Application Note

Imaging the 4D Microstructure Evolutionof a Commercial 18650 Li-ion BatteryZEISS Xradia 520 Versa

Application Note

2

Imaging the 4D Microstructure Evolutionof a Commercial 18650 Li-ion BatteryZEISS Xradia 520 Versa

Author: Jeff Gelb, Senior Applications Engineer ZEISS Microscopy, CA, USA

Date: May 2017

Lithium-ion batteries are of increasing importance in consumer technologies. In parallel with understanding the

performance characteristics throughout a battery’s life, it is equally important to understand when, why, and

how the batteries ultimately fail. Recent research has pointed toward microscopy as a beneficial tool for study-

ing battery microstructures to understand what – if any – aspects of the microstructure lead to certain perfor-

mance or failure characteristics. In order to better develop an image-based microstructure investigative work-

flow, this study utilized non-destructive X-ray microscopy (XRM) to perform a 4D imaging experiment over a

battery’s lifetime. A fresh battery was studied intact, then subjected to aging until failure and imaged again.

The results of electrochemically aged batteries were compared to those of temporally aged batteries, which

showed a systematic microstructural change uniquely in response to the electrochemical aging routine.

Introduction

Energy storage solutions are ubiquitous in the modern

world. From portable electronic devices to electric vehicles

and stationary power supplies, the demand for energy stor-

age devices is expected to increase for many years to come.

Many manufacturers are now using lithium-ion batteries to

satisfy consumer requirements, with technological

advancements being made at a rapid pace [1]. In spite

of this, the performance and degradation mechanisms

of Li-ion batteries remain sparsely understood [2].

Creating a successful battery product relies on working

within a few key constraints. The batteries should be

powerful enough to serve the intended applications,

safe to operate, and reliable over their expected lifetimes.

These high-level challenges, however, may have causes root-

ed in microstructure, as the particle/pore interactions may

influence the capacity and discharge characteristics [3,4].

In addition, cell safety engineering has been shown to

begin with understanding the real-time mechanisms

of failure [5] and cell reliabilities are difficult, if not

impossible, to predict using conventional means [6].

Recent work has identified X-ray microscopy (XRM) as a

viable solution for visualizing the interior microstructures

of Li-ion batteries [7]. X-ray imaging has the unique

advantage of being entirely non-destructive, which allows

battery interiors to be revealed in 3D without sectioning or

opening the packaging of the specimen [8]. This technique

has shown a unique versatility for characterizing battery

specimens, including cathode [9], anode [10], and separator [11],

as well as offering the advantage of probing evolutionary

characteristics [12,13] and generating microstructural

models for analysis via computer simulation [14].

In this study, an ensemble of commercial Li-ion batteries

were studied intact using ZEISS Xradia 520 Versa. Using the

non-destructive power of X-rays, coupled with the high-

resolution and flexibility of Xradia 520 Versa, the same re-

gions of the same specimens were imaged before and after

aging to failure, revealing microstructural changes as a func-

tion of battery operation. The results were placed alongside

similar results on unaged batteries, to establish a control

group for comparison. This 4D study points to valuable infor-

mation provided by XRM, particularly with respect to how

the structure of a battery changes after operation. These

Application Note

3

5 mm

Current – collecting tab

Active materials+ current collectors

X-ray source

Specimen

X-ray detector

Metalhousing

Cathode

+ Currentcollector (Al)

– Currentcollector (Cu)

Anode + Separator

Densecentral ring

5 mm

3 mm 100 µm

results further reinforce the power of image-based charac-

terization for understanding degradation mechanisms within

a battery and open the door to future studies where specific

battery systems and/or battery aging characteristics may be

of interest.

Methods & Results

The specimens used for this investigation were commercially-

sourced 18650 Li-ion batteries from a reputable manufactur-

er. All investigations were performed without opening – or

otherwise disturbing – the packaging, i.e., non-destructively

imaged (Figure 1).

3D Imaging Protocols

Initial inspection with 3D X-ray microscopy was performed

with 22 µm voxel size in order to capture the entire

diameter of the battery (Figure 2). Subsequent stitching of

several fields of view enabled the battery to be captured in

its entirety using the XRM technique, with a total acquisition

time of 5 hours.

The initial survey enabled identification of bulk components,

such as the central pin, the current-collecting tab, and the

different active layers within the spiral winding. One key

architectural feature of the Xradia Versa family, is the ability

to non-destructively zoom into a specific region of interest

(ROI) by changing the objective lens to provide additional

optical magnification. Applying this “scout and zoom”

approach to one region with higher resolution, (Figure 3)

enabled the individual layers to be identified, including the

cathode, anode, and current collectors.

Figure 1 Battery specimen installed in ZEISS Xradia 520 Versa, for initial inspection.

Figure 2 The entire 18650 Li-ion battery was imaged using ZEISS Xradia 520 Versa with 22 µm voxel size and visualized in ORS Dragonfly Pro. This ren-dering shows a virtual cutaway of the full 3D dataset, revealing the interior structure. The results showed the overall assembly of the battery and the various layers within the jelly-roll.

Figure 3 ZEISS Xradia 520 Versa was used with 1.8 µm voxel size to optically enlarge a smaller section within the battery, allowing a more detailed inspection of the cell construction and identification of the different layers. These images represent virtual cross sections through the reconstructed 3D datasets.

Application Note

4

1 mm

3.0

2.5

2.0

1.5

1.0

0.5

0 10 20 30 40 50

Sample 1Sample 2Sample 3

Cycle Number

Cap

acit

y (A

h)

A region in the center of the package was selected for non-

destructive, high-resolution investigation with 1.8 µm voxel

size. The results (Figure 4) showed the different layers of the

structure and no obvious defects were noted.

4D X-Ray Microscopy

Once the 3D imaging protocols had been established for

the specimen’s geometry, the study was extended to a 4D

investigation of the microstructure evolution in response

to aging. Six batteries were studied in total, within the con-

text of the present investigation. Three of those batteries

were imaged as received, cycled until failure at 1.5 oC, and

then imaged again. The other three batteries were imaged

as received, set next to the cycling apparatus (but not cy-

cled), and then imaged again in sequence with the other

three batteries. Thus, the study was divided into two groups:

an experimental group and a control, respectively. The

capacity vs. cycle number was logged for each battery

in the experimental group, which consistently exhibited

capacity fade from 3.2 V to 2.5 V (Figure 5).

Each battery exhibited a unique aging behavior, as may

be expected from more exhaustive studies documented

in the literature [6], but the aging routines were halted

when the battery either stopped charging or faded to

2.5 V capacity, whichever came first.

Figure 4 The XRM results non-destructively showed the different layers within the battery, as visualized here using ORS Dragonfly Pro. The different layers included the Al current collector (purple), cathode (orange), anode and separator (blue), and Cu current collector (white).

Figure 5 The capacity vs. cycle number was logged for each charge cycle of the experimental group. A capacity fade from 3.2 V to 2.5 V was observed over the lifetime of each battery.

Application Note

5

200 µm

300 µm

The non-destructive 3D interior tomograms were then

compared to each other, using dispersed internal inclusions/

defects that were observed within the battery as registration

points to correlate the fresh and aged results. A path was

drawn along the same curved section of the jelly-roll in

the fresh and aged datasets from which the wraps of

positive electrode foil were virtually “unrolled”, enabling

visual inspection of each planar electrode (Figure 6). The im-

ages were examined for the presence of cracks or other mi-

crostructural defects which might affect the performance of

the battery. These cracks were identified with colored arrows

(Figure 7) and data interpretation was performed based on

the presence, appearance, and/or disappearance of cracks in

the 4D datasets. Cracks were observed to systematically de-

crease in detectability for the aged specimens (i.e., the ex-

perimental group) (Figure 7). In contrast, the control group,

did not show any significant or systematic microstructural

change.

While the microstructure observations were the primary

focus of this research project, a secondary question arose

with respect to the reasons or mechanisms for cell failure.

The cells were charged and discharged without any special

considerations; however, an in situ thermal monitoring

system revealed that the batteries routinely experienced

temperatures in excess of 50 °C. To understand what –

if any – impact this might have had on the reliability of

the batteries, the top cap assembly was inspected with

low-resolution XRM. A battery from the experimental

group was compared to one from the control group,

which resulted in a 3D visualization of the current

interrupt device (CID) engagement in the experimental

group battery (Figure 8). Thus, these results suggested that

thermal overload may have been the primary mechanism for

cell failure, though further research is needed to establish

this conclusion.

Figure 6 Example virtual slice, showing the “path“ (green curve) along which the battery foils were virtually unrolled.

Figure 7 Example virtual slices from the experimental (top) and control (bottom) group datasets. The experimental group batteries systematically showed a reduction in the presence/appearance of cracks, whereas the control group did not show a significant change between its initial and final images (to be expected, since there were no significant treatments applied to the control group specimens).

Experimental Group – Fresh

Control Group – Initial

Experimental Group – Aged

Control Group – Final

Application Note

6

2 mm

Summary

A 4D study has been performed on a set of commercial Li-

ion batteries in 18650 cell geometries. The batteries were

first inspected in their entirety with low-resolution 3D X-ray

microscopy, to survey for bulk defects and to identify an ROI

for higher-resolution investigation. Without disturbing the

battery, a smaller region from each specimen was then opti-

cally enlarged using the non-destructive nature of X-rays

along with the uniquely high-resolution capabilities of ZEISS

Xradia 520 Versa. This multi-scale imaging approach was

then repeated after each battery had been cycled to failure,

and the results aligned to their corresponding “fresh” states.

Subsequent failure analysis scans focused on the current in-

terrupt device, to investigate the mechanism for cell failure.

The XRM results indicated that microstructural changes re-

sulted after aging in a systematic fashion, and that thermal

overload did, indeed, lead to the engagement of a current

interrupt device. This study represents an important path to-

ward understanding how batteries change with operation,

to elucidate the relationship between microstructure,

performance, and failure modes.

Acknowledgements

The author wishes to thank Dr. Paul Shearing and Prof. Dan

Brett at University College London, as well as, Dr. Donal

Finegan at National Renewable Energy Laboratory, for their

help with battery cycling and data interpretation. Funding

from ZEISS and research guidance from Dr. Melanie McNeil

and Dr. Craig England at San Jose State University is also

gratefully acknowledged.

Figure 8 Examining the top cap of (top) one control group battery and comparing it to (bottom) an experimental group battery, a microstructural difference was observed that corresponds to engagement of the CID, shown with an arrow on the figure. This indicated that thermal overload may have been a fundamental cause for early cell failure.

Application Note

7

References

[1] J.M. Tarascon and M. Armand, “Issues and challenges facing rechargeable lithium batteries,” Nature. (2001).

[2] P.R. Shearing, D.S. Eastwood, R.S. Bradley, J. Gelb, S. Cooper, F. Tariq, et al., “Exploring electrochemical devices using X-ray microscopy:

3D micro-structure of batteries and fuel cells,” Microscopy and Analysis. 1–4 (2013).

[3] D. Kehrwald, P.R. Shearing, N.P. Brandon, P.K. Sinha and S.J. Harris, “Local Tortuosity Inhomogeneities in a Lithium Battery Composite

Electrode,” J. Electrochem. Soc. 158 A1393–7 (2011).

[4] M. Ebner, D.-W. Chung, R.E. Garcia and V. Wood, “Tortuosity Anisotropy in Lithium-Ion Battery Electrodes,” 4 n/a–n/a (2013).

[5] D.P. Finegan, M. Scheel, J.B. Robinson, B. Tjaden, M. Di Michiel, G. Hinds, et al., “Investigating lithium-ion battery materials during over

charge-induced thermal runaway: an operando and multi-scale X-ray CT study,” Phys Chem Chem Phys. (2016).

[6] S.J. Harris, D.J. Harris and C. Li, “Failure statistics for commercial lithium ion batteries: A study of 24 pouch cells,” Journal of Power

Sources. 342 589–597 (2017).

[7] P.R. Shearing, L.E. Howard, P.S. Jørgensen, N.P. Brandon and S.J. Harris, “Characterization of the 3-dimensional microstructure of a

graphite negative electrode from a Li-ion battery,” Electrochemistry Communications. 12 374–377 (2010).

[8] A.P. Merkle and J. Gelb, “The Ascent of 3D X-ray Microscopy in the Laboratory,” Micros. Today. 21 10–15 (2013).

[9] C. Weisenberger, A. Kopp, T. Bernthaler, V. Knoblauch, G. Schneider, H. Stegmann, et al., “Multi-scale characterization of a lithium ion

battery cathode material by correlative X-ray and FIB-SEM microscopy,” Microscopy and Analysis. 17–19 (2015).

[10] F. Tariq, V. Yufit, M. Kishimoto, P.R. Shearing, S. Menkin, D. Golodnitsky, et al., “Three-dimensional high resolution X-ray imaging and

quantification of lithium ion battery mesocarbon microbead anodes,” Journal of Power Sources. 248 1014–1020 (2014).

[11] D.P. Finegan, S.J. Cooper, B. Tjaden, O.O. Taiwo, J. Gelb, G. Hinds, et al., “Characterising the structural properties of polymer separators

for lithium-ion batteries in 3D using phase contrast X-ray microscopy,” Journal of Power Sources. 333 184–192 (2016).

[12] J. Nelson, S. Misra, Y. Yang, A. Jackson, Y. Liu, H. Wang, et al., “In Operando X-ray Diffraction and Transmission X-ray Microscopy of

Lithium Sulfur Batteries,” J. Am. Chem. Soc. 134 6337–6343 (2012).

[13] Y.-C.K. Chen-Wiegart, P. Shearing, Q. Yuan, A. Tkachuk and J. Wang, “3D morphological evolution of Li-ion battery negative electrode

LiVO2 during oxidation using X-ray nano-tomography,” Electrochemistry Communications. 21 58–61 (2012).

[14] B. Yan, C. Lim, L. Yin and L. Zhu, “Simulation of heat generation in a reconstructed LiCoO2 cathode during galvanostatic discharge,”

zElectrochimica Acta. 100 171–179 (2013).

Carl Zeiss Microscopy GmbH 07745 Jena, Germany microscopy@zeiss.com www.zeiss.com/microscopy

EN_4

4_01

3_04

4 | C

Z 05

-201

7 | D

esig

n, s

cope

of

deliv

ery

and

tech

nica

l pro

gres

s su

bjec

t to

cha

nge

with

out

notic

e. |

© C

arl Z

eiss

Mic

rosc

opy

Gm

bH

Not

for

the

rape

utic

, tre

atm

ent

or m

edic

al d

iagn

ostic

evi

denc

e. N

ot a

ll pr

oduc

ts a

re a

vaila

ble

in e

very

cou

ntry

. Con

tact

you

r lo

cal Z

EISS

rep

rese

ntat

ive

for

mor

e in

form

atio

n.