All chapters
Atomic emissionBragg’s law and XRDNeutron diffractionRaman I_D/I_GEIS

Every claim in this series is measurable, and almost every measurement is a wavelength measurement. That is not a coincidence — energy levels are quantised, and quantised energy differences emit and absorb at specific wavelengths.

It is worth appreciating one consequence before starting. The crimson of a lithium flame test is you directly observing the 2s valence electron that the entire battery industry depends on.

11.1Atomic emission — elemental fingerprints

When an excited electron falls from a higher to a lower energy level, it emits a photon of energy exactly equal to the gap.

E = hν = hc/λ

h = Planck’s constant, 6.626×10⁻³⁴ J·s; c = speed of light; λ = wavelength.

Because energy levels are quantised and unique to each element, the emitted wavelengths are an unforgeable fingerprint.

ElementCharacteristic lineTransition
Lithium670.8 nm (crimson)2p → 2s
Sodium589.0 / 589.6 nm (yellow doublet)3p → 3s
Potassium766.5 / 769.9 nm (lilac)4p → 4s
Copper324.8 nm
Cobalt228.6 nm

Why this matters

Note the trend: Li 670.8 nm → Na 589.0 nm → K 766.5 nm. The emission wavelength directly reports the valence energy-level spacing, which is the ionisation-energy trend from chapter 1 made visible. The periodic table is not an abstraction here — you can see it.

400450500550600650700750Li670.8 nmNa589.0 nmK766.5 nmCu578.2 nmATOMIC EMISSION LINES / E = hc/λLi 670.8 nm is the 2s valence electron — the same electron the entire battery industry moves.WAVELENGTH / nm
Figure 11.1E = hν = hc/λ. When an excited electron falls to a lower level it emits a photon of energy exactly equal to the gap. Because energy levels are quantised and unique to each element, the wavelengths are a fingerprint. Note the trend across Li 670.8 → Na 589.0 → K 766.5 nm: the emission wavelength directly reports valence-level spacing, which is the ionisation-energy trend of Plate 04 made visible to the eye. This principle drives ICP-OES for verifying cathode stoichiometry, ICP-MS for finding the ppm-level Fe and Cr particles that cause micro-shorts, and LIBS for sorting cells on a recycling line.

11.1.1The instruments built on this

  • ICP-OES / ICP-MS — plasma-excited emission or mass spectrometry. The standard for verifying cathode stoichiometry and detecting ppm-level contaminants. Iron, chromium and nickel particles are the metallic contaminants that cause micro-shorts and self-discharge failures.
  • LIBS — laser-induced breakdown spectroscopy. A laser pulse ablates a microscopic spot and the emission is read. Used for rapid sorting of cells and scrap on recycling lines.
  • AAS — atomic absorption spectroscopy, measuring absorption at the same characteristic wavelengths.

11.2X-ray diffraction — measuring the lattice

nλ = 2d sin θ

λ = X-ray wavelength, d = spacing between crystal planes, θ = incidence angle, n = integer.

The standard laboratory source is Cu Kα at 1.5406 Å, chosen because it is comparable to interatomic spacings of a few ångström, so the diffraction angles are conveniently measurable.

  • Lattice parameters — confirm the phase is what you ordered.
  • Peak shift on cycling — a direct measurement of lattice expansion and contraction, such as graphite’s 3.35 → 3.70 Å from chapter 6.
  • The (003)/(104) intensity ratio in layered oxides — a quantitative measure of cation mixing. Above about 1.2 indicates good layering; below indicates Ni²⁺ has invaded the lithium sites, exactly as described in chapter 4.
  • Peak broadening — crystallite size via the Scherrer equation, D = Kλ/(β cos θ).
  • In-situ / operando XRD — watching phase transitions happen during a real charge.

11.3Why neutrons, not X-rays, for lithium

X-rays scatter off electrons. Lithium has three. Its X-ray scattering factor is tiny, and in a lattice full of cobalt with 27 electrons and oxygen with 8, lithium is effectively invisible to XRD.

Neutrons scatter off atomic nuclei, and the scattering length is unrelated to atomic number. Lithium is readily detectable — specifically ⁷Li, while ⁶Li has a negative scattering length.

Important

Neutron diffraction is therefore the only routine technique that directly locates lithium in a crystal structure and quantifies Li/Ni site exchange. This is why cathode research clusters around a handful of neutron sources worldwide — a genuine physical constraint on how fast the field can move.

dθINCIDENT X-RAY λ = 1.5406 Å (Cu Kα)DIFFRACTEDnλ = 2d sin θλ chosen to match interatomic spacing, so θ is measurableWHAT XRD TELLS YOUPeak position → lattice parametersPeak shift → expansion on cyclingI(003)/I(104) → cation mixingPeak width → crystallite size (Scherrer)Operando → phase transitions liveWHY NEUTRONS FOR LITHIUMX-rays scatter off electrons.Lithium has three.Beside Co (27) and O (8) it iseffectively invisible to XRD.Neutrons scatter off nuclei — Li shows up.
Figure 11.2The (003)/(104) intensity ratio is the standard quantitative measure of layering quality in a layered oxide: above about 1.2 the structure is well ordered; below it, nickel has invaded the lithium sites. It is one number, read off a routine diffractogram, that tells you whether your cathode will deliver its rated capacity.

11.4Raman — bonds and carbon quality

The Raman effect is inelastic scattering of light in which the scattered photon shifts in energy by the amount of a vibrational quantum of the material, reported as a Raman shift in wavenumbers.

For carbon materials, two bands dominate:

  • G band, ~1580 cm⁻¹ — in-plane stretching of sp²-bonded carbon pairs. The “graphitic” band.
  • D band, ~1350 cm⁻¹ — a defect-activated breathing mode, forbidden in perfect graphite.

In plain English

The I_D/I_G ratio is the standard quantitative measure of graphitic order. A low ratio means well-ordered graphite — high capacity, flat profile, low surface area. A high ratio means disordered or hard carbon. That one number screens incoming anode material and characterises LFP carbon coatings.

11.5XPS — surface chemistry and oxidation state

X-ray photoelectron spectroscopy irradiates a sample with X-rays of known energy and measures the kinetic energy of the ejected photoelectrons.

E_binding = hν − E_kinetic − φ

φ is the work function.

Binding energy is characteristic of both element and oxidation state, so XPS distinguishes Mn³⁺ from Mn⁴⁺ — directly relevant to the Jahn–Teller and dissolution problems of chapter 4 — and identifies LiF against Li₂CO₃ against ROCO₂Li within an SEI.

Sampling depth is only 5 to 10 nm, which makes it perfect for surface films and useless for bulk composition.

11.6EIS — frequency instead of wavelength

Electrochemical impedance spectroscopy applies a small sinusoidal voltage, typically 5 to 10 mV, across a wide frequency range from millihertz to 100 kHz, and measures the complex impedance response.

Different physical processes respond on different timescales, so they separate in frequency — the same logic as separating elements by wavelength.

Frequency rangeProcess observed
> 10 kHzBulk ohmic resistance — electrolyte, foils, welds
1–10 kHzSEI film resistance
1 Hz – 1 kHzCharge-transfer resistance (the semicircle in a Nyquist plot)
< 1 HzSolid-state diffusion (Warburg, the 45° line)

Technical framing

The practical value is that EIS separates why a cell’s resistance grew. SEI thickening looks different from lost contact, which looks different from electrolyte depletion.

The 1 kHz ACIR reading from a handheld tester is a single point on this spectrum, chosen deliberately because it captures ohmic resistance with minimal interference from the slower processes. It is a useful number and a very incomplete one.

Z′ — REAL IMPEDANCE−Z″R_ohmic>10 kHzSEI FILM1–10 kHzCHARGE TRANSFER R_ct1 Hz – 1 kHzWARBURG — SOLID-STATE DIFFUSION<1 Hz, 45° lineNYQUIST PLOT — HIGH FREQUENCY AT LEFT, LOW AT RIGHT1 kHz ACIR reads here
Figure 11.3Different physical processes respond on different timescales, so they separate in frequency. This is why EIS can tell you why resistance grew — SEI thickening looks different from lost contact, which looks different from electrolyte depletion. The single 1 kHz number your handheld tester reports is one point on this curve, chosen deliberately to capture ohmic resistance with minimal interference from the slower processes.

11.7The rest of the toolkit

  • SEM / TEM — electron microscopy for particle morphology, cracking, coating uniformity and dendrites.
  • EDS / EDX — elemental mapping within an electron microscope.
  • BET — nitrogen adsorption, measuring specific surface area in m²/g, which predicts SEI formation and first-cycle loss.
  • Mercury intrusion porosimetry and X-ray nano-CT — direct measurement of the porosity and tortuosity from chapter 9 in real electrodes.
  • DSC — differential scanning calorimetry, measuring heat flow against temperature. It identifies exothermic onset temperatures and is the basis of thermal-stability comparison between LFP and NMC.
  • ARC — accelerating rate calorimetry, adiabatic thermal runaway characterisation. It produces the onset, self-heating and runaway temperatures quoted in safety datasheets.
  • GC-MS — analysing gases evolved during formation or abuse testing.

Quick check: test yourself

1.You want to know whether nickel has invaded the lithium layer in an NMC811 powder. Which measurement, and what are you looking for?

Show answer
XRD, and the (003)/(104) peak intensity ratio. Above about 1.2 indicates good layering; below suggests Ni²⁺ occupying lithium sites. It is a standard incoming-quality check on nickel-rich cathode powder.

2.Why can’t XRD tell you where the lithium is?

Show answer
Because X-rays scatter off electrons and lithium has three, so in a lattice containing cobalt (27) and oxygen (8) it is effectively invisible. Neutrons scatter off nuclei with a scattering length unrelated to atomic number, which is why neutron diffraction is the only routine way to locate lithium and quantify Li/Ni site exchange.

3.A cell’s internal resistance has doubled. Why is a 1 kHz ACIR reading not enough to diagnose it?

Show answer
Because 1 kHz mostly captures bulk ohmic resistance. SEI film resistance appears at 1–10 kHz, charge transfer at 1 Hz–1 kHz and diffusion below 1 Hz. A full EIS sweep separates SEI thickening from lost contact from electrolyte depletion; one frequency cannot.

Chapter summary

Frequently asked questions

Why can’t X-ray diffraction see lithium?+

Because X-rays scatter off electrons and lithium has three. Its X-ray scattering factor is tiny, and in a lattice full of cobalt with 27 electrons and oxygen with 8, lithium is effectively invisible. Neutrons scatter off atomic nuclei instead, and the scattering length is unrelated to atomic number, so lithium is readily detectable — ⁶Li even has a negative scattering length. Neutron diffraction is therefore the only routine technique that directly locates lithium in a crystal structure and quantifies Li/Ni site exchange, which is why cathode research clusters around a handful of neutron sources worldwide.

What does the (003)/(104) peak ratio tell you?+

It is a quantitative measure of cation mixing in a layered oxide. A ratio above roughly 1.2 indicates good layering — lithium and metal are staying in their own sheets. Below that indicates Ni²⁺ has invaded the lithium sites, which blocks diffusion pathways and costs capacity and rate. Since the two ions are nearly the same size at 69 and 76 pm, this single XRD ratio is one of the standard incoming-quality checks on nickel-rich cathode powder.

What is the I_D/I_G ratio in Raman spectroscopy used for?+

It is the standard quantitative measure of graphitic order. The G band near 1580 cm⁻¹ comes from in-plane stretching of sp²-bonded carbon pairs; the D band near 1350 cm⁻¹ is a defect-activated breathing mode that is forbidden in perfect graphite. A low ratio means well-ordered graphite, which means high capacity, a flat voltage profile and low surface area. A high ratio means disordered or hard carbon. That one number screens incoming anode material and characterises LFP carbon coatings.

What does EIS tell you that a resistance reading does not?+

Why the resistance grew. Different physical processes respond on different timescales, so they separate in frequency: above 10 kHz you see bulk ohmic resistance from electrolyte, foils and welds; 1 to 10 kHz shows SEI film resistance; 1 Hz to 1 kHz gives charge-transfer resistance as a semicircle in the Nyquist plot; and below 1 Hz solid-state diffusion appears as a 45-degree Warburg line. SEI thickening therefore looks different from lost contact, which looks different from electrolyte depletion. The 1 kHz ACIR reading from a handheld tester is one deliberately chosen point on that spectrum.

Reviewed by

SG

Sahil Goyal

Co-founder, Wingzman

LinkedIn
SG

Sourabh Goyal

Co-founder, Wingzman

LinkedIn

The Periodic Table of the EV is an original educational series on the materials science of electric vehicles. All values are standard-condition literature figures for representative materials, not measured data from a specific product, and sources differ on several of them. Always verify against the specific material datasheet in use before making design, purchasing or certification decisions.