Light Engineering in New Materials How material science can help in - PowerPoint PPT Presentation

Light Engineering in New Materials How material science can help in realizing new efficient incoherent and coherent light sources Anna Vinattieri, Dept. Physics and Astronomy, UNIFI Lights of Tuscany 2019 April 15, 2019 1

Outline  A little bit of history  Band-gap engineering  Epitaxial growth  Chemical synthesis  Perovskites  A look to the future Lights of Tuscany 2019 April 15, 2019 2

A little bit of history: 150 years from the Mendeleev Table The semiconductor tree The Mendeleev table is the tree principal root Lights of Tuscany 2019 April 15, 2019 3

The recognition of material science relevance The nanostructures and the superconductors Lights of Tuscany 2019 April 15, 2019 4

Heterostructures and integrated circuit Lights of Tuscany 2019 April 15, 2019 5

The CCD Lights of Tuscany 2019 April 15, 2019 6

The Graphene: the perfect lattice Lights of Tuscany 2019 April 15, 2019 7

The blue/white laser diodes Lights of Tuscany 2019 April 15, 2019 8

The materials for opto-devices Inorganic Inorganic Ordered structures Ordered structures ( bulk crystals and ( bulk crystals and nanostructures) nanostructures) Semiconductors Semiconductors Amorphous Amorphous Organic Organic Lights of Tuscany 2019 April 15, 2019 9

The Rosetta Stone of Semiconductors Lights of Tuscany 2019 April 15, 2019 10

The King of Semiconductors for electronics: Silicon 32 inch , 80 cm Unfortunately Si is very bad for light emitters being an indirect band-gap semiconductor, so it is hard to have electrons and holes recombining in a radiative way ! Lights of Tuscany 2019 April 15, 2019 11

The Kings of Semiconductors for light emitters : III-V and III-Nitrides In,Ga,Al Arsenide and Nitride Alloys with a band gap value ranging from 0.8 to 4 eV We can realize a device emitting light from 1.55 µm ( perfect for telecomm) and 300 nm ( perfect for UV- lithography and blue-rays, UV curing, etc.). White LEDs are realized using nitride alloys. Lights of Tuscany 2019 April 15, 2019 12

How materials are realized?  Czochralski growth (1916): very fast for big crystals : O and C common contaminants  growth speed  Epitaxial growth developed in late 1960s in ~ 2 - 3 Bell Labs mm/minute Lights of Tuscany 2019 April 15, 2019 AA 2018/19 Fisica dei Semiconduttori:Teoria e Applicazioni 13

Molecular Beam Epitaxy Lights of Tuscany 2019 April 15, 2019 14

Epitaxy can be by Molecular Beam, Chemical Vapour, Liquid Phase Epitaxy means in “ growth in an ordered way”: epitaxial growth is therefore a growth with an high control of the deposition, at the atomic level It requires same crystallographic lattice and “same” lattice constant to avoid extended defects which are detrimental for the device operation Lights of Tuscany 2019 April 15, 2019 15

But Nature can help: Self-assembled growth Lights of Tuscany 2019 April 15, 2019 16

Some results A superlattice Quantum dots Lights of Tuscany 2019 April 15, 2019 17

What’s the meaning? We can modify the electronic properties ( energy, band dispersion, band gap … ) at the nanoscale ( few Angstrom up to tens of nm) Quantum effects dominate and we can tailor the material properties the way we prefer Lights of Tuscany 2019 April 15, 2019 18

So very nice results but … .. Highly expensive techniques As alternative: Chemical synthesis Lights of Tuscany 2019 April 15, 2019 19

Advantages of chemical synthesis  Low cost synthesis and processing  High tunability of band-gap  Easier integration in photonic structures Lights of Tuscany 2019 April 15, 2019 20

The promising class of materials: Perovskites Same class of material as CaTiO 3 A:Ca 2+ B:Ti 4+ X:O - Depending on A, they can be hybrid with organic cation or fully inorganic A: CH 3 NH 3 , Cs B:Pb,Sn X:Cl,Br,I Lights of Tuscany 2019 April 15, 2019 21

Application Fields Energy Harvesting Lights of Tuscany 2019 April 15, 2019 22

LEDs and Lasers Lights of Tuscany 2019 April 15, 2019 23 https://doi.org/10.1016/j.mattod.2017.03.021

Different material nanostructuring means different behavior From bulk to plates, wires, dots Lights of Tuscany 2019 April 15, 2019 24

Here in Florence • Synthesis • Morphological and structural characterization (XRD, SEM) • High resolution optical spectroscopy in space and time Lights of Tuscany 2019 April 15, 2019 25

Synthesis of nanostructures/thin films Lights of Tuscany 2019 April 15, 2019 26

Then precursors are spin-coated on a substrate …… ..solvent evaporation produces perovskite Lights of Tuscany 2019 April 15, 2019 27

For crystals Lights of Tuscany 2019 April 15, 2019 28

At the optical microscope: spin-coated samples Lights of Tuscany 2019 April 15, 2019 29

High resolution Imaging SEM AFM Lights of Tuscany 2019 April 15, 2019 30

High resolution SEM Lights of Tuscany 2019 April 15, 2019 31

Bulk crystals Using an optical microscope Using a SEM Lights of Tuscany 2019 April 15, 2019 32

What spectroscopy on such samples? ➢ We need to determine the bandgap: standard cw spectroscopy ( transmission/ absorption/ photoluminescence) ➢ We need to understand the non-radiative recombination paths: temperature-dependent measurements ➢ We need to understand the carrier interactions and the recombination kinetics: time-resolved experiments, typically at the picosecond time-scale Lights of Tuscany 2019 April 15, 2019 33

Moreover: optical spectroscopy at the macro or micro scale? Macro : we probe macroscopic portions of the sample ( ≥ 100 µm 2 ) Micro : Spatial resolution at the diffraction limit (≈µm in the vis) Near Field detection: spatial resolution limited by the fiber tip ( ≈ 100 nm) Different informations can be extracted! Lights of Tuscany 2019 April 15, 2019 34

Here in the labs Lights of Tuscany 2019 April 15, 2019 35

MicroPL Lights of Tuscany 2019 April 15, 2019 36

SNOM Lights of Tuscany 2019 April 15, 2019 37

At the end Lights of Tuscany 2019 April 15, 2019 38

Some examples Superlinear emission: ASE or lasing? 1.60 1.61 1.62 1.63 1.64 1.61 1.62 1.63 1.64 1.65 P=140 I 0 P = 5.4 I 0 P=85 I 0 P = 3.4 I 0 PL Intensity (a.u.) PL Intensity (a.u.) P=60 I 0 P = 2.3 I 0 P=5 I 0 P = 1.6 I 0 P= I 0 P = I 0 1.60 1.61 1.62 1.63 1.64 1.61 1.62 1.63 1.64 1.65 Energy (eV) Energy (eV) Lights of Tuscany 2019 April 15, 2019 39

Normalized PL Intensity Decay kinetics is similar Experimental response 1 SE SPE but polarization  <30ps  400ps properties are different -1 10  400ps y -2 10 0.0 0.2 0.4 0.6 0.8 1.0 Time (ns) A 90 B 120 60 SPE 150 30 x 90 120 60 SPE 180 0 150 30 210 330 A 180 0 PL Intensity (a.u.) B 240 300 270 210 330 1.634 1.636 1.638 1.640 240 300 Energy (eV) 270 Lights of Tuscany 2019 April 15, 2019 40 1.62 1.63 1.64 1.65 1.66 Energy (eV)

Surface states role Integrated PL Intensity 293K 250K 230K Normalized PL Intensity A PL Intensity (a.u.) 1.0 210K 180K C 0.8 T = 10 K 150K A 130K 0.6 110K B 90K 0.4 D 70K 50K 0.00 0.02 0.04 0.06 0.08 0.10 0.2 -1 ) 1/Temperature (K 30K A C B D 11K 0.0 2.3 2.4 2.5 2.6 2.2 2.3 2.4 2.5 2.6 2.7 Energy (eV) Energy (eV) Recombination dynamics slows 293 K 1 Norm. PL Intensity 210 K down increasing the temperature! P 11 K -1 10 11 K The smaller the nanocrystals, B A A P the larger the effect 1 293K Norm. PL Intensity 293K P -1 10 11K -2 10 D C Lights of Tuscany 2019 April 15, 2019 41 P 11K 1ns Time 1ns Time

The increase of lifetime with T is counterintuitive ! Usually non radiative channels are more effective as T increases But … . Large crystals or high temperature (smaller barrier, less surface defects) Small crystals or low temperature (larger barrier, more surface defects) CB Surface state VB Surface states act as a reservoir an can release population as T increases Lights of Tuscany 2019 April 15, 2019 42

Thermally activated transfer − 𝑭 𝑪 𝒍 𝑪 𝑼 𝑸 𝑼 = 𝑩𝑼 + 𝑪𝒇 A P 0 = P(T=10K) Band A (K -1 ) B (10 -3 K -2 ) E B (meV) B A 0.18 33 C 10 c D B 0.48 39 P/P 0 ≈ 10 -3 C 2.6 67 D 25 130 1 0 50 100 150 200 250 300 Temperature (K) Lights of Tuscany 2019 April 15, 2019 43

Non-linearities with two beams Non resonant bias CW @ 405 nanop2018-Rome October 1-3 nm + ps pulse @ 370 nm I pb - (I p +I b ) vs time 44 Lights of Tuscany 2019 April 15, 2019 44

Negative signal: bleaching related to localized/bound states Positive signal: Superlinearity from excitons formation E 0 = zero crossing energy 0 0 I pb -(I p +I b ) I pb -(I p +I b ) 56.8ps 0ps 85.2ps 113.6ps 14.2ps 142ps 28.4ps 184.6ps 42.6ps 227.2ps 269.8ps 2.26 2.28 2.3 2.32 2.34 2.36 2.26 2.28 2.3 2.32 2.34 2.36 Energy(eV) Energy(eV) Lights of Tuscany 2019 April 15, 2019 45