Hydrogen-Related Ionic Transport in Solids: Proton, Hydride, and Oxide-Ion Conduction
Hydrogen-Related Ionic Transport in Solids: Proton, Hydride, and Oxide-Ion Conduction
Hydrogen-related transport in solids is not a single mechanism. Depending on the material structure and chemical environment, hydrogen can participate in transport as a positively charged protonic defect (H+), as a negatively charged hydride ion (H−), or as part of a neutral species. In oxide ceramics, oxygen ions (O2−) also move through the anion sublattice and are often central to the same electrochemical systems.
These carriers differ in charge, local bonding, defect chemistry, migration pathway and stability conditions. A proton in an oxide is generally associated with a lattice oxygen as an OH defect; a hydride ion is an anion occupying a crystallographic site in a hydride or oxyhydride; and an oxide ion migrates through the oxygen sublattice, commonly by a vacancy-mediated process.
This article compares the three transport mechanisms, explains how to interpret conductivity measurements, and outlines experiments needed to identify the dominant mobile species rather than infer it from impedance alone.
1. The three carriers: charge and chemical identity
| Carrier | Effective charge | Typical solid-state environment | Representative materials |
|---|---|---|---|
| Protonic defect | +1 | Proton bound to lattice oxygen, commonly written OH_O• | Acceptor-doped BaZrO3- and BaCeO3-based oxides |
| Hydride ion | −1 | H− occupying an anion site in hydrides or oxyhydrides | BaH2, La2LiHO3 and related oxyhydrides |
| Oxide ion | −2 | O2− moving on the oxide anion sublattice | YSZ, doped ceria and many perovskite oxides |
The symbols H+, H− and O2− describe effective ionic charge states in the solid. They do not imply that the carriers behave as isolated gas-phase ions. Their mobility is controlled by the local crystal structure, defect population, bonding and lattice dynamics.
A crucial distinction is that proton conduction in an oxide is not equivalent to hydride-ion conduction in an oxyhydride. Both involve hydrogen nuclei, but the charge state and transport chemistry are fundamentally different.
2. Proton conduction in oxides
2.1 Proton incorporation by hydration
In many acceptor-doped perovskite oxides, acceptor substitution creates oxygen vacancies. Water vapour can be incorporated by hydration:
\[ \mathrm{H_2O(g)+V_O^{\bullet\bullet}+O_O^x \rightleftharpoons 2OH_O^{\bullet}} \]
Here, V_O•• is an oxygen vacancy, O_O× is an oxygen ion on a normal oxygen site, and OH_O• represents a proton associated with a lattice oxygen.
The hydration reaction consumes oxygen vacancies and produces protonic defects. The equilibrium depends on temperature, water-vapour partial pressure, vacancy concentration and the composition of the oxide.
2.2 Proton migration
The proton is commonly bonded to an oxygen ion as an O–H group. Long-range transport involves coupled local processes:
- Reorientation of the O–H bond around its host oxygen.
- Transfer of the proton to a neighbouring oxygen.
- Repetition of these steps to produce long-range proton displacement.
This mechanism is often described using a Grotthuss-type picture. It is not simply the diffusion of a free proton through an empty crystallographic channel.
The proton conductivity can be expressed as
\[ \sigma_{\mathrm{H^+}}=n_{\mathrm{H^+}}q\mu_{\mathrm{H^+}} \]
where n is the mobile proton concentration, q is the carrier charge magnitude and μ is the mobility. A high proton concentration does not guarantee high conductivity if the migration mobility is low.
2.3 Atmosphere dependence
Proton concentration is strongly affected by water activity. Drying a proton-conducting oxide can reduce hydration and lower proton conductivity. At sufficiently high temperature, dehydration may become significant even when water vapour is present.
Electronic defects and oxygen vacancies can coexist with protons. Consequently, a measured total conductivity may contain protonic, oxide-ion and electronic contributions.
3. Hydride-ion conduction in hydrides and oxyhydrides
3.1 What is a hydride ion?
A hydride ion is hydrogen in the formal negative oxidation state, H−. In a solid hydride or oxyhydride, it occupies an anion site and participates in the anion framework.
This differs from a protonic defect in an oxide: H− is not an OH group attached to an oxide ion. Its migration involves movement between available anion sites or through pathways defined by the host structure.
3.2 Hydride migration
Hydride-ion transport can occur through thermally activated jumps between crystallographic sites. The migration pathway depends on anion ordering, lattice dimensions, vacancy concentration and the local electrostatic environment.
In oxyhydrides, oxide ions and hydride ions may share or order over anion sublattices. Structural disorder or suitable vacancies can create connected pathways for H− migration.
Representative material families include alkaline-earth hydrides and layered or perovskite-related oxyhydrides. La2LiHO3 is an important example of a crystalline hydride-ion conductor. Other oxyhydride structures demonstrate that anion ordering and vacancy engineering can strongly influence hydride mobility.
3.3 Hydride defect notation
A hydride ion on a normal hydride site can be represented as H_H× in Kröger–Vink notation, where the reference sublattice is explicitly the hydride sublattice. A vacant hydride site may be written V_H• relative to that reference.
The notation must be defined for the chosen host lattice. One should not reuse oxygen-sublattice notation without specifying the crystallographic site and effective charge reference.
Hydride conductors require careful control of chemical potential and atmosphere. Exposure to oxidizing or moisture-containing environments may alter composition, surface chemistry or phase stability. Handling and measurement conditions must be selected for the particular compound.
4. Oxide-ion conduction
4.1 Vacancy-mediated migration
In many oxide-ion conductors, oxygen vacancies provide the sites into which neighbouring oxide ions can hop. A simplified lattice process is
\[ \mathrm{O_O^x+V_O^{\bullet\bullet} \rightleftharpoons V_O^{\bullet\bullet}+O_O^x} \]
This expression indicates an exchange of an occupied oxygen site and a vacancy; it is schematic rather than a full elementary reaction with distinct site labels.
The oxide-ion mobility depends on vacancy concentration, migration barrier, local coordination, lattice distortion and defect association.
4.2 Common material families
- Yttria-stabilized zirconia (YSZ): a widely used oxygen-ion conductor at elevated temperature.
- Doped ceria: often exhibits high oxide-ion conductivity under oxidizing conditions, with possible electronic contributions under reducing conditions.
- Perovskite oxides: can support oxide-ion transport through oxygen vacancies and may also exhibit electronic or protonic conduction depending on composition and atmosphere.
4.3 Oxygen partial pressure
The oxygen chemical potential affects vacancy and electronic-defect concentrations. Under reducing conditions, oxygen loss can be represented by
\[ \mathrm{O_O^x \rightleftharpoons \tfrac{1}{2}O_2(g)+V_O^{\bullet\bullet}+2e'} \]
This equilibrium couples oxygen nonstoichiometry to electronic conductivity. In materials with significant electronic conduction, total conductivity is not equal to oxide-ion conductivity.
5. Comparing the migration mechanisms
| Feature | Protonic defect (H+) | Hydride ion (H−) | Oxide ion (O2−) |
|---|---|---|---|
| Typical host | Hydrated oxide | Hydride or oxyhydride | Oxide ceramic |
| Local chemical state | O–H group | Hydride anion site | Oxide anion site |
| Common migration picture | Reorientation and proton transfer | H− jumps between anion sites | Vacancy-mediated oxygen jumps |
| Key chemical control | Water activity and hydration | Hydride chemical potential and anion-site structure | Oxygen partial pressure and vacancy chemistry |
| Important structural factors | O–O separation, symmetry, hydrogen bonding | Anion ordering, site connectivity, vacancies | Oxygen coordination, vacancy distribution, migration bottlenecks |
| Common experimental complication | Mixed protonic/oxide-ion/electronic conduction | Chemical instability and phase changes | Electronic leakage and grain-boundary resistance |
The table summarizes common mechanisms, not universal rules. Specific compounds may exhibit more complex pathways, multiple mobile species or changes in dominant transport with temperature and atmosphere.
6. Conductivity and Arrhenius analysis
For a thermally activated transport process, conductivity is often analysed using an Arrhenius-type relation:
\[ \sigma T=\sigma_0 \exp\left(-\frac{E_a}{k_BT}\right) \]
where σ0 is a pre-exponential factor, Ea is the apparent activation energy, kB is Boltzmann’s constant and T is absolute temperature.
A plot of ln(σT) against 1/T may be approximately linear over a restricted temperature range. The slope is related to the apparent activation energy:
\[ E_a=-k_B\,\frac{d\ln(\sigma T)}{d(1/T)} \]
For molar energy units, R replaces kB when the conductivity expression and activation energy are written consistently per mole.
The extracted activation energy may combine contributions from carrier formation and carrier migration. It should not automatically be interpreted as a pure migration barrier. Changes in hydration, defect association, phase transitions or carrier concentration can produce curvature or multiple regimes.
7. Why impedance spectroscopy alone cannot identify the carrier
Electrochemical impedance spectroscopy (EIS) measures the frequency-dependent electrical response of a sample. It can help separate bulk, grain-boundary and electrode contributions, but a semicircle or fitted resistance does not by itself identify whether the mobile species is H+, H− or O2−.
Different ionic and electronic carriers can produce similar-looking impedance spectra. Equivalent-circuit fitting is therefore not a direct chemical identification method.
A carrier assignment should be supported by complementary experiments, such as:
- Controlled variation of water-vapour partial pressure;
- Controlled variation of oxygen or hydrogen partial pressure;
- Hydrogen/deuterium isotope substitution;
- Electromotive-force measurements using suitable concentration cells;
- Hebb–Wagner polarization or other electronic-conductivity measurements;
- Transference-number measurements;
- Thermogravimetry or chemical analysis of nonstoichiometry;
- Isotope tracing and depth profiling where appropriate;
- Structural analysis before and after controlled atmosphere treatment.
The strongest interpretation combines electrical measurements with controlled chemical perturbations and independent evidence of composition or isotope movement.
8. Isotope effects: H versus D
Replacing hydrogen with deuterium changes the mass of the moving nucleus while leaving its formal charge unchanged. A measurable isotope effect can therefore provide evidence about the role of hydrogen motion.
For a simple classical vibrational picture, characteristic frequencies scale approximately with the inverse square root of mass. Real proton-conducting solids involve coupled lattice vibrations, hydrogen bonding and quantum effects, so the conductivity isotope effect is material-dependent.
A comparison between humidified H2O and D2O atmospheres can help investigate proton incorporation and migration. However, the experiment must control water activity, gas flow, equilibration time and possible isotope exchange with the apparatus.
An isotope effect alone is not sufficient to establish a unique mechanism. It should be interpreted together with defect chemistry and other transport measurements.
9. Determining ionic and electronic transference
The total conductivity may contain several contributions:
\[ \sigma_{\mathrm{total}}= \sigma_{\mathrm{H^+}}+ \sigma_{\mathrm{H^-}}+ \sigma_{\mathrm{O^{2-}}}+ \sigma_{\mathrm{e}}+ \sigma_{\mathrm{h}} \]
This is a general bookkeeping expression for a material in which these carriers are actually present. Most materials do not contain all five contributions at significant levels under one set of conditions.
The transference number of carrier i is
\[ t_i=\frac{\sigma_i}{\sigma_{\mathrm{total}}} \]
and, when all relevant contributions are included,
\[ \sum_i t_i=1 \]
A high total conductivity does not necessarily mean high conductivity of the desired ion. For an electrolyte, the relevant target is high ionic transference and low electronic leakage. For a mixed conductor, both ionic and electronic pathways may be intentionally required.
10. Practical experimental strategy to distinguish carriers
A robust experimental programme should use a controlled matrix of temperature and atmosphere.
Step 1: Establish phase and composition
Use X-ray diffraction to identify crystalline phases before and after testing. Where hydrogen occupancy or light-element positions are central to the structural model, conventional laboratory XRD may be insufficient; neutron diffraction or complementary chemical methods may be required.
Step 2: Measure impedance under controlled atmospheres
Record EIS over a suitable frequency range while controlling temperature, gas composition, humidity and equilibration time. Use identical electrode geometry and sample dimensions when comparing conditions.
Step 3: Change one chemical variable at a time
For protonic transport, vary water-vapour partial pressure while keeping other conditions fixed. For oxide-ion transport, vary oxygen partial pressure. For hydride conductors, control the hydride chemical potential and ensure that the selected atmosphere does not decompose or oxidize the sample.
Step 4: Quantify electronic conductivity
Use a suitable polarization method or independent DC measurement, taking account of electrode reactions and sample stability. A high electronic contribution can obscure the ionic response in a simple impedance measurement.
Step 5: Use isotope substitution where appropriate
Compare H-containing and D-containing conditions after sufficient equilibration. Verify that isotope exchange has occurred and that the sample has not changed phase.
Step 6: Confirm transport with independent evidence
Where feasible, use concentration cells, tracer experiments, gas analysis, thermogravimetry or post-test composition analysis. Avoid assigning a carrier based on a single fitted resistance.
11. Worked calculation: conductivity and carrier mobility
For a single dominant carrier, the Nernst–Einstein relation provides an approximate connection between conductivity and diffusivity:
\[ \sigma=\frac{nq^2D}{k_BT} \]
where n is carrier number density, q is carrier charge magnitude and D is the tracer or self-diffusion coefficient under the assumptions of the model.
Consider an illustrative material at 600 K with:
- Carrier number density \(n=1.0\times10^{26}\ \mathrm{m^{-3}}\);
- Carrier charge magnitude \(q=1.602\times10^{-19}\ \mathrm{C}\);
- Diffusion coefficient \(D=1.0\times10^{-10}\ \mathrm{m^2\,s^{-1}}\).
Then
\[ \sigma= \frac{(1.0\times10^{26})(1.602\times10^{-19})^2 (1.0\times10^{-10})} {(1.381\times10^{-23})(600)} \]
\[ \boxed{\sigma\approx3.10\times10^{-2}\ \mathrm{S\,m^{-1}}} \]
This estimate assumes a single carrier population and ideal uncorrelated transport. Correlation factors, defect association, multiple carriers and the distinction between tracer and chemical diffusion can modify the relationship. It is a consistency estimate, not a substitute for measuring conductivity.
12. Common interpretation errors
- Calling every hydrogen-related defect a proton: H+ and H− have opposite charge and different lattice chemistry.
- Equating OH_O• with a hydride ion: a protonic hydroxyl defect is not H− on an anion site.
- Assuming EIS identifies the mobile species: impedance resolves electrical response, not carrier identity by itself.
- Using total conductivity as ionic conductivity: electronic contributions must be quantified.
- Ignoring atmosphere: humidity, oxygen partial pressure and hydrogen chemical potential change defect populations.
- Treating activation energy as a pure migration barrier: measured values can include carrier-formation and association contributions.
- Ignoring grain boundaries: ceramic microstructure can strongly affect measured conductivity.
- Assuming isotope substitution is automatically conclusive: equilibration and chemical stability must be verified.
- Applying the Nernst–Einstein relation without checking assumptions: correlated motion and multiple carriers can invalidate a simple estimate.
- Comparing values measured under different atmospheres or hydration states: the carrier concentration may not be comparable.
13. Selecting the right material for an application
The desired mobile species depends on the electrochemical function.
- Protonic ceramics: useful for protonic ceramic fuel cells, steam electrolysis, hydrogen pumping and hydrogen separation when the membrane/electrode architecture is designed for proton transport.
- Hydride-ion conductors: an emerging class for solid-state electrochemical devices in which H− is the intended mobile anion. Chemical stability and compatible electrodes remain major development issues.
- Oxide-ion conductors: established in solid oxide fuel cells, electrolysis cells, oxygen sensors and oxygen-separation systems.
A material should be selected based on operating temperature, atmosphere, chemical stability, ionic transference, electrode compatibility and the required device reaction—not simply the largest conductivity value reported in the literature.
Further reading
- K. D. Kreuer, “Proton-Conducting Oxides,” Annual Review of Materials Research 33 (2003), 333–359. https://doi.org/10.1146/annurev.matsci.33.022802.091825.
- R. Merkle, M. F. Hoedl, G. Raimondi, R. Zohourian and J. Maier, “Oxides with Mixed Protonic and Electronic Conductivity,” Annual Review of Materials Research 51 (2021), 461–493. https://doi.org/10.1146/annurev-matsci-091819-010219.
- “A focused review on structures and ionic conduction mechanisms in inorganic solid-state proton and hydride anion conductors,” Materials Advances 4 (2023), 389–407. https://doi.org/10.1039/D2MA01003H.
- Y. Iwasaki et al., “Synthesis, crystal structure, and ionic conductivity of hydride ion-conducting Ln2LiHO3 (Ln = La, Pr, Nd) oxyhydrides,” Journal of Materials Chemistry A (2018). https://doi.org/10.1039/C8TA06880A.