Elemental composition analysis of Nickel-Manganese-Cobalt cathodes and their precursor materials using XRF

Introduction

In recent years, lithium-ion batteries have revolutionized the energy storage landscape by powering portable electronics, electric vehicles, and renewable energy storage systems. Among the various types of lithium-ion batteries, lithium nickel manganese cobalt oxide (Li-NMC) batteries have emerged as a prominent choice due to their high energy density, improved stability, and established industrial-scale production. The chemical composition of Li-NMC cathode materials plays a critical role in determining their performance, making accurate elemental analysis of these materials essential for battery development and production.

There are many analytical techniques that can analyze elemental composition, of which ICP and XRF are of most significance. Comparing X-ray fluorescence (XRF) spectrometry to other elemental analysis techniques like inductively coupled plasma (ICP) spectroscopy, XRF spectrometers enable simpler and faster analysis while delivering high quality results in terms of precision and accuracy. This makes XRF a practical solution for process and quality control in battery cathode and precursor production and in battery recycling.

NCM CRMs Package

XRF can analyze elemental composition in two ways. The first is the standardless screening of input materials, which provides detection and semiquantitative estimate of the elemental composition. However, for the accuracy desired in cathode materials production process and quality control, the second method involving calibration using standard samples needs to be deployed. In the latter case, the accuracy of XRF results relies on the availability of high-quality calibration materials, and there is a clear lack of commercially available calibration standards for battery cathode materials. Malvern Panalytical has designed and produced a set of Nickel-Cobalt-Manganese (NCM) Certified Reference Materials (CRMs) for its XRF calibration, along with using our sample preparation systems and expertise, which can deliver highly accurate and reliable results on NCM cathode materials.

The NCM CRMs package compromises 12 synthetic mixes specifically designed for preparing XRF fused bead specimens. The package also includes a fusion recipe and an XRF application method template. The CRMs are made from pure chemicals using a gravimetric approach for metrological traceability and adherence to ISO 17034. The elemental composition of the NCM CRMs and corresponding minimum and maximum concentrations are given in Table 1.

The NCM CRMs package is also suitable for cathode materials such as Lithium Nickel Cobalt Aluminum Oxide, Lithium Cobalt Oxide, and Lithium Manganese Oxide and their precursors. For high accuracy and repeatability, it is recommended to use fusion sample preparation method. However, these CRMs can also be used for making secondary calibration standards in the form of pressed pellets. 

Table 1 minimum and maximum concentrations of NCM CRMs

	Li (%)*	Mn (%)	Co (%)	Ni (%)	Al (%)	Ca (%)	Zr (%)	Na (%)	S (%)
Lowest cal. point **	5.70	3.00	3.00	10.00	0	0	0	0	0
Highest cal. point	9.00	27.00	27.00	55.00	2.00	0.10	2.00	1.00	0.40

* Lithium or Lithium oxide cannot be measured directly by XRF instruments, but they are added to the composition to simulate batteries cathodes mixes.
** The lowest calibration point should not be considered as the minimum concentration that can be reported. Instead, the Limit of Quantification (LOQ) is used for this purpose. LOQ depends on sample preparation, XRF instrument, measurement conditions, and measurement time. LOQ values for each element of interest are calculated during the instrument calibration process. Typically, LOQ values for the elements listed in Table 1 are in the range of 50 – 300 ppm.

Instrument

In this application study, Malvern Panalytical Eagon 2 and Zetium are used respectively for sample preparation and XRF measurement. The Zetium configuration is detailed in Table 2. Depending on specific analytical needs, alternative configuration options and additional features can be chosen.

Table 2 The Zetium configuration used in this application note

Zetium Spectrometer	Description
Tube	4kW SST R-mAX tube with ZETA technology
Tube anode	Rh
Tube window	50um Be
Tube coating	CHI-BLUE coating
Filters	Al (200um)
Collimator mask	27 mm
Collimators	150um, 300um
Crystals	PX10, Ge 111 curved, PE 002 curved, PX1, LiF220
Detectors	Duplex detector HiPer Scintillation detector

Sample Preparation

Specimens of calibration standards and a validation CRM were prepared as diameter 32 mm fused beads with lithium borate fusion technique. A dilution ratio of 1 mass part of a sample to 10 mass parts of lithium borate flux was used for preparation with an integrated non-wetting agent. Fusion of beads was done at 1100°C with the Eagon 2 automatic fusion machine. Total fusion time of “cold-to-cold” operation cycle takes about 30 minutes. Proven fusion recipes are also available for LeNeo and FORJ automatic fusion machines.

Measurement Conditions

During the application study, the measurement time for one specimen was approximately 8 minutes, covering the analysis of 8 compounds present in NCM CRMs. Lithium oxide in finished cathode materials could not be measured directly by XRF technique, but an average expected value is sufficient to obtain accurate results for the other measurable elements. 

Calibration Results

The examples of obtained calibrations for Co, Mn, Ni and S are given in the Figures 1 - 4. Their RMS and K-factor values are summarized in Table 3. The accuracy of the calibrations is indicated by RMS (absolute error) and K-factor (weighted error) values. The lower the RMS and K-factor values, the better the calibration.

Table 3 Summarized RMS and K-factors values in figures 1-4

	CoO	MnO	NiO	SO3
RMS (wt%)	0.0771	0.0905	0.1215	0.0050
K-factor*	0.0170	0.0189	0.0203	0.0096

*Since the K-factor is independent of the calibrated concentration range (unlike the RMS value), it gives a more reliable indication of calibration quality. The lower the K-value, the better the calibration, with values below 0.05 indicative of good calibrations.

Accuracy

The final validation of the method trueness was conducted by measuring fused beads of a commercially available CRM BAM-S014, a Li-NMC 111 cathode material. 7 beads of this CRM were prepared in 3 different days, their results are shown in Table 4.

The XRF measurement results obtained using NCM calibrations closely match the certified values of BAM-S014, with notably lower standard deviation values. The observed differences between the measured results and certified values are remarkably small and fall within the allowed difference range calculated according to ISO Guide 35 requirements. Furthermore, the consistency of results across 3 different days indirectly underscores the high reproducibility of the sample preparation method and the stability of Eagon 2 and Zetium.

Table 4 Accuracy validation using a commercially available CRM (BAM-S014) made into 7 beads

	Ni (%)	Co (%)	Mn (%)	S (%)
BAM-S014_day 1_bead 1	19.697	19.805	18.161	0.148
BAM-S014_day 1_bead 2	19.669	19.771	18.145	0.147
BAM-S014_day 2_bead 1	19.753	19.843	18.212	0.148
BAM-S014_day 2_bead 2	19.743	19.832	18.186	0.146
BAM-S014_day 2_bead 3	19.806	19.873	18.222	0.147
BAM-S014_day 3_bead 1	19.767	19.867	18.214	0.144
BAM-S014_day 3_bead 2	19.794	19.881	18.234	0.147
Measured mean	19.747	19.839	18.196	0.1467
*St.dev. of mean	0.050	0.040	0.033	0.0014
Certified value	19.76	19.80	18.22	0.1421
Between labs st.dev. of cert. value	0.21	0.20	0.25	0.0123
Uncertainty of cert. value	0.13	0.12	0.14	0.0070
Actual difference	-0.013	0.039	-0.023	0.0046
Allowed difference**	0.135	0.123	0.142	0.007

* St.dev. stands for standard deviation.
**Allowed differences calculated according to the ISO Guide 35 requirements.

Precision and Instrument Stability

Method precision, including sample preparation and measurement errors, was estimated by preparing and measuring 10 replicate beads of the same sample in one day. The results are shown in the Table 5. The small standard deviation values here demonstrate the high precision of this method.

Table 5 Precision validation for 10 replicate beads within 1 day

Sample ID	Ni (%)	Co (%)	Mn (%)	S (%)
BAM-S014_ repeat_bead_01	19.725	19.787	18.19	0.145
BAM-S014_ repeat_bead_02	19.705	19.792	18.185	0.143
BAM-S014_ repeat_bead_03	19.716	19.773	18.161	0.144
BAM-S014_ repeat_bead_04	19.751	19.828	18.206	0.143
BAM-S014_ repeat_bead_05	19.676	19.759	18.158	0.145
BAM-S014_ repeat_bead_06	19.729	19.795	18.199	0.143
BAM-S014_ repeat_bead_07	19.725	19.789	18.181	0.146
BAM-S014_ repeat_bead_08	19.777	19.837	18.233	0.143
BAM-S014_ repeat_bead_09	19.642	19.745	18.135	0.143
BAM-S014_ repeat_bead_10	19.602	19.683	18.092	0.144
Measured mean	19.705	19.779	18.174	0.144
St.dev. of mean	0.052	0.044	0.040	0.001

Conclusion

This application study demonstrates that utilizing NMC CRMs package along with Malvern Panalytical fusion machine Eagon 2 and XRF spectrometer Zetium can deliver highly accurate and precise elemental analysis results for Li-NMC battery cathode materials and their precursors. This turn-key solution enables high-throughput and seamless elemental analysis, eliminating the need for extensive sample preparation, specialized analytical skills, and the use of strong acid chemicals typically required for ICP analysis.