Comparison of Determining the 10B and 6Li Depth Profiles based on
NDP and SIMS Analytical Methods
Hoang Sy Minh Tuan1, *
1

Institute of Applied Technology - Thu Dau Mot University, 6, Tran Van On, Phu Hoa
Ward, Thu Dau Mot City, Binh Duong, Vietnam, 820000
*

ABSTRACTS
Depth resolution and probing depth for 6Li in lithium thin-film batteries achievable using
different ion beam analytical techniques were investigated. In this study, the comparison of the
analytical results between SIMS (Secondary-ion mass spectrometry) and NDP (Neutron Depth
Profiling) methods have been carried out with LiCoO2 and BSi samples. The NDP is an
analytical method to analyze the component nuclide concentration versus depth distribution in a
sample by detecting the charged particles emitted after the neutrons are absorbed. The
6

10

B and

Li depth profiles in the BSi and LiCoO2 samples were also analyzed using a CAMECA IMS 7f

SIMS instrument at the National Nanofab Center (Republic of Korea). The results from NDP
analysis have been performed at the NDP system (HANARO, Republic of Korea). In comparison
results for the samples, the peak depths, peak concentrations, and total dose of the NDP results
are consistent with the SIMS results within 2, 6, and 11 %, respectively. The NDP is useful for
analyzing light elements with high neutron cross-sections for particle-producing reactions.
Keywords: NDP; SIMS; Depth Profiling.

Tóm tắt
Đánh giá đặc tính theo độ sâu của 6Li trong pin màng lithium mỏng có thể đạt được bằng cách sử
dụng các kỹ thuật phân tích chùm ion khác nhau. Trong nghiên cứu này, việc so sánh các kết quả
phân tích giữa kỹ thuật SIMS (Secondary-ion mass spectrometry) và NDP (Neutron Depth
Profiling) được thực hiện với các mẫu LiCoO2 và BSi. Kỹ thuật NDP là một kỹ thuật phân tích
để phân tích hàm lượng nguyên tố thành phần so với phân bố độ sâu trong một mẫu bằng cách
phát hiện các hạt tích điện phát ra sau khi neutron được hấp thụ. Các phân bố độ sâu 10B và 6Li
trong các mẫu BSi và LiCoO2 được phân tích bằng cách sử dụng thiết bị SIMS CAMECA IMS
7f tại Trung tâm Nanofab Quốc gia (Hàn Quốc). Kết quả từ phân tích NDP đã được thực hiện tại
170


hệ thống NDP (HANARO, Hàn Quốc). Trong kết quả so sánh cho các mẫu, độ sâu đỉnh, nồng độ
cao nhất và tổng liều của kết quả NDP phù hợp với kết quả SIMS trong vòng 2, 6 và 11%, tương
ứng. Kỹ thuật NDP rất hữu ích để phân tích các nguyên tố ánh sáng có mặt cắt ngang neutron
cao cho các phản ứng tạo ra hạt.

1. Introduction
Lithium-ion rechargeable batteries are commonplace in consumer gadgets such as cell
phones and laptop computers. They are based on the movement of lithium ions through an
electrolyte between a positive electrode (cathode), which is a lithium-containing material, and a
negative electrode (anode), which is generally a porous substance. The efficiency, capacity, and
durability of batteries are all dependent on lithium movement during charge and discharge.
Energy storage technology is essential for a sustainable energy transition. Understanding the
regulating mechanisms is critical to meeting the ever-increasing energy storage need. Different
ion-beam analytical techniques were used to explore the depth resolution and probing depth for
Li in lithium thin-film batteries.
SIMS (Secondary Ion Mass Spectrometry) detects very low dopant and impurity
concentrations. A beam of intense heavy ions erodes the surface of the material under
investigation in SIMS, while secondary ions created during the sputtering process are mass

assessed and recognized [1]. As a result, it is no wonder that depth profiling is one of SIMS's
most popular applications. The method generates elemental depth profiles over a wide range of
depths, from a few angstroms to tens of micrometers. A beam of primary ions (typically O2+ or
Cs+) sputters/etches the sample surface, while secondary ions generated during the sputtering
process are collected and studied using a mass spectrometer (quadrupole, magnetic sector, or
Time-of-flight). The concentrations of secondary ions might range from matrix levels to subppm trace levels. SIMS methodology is limited for operando monitoring of lithium-ion batteries
due to the intrinsic difficulties of studying light ions with traditional techniques. As a result,
SIMS depth profiling has the disadvantage of providing no information on the sample's atomic
structure and is damaging. On the other hand, SIMS depth profiling has the disadvantage of
providing no information on the sample's atomic structure and is destructive. As a result, an
examined spot can no longer be used as an active device, and the SIMS results and laser
performance cannot be linked.
171


Real-time in-situ visualization and quantification of the changing lithium distribution in
batteries during charging and discharging is beneficial information for their development,
especially for a better understanding of the mechanisms restricting charge and discharge rates.
Moreover, at the present, only one technique, Neutron Depth Profiling (NDP), can accomplish
such a difficult task [2]. During (dis)charge, this approach offers information on the geographical
and temporal lithium concentration. NDP is employed in this study to shed light on critical
lithium-ion battery issues. The findings give a clear picture of electrode properties such as
tortuosity and Li-ion transport (Figure 1). This provides for a reduction in charge times by
lowering battery internal resistance or increasing the charging current.

Figure 1. The principle of using Neutron Depth Profiling for Li characterization in Li-ion battery.

NDP has long been used to determine the depth of lithium in a variety of materials.
Oudenhoven et al. from the Eindhoven University of Technology in the Netherlands performed
in-situ NDP on thin-film solid-state micro-batteries, seeing the lithium concentration profile

evolve [3]. C. Co et al. from Ohio State University in the United States collected one NDP
spectrum every 5 minutes during charge and discharged to perform real-time in-situ
quantification of Li transport in a Li-ion cell [4]. This offered previously unobtainable
information on lithium incorporation rates and losses in various battery regions, establishing
operando NDP as a viable tool for understanding Li transport in Li-ion batteries. The lowering of
battery weight was one of the issues that were eventually overcome. A copper sheet, which is
heavy and expensive, is used to capture the current in the anode. Aluminum is substantially
172


lighter and less expensive than steel. Aluminum's application in Li-ion batteries has been
hampered so far by a reaction with lithium that removes the lithium ions from the anode. With
operando NDP in a cell where the aluminum was covered with a protective 500 nm thick Sn
layer, the viability of Al as the anode current collector in a lithium-ion cell was demonstrated [5].
The Sn worked as a protective layer that did not react with the aluminum by integrating some
lithium. Yuping et al. from Binghamton University in the United States used a neutron beam
with an 11 mm2 aperture to measure the Li depth profile for each pixel on the surface of a Li-ion
cell with a LeFePO4 electrode [6]. The 3D lithium distribution for the total sample is obtained by
combining all depth profiles. They compared electrodes with only one charge/discharge cycle to
electrodes that had 5000 charge/discharge cycles. The Li concentration dropped by roughly 40%
after 5000 cycles, whereas lateral inhomogeneity rose considerably. This is a direct observation
of one of the causes of battery degeneration and failure.

2. Method
If the target isotope captures the neutron, the compound nucleus emits a charged particle,
and the leftover nucleus recoils in 1 ps. Depending on the isotope, these reactions create a proton
or an alpha particle and a recoiling nucleus. Figure 2 depicts the process of neutron captures in a
sample deposited to a thickness of x, as well as the emission of charged particles (a). Some
charged particles created isotopically in the sample pass through the sample material, reach the
sample surface, and then pass through a charged particle detector. Figure 2 depicts the NDP

spectrum acquired as a result of the procedure (b). A neutron beam illuminates the planar surface
in a vacuum chamber at an angle of θ1 to the surface normal. The detector should have an areaplane surface with a normal angle θ2 away from the sample normal.

173


Figure 2. Schematic representation of the neutron-induced charged particle emission process (a), and the
resultant spectrum (b).

Then the depth of target isotope is,

x = s.cos 2

(1)

When the average stopping power in the path length is I, the energy loss of charged particle
in the path length I, E is the same as s.S , then (2-9) is,

x=

E.cos  2
S

(2)

The uncertainty of x can be defined as depth resolution, where it can describe the uncertainty
of energy loss ΔE.
Using the TRIM software [7], the energies of the spectrum were converted to depths based
on the residual energy versus path length of the alpha particle. Each channel's count rate was
translated to a 6Li concentration. When the measured spectrum has no energy broadening, the

concentration profile C(x) at depth x can be written as:
C ( x) =

N ( E ( s )  S ( E ( s ))
   d  An  cos  s
f   C  C
cos  n

(3)

Where E(s) is the residual energy for particles born at depth x that travels a distance s before
escaping the sample surface; N(E) is the number of ions with energy E measured by the detector
174


per unit time; S(E) is the stopping power of ions with energy E; f is the fractional yield for the
particle of interest per charged particle production reaction; sc is the microscopic neutron crosssection for charged particle production; ϕc is the neutron flux; ε is the detection efficiency; Ωd is
the solid detection angle. The cold neutron beam with the cross-sectional area of An illuminates
the surface of the sample at an angle θn to the surface normal, and the detector is assumed to
have a plane surface whose normal is at an angle θs to the sample surface normal.
Figure 3 depicts a schematic of the NDP system used in this investigation. A cold neutron
beam is delivered to a target chamber through a cold neutron guide. The LiF beam slit in the
guide is used to collimate the neutron beam, and the Cd collimator in the beam inlet port with a
10 mm diameter aperture is used to define the beam region at the sample point. The average
wavelength of the cold neutron beam is 5.1 A, and the actual integrated neutron flux at the
sample point is 7.37  106 n/cm2s, as determined by an activation analysis of gold foil. The target
chamber is kept at a vacuum of less than 2 9 10-4 Pa during operation.

Figure 3. The schematic of the NDP system at HANARO used in this study.


Lithium-coated silicon samples were created for deposition thickness measurements. The
samples were made at Seoul National University's Electronic Materials Laboratory. When 6Li
absorbs a neutron, it undergoes the nuclear reaction shown below.
6

Li + n → 3H (2727.88keV ) + 4 H (2055.51keV )

(4)

Ion-beam sputtering of a LiCoO2 target with Ar+ ions was used to deposit LiCoO2 on an
oxidized (SiO2 thickness: 0.1 m silicon wafer. The silicon wafer had a thickness of 600 m. By
varying the sputtering conditions, two types of samples (LCO-1 and LCO-2) were created. The
RF power, deposition time, substrate temperature, and argon gas flow rate were 200 W, 4 h, 350
175


o

C, and 45 sccm for LCO-1; and 120 W, 9 h, room temperature, and 15 sccm for LCO-2. SIMS

was used to compare the

10

B and 6Li depth profiles in the BSi and LCO samples with the NDP

analysis, utilizing a CAMECA IMS 7f SIMS equipment [8].
Samples were inserted in the target chamber by connecting them to the sample holder's 80
m thick Teflon tape. The sample stage, which is placed above the target chamber, was used to
fine-tune the location of the sample as well as the angle between the sample and the neutron

beam. Figure 4 depicts the boron-implanted silicon wafer, the lithium-coated silicon wafer, and
the lithium-ion battery and its electrodes.

Figure 4. Pictures of (a) boron implanted silicon wafer, (b) lithium-coated silicon wafer, and (c) the
lithium-ion battery and its electrodes.

3. Results and Discussions
Figure 5 depicts the analysis findings for the BSi, LCO, and electrode samples. SIMS
analysis findings are given in the same figures for comparison. The NDP error bars show the
error from data smoothing with Poisson counting statistics obtained by error propagation. The
positive error bars are shown in the figures for clarification. 10B is dispersed to a depth of 0.5 m
176


for the BSi-1 sample and 0.3 m for the BSi-2 sample, as shown in Fig. 5a. The analytical
findings for the BSi-3 sample are shown in Figure 5b. Three results are highly agreed upon in the
region up to 0.3 m of the sample. In contrast, when the sample-detector angle rises, the
concentration of

10

B increases in the area deeper than 0.3 m. This is due to an increase in the

travel length of the alpha particles released by

10

B. The standard deviation of energy widening

owing to energy straggling grows as the route length of alpha particles increases. In this case, as

explained in Eqs. (3) the depth profile C(x) at depth x can be raised. The NDP results' peak
depth, concentration, and total dosage are within 2%, 6%, and 11% of the SIMS values,
respectively.

Figure 5. 10B depth profiles of the BSi-1, BSi-2 (a), and BSi-3 samples (b). Relative depth profiles of 6Li
in the LCO-1 and LCO-2 samples (c). 6Li depth profiles of the electrode samples from the lithium ion
battery (d).

The SIMS technique requires a reference material with the same matrix as the sample in
177


order to measure the isotope concentration. However, because there was no standard material for
LCO samples, SIMS gave only the relative concentration of 6Li. The NDP technique findings for
the LCO-1 and LCO-2 samples, normalized to the SIMS results, are presented in Figure 5c. The
NDP and SIMS findings for LCO-1 concur well up to 0.56 m. However, at the region deeper
than 0.56 m, the NDP result contradicts the SIMS result. In contrast, the concentration of 6Li in
the SIMS result improves from 0.56 to 0.72 m, it continues to fall in the NDP result. The
highest depths of 6Li in the SIMS and NDP samples for the LCO-2 sample match the 0.4 m.
However, the NDP-determined 6Li concentration at each level contradicts the SIMS conclusion.
The LiCoO2 has been deposited on the Si + SiO2 substrate in these samples. The SIMS approach
may have exaggerated the concentration of 6Li due to the lack of information regarding the
matrix of the SiO2 layer. The aerial densities of 6Li in the LCO-1 and LCO-2 samples were
found to (1.7 ± 0.3)  1017 and (7 ± 1)  1016 atoms/cm2, respectively, using the NDP technique.
The uncertainty is expressed as a 95% confidence interval that incorporates random
measurement errors and systematic errors. Over the entire sample spectrum, the systematic error
is estimated to be 10%. The concentrations of 6Li in the plateau area LCO-1 and LCO-2 samples
were determined to be (3.03 ± 3.12)  1021 and (1.36 ± 2.23)  1021 atoms/cm3, respectively. As
a result, the region of channel numbers 2000  2700, where only triton contribution existed, was
utilized for the conversion to concentration distribution. Figure 5d depicts depth profiles of 6Li in

electrode samples. The concentrations of 6Li in the cathode and anode samples were 3.5  1021
and 6.7  1020 atoms/cm3, respectively).

4. Conclusions
The relative concentrations of 6Li obtained by NDP and SIMS were discordant in several
sections of the samples in this investigation with the case of the lithium-coated Si samples. The
SIMS profiles for the lithium-coated samples are questionable since the tendency toward an
apparent rise in lithium content towards the surface is highly unusual for sputter-deposited
samples. In the future, this issue will necessitate in-depth research and the use of a different
analysis approach. Additional test samples were made by removing cathode and anode material
from an old lithium-ion battery. The concentration of 6Li in the cathode sample of a completely
drained lithium-ion battery was five times greater than in the anode sample. Both samples' layer
thicknesses were determined to be more than 22 m, which is the range of tritons in LiCoO2.
178


References
[1] Magee, C. W., & Honig, R. E. (1982). Depth profiling by SIMS—depth resolution,
dynamic range and sensitivity. Surface and Interface Analysis, 4(2), 35-41.
[2] Verhallen, T. W., Lv, S., & Wagemaker, M. (2018). Operando neutron depth profiling to
determine the spatial distribution of Li in Li-ion Batteries. Frontiers in Energy Research, 6,
62.
[3] Oudenhoven, J. F. M., Labohm, F., Mulder, M., Niessen, R. A. H., Mulder, F. M., &
Notten, P. (2011). In situ neutron depth profiling: A powerful method to probe lithium
transport in micro‐batteries. Advanced Materials, 23(35), 4103-4106.
[4] Liu, D. X., Wang, J., Pan, K., Qiu, J., Canova, M., Cao, L. R., & Co, A. C. (2014). In situ
quantification and visualization of lithium transport with neutrons. Angewandte Chemie,
126(36), 9652-9656.
[5] Liu, D. X., Cao, L. R., & Co, A. C. (2016). Demonstrating the feasibility of Al as anode
current collector in li-ion batteries via in situ neutron depth profiling. Chemistry of

Materials, 28(2), 556-563.
[6] He, Y., Downing, R. G., & Wang, H. (2015). 3D mapping of lithium in battery electrodes
using neutron activation. Journal of Power Sources, 287, 226-230.
[7] McDaniel, F. D., Anthony, J. M., Renfrow, S. N., Kim, Y. D., Datar, S. A., & Matteson, S.
(1995). Depth profiling analysis of semiconductor materials by accelerator mass
spectrometry. Nuclear Instruments and Methods in Physics Research Section B: Beam
Interactions with Materials and Atoms, 99(1-4), 537-540.

179