li S., 32Lipovskii A. A., 66Lukyanchuk I., 77, 80Lutsenko E. V., 19Medard F., 28Martelli F., 62Matteini F., 55, 60Maximov I., 25Mazlin V. A., 68Mendis B., 26, 40Messanvi A., 18, 45Mikheykin A., 77Monier G., 4, 28Morassi M., 50Morina S., 79Mukhin I., 49Mukhortov V., 77Nadvornık L., 24Nemec H., 24, 75Nemec P., 24Neplokh V., 18, 49, 50Nevinskas I., 73Nikitina E. V., 5Nilsson N., 25Nogues G., 41Novak V., 24Nys J.-P., 42O’Keeffe P., 62Oehler F., 52Olejnık K., 24Osipov A., 14

84

Author Index 85

Osipov A. V., 53Osipov S. A., 5Ostatnicky T., 75Otnes G., 25Palaferri D., 74Pantzas K., 52Patriarche G., 52Petty M., 26, 40Piazza V., 18, 50Plissard S., 42Plissard S. R., 47Portnoi M. E., 23, 67, 70, 78Potts H., 60Potts H. A., 55Priante G., 52Pushkarev V., 75Qu B., 35Quach P., 50Reveret F., 28Ramdani R., 4Razumnaya A., 77Reduto I., 66Reig Escale M., 81Retuerto M., 64Reznik R. R., 5

Rigutti L., 18Roche E., 28Roddaro S., 8Rong X., 19Ross F. M., 1Rossella F., 8Saroka V. A., 67, 78Sergeev A., 81Shahnazaryan V., 79Shelykh I., 79Sibirev N.V., 34, 39Sibirev N., 26, 46Sirtori C., 74Skiadopoulou S., 64Skoromets V., 24Sorba L., 8, 56, 58Soshnikov I. P., 5Stanionyte S., 73Tutuncuoglu G., 42, 55, 60Tang Z., 21Tarasenko S., 79Tavakoli Dastjerdi H., 7Tchernycheva M., 18, 45, 49, 50Telyatnik R. S., 53Tian L., 62

Timofeeva M., 81Todorov Y., 22, 74Toropov A. A., 19Trassoudaine A., 4, 28Tredicucci A., 8Trojanek F., 24Turchina M. A., 47Turchini S., 62

Viheriala J., 83Virtanen H., 83Vukajlovic Plestina J., 55

Wang X., 19Wu H., 35Wunderlich J., 24

Xu J., 8Xu T., 42, 47

Yan J., 45Yazdi S., 7Yuzyuk Y., 77

Zacharias M., 75Zannier V., 56, 58Zeze D., 26, 40, 69, 71Zhang H., 18, 45Zhang J., 35

LIST OF PARTICIPANTS

Agruzov Petr RussiaPeter the Great St PetersburgPolytechnic [email protected]

Akkuzina Alina RussiaDmitry Mendeleev Universityof Chemical [email protected]

Alam A B M Khairul FinlandUniversity of [email protected]

Aldana Rodriguez Andres UKNewcastle [email protected]

Andre Yamina FranceUniversite Blaise Pascal, Clermont-Ferrand,ITMO University, St Petersburg, [email protected]

Anurova Mariia RussiaMendeleev University of Chemical Technology,Moscowmaria [email protected]

Arakawa Yasuhiko JapanThe University of [email protected]

Arseneva Julia RussiaSt Petersburg Academic Universityyulia [email protected]

Baldycheva Anna UKUniversity of [email protected]

Balocco Claudio UKDurham [email protected]

Bej Subhajit FinlandUniversity of [email protected]

Berdnikov Yury RussiaSt Petersburg Academic [email protected]

Bondarenko Sergey RussiaInstitute of Chemistry, Saratov State University

Borie Sylvain FranceUniversite Blaise Pascal, Clermont-Ferrand,ITMO University, St Petersburg, [email protected]

Butte Raphael Switzerland

Ecole Polytechnique Federale de [email protected]

Chandramohan Abhishek UKDurham [email protected]

Chen Xiao Ming ChinaDalian University of Technologychen [email protected]

Chervinskii Semen RussiaUniversity of Eastern Finland, Peter the GreatSt Petersburg Polytechnic [email protected]

Cirlin George RussiaSt Petersburg Academic [email protected]

Clark Stewart John UKDurham [email protected]

Collier Thomas Pierre UKUniversity of [email protected]

Cremel Thibault FranceUniversite Grenoble Alpes, Grenoble, CEA,INAC-SP2M, [email protected]

Dıaz Alvarez Adrian FranceISEN-CNRS, [email protected]

Dagnet Thibault FranceInstitut Pascal, [email protected]

Dong Zhenning FranceInstitut Pascal, [email protected]

Dubrovskii Vladimir RussiaSt Petersburg Academic [email protected]

Ercolani Daniele ItalyNEST, Istituto Nanoscienze-CNR and ScuolaNormale [email protected]

Fontcuberta i Morra Anna Switzerland


Friedl Martin Switzerland


Gallant Andrew UKDurham [email protected]

Gayral Bruno FranceCEA-Grenoble, [email protected]

GERARD Jean-Michel FranceCEA, Institute for Nanoscience and [email protected]

Gerasimenko Alexander RussiaNational Research University of ElectronicTechnology, Institute of Nanotechnology ofMicroelectronics RAS, Kotelnikov Instituteof Radio-Engineering and Electronics RAS,Scientific-Manufacturing Complex “Technologi-cal Centre’’[email protected]

Gil Evelyne FranceInstitut Pascal CNRS, Universit’e Blaise Pas-cal – ITMO [email protected]

Glas Frank FranceCNRS – Laboratoire de Photonique et [email protected]

Gordeychuk Dmitrii RussiaInstitute for Analytical Instrumentation RAS

Grandidier Bruno FranceISEN-CNRS, [email protected]

Grecenkov Jurij RussiaSt Petersburg Academic [email protected]

Guan Nan FranceIEF – University Paris [email protected]

Gubaydullin Azat RussiaSt Petersburg Academic [email protected]

Hajji Maryam UKDurham [email protected]

Hassanieh Atefeh SwedenLund [email protected]

Huang Hui ChinaDalian University of [email protected]

Ivanov Sergey RussiaIoffe Institute, St [email protected]

Jagadish Chennupati AustraliaAustralian National [email protected]

Jho Young-Dhal South KoreaGwangju Institute of Science and Technology(GIST)[email protected]

Jiang Yuan ChinaEngineering University of CAPF, [email protected]

Johansson Sten Jonas SwedenLund [email protected]

Kaliteevskaya Natalia UKDurham [email protected]

Kaliteevskii Mikhail RussiaITMO University, St [email protected]

Karlsen Peter UKUniversity of [email protected]

King Jennifer Anne UKDurham [email protected]

Klein Andreas Kurt UKDurham [email protected]

Kondovych Svitlana FranceUniversity of Picardie Jules Verne, [email protected]

Koryakin Alexandr RussiaSt Petersburg Academic [email protected]

Kozina Olga RussiaKotel’nikov Institute of Radio-Engineering andElectronics RAS, [email protected]

Kratzer Peter GermanyUniversity [email protected]

87

88 List of participants

Krichevsky Denis RussiaNational Research Nuclear University “MEPhI”,Lomonosov Moscow State University,Prokhorov General Physics Institute RAS,Institute of Physiologically Active [email protected]

Kucherik Alexey RussiaStoletov Vladimir State University, [email protected]

Kukushkin Sergey RussiaInstitute of Mechanical Engineering RAS, St [email protected]

Kuraptsev Alexey RussiaPeter the Great St PetersburgPolytechnic [email protected]

Kutrovskaya Stella RussiaStoletov Vladimir State University, National Re-search University [email protected]

Kuzel Petr Czech RepublicInstitute of Physics, Czech Academy of [email protected]

LaPierre Ray CanadaMcMaster [email protected]

Leila Ahmadi FinlandUniversity of Eastern [email protected]

Leshchenko Egor RussiaSt Petersburg State [email protected]

Markova Nadezhda RussiaSt Petersburg Academic [email protected]

Maximov Ivan SwedenLund [email protected]

Mayakova Mariya RussiaProkhorov General Physics Institute RAS,Mendeleev University of Chemical Technology,Lomonosov Moscow State [email protected]

Mikhaylina Anna RussiaInstitute of Geology Karelian Research CentreRAS, [email protected]

Mitetelo Nikolay RussiaLomonosov Moscow State University, [email protected]

Mostafavi Kashani Seyed MohammadGermanySiegen [email protected]

Neplokh Vladimir FranceIEF, University [email protected]

Nepomniashchaia Elina RussiaPeter the Great St PetersburgPolytechnic [email protected]

Nevinskas Ignas LithuaniaCentre for Physical Sciences and Technology,[email protected]

Osipov Andrey RussiaInstitute of Mechanical Engineering RAS, St [email protected]

Osipov Anton RussiaStoletov Vladimir State University, Russia,Laser Zentrum Hannover, Germany

Palaferri Daniele FranceUniversity Paris Diderot [email protected]

Panciera Federico UKUniversity of Cambridge, IBM, [email protected]

Petrov Pavel RussiaITMO University, St [email protected]

Petrova Olga RussiaDmitry Mendeleev University of Chemical [email protected]

Piazza Valerio FranceUniversite Paris [email protected]

Platonov Vyacheslav RussiaInstitute of Electrophysics UB RAS, Zuev Insti-tute of Atmospheric Optics SB RAS, NationalResearch Tomsk Polytechnic [email protected]

Polokhin Alexander RussiaNational Research University of ElectronicTechnology, Institute of Nanotechnology ofMicroelectronics RAS, Kotelnikov Instituteof Radio-Engineering and Electronics RAS,Scientific-Manufacturing Complex “Technologi-cal Centre”[email protected]

Portnoi Mikhail UKUniversity of [email protected]

Priante Giacomo FranceCNRS – Laboratoire de Photonique et [email protected]

Pukhov Konstantin RussiaLaser Materials and Technology Centre, Gen-eral Physics Institute of the RAS (GPI RAS),[email protected]

Pushkarev Vladimir Czech RepublicInstitute of Physics of the Czech Academy [email protected]

Razumnaya Anna FranceUniversity of Picardy Jules [email protected]

Redkov Alexey RussiaSt Petersburg Academic [email protected]

Reduto Igor FinlandUniversity of Eastern [email protected]

Reynes Laure FranceUniversite Blaise Pascal, Clermont-Ferrand,

ITMO University, St Petersburg, [email protected]

Reznik Rodion RussiaPeter the Great St PetersburgPolytechnic [email protected]

Ritter Dan IsraelTechnion-Israel Institute of Technology, [email protected]

Ross Frances Mary USAIBM T. J. Watson Research Center, YorktownHeights, New [email protected]

Rumyantsev Vladimir UkraineGalkin Institute for Physics and Engineer-ing, Mediterranean Institute of FundamentalPhysics, [email protected]

Rylkova Marina RussiaSt Petersburg Academic [email protected]

Saroka Vasil UKUniversity of [email protected]

Sazonkin Stanislav RussiaBauman Moscow State Technical [email protected]

Shahnazaryan Vanik RussiaUniversity of Iceland, ITMO University, St [email protected]

Sibirev Nickolay RussiaSt Petersburg Academic [email protected]

Sibireva Elena RussiaSt Petersburg Academic [email protected]

Skiadopoulou Stella Czech RepublicInstitute of Physics, Czech Academy of [email protected]

Sokolova Zhanna RussiaSt Petersburg Academic Universitysokolova [email protected]

Sonnada Math Shivaprasad IndiaJawaharlal Nehru Centre for Advanced Scien-tific [email protected]

Sorba Lucia ItalyNEST, Istituto Nanoscienze-CNR and ScuolaNormale [email protected]

Starovoytov Anton RussiaITMO University, St [email protected]

Stepanova Irina RussiaDmitry Mendeleev University of Chemical [email protected]

Tang Zhenan ChinaDalian University of [email protected]

Tchernycheva Maria FranceIEF-CNRS, University Paris [email protected]

Telyatnik Rodion RussiaInstitute of Mechanical Engineering RAS, St [email protected]

Tian Huiping ChinaBeijing University of Posts and [email protected]

Tian Lin ItalyCNR-IMM, [email protected]

List of participants 89

Tiburzi Giorgio UKUniversity of [email protected]

Timofeeva Mariia SwitzerlandETH [email protected]

Todorov Yanko FranceUniversity Paris [email protected]

Turchina Mariia RussiaSt Petersburg Academic [email protected]

Uskov Alexander RussiaLebedev Physical Institute RAS, [email protected]

Vilejshikova Elena RussiaBelarusian National Technical University, Be-larus, NITIOM Vavilov State Optical [email protected]

Virtanen Heikki FinlandOptoelectronic Research Centre, Tampere Uni-versity of Technology,[email protected]

Vitkin Vladimir RussiaNITIOM Vavilov State Optical Institute, Russia,Belarusian National Technical University, Be-larus, ITMO [email protected]

Vukajlovic Plestina Jelena Switzerland

Ecole Polytechnique Fe de rale de [email protected]

Wang Jianguo ChinaShanghai Institute of Optics and Fine Mechan-ics [email protected]

Yablonovitch Eli USAUniversity of California, [email protected]

Zannier Valentina ItalyNEST Scuola Normale Superiore and IstitutoNanoscienze – CNR, [email protected]

Zeze Dagou UKDurham [email protected]

Zwiller Valery [email protected]

event.itmo.infoevent.itmo.info/images/pages/45/nsp16june27.pdf · physics and technology of...

Documents