First Principles Calculations and NMR Spectroscopy of ... · First Principles Calculations and NMR Spectroscopy of Electrode ... • Performance (high energy density) ... computations

First Principles Calculations and NMR Spectroscopy of Electrode

Materials

Professor Clare Grey University of Cambridge

6/17/2014 Project ID

es055

This presentation does not contain any proprietary, confidential, or otherwise restricted information

2

• Project start date: 1/1/13 • Project end date: 12/31/16 • Percent complete: 33%

• Life (capacity fade) • Performance (high energy density) • Rate

• Total project funding: $1.1M • Funding received in FY13:

$265,913 • Funding for FY14: $272,197

Timeline

Budget

Barriers

• Brett Lucht • Jordi Cabana • Kristin Persson • Guoying Chen • Stan Whittingham

BATT collaborators

Overview

Relevance: Objectives - 2013/14 •Identify major solid electrolyte interphase (SEI) components, and their spatial proximity, and how this changes with cycling (capacity fade) •Complete structural/mechanistic studies of Si (performance) •Investigate local structural changes of high voltage/high capacity electrodes on cycling (performance/capacity fade) 2015/16 •Contrast SEI formation on Si vs. graphite and high voltage cathodes (capacity fade) •Correlate Li+ diffusivity in particles and composite electrodes with rate

3

• Optimizing Si performance – Structures formed on cycling

– Reducing overpotential – Building a better SEI

• SEI studies

– NMR studies of local structure as a function of cycling

• Improving rate performance

(electrode tortuosity studies)

• High voltage spinels

Approach/Strategy

– Development of new platform for in situ studies.

– Li and NMR studies of structure – NMR and electrochemical studies

of Si coatings/surface treatments

– 13C NMR studies of 13C enriched electrolytes to study SEI organic components; 19F and 31P studies of inorganics

– Develop pulse field gradient (PFG) approach to study electrode tortuosity (LiCoO2 current model compound)

– Development of in situ methods to study phase transformations

– In-situ and ex-situ NMR studies of Li+ transport and structural changes

4

Milestones • Identify major components (LiF, phosphates, carbonates and organics) in Si

SEI by NMR methods. (Dec-13). Complete • Correlate presence of SEI components with cycle number and depth of

discharge of Si. Complete preliminary TOF-SIMS measurements to establish viability of approach. (Mar-14) Ongoing. Difficulties encountered with sample reproducibility and Si cracking (TOF-SIMS)

• Identify SEI components in the presence of FEC and VC in Si and determine how they differ from those present in the absence of additives. (Jun-14) Ongoing

• Go/No-Go: Stop Li+ PFG diffusivity measurements of electrodes. Criteria: If experiments do not yield correlation with electrochemical performance. (Sep-14) PFG studies initiated of LiCoO2.

Approach/Strategy (cont.)

Again small clusters only seen in 2nd

process Salager et al. in prep.

Optimizing Si performance: I • Used 29Si NMR and Li NMR to study Si cluster formation and bond breakage as

function of state of charge - Showed that mechanisms for lithiation are different in smaller Si particles

7Li NMR 29Si NMR

200 nm Si 15 nm

• Investigated (Li) defects in Si by DFT-based computations

A.J. Morris et al., Phys Rev B (2012)

Technical Accomplishments and Progress

CF+50nm-Au+SiNWs

CF

SiNWs

Cross sectional SEM images

K. Ogata C. Kerr S. Hoffman

Optimizing Si performance II: Studying nanoparticles by in-situ NMR

Inspired by: C. K. Chan. … R. A. Huggins, Y. Cui, Nature Nanotechnology, 3, 31 (2008)

In-situ NMR of Si Nanowires:

Ideal Model Systems for Studying Mechanisms – allow GITT and PITT

experiments to be followed in situ

P-doped Si

Li15+xSi4 Clusters

Galvanostatic mode

PITT

Detect Si=Si defects in Li15Si4 => Set overpotential voltage on charge

Ogata et al., Nat. Commun. (2014)

SEI Studies: Si

SEI Layer ~ 10 - 100 nm thick

Silicon

1H and 13COrganics

OOOO

O

O

Li

Li

O O

O

Li

LiF19F & 7Li

Li2CO3

7Li & 13C

31P & 19FPF3O

13C

O

OO

13CO

O

O13CH313CH3

Structure of the interface?

System: C + Si, 1 : 1

•  SEI as a function of voltage? •  SEI as a function of cycle

number?

5

10

15

-300-200-10005

10

15

19F shift / ppm[ppm] 10 5 0 - 5

*

-71.8 ppm LiPF6

-203 ppm LiF

-0.3 ppm

4.6 ppm EC

1.2 ppm CH*3CH2OCO2Li

3.8 ppm LEDC

7Li

1H

Correlate 1H-7Li nuclei in close proximity using 2D experiment

Decomposition product of EC

(-0.43, 3.8)LEDC

7Li

1H

LiO OO OLi

O

O

60 kHz, 700 MHz

-20-10010200100

7Li Shift / ppm

-10-551015 1H Shift / ppm

19F* ****

Decompositionproducts of EC

0

8

40

2

Decomposition product a result of reactions with H2O

1

0 50 100 1500

1

2

3

Volta

ge (V

)

0 50 100 150−200

−100

0

100

Time (hours)

Curre

nt ( μ

A)

0 500 1000 1500 2000 2500

0.5

1

1.5

2

2.5

3

Capacity (mAh/g)

Volta

ge (V

)

C/75

HoldC/75 Irreversible Capacity ~ 920 mAh/gRest

Discharge Capacity ~ 2880 mAh/g

Hold

**

**** **

13C shift / ppm10 kHz MAS

13C shift / ppm10 kHz MAS

250 200 150 100 50 0 250 200 150 100 50 0

13C EC Enriched (RED) 13C DMC Enriched (BLUE)

* *

* *1H - 13C

7Li - 13C

13C

CP ct = 5 ms

CP ct = 500 µs

HEcho

200 100

10

5

0

0 4.5 ppm 1H

4.5 ppm

250 200 150 100 50 0

(EC)67.3 (4.5 1H)

(EC) + (SEI)159 / 161.5

Projection1D cp

13C shift / ppm

13 C

O

OO

ct = 500 µs

Proje

ction

Projection (+)

1D cp

180 160

20

0

- 20

SpinningSidebands(2 Pairs)

3.5 ppm 1H

4.5 ppm 1H

4.5 ppm

3.5 ppm

180 170 160

10

5

0

180 170 160

180 170 160

(EC)159.3 (4.5 1H)(SEI)

LEDC (CH2OCO2Li)2 161.5 (~3.5 1H)

*

13C shift / ppmct = 500 µs

1 H s

hift

/ ppm

(larg

er s

cale

)

13C shift / ppm 13C shift / ppm

1 H s

hift

/ ppm

1 H s

hift

/ ppm

16

1

6

16

****

(SEI)LEDC (CH2OCO2Li)2

161.5 (~3.5 1H)

13C Solid State NMR – SEI Composition

13C Enriched EC DMC

Use correlation NMR experiments (cross-polarization (CP)) to establish spatial proximity between atoms

13C

O

OO

13CO

O

O13CH313CH3

EC: 2D experiments confirm assignments of LEDC (similar experiments performed for DMC)

Collaboration with B. Lucht: Structures of reduced VC and FEC

VC FEC

185 = RCO2 128 = C=C 69 – 71 = OCH2

Very similar polymers .. Now we can compare spectra with those obtained when using VC and FEC as additives

Reducing SEI Formation: Si surface coatings improve capacity retention

APTMS                                                                          TEOOS

                 PMVEMA                    (3-­Glycidyloxypropyl)trimethoxysilane

0 5000 10000 15000 20000 25000 30000

0.0

0.5

1.0

1.5

2.0

2.5

3.0 Theoretical capacity: 3750 mAhg-1

Non-coated SNP (cell #1)

APTES-coated SNP (cell #5)

APTES-coated SNP (cell #6)Rate: C/100

Capacity (mAh/g)

Pot

entia

l vs

Li/L

i+ (V

)

5 10 15 20 250

1000

2000

3000

4000

Cycle number

Cap

acity

(mA

h/g)

Comparison: Capacity RetentionDischarge CapacityCharge Capacity

5 10 15 20 25

0.8

1

Cou

lom

b Ef

ficie

ncy

Cycle number

Comparison: Coulombic Efficiency

Discharge CapacityCharge Capacity

APTES-coated

APTES-coated

non-coated

non-coated

-  Echem performance improvement: APTMS> Si(CH2)2CN>TEOOS> PMVEMA> >Si(CH2)7CH3 > non-coated

Si(OCH2CH3 )2CH2CH2CN

NMR studies in progress to study nature of grafting

Development of In situ NMR method for Paramagnetic Cathode Materials:

Li1.08Mn1.92O4 Electrode Orientation = 54.7 Degrees

1st cycle

7Li

Relaxation (T2) effects – and thus spectral intensity - are strongly affected by Li motion

0 50 100 150 200 250 3002.83.03.23.43.63.84.04.24.44.6

Capacity(mAh/g)

Volta

ge(V

)

0

20

40

60

80

100

120(118mAh/g)

In-situ T2 (µs)

0 50 100 150 200 2502.83.03.23.43.63.84.04.24.44.6

Capacity(mAh/g)

Volta

ge(V

)

0.00.10.20.30.40.50.60.70.8

Intensity

0 50 100 150 200 2502.83.03.23.43.63.84.04.24.44.6

Capacity(mAh/g)

Volta

ge(V

)

0.00.20.40.60.81.01.21.41.61.82.0 Corrected Intensity via T

2

tau tau - detect

π/2 π

Spin-spin relaxation (T2) measurements:

Use In-situ T2 Measurements to Study Li Dynamics as a Function of Temperature

• Rapid Li+ and electronic motion in partially charged samples • Clear evidence for solid solution from 1 Li to 0.5 Li • Ordering tendency @ 0.5 Li -

• Method can be used to study dynamics during (de)lithiation

0.0 0.5 1.0 1.5 2.0 2.5 3.0

20406080

100120140

700750

in s

itu T

2 (µs

) RT 40T 60T 70T

(ppm

)

°C

°C °C

Solid solution

2 phase

Cation ordering

21 mV

High Voltage Spinels (Li(Ni0.5Mn1.5O4) - Ordering affects the electrochemistry @ the 4.8 V process

9 Mn4+ 3 Ni2+

8 Mn4+ 4 Ni2+ 10 Mn4+ 2 Ni2+

6 Li NMR

946 ppm

P4332: Ni2+ and Mn4+ ordering

Fd-3m: Ni2+ and Mn4+ randomly distributed

DISORDERED

ORDERED

50 mV

50 mV

With J Cabana (LBNL/UIC) C Kim (LBNL)

In-situ 7Li NMR – Mechanisms of (de)lithiation

Disordered 900 C spinel

Increased Li mobility Solid solution

Intensity T2 corrected

2 - phase “T2” Relaxation studies show solid solution followed by 2 phase

Cation ordering @ x = 0.5

Intensity vs. cycling

In-situ 7Li NMR – Lithium & electrolyte region: Follow Li

dendrite formation and electrolyte decomposition in

situ

Electrolyte decomposition

Dendrite formation

Ordered

18

Summary • Silicon – structural work essentially complete. • Used 29Si NMR to study cluster formation on lithiation • Developed in-situ NMR method for studying nanowires • Identified source over large “overpotential” on charging fully lithiated

Li15Si4 phase • Si SEI – NMR methodology has been developed. • Clear NMR signatures of many SEI components identified • Si Coatings – Grafting of small molecules helps improve

capacity retention. • Tortuosity – PFG method demonstrated • High Voltage spinels – Demonstrated novel NMR methodology

to study paramagnetic materials • Approach can be used to study Li ion dynamics, cation ordering,

solid solution vs. 2-phase behaviour • Used to study nature of electrode reactions for the high voltage

spinel Li(Ni0.5Mn1.5)O4, electrolyte decomposition and Li dendrite formation

• Ex-situ NMR highly sensitive to cation (Ni/Mn) order

Future Work •Silicon structure Future work will focus on (i) Reverse Monte Carlo simulations of amorphous phases (with pair distribution function analysis) to quantify Si amorphous structures and (ii) the effect of P doping on rate, overpotential (energy efficiency) and lithiation mechanisms. •Si SEI •Further 2D NMR experiments of fully enriched EC will be performed for more detailed structure solution of decomposition products (including reduced FEC/VC) •Future work will focus on cycling studies, and structure f(voltage) and VC/FEC •C-Si experiments will be initiated to investigate SEI/Si interface •Si Coatings •Detailed NMR studies will be performed to understand nature of grafting and how that changes with cycling. Use results/understanding to investigate different molecules

•Tortuosity •PFG method will be applied to cycled/discharged samples to explore how SEI growth affects tortuosity (and rate) •Investigation of Si/conducting polymer samples prepared by Gao Liu (LBNL) – particularly to investigate conductivity in pores and binder – are planned •High Voltage spinels •Separate equilibrium from non-equilibrium processes in NMR •Initiate work on SEI of HV spinels. Combine with graphite carbon for full cell studies. •Extend method to other high voltage systems.

20

Collaboration and Coordination with Other Institutions

• Dupont CR&D (E. McCord and W. Holstein) – Investigation of electrolyte stability and SEI formation

• Brett Lucht (Rhode Island) – Investigation of SEI

• Jordi Cabana (UI Chicago) – synthesis, XRD of high voltage spinels

• Stanley Whittingham (Binghamton) – magnetism of spinels

• Stephan Hoffman (Cambridge) – Si nanostructures

• Andrew Morris (Cambridge) – DFT structures of LixSi

• Guoying Chen (LBNL) – synthesis of high voltage spinesl

• N/A – not reviewed last year

Responses to Previous Year Reviewers’ Comments

21

Technical Back-Up Slides

22

Publications and Presentations • “Structure of aluminum fluoride coated Li[Li1/9Ni1/3Mn5/9]O2 cathodes for secondary lithium-ion batteries”, K.J.

Rosina, M. Jiang, D. Zeng, E. Salager, A.S. Best, C.P. Grey, J. Mat. Chem., 22, 20602-20610, (2012). • “Scanning x-ray fluorescence imaging study of lithium insertion into copper based oxysulfides for Li-ion batteries”,

R. Robert, D. Zeng, A. Lanzirotti, P. Adamson, S.J. Clarke, and C.P. Grey, Chem. Mat., 24, 2684-2691, (2012). • “Sidorenkite (Na3MnPO4CO3): A new intercalation cathode material for Na-ion batteries”, H. Chen, Q. Hao, O.

Zivkovic, G. Hautier, L.-S. Du, Y. T, Y.-Y. Hu, X. Ma, C.P. Grey, and G. Ceder, Chem. Mat., 25, 2777-2786 (2013). • “Study of the transition metal ordering in layered NaxNix/2Mn1–x/2O2 (2/3 ≤ x ≤ 1) and consequences of Na/Li

Exchange”, J. Cabana, N.A. Chernova, J. Xiao, M. Roppolo, K.A. Aldi, M. Stanley Whittingham, and C. P. Grey, Inorg. Chem., 52, 8540-8550 (2013).

• “Paramagnetic electrodes and bulk magnetic susceptibility effects in the in situ NMR studies of batteries: Application to Li1.08Mn1.92O4 spinels”, L. Zhou, M. Leskes, A.J. Ilott, N.M. Trease, and C.P. Grey, J. Mag. Res., 234, 44-57 (2013).

• “Lithiation of silicon via lithium Zintl-defect complexes from first principles”, A.J. Morris, R.J. Needs, E. Salager, C.P. Grey, C.J. Pickard, Phys. Rev. B, 87, 174108-1 – 174108-4, (2013).

• “Revealing the kinetics of key LixSi phase transformations in nano-structured Si based Li-ion batteries via in situ NMR”, K. Ogata, E. Salager, C. J. Kerr, A. J. Morris, A. Fraser, C. Ducati, S. Hofmann, C. P. Grey, Nature Communications, 5:3217 | DOI: 10.1038/ncomms4217 (2014). “Following Function in Real Time: New NMR and MRI Methods for Studying Structure and Dynamics in Batteries and

Supercapacitors” Talk (or related talk) given at the following meetings in 2013: iNano Opening, Plenary Talk, Billund, January; Chemistry Department, ENS Lyon, January RS, Theo Murphy International Scientific Meeting, 28 & 29 January; Chemistry Department, University of Wisconsin, February; IBA2013, Research Award Address, Barcelona, March; ACS, New Orleans, March 54th ENC, Laukien Award Address, Asilomar, CA, April; ISMAR 2013, Rio de Janeiro, Brazil, May University of Basel, Basel, May; SSI-19 Conference, Kyoto, June RSC MC11, Warwick University, July; ICMRM, Cambridge, August Electrochem2013, Southampton University, Southampton, September; RSC ISACS12, Cambridge, September 8th Alpine Meeting on Solid State NMR, Chamonix, September; LG Chemicals, Daejeon, Korea, October Korea Basic Science Institute, Daegu, Korea, October; Department of Energy Science, Sungkyunkwan University, Korea, October Samsung Advanced Institute of Technology (SAIT), Seoul, Korea, October Institute of Energy & Climate Research, Forschungszentrum Juelich, Germany, November MRS, December (2013).

• Susceptibility effects and dipolar broadening are significant for paramagnetic samples

2000 1500 1000 500 0 -500 -1000 -1500

ppm

Rotation angle α =90o(Horizontal) Rotation Angle α =54.7o(Magic Angle) Rotation Angle α =0o(Vertical)

-159

527

837 LiMn2O4 vertical

horizontal

Challenges for In-situ NMR: Paramagnetic Cathode Materials

Li

Mn θ LiMn2O4

B0

L. Zhou, et al. JMR 2013

-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

-200

0

200

400

600

800

1000

Spi

nel P

eak

Pos

ition

(ppm

)

Rotation Angles(Radians)

x-rotation y-rotation z-rotation

δiso= 520 ppm

Li1.1Mn1.9O4 electrode: rotation plot

54.7° - the magic angle..

Orientation of battery film at the magic angle minimizes susceptibility effects

In-situ 7Li NMR – Ordered vs Disordered

Ordered Disordered

Increased Li mobility Solid solution Limited range of

Solid solution

Intensity T2 corrected

Intensity T2 corrected

hysteresis reversible

More 2-phase like

(331)

OO700

(331)

Charge

Discharge

A900

Chunjoong Kim, Jordi Cabana In-situ XRD: Ordered vs. Disordered

y

z

x

•  Samples are saturated with electrolyte solvent (DMC) and diffusion is measured along the z-axis and length of sample

Silicon Composite Electrode

Substrate

Surface of pristine electrode prior to saturation with DMC

Quantify tortuosity using Do=τDeff

PFG NMR Experiments Si and LiCoO2

FinalInitial

<(x’-x)2> = 2Dt

δ/2

∆ Diffusion Time

gi

δ1 δ2

RF

g(t)

S(g) = S(0)exp[-D(γδgi)2(∆-δ/3)]

= S(0)exp[-bD]Encode Decode

0.1

1

Inte

grat

ed S

igna

l Int

ensi

ty (a

.u.) DMC Saturated Sample (b)

Deff = 1.6x10-9 m2/s (*)

Free DMC (**) Do = 2.0x10-9 m2/s

Normalized Gradient Strength

(��G)2����/3) (108 s/m2) 0 5 10 15 20•  Unrestricted diffusion distance (2DΔ)1/2 >>

average pore size in diffusion time Δ = 20 ms

•  Apparent tortuosity given by τ = Do/Deff = 1.3 (room temperature)