Nanobatteries Part 2 J Seminario, D Galvez, V Ponce, L Selis, C - - PowerPoint PPT Presentation

nanobatteries part 2
SMART_READER_LITE
LIVE PREVIEW

Nanobatteries Part 2 J Seminario, D Galvez, V Ponce, L Selis, C - - PowerPoint PPT Presentation

Atomistic Classical and Quantum Simulations of Nanobatteries Part 2 J Seminario, D Galvez, V Ponce, L Selis, C Roman, F Gallo, M Gamero Dept. of Chemical Engineering Dept. of Electrical & Computer Engineering Dept. of Materials Science


slide-1
SLIDE 1

Cartagena, Colombia ICTP School 6/1/2019

Atomistic Classical and Quantum Simulations of Nanobatteries – Part 2

J Seminario, D Galvez, V Ponce, L Selis, C Roman, F Gallo, M Gamero

  • Dept. of Chemical Engineering
  • Dept. of Electrical & Computer Engineering
  • Dept. of Materials Science & Engineering

Texas A&M University

slide-2
SLIDE 2

Pristine boxes: Li-metal & Li9N2Cl3 solid- state electrolyte Interfacial box Li9N2Cl3 (002) and (001) versus (001) Li-metal

Li9N2Cl3 (002) and (001) planes

atoms on top highlighted. Lim (red) Li (purple) N (brown) Cl (green)

slide-3
SLIDE 3
  • 415.58
  • 415.57
  • 415.56
  • 415.55

2 4 6 8 10 12 14 eV/atom t/ps Experimental (space group 225) cubic crystal structure of Li9N2Cl3. Crystal Structure after 15 ps of equilibration. Li (purple), N (brown) Cl (light green) Energy per atom curve along the equilibration.

slide-4
SLIDE 4
  • 122.36
  • 122.35
  • 122.34
  • 122.33

2 4 6 8 10 12 14 E/eV/atom t/ps

Experimental BCC (space group 225) crystal structure of Li-metal Energy per atom (eV) curve along the equilibration Li(s) (red) after 15 ps of AIMD equilibration

slide-5
SLIDE 5
  • 0.06
  • 0.04
  • 0.02

0.02 0.04 0.06 0.08 10 20 30 40 50 60 q/e atom labels

Li-metal trajectories & final Bader charges

slide-6
SLIDE 6

Method a b c volume Density Experimental49 10.814 10.814 10.814 1264 1.65 AIMD 10.205 11.406 10.296 1198 1.74 Difference [%] 5.60 5.47 4.79 5.22 5.45

Method a b c volume Density Experimental 10.53 10.53 10.53 1167 0.53 AIMD 9.96 11.38 10.91 1237 0.50 Difference [%] 5.41 8.07 3.61 5.99 5.66 Experimental and AIMD lattice parameters (Å), and density (g/cm3) for the Li metal crystal structure. Experimental ( = 0.5 fs, NPT ensemble) for Li9N2Cl3 AIMD shows parameters averages

slide-7
SLIDE 7
  • 2.50
  • 2.00
  • 1.50
  • 1.00
  • 0.50

0.00 0.50 1.00 20 40 60 80 100 q/e Atom labels

Li9N2Cl3 diffusion trajectories during 15 ps of AIMD with a  = 0.5 fs under NPT ensemble. (d) Atomic Bader charges after equilibration.

N (brown) Cl (green) Li (purple).

slide-8
SLIDE 8

Input potential (red line) Ep profile ℰ = 0.5 V/Å @ 0 ps (black line) Ep with ℰ = 0 @ 0 ps (yellow line) Ep profile @ 20 ps ℰ = 0.5 V/Å

slide-9
SLIDE 9

AIMD Atomic density distributions at 0, 10, 20

ps

slide-10
SLIDE 10
  • 3
  • 2
  • 1

1 5 10 15 20 25

q/e z/Å

Li(m) (red) Li (violet) Cl (green) N (brown)

Charge evolution

q vs z-distance at 20 ps

slide-11
SLIDE 11

𝑒2 𝑢 = 1 𝑁𝑂 ෍

𝜐=1 𝑁

𝑜=1 𝑂

𝑦𝑜 𝜐 + 𝑢 − 𝑦𝑜 𝜐

2

d 2(t) ≡ MSD

𝐸(𝑢) = 1 6 ൯ 𝑒2(𝑢 𝑢 𝐸 𝑢 = 1 6𝑁𝑂𝑢 ෍

𝜐=1 𝑁

𝑜=1 𝑂

𝑦𝑜 𝜐 + 𝑢 − 𝑦𝑜 𝜐

2

slide-12
SLIDE 12

2 4 6 8 5 10 15 20 MSD(t) /Å2 t/ps Cl Li(se) N Li(s) 5 10 0.001 0.01 0.1 1 10 MSD (Å2) t/ps Cl Li(se) N Li(s) 0.2 0.4 0.001 0.01 0.1 1 10 MSD (Å2) t/ps

Li(se) N Cl Li(s)

10 20 30 5 10 15 20 D(t)/10-7 cm2/s t/ps Cl Li(se) N Li(s)

N (brown), Cl (green), Li(se) (purple) and Li(s) (red)

MSD vs t, linear-scale MSD vs t, log-scale exploded MSD(t) t-dependent diffusion D(t)

Li-metal/SSE/Li-metal cell

slide-13
SLIDE 13

4.4 4.6 4.8 5 2 4 6 8 10 12 14 16 18 20 t/ps

Cl-Li cn

6.6 6.8 7 7.2 2 4 6 8 10 12 14 16 18 20 t/ps

N-Li cn

slide-14
SLIDE 14

5 10

Cl (7)

5 10

Cl(17)

5 10

Cl(2)

5 10 5 10 t/ps

Cl(5)

Coordination number of Cl and N at the interphase during AIMD simulation

5 10

N(82)

5 10

N(84)

5 10

N(89)

5 10 5 10 t/ps

N(85)

slide-15
SLIDE 15

Li9N2Cl3/Li interface

atomic density profiles along z

slide-16
SLIDE 16

Li Cl Lim N Li(i)

AIMD DFT PBE, τ = 0.5 fs, tAIMD = 50 ps, Ecutoff = 40 Ry (λcutoff = 0.5 Å), ℰ = 0.75 V/Å

Nano-battery model: Li-metal/Li9N2Cl3/NMC

Ab-Initio MD model to study an take parameters of interfacial behavior Anode/Electrolyte

1 2

Nanobattery MD model is developed using parameters from Ab-Initio MD simulation

1 14 19 24 e z (Å) 5 10 15 30 60 90

# atoms index

Cl N 5 10 15 22 24 26

# atoms index

Cl N Li(s) 5 15 25 10 20

# atoms index

Li(s)

Li-ions Li(i) Li(s) Li(i) Li-ions Li(s)

10.14 × 11.41 × 22.41 Å

ML classifier is trained with data obtained from AIMD to develop a full Nano-battery model considering: Bader charge analysis Atoms neighborhood in an sphere with r = 5 Å / 3 Li types are recognized

21.62 × 21.62 × 106 Å LAMMPS: MD, NVT, τ = 0.01 fs, tMD = 1.2 ns, ℰ = 0.75 V/Å, ML algorithm every 1 fs

ℰ = 0.75 V/Å

0.2 0.4 0.6 0.8 1 20 25 30 q z / Å Charge evolution of a Li traveling from electrolyte through the interphase to the anode. Interphase Electrolyte Anode Li-metal Anode – Reax potential Li(i) – ML and LJ potential Electrolyte – MEAM and LJ potential Other interactions – LJ potential

Li-metal/Li9N2Cl3/NMC

Electrolyte: Li365N78Cl114 Anode: Li145 Cathode: Ni53Mn8C8O136Li62

slide-17
SLIDE 17

Nano-battery model: Li-metal/Li9N2Cl3/NMC

10 20 30 40 50 60 0 ps 200 ps 400 ps 600 ps % Other FCC HCP BCC ICO SC 0 ns 0.5 ns 1 ns 1.5 ns

Polyhedral template matching algorithm is used to identify crystallographic structures. Li-metal anode shows two clear phases during lithiation: BCC/Amorphous 0 ps 500 ps 1 ns 1.5 ns

BCC Li Amorphous Amorphous Amorphous Amorphous BCC Li BCC Li BCC Li

Lithiation due to an ℰ = 0.75 V/Å

5 10 15 1 2 3 4 5 Li-Li (0 ns) Li-Li (0.5 ns) Li-Li (1 ns) Li-Li (1.5 ns)

slide-18
SLIDE 18

Nano-battery model: Li-metal/Li9N2Cl3/NMC

Lithiation due to an ℰ = 0.75 V/Å

120 140 160 180 200 220 240 0.5 1 1.5 # of Li t/ns Total Li(anode) Li(s) Li(i) 360 380 400 420 0.5 1 1.5

# of Li t/ns

2000 2200 2400 2600 2800 3000 3200 0.2 0.4 0.6 0.8 1 1.2 1.4 Å3 t/ns

slide-19
SLIDE 19

5 10 15 20 25 1 2 3 4 5 Displacement / Å time / ps

Nano-battery model: Li-metal/Li9N2Cl3/NMC

Trajectory of a Li-ion through the electrolyte (5 ps)

Distance / Å Distance / Å

Cl47 Cl67 Cl96Cl111Cl61 Cl105 Cl66 Cl60 N488 N522N507N487 Cl31N521 Cl110 Cl65 N486 N516 N478

Cl47 Cl67 Cl96 Cl111 Cl65 Cl105 Cl66 Cl60 Cl110 N488 N522 N507 N487 Cl31 N521 N486 N516 N478 Cl61

Cl47 Cl67 Cl96 Cl111 Cl61 Cl66 Cl105 Cl110 Cl60 Cl65 Cl31

Distance / Å

N488 N522 N507 N487 N521 N486 N516 N478

time / ps

NN Li-ion

slide-20
SLIDE 20

Nano-battery model: Li-metal/Li9N2Cl3/NMC

Trajectory of a Li-ion through the electrolyte (5 ps)

slide-21
SLIDE 21

Conclusions ▪ Li closest to interface shows major diffusion (B) ▪ new bonds with N and Cl of the SSE (B) ▪ SSE shows stability at the interface with Li-metal (B) ▪ Initial reactions on the SSE(002) plane (of Li9N2Cl3) (NE) ▪ Reactions don't modify crystallographic structure of electrolyte (B) ▪ Li(s) density increases at both interfaces, (001) and (002) (NE) ▪ Major density change at (002) interface due to negative N and Cl (NE) ▪ N and Cl have 3-4 Li NNs at the beginning of the simulation (NE) ▪ To complete 7-8 Li NNs of Li9N2Cl3, N and Cl take Li(s) from anode ▪ Li(s) diffusion anode to electrolyte at (001) lower than at (002) (NE) ▪ Because Cl and N already have at least 7 Li at the Li9N2Cl3(001) no need to take Li(s) from the anode. (NE)

slide-22
SLIDE 22

Conclusions ▪ Li-BCC structure changes into HCP (E) ▪ But BCC and HCP energies difference ≤ 10 meV ▪ Thus, @ 300 K (26 meV), transitions are highly possible (E) ▪ but not necessarily due to the electric field. ▪ (001) better than (002) of SSE to reduce deformation of anode (NE) ▪ No new phase formed due to interfacial interactions (B) ▪ no expected, if we extend the simulation times based on MSD trends and charge stability. (B) ▪ Li9N2Cl3 remains very stable in contact with Li-metal (B) ▪ Theory correctly model metallic-insulator interface (B)

slide-23
SLIDE 23