Axial Pressure Optimization for Solid State Cells

I investigate how axial pressure controls the coupled electrochemical and mechanical response of a solid state cell. In a solid state cell, the electrode-electrolyte interface is a solid-solid contact rather than a liquid-solid contact. This solid-solid contact cannot spontaneously wet rough surfaces, so the real contact area is usually smaller than the nominal area. The resulting contact loss increases interfacial polarization, reduces the effective reaction area, and limits capacity utilization. At the same time, lithium insertion and extraction change the volume of active materials, which generates internal stress, interface stress mismatch, and possible contact degradation. Therefore, axial pressure is not merely an external mechanical boundary condition. It is a key operating variable that affects interfacial kinetics, ion transport, mechanical reliability, and stack design. I develop a coupled electrochemical-mechanical model for a solid state cell and use it to define a multi-criteria pressure window. The model treats pressure-dependent contact as a coupling path between interfacial reaction kinetics and interfacial mechanical response. I then optimize axial pressure by considering usable areal capacity, interfacial contact factor, interfacial stress difference, and pressurization cost.

Model scope and geometry

I treat the solid state cell as a one-dimensional layered domain along the thickness direction because the in-plane dimensions are much larger than the thickness in an ideal stack. I neglect edge effects, in-plane transport, and in-plane potential gradients. The solid state cell consists of a positive electrode, a solid polymer electrolyte, and a negative electrode. I denote the positive electrode, solid polymer electrolyte, and negative electrode by the subscripts \(\mathrm{p}\), \(\mathrm{e}\), and \(\mathrm{n}\). The natural thicknesses are \(l_{\mathrm{p}}\), \(l_{\mathrm{e}}\), and \(l_{\mathrm{n}}\). The cell is isothermal. Each layer is treated as a homogeneous continuum for the electrochemical and mechanical fields. This ideal layered solid state cell provides a controlled platform to analyze how axial pressure changes interfacial contact and coupled response.

Electrochemical transport and interfacial reaction in the solid state cell

I describe lithium transport in the positive and negative electrodes with mass conservation and a stress-coupled flux. For \(k\in\{\mathrm{p},\mathrm{n}\}\),

$$ \frac{\partial c_k}{\partial t}+\frac{\partial j_k}{\partial x}=0 $$

$$ j_k=-D_k\left(1-\frac{c_k}{c_k^{\max}}\right)\left(\frac{\partial c_k}{\partial x}-\frac{\Omega_k c_k}{RT}\frac{\partial \sigma_k}{\partial x}\right) $$

Here, \(c_k\) is the lithium concentration, \(c_k^{\max}\) is the maximum lithium concentration, \(D_k\) is the lithium diffusion coefficient, \(\Omega_k\) is the partial molar volume, \(\sigma_k\) is the hydrostatic stress, \(R\) is the gas constant, and \(T\) is temperature. The stress term couples mechanical pressure and chemical expansion into lithium transport. In the solid polymer electrolyte, I use a modified Nernst-Planck description for lithium ions:

$$ \frac{\partial c_{\mathrm{e}}}{\partial t}+\frac{\partial j_{\mathrm{e}}}{\partial x}=0 $$

$$ j_{\mathrm{e}}=-D_{\mathrm{e}}\left(\frac{\partial c_{\mathrm{e}}}{\partial x}-\frac{\Omega_{\mathrm{e}} c_{\mathrm{e}}}{RT}\frac{\partial \sigma_{\mathrm{e}}}{\partial x}+\frac{F c_{\mathrm{e}}}{RT}\frac{\partial \Phi_{\mathrm{e}}}{\partial x}\right) $$

where \(c_{\mathrm{e}}\) is the lithium concentration in the electrolyte, \(D_{\mathrm{e}}\) is an effective diffusion coefficient, \(\Omega_{\mathrm{e}}\) is an effective partial molar volume, \(\Phi_{\mathrm{e}}\) is the electrolyte potential, and \(F\) is Faraday’s constant. I define the effective electrolyte transport properties as

$$ D_{\mathrm{e}}=\frac{2D_{+}D_{-}}{D_{+}+D_{-}} $$

$$ \Omega_{\mathrm{e}}=\Omega_{+}+\Omega_{-} $$

where \(D_{+}\) and \(D_{-}\) are cation and anion diffusion coefficients, and \(\Omega_{+}\) and \(\Omega_{-}\) are their partial molar volumes. Charge conservation in the solid polymer electrolyte is written as

$$ \frac{\partial i_{\mathrm{e}}}{\partial x}=0 $$

At each electrode-electrolyte interface, I use Butler-Volmer kinetics. For a symmetric charge-transfer coefficient, the interfacial current density is

$$ i_k=2i_{k,0}\sinh\left(\frac{F\eta_k}{2RT}\right),\quad k\in\{\mathrm{p},\mathrm{n}\} $$

The exchange current densities are

$$ i_{\mathrm{p},0}=Fk_{\mathrm{p}}c_{\mathrm{e}}^{\beta_{\mathrm{p}}}\left(c_{\mathrm{p}}^{\max}-c_{\mathrm{p},s}\right)^{1-\beta_{\mathrm{p}}}c_{\mathrm{p},s}^{\beta_{\mathrm{p}}} $$

$$ i_{\mathrm{n},0}=Fk_{\mathrm{n}}c_{\mathrm{e}}^{\beta_{\mathrm{n}}}\left(c_{\mathrm{n}}^{\max}-c_{\mathrm{n},s}\right)^{1-\beta_{\mathrm{n}}}c_{\mathrm{n},s}^{\beta_{\mathrm{n}}} $$

where \(k_{\mathrm{p}}\) and \(k_{\mathrm{n}}\) are reaction rate constants, \(\beta_{\mathrm{p}}\) and \(\beta_{\mathrm{n}}\) are charge-transfer coefficients, and \(c_{\mathrm{p},s}\) and \(c_{\mathrm{n},s}\) are surface concentrations. I define state of charge for the solid state cell as

$$ \mathrm{SOC}(t)=\frac{\int_{l_{\mathrm{p}}+l_{\mathrm{e}}}^{l_{\mathrm{p}}+l_{\mathrm{e}}+l_{\mathrm{n}}}c_{\mathrm{n}}(x,t)\,dx}{\int_{l_{\mathrm{p}}+l_{\mathrm{e}}}^{l_{\mathrm{p}}+l_{\mathrm{e}}+l_{\mathrm{n}}}c_{\mathrm{n}}^{\max}\,dx} $$

I decompose the terminal voltage of the solid state cell into reference, concentration, stress, interfacial overpotential, and electrolyte potential contributions:

$$ V_{\mathrm{cell}}=V_{\mathrm{ref}}+V_{\mathrm{con}}+V_{\mathrm{str}}+V_{\mathrm{ove}}+V_{\mathrm{ele}} $$

Voltage component Expression Physical role
Reference voltage \(V_{\mathrm{ref}}=3.9\ \mathrm{V}\) Baseline thermodynamic level
Concentration contribution \(V_{\mathrm{con}}=\frac{RT}{F}\ln\left(\frac{c_{\mathrm{n},s}(c_{\mathrm{p}}^{\max}-c_{\mathrm{p},s})}{c_{\mathrm{p},s}(c_{\mathrm{n}}^{\max}-c_{\mathrm{n},s})}\right)\) Composition-dependent open-circuit shift
Stress contribution \(V_{\mathrm{str}}=\frac{1}{F}\left[\Omega_{\mathrm{p}}\sigma_{\mathrm{p},s1}-\Omega_{\mathrm{n}}\sigma_{\mathrm{n},s2}+\Omega_{\mathrm{e}}(\sigma_{\mathrm{e},s2}-\sigma_{\mathrm{e},s1})\right]\) Mechanical contribution to voltage
Interfacial overpotential \(V_{\mathrm{ove}}=\eta_{\mathrm{p}}-\eta_{\mathrm{n}}\) Kinetic polarization at interfaces
Electrolyte potential \(V_{\mathrm{ele}}=\Phi_{\mathrm{e},s1}-\Phi_{\mathrm{e},s2}\) Potential drop in electrolyte

I apply zero lithium flux at the outer current collectors. At the positive electrode-electrolyte interface and the negative electrode-electrolyte interface, the lithium flux is linked to the interfacial current density. The electrolyte current density is continuous with the interfacial reaction current. These boundary conditions close the electrochemical problem for the solid state cell.

Mechanical model and chemical strain

I assume mechanical equilibrium in each layer:

$$ \nabla\cdot\boldsymbol{\sigma}+\mathbf{b}=0 $$

Axial pressure is applied along the thickness direction. Because a solid state cell in a constrained stack is often restricted in the in-plane direction, I impose zero in-plane strain and prescribe the thickness-direction stress by the axial pressure:

$$ \sigma_{xx}=-P,\quad \epsilon_{yy}=\epsilon_{zz}=0 $$

The total strain is decomposed into elastic and chemical parts:

$$ \boldsymbol{\epsilon}=\boldsymbol{\epsilon}^{\mathrm{el}}+\boldsymbol{\epsilon}^{\mathrm{chem}} $$

For each layer, I use the chemical strain

$$ \epsilon_k^{\mathrm{chem}}=\frac{1}{3}\Omega_k(c_k-c_k^0),\quad k\in\{\mathrm{p},\mathrm{e},\mathrm{n}\} $$

where \(c_k^0\) is the initial lithium concentration. The constitutive relation gives the stress state under the combined action of axial pressure and concentration change. For the positive electrode, electrolyte, and negative electrode, I compute the layer strain and then the total thickness change:

$$ \Delta l(P,t)=l_{\mathrm{p}}\epsilon_{\mathrm{p}}(P,t)+l_{\mathrm{e}}\epsilon_{\mathrm{e}}(P,t)+l_{\mathrm{n}}\epsilon_{\mathrm{n}}(P,t) $$

I define the maximum thickness variation during operation as

$$ H(P)=\max_t \Delta l(P,t)-\min_t \Delta l(P,t) $$

This quantity measures the breathing amplitude of the solid state cell under a given axial pressure. It is useful for checking whether thickness variation itself should limit the pressure window.

Pressure-dependent interfacial contact

The solid-solid interface in a solid state cell is rough at the microscale. The real contact area \(A\) is smaller than the nominal contact area \(A_0\). I introduce the interfacial contact factor

$$ \gamma(P)=\frac{A}{A_0} $$

I describe \(\gamma(P)\) using a Persson-type contact model. The contact factor increases with axial pressure and saturates as the real contact area approaches the nominal area. I write the contact factor as

$$ \gamma(P)=\int_{q_L}^{q_0}\operatorname{erf}\left(\sqrt{\frac{P}{E^*}}\frac{q}{\xi}\right)C(q)\,dq $$

where the effective modulus is

$$ \frac{1}{E^*}=\frac{1-\nu_1^2}{E_1}+\frac{1-\nu_2^2}{E_2} $$

and the roughness power spectral density is represented as

$$ C(q)=10^{m}q^{I} $$

with lower wave-number cutoff

$$ q_L=\frac{2\pi}{A_0} $$

and a diffusion-related amplification parameter

$$ \xi=\sqrt{\frac{4A_0}{D_{\mathrm{e}}t}} $$

The parameters \(m\) and \(I\) characterize the roughness of the interface. I treat the positive electrode-electrolyte interface as a film-type contact and the negative electrode-electrolyte interface as a bulk-type contact. I then couple the contact factor to the interfacial exchange current density:

$$ i_{\mathrm{p},0}^{\mathrm{eff}}=\gamma_{s1}i_{\mathrm{p},0} $$

$$ i_{\mathrm{n},0}^{\mathrm{eff}}=\gamma_{s2}i_{\mathrm{n},0} $$

This coupling means that a lower real contact area reduces the effective exchange current density and increases interfacial polarization in the solid state cell. I also use the contact factor to correct the elastic load-bearing part of the interfacial stress. The effective stresses at the positive interface and negative interface are written as

$$ \sigma_{\mathrm{p},s1}^{\mathrm{eff}}=\frac{P}{\gamma_{s1}}+\sigma_{\mathrm{p},s1}^{\mathrm{chem}} $$

$$ \sigma_{\mathrm{e},s1}^{\mathrm{eff}}=\frac{P}{\gamma_{s1}}+\sigma_{\mathrm{e},s1}^{\mathrm{chem}} $$

$$ \sigma_{\mathrm{e},s2}^{\mathrm{eff}}=\frac{P}{\gamma_{s2}}+\sigma_{\mathrm{e},s2}^{\mathrm{chem}} $$

$$ \sigma_{\mathrm{n},s2}^{\mathrm{eff}}=\frac{P}{\gamma_{s2}}+\sigma_{\mathrm{n},s2}^{\mathrm{chem}} $$

I define the interfacial stress differences as

$$ \Delta\sigma_{s1}^{\mathrm{eff}}=\left|\sigma_{\mathrm{p},s1}^{\mathrm{eff}}-\sigma_{\mathrm{e},s1}^{\mathrm{eff}}\right| $$

$$ \Delta\sigma_{s2}^{\mathrm{eff}}=\left|\sigma_{\mathrm{n},s2}^{\mathrm{eff}}-\sigma_{\mathrm{e},s2}^{\mathrm{eff}}\right| $$

These stress differences represent the mechanical mismatch across the solid-solid interfaces. They are important because they indicate where contact damage, local yielding, or interfacial degradation may begin in a solid state cell.

Parameters and numerical implementation

I solve the coupled model in a finite-element framework. I use a fully coupled solver and a direct linear solver. The relative tolerance is \(0.005\), and the tolerance factor is \(0.1\). For pressure scanning, I use a step of \(0.3\ \mathrm{MPa}\) below \(4\ \mathrm{MPa}\) and \(1\ \mathrm{MPa}\) above \(4\ \mathrm{MPa}\). The solid state cell is charged at \(1\ \mathrm{A}/\mathrm{m}^2\) to \(4.03\ \mathrm{V}\), and then discharged at the same current density until \(\mathrm{SOC}=0.05\). The main parameters are listed in the table.

Parameter Symbol Positive electrode Solid polymer electrolyte Negative electrode
Thickness \(l\) \(550\ \mathrm{nm}\) \(2000\ \mathrm{nm}\) \(200\ \mathrm{nm}\)
Maximum concentration \(c^{\max}\) \(23300\ \mathrm{mol}/\mathrm{m}^3\) \(1500\ \mathrm{mol}/\mathrm{m}^3\) \(30555\ \mathrm{mol}/\mathrm{m}^3\)
Initial concentration \(c^0\) \(22135\ \mathrm{mol}/\mathrm{m}^3\) \(1200\ \mathrm{mol}/\mathrm{m}^3\) \(1527.75\ \mathrm{mol}/\mathrm{m}^3\)
Reaction rate constant \(k\) \(1.53\times10^{-11}\) — \(5.03\times10^{-11}\)
Diffusion coefficient \(D\) \(1.76\times10^{-15}\ \mathrm{m}^2/\mathrm{s}\) \(D_{+}=2.5\times10^{-13}\), \(D_{-}=3.0\times10^{-13}\ \mathrm{m}^2/\mathrm{s}\) \(8.0\times10^{-14}\ \mathrm{m}^2/\mathrm{s}\)
Partial molar volume \(\Omega\) \(1.92\times10^{-6}\ \mathrm{m}^3/\mathrm{mol}\) \(\Omega_{+}=3.95\times10^{-6}\), \(\Omega_{-}=1.46\times10^{-4}\ \mathrm{m}^3/\mathrm{mol}\) \(4.17\times10^{-6}\ \mathrm{m}^3/\mathrm{mol}\)
Charge-transfer coefficient \(\beta\) \(0.5\) — \(0.5\)
Young’s modulus \(E\) \(70\ \mathrm{GPa}\) \(0.5\ \mathrm{GPa}\) \(10\ \mathrm{GPa}\)
Poisson ratio \(\nu\) \(0.20\) \(0.40\) \(0.30\)

I use a temperature of \(300\ \mathrm{K}\) and a reference concentration of \(1\ \mathrm{mol}/\mathrm{m}^3\). The nominal contact area is \(1\ \mathrm{cm}^2\). I assign \(m=-2.77\) and \(I=-9.35\) to the positive interface, and \(m=-3.08\) and \(I=-6.61\) to the negative interface. These values represent different roughness characteristics in the solid state cell.

Model verification and consistency

I verify the electrochemical behavior against a reference voltage-SOC response for the same layered material system when the contact factor is unity. The comparison gives a coefficient of determination of \(0.9931\) and a maximum absolute voltage error of \(0.0310\ \mathrm{V}\). This agreement indicates that the electrochemical part of my solid state cell model captures the main charge-discharge behavior. I also compare the area-scaling exponent of interfacial impedance. My linear coupling of contact factor to exchange current density gives an area coefficient of \(-1.00\). Experimental studies report values between \(-1.13\) and \(-1.05\), so the response order is consistent. For the pressure sensitivity of charge-transfer resistance, I compute a normalized resistance of \(0.7041\) when the pressure ratio is \(1.2/0.43\). The corresponding experimental value is \(0.7326\). The difference is small enough to support the pressure-contact coupling used in my model.

Verification item My model result Reference behavior
Voltage-SOC coefficient of determination \(0.9931\) Close to unity
Maximum absolute voltage error \(0.0310\ \mathrm{V}\) Acceptable for coupled modeling
Area-scaling exponent of interfacial impedance \(-1.00\) \(-1.13\) to \(-1.05\)
Normalized charge-transfer resistance at pressure ratio \(1.2/0.43\) \(0.7041\) \(0.7326\)

Electrochemical response to axial pressure

Axial pressure affects the solid state cell mainly through interfacial contact. As pressure increases, the contact factor at both electrode-electrolyte interfaces first rises rapidly and then saturates. The positive interface tends to reach good contact at lower pressure, while the negative interface requires higher pressure to approach full contact. I summarize representative contact behavior in the table.

Axial pressure \(P\) (\(\mathrm{MPa}\)) Positive interface \(\gamma_{s1}\) Negative interface \(\gamma_{s2}\) Minimum contact factor \(\gamma_{\min}\)
0.1 0.22 0.05 0.05
1.0 0.63 0.20 0.20
4.0 0.95 0.45 0.45
10.0 0.996 0.697 0.697
12.2 0.998 0.747 0.747
15.8 0.999 0.808 0.808
20.0 0.999 0.860 0.860
50.0 1.000 0.980 0.980

The discharge voltage-capacity response shifts toward higher voltage and higher capacity as the contact factor increases. When the contact factor is small, the effective reaction area is limited. Under a fixed current density, the same current must pass through a smaller interfacial area, so the interfacial charge-transfer overpotential increases. The terminal voltage drops earlier, and the accessible capacity decreases. When the contact factor is large, the effective reaction area increases, the exchange current density increases, and the interfacial overpotential decreases. As a result, the solid state cell can maintain a higher terminal voltage over a wider SOC range. This is the main electrochemical benefit of axial pressure.

I decompose the terminal voltage into \(V_{\mathrm{con}}\), \(V_{\mathrm{str}}\), and \(V_{\mathrm{ove}}\) to identify the dominant pressure effect. The concentration contribution is mainly controlled by SOC and changes little with pressure at a fixed SOC. The stress contribution increases slightly with pressure and SOC, but its magnitude remains a correction term. The interfacial overpotential contribution is the most pressure-sensitive component. As pressure increases, contact improves, the effective exchange current density increases, and \(V_{\mathrm{ove}}\) decreases significantly. Therefore, the improvement in voltage and capacity utilization mainly comes from reduced interfacial polarization. The table gives representative shares of voltage components.

SOC Axial pressure \(P\) (\(\mathrm{MPa}\)) \(V_{\mathrm{con}}\) share \(V_{\mathrm{str}}\) share \(V_{\mathrm{ove}}\) share
0.20 0.1 20% 5% 75%
0.20 10.0 45% 8% 47%
0.20 50.0 60% 10% 30%
0.45 0.1 18% 7% 75%
0.45 10.0 38% 12% 50%
0.45 50.0 52% 15% 33%

At higher SOC, the overpotential share is larger because the electrochemical and transport states of the solid state cell become more demanding. The stress contribution also becomes more visible at higher SOC because chemical strain grows with lithium concentration change. However, the stress contribution remains smaller than the concentration and overpotential contributions. The key result is that axial pressure improves the solid state cell primarily by increasing real interfacial contact and reducing interfacial overpotential, not by directly changing the concentration contribution.

Mechanical response to axial pressure

The internal stress distribution depends strongly on SOC. As SOC increases, the positive electrode tends toward higher tensile stress, while the negative electrode develops larger compressive stress. The electrolyte layer shows a smaller stress amplitude. This behavior results from the different elastic moduli, partial molar volumes, and concentration changes of the layers. At a fixed SOC, the direct effect of axial pressure on the through-thickness stress distribution is moderate. The stress state is therefore not controlled only by external pressure; it is controlled by the combination of external pressure, chemical strain, and material property mismatch.

The maximum interfacial stress difference increases with axial pressure. This increase is non-linear. At low pressure, the stress difference rises quickly because real contact is incomplete and the external load is carried by a small contact area. At higher pressure, contact improves and the growth of the stress difference slows. Nevertheless, the stress difference continues to increase because the pressure itself rises and because the solid state cell can access higher SOC states. The positive electrode-electrolyte interface generally shows a larger stress difference than the negative interface. This makes the positive interface the more critical location for mechanical mismatch in my model. The table summarizes representative values.

Axial pressure \(P\) (\(\mathrm{MPa}\)) Usable areal capacity \(Q\) (\(\mathrm{Ah}/\mathrm{m}^2\)) Minimum contact factor \(\gamma_{\min}\) Maximum interfacial stress difference \(S\) (\(\mathrm{MPa}\)) Maximum thickness variation \(H\) (\(\mathrm{nm}\))
0.1 0.082 0.05 210 4.0
1.0 0.108 0.20 320 6.5
4.0 0.123 0.45 480 8.0
10.0 0.1316 0.697 569.51 9.8
12.2 0.1342 0.747 574.93 10.1
15.8 0.1368 0.808 579.25 10.4
20.0 0.1380 0.860 583 10.8
50.0 0.1395 0.980 592 11.5

The maximum thickness variation \(H\) increases strongly at low pressure and then flattens at higher pressure. Low pressure limits the accessible SOC range, so the lithium concentration change and chemical strain are small. As pressure improves contact and reduces polarization, the solid state cell operates over a wider SOC range, and the thickness variation increases. Once contact and capacity utilization approach saturation, the additional pressure produces little increase in thickness variation. The absolute magnitude of \(H\) remains on the nanometer scale compared with the total cell thickness. Therefore, thickness breathing is a useful response indicator, but it is not the main limiting factor at high pressure. The interfacial stress difference is a more relevant mechanical constraint for the solid state cell.

Multi-criteria pressure window and optimization formulation

I define four normalized indicators for axial pressure optimization: usable areal capacity, minimum interfacial contact factor, maximum interfacial stress difference, and pressurization cost. The capacity is computed from the discharge current density and cutoff time:

$$ Q(P)=\frac{i_{\mathrm{app}}t_{\mathrm{cut}}(P)}{3600} $$

I normalize capacity as

$$ Q_n(P)=\frac{Q(P)-Q_{\min}}{Q_{\max}-Q_{\min}} $$

The minimum contact factor is

$$ \gamma_{\min}(P)=\min\left(\gamma_{s1}(P),\gamma_{s2}(P)\right) $$

and its normalized form is

$$ \gamma_n(P)=\frac{\gamma_{\min}(P)-\gamma_{\min}^{\min}}{\gamma_{\min}^{\max}-\gamma_{\min}^{\min}} $$

The maximum interfacial stress difference is

$$ S(P)=\max\left(\Delta\sigma_{s1}^{\mathrm{eff}}(P),\Delta\sigma_{s2}^{\mathrm{eff}}(P)\right) $$

with normalized form

$$ S_n(P)=\frac{S(P)-S_{\min}}{S_{\max}-S_{\min}} $$

The pressurization cost is represented by the normalized pressure:

$$ P_n(P)=\frac{P-P_{\min}}{P_{\max}-P_{\min}} $$

I set \(P_{\min}=0.1\ \mathrm{MPa}\) and \(P_{\max}=50\ \mathrm{MPa}\). I then define a composite loss function:

$$ J(P)=J_Q+J_\gamma+J_S+J_P $$

The components are

$$ J_Q=\omega_Q\left[1-Q_n(P)\right]^2 $$

$$ J_\gamma=\omega_\gamma\left[1-\gamma_n(P)\right]^2 $$

$$ J_S=\omega_S S_n(P) $$

$$ J_P=\omega_P P_n(P) $$

where \(\omega_Q\), \(\omega_\gamma\), \(\omega_S\), and \(\omega_P\) are weight coefficients. I use three weight schemes to represent different engineering preferences. The performance-priority scheme emphasizes capacity and contact recovery. The balanced scheme treats capacity, contact, stress risk, and pressurization cost together. The conservative scheme places more weight on interfacial stress difference and pressure cost. The weights are listed in the table.

Scheme \(\omega_Q\) \(\omega_\gamma\) \(\omega_S\) \(\omega_P\)
Performance priority 0.35 0.40 0.15 0.10
Balanced 0.30 0.35 0.20 0.15
Conservative 0.25 0.30 0.25 0.20

The optimal pressure for a given scheme is

$$ P^*=\arg\min_{P_{\min}\le P\le P_{\max}}J(P) $$

Because the loss function depends on the full coupled solid state cell model, I use a derivative-free particle swarm algorithm for bounded search. Each candidate pressure is evaluated by solving the full model. I also use a deterministic method combining coarse full-range scanning with one-dimensional bounded optimization to confirm the result. The particle swarm update equations are

$$ v_i^{k+1}=\omega v_i^k+c_1r_1\left(p_i^k-P_i^k\right)+c_2r_2\left(g^k-P_i^k\right) $$

$$ P_i^{k+1}=P_i^k+v_i^{k+1} $$

where \(P_i^k\) is the pressure position of particle \(i\), \(v_i^k\) is its velocity, \(p_i^k\) is its personal best, and \(g^k\) is the global best. I use 15 particles, 9 iterations, inertia weight \(0.65\), individual and group learning factors \(1.45\), and a velocity limit of \(8\ \mathrm{MPa}\). For the deterministic method, I use shape-preserving piecewise cubic Hermite interpolation for the normalized capacity, contact, and stress-difference curves. I perform a coarse scan from \(0.1\ \mathrm{MPa}\) to \(50\ \mathrm{MPa}\) with a maximum interval of \(1\ \mathrm{MPa}\). Around each local minimum candidate, I apply a one-dimensional bounded optimizer. The pressure convergence tolerance is \(10^{-3}\ \mathrm{MPa}\).

Optimization results and pressure window

The normalized capacity and contact factor both increase with pressure, but their growth rates differ. Capacity rises rapidly at low pressure and flattens after about \(10\ \mathrm{MPa}\). The minimum contact factor continues to increase over a wider pressure range because the poorer interface requires higher pressure to recover contact. The normalized stress difference and pressure cost increase monotonically. Therefore, increasing pressure cannot simultaneously maximize electrochemical benefit and minimize mechanical or engineering cost. The optimized pressures are shown in the table.

Scheme Particle swarm optimum (\(\mathrm{MPa}\)) Direct scan optimum (\(\mathrm{MPa}\)) Usable capacity \(Q\) (\(\mathrm{Ah}/\mathrm{m}^2\)) \(\gamma_{\min}\) \(S\) (\(\mathrm{MPa}\))
Performance priority 15.8 15.7 0.1368 0.8079 579.25
Balanced 12.2 12.4 0.1342 0.7471 574.93
Conservative 10.0 9.8 0.1316 0.6971 569.51

The two optimization methods differ by no more than \(0.2\ \mathrm{MPa}\) for all three schemes. This confirms that the result is not strongly dependent on the specific search algorithm. The composite loss function reaches a minimum in an intermediate pressure range. At low pressure, capacity and contact are insufficient, so \(J_Q\) and \(J_\gamma\) dominate. As pressure increases, \(J_Q\) and \(J_\gamma\) decrease rapidly. At higher pressure, capacity and contact gains saturate, while the interfacial stress-difference term \(J_S\) and the pressurization-cost term \(J_P\) continue to increase. The balanced scheme has a clear minimum near \(12.2\ \mathrm{MPa}\). The performance-priority scheme shifts the optimum to \(15.8\ \mathrm{MPa}\), and the conservative scheme shifts it to \(10.0\ \mathrm{MPa}\). Therefore, I define the trade-off axial pressure window as \(10.0\) to \(15.8\ \mathrm{MPa}\), with \(12.2\ \mathrm{MPa}\) as the balanced optimum for the solid state cell.

I also evaluate marginal benefit and marginal cost. I define the electrochemical benefit as the average of normalized capacity and normalized contact recovery:

$$ B_n(P)=\frac{Q_n(P)+\gamma_n(P)}{2} $$

I define the mechanical and pressure cost as the average of normalized stress difference and normalized pressure:

$$ C_n(P)=\frac{S_n(P)+P_n(P)}{2} $$

I compute the changes in \(B_n\) and \(C_n\) for an additional \(1\ \mathrm{MPa}\) of axial pressure. At low pressure, the marginal electrochemical benefit is large. As pressure increases, the marginal benefit decreases rapidly. The marginal cost remains positive over a wide pressure range. Near \(12\ \mathrm{MPa}\), the marginal benefit and marginal cost are close. Above about \(16\ \mathrm{MPa}\), the marginal cost exceeds the marginal benefit. This supports the conclusion that further pressure increase is not efficient for the solid state cell. The pressure window \(10.0\) to \(15.8\ \mathrm{MPa}\) captures the region where electrochemical gains and mechanical or engineering costs are reasonably balanced.

Sensitivity of the optimal pressure window

I test whether the pressure window is robust under different weight combinations, normalization intervals, loss function forms, and current densities. First, I sample many weight combinations from a Dirichlet distribution and retain those that satisfy broad bounds for the four weights. I evaluate the optimal pressure for 5000 valid samples. The distribution is centered in the intermediate pressure range. The percentiles are listed in the table. About \(63.56\%\) of the samples fall inside the \(10.0\) to \(15.8\ \mathrm{MPa}\) window. The median optimum is \(11.10\ \mathrm{MPa}\), which is close to the balanced optimum of \(12.2\ \mathrm{MPa}\).

Percentile of optimal pressure Value (\(\mathrm{MPa}\))
5% 7.12
25% 9.10
50% 11.10
75% 13.81
95% 17.92
Fraction inside \(10.0\) to \(15.8\ \mathrm{MPa}\) 63.56%

Second, I change the normalization interval. The original interval is \(0.1\) to \(50\ \mathrm{MPa}\). I also test an upper-limit extension of \(20\%\) and a \(5\%\) to \(95\%\) percentile interval. The resulting windows remain in the intermediate pressure range. The table summarizes the results.

Normalization interval Trade-off optimal pressure window
\(0.1\) to \(50\ \mathrm{MPa}\) \(10.0\) to \(15.8\ \mathrm{MPa}\)
\(0.1\) to \(60\ \mathrm{MPa}\) \(10.8\) to \(17.1\ \mathrm{MPa}\)
\(2.6\) to \(47.5\ \mathrm{MPa}\) \(11.8\) to \(17.9\ \mathrm{MPa}\)

Third, I test different loss function forms. I use the original form, a fully quadratic form, a normalized Huber form with threshold \(0.5\), and an exponential form. The optimal windows shift slightly but remain in the intermediate pressure range and overlap with the original window. The table gives the results.

Loss function form Trade-off optimal pressure window
Original form \(10.0\) to \(15.8\ \mathrm{MPa}\)
Fully quadratic form \(11.5\) to \(16.7\ \mathrm{MPa}\)
Normalized Huber form \(12.8\) to \(17.5\ \mathrm{MPa}\)
Exponential form \(11.0\) to \(18.6\ \mathrm{MPa}\)

Fourth, I vary the charge and discharge current density. Lower current density shifts the optimal window slightly toward lower pressure, but the values remain in the intermediate range. The table lists the current-density sensitivity.

Current density (\(\mathrm{A}/\mathrm{m}^2\)) Trade-off optimal pressure window
1.000 \(10.0\) to \(15.8\ \mathrm{MPa}\)
0.135 \(8.8\) to \(14.0\ \mathrm{MPa}\)
0.270 \(9.3\) to \(14.9\ \mathrm{MPa}\)
0.406 \(9.6\) to \(15.4\ \mathrm{MPa}\)
0.541 \(9.8\) to \(15.6\ \mathrm{MPa}\)
0.744 \(9.9\) to \(15.8\ \mathrm{MPa}\)

These sensitivity results show that the optimal pressure window is not an artifact of one weight set or one numerical choice. The solid state cell consistently favors an intermediate pressure range where contact is substantially improved and electrochemical gains are still meaningful, while excessive stress mismatch and pressurization cost are avoided.

Engineering load implication for a solid state cell stack

For a large-format solid state cell, the pressure window must be translated into a stack preload. I consider a cell effective area of \(120\ \mathrm{mm}\times80\ \mathrm{mm}\), which is \(0.0096\ \mathrm{m}^2\). The total preload is

$$ F=PA $$

The pressure window \(10.0\) to \(15.8\ \mathrm{MPa}\) corresponds to \(96.0\) to \(151.7\ \mathrm{kN}\). The balanced pressure \(12.2\ \mathrm{MPa}\) corresponds to \(117.1\ \mathrm{kN}\). These values are large and place demands on end plates, fasteners, and fixture stiffness. However, they are within the range used in high-pressure solid state cell fixtures. A uniformity indicator for cell surface pressure can be defined as

$$ H_u=\frac{\Delta P_{\mathrm{std}}}{P_{\mathrm{op}}} $$

where \(\Delta P_{\mathrm{std}}\) is the standard deviation of surface pressure and \(P_{\mathrm{op}}\) is the operating pressure. For a well-designed fixture in the \(10\) to \(30\ \mathrm{MPa}\) range, the average pressure uniformity indicator can be about \(2.16\%\). This suggests that the pressure window is practically meaningful if the end plate and clamping structure are designed properly. Nevertheless, my result is based on a material parameter set, an ideal layered structure, and an isothermal assumption. Application to a full automotive solid state cell or module requires further structural design and validation.

Quantity Value
Cell effective area \(120\ \mathrm{mm}\times80\ \mathrm{mm}=0.0096\ \mathrm{m}^2\)
Pressure window \(10.0\) to \(15.8\ \mathrm{MPa}\)
Total preload window \(96.0\) to \(151.7\ \mathrm{kN}\)
Balanced pressure \(12.2\ \mathrm{MPa}\)
Balanced total preload \(117.1\ \mathrm{kN}\)
Typical pressure uniformity indicator \(2.16\%\)

Discussion

My results show that axial pressure has a dual role in a solid state cell. On the electrochemical side, pressure improves solid-solid contact, increases the effective reaction area, raises the effective exchange current density, reduces interfacial overpotential, and increases usable capacity. On the mechanical side, pressure increases interfacial stress mismatch and pressurization cost. The benefits saturate, while the costs continue to accumulate. This is why a pressure window rather than a single high pressure is the appropriate design target. The solid state cell should be operated at a pressure that is high enough to maintain contact and low enough to avoid excessive mechanical and structural burden.

The positive electrode-electrolyte interface is the more critical interface in my model. Its contact factor recovers at lower pressure than the negative interface, but its interfacial stress difference is larger and continues to grow at higher pressure. This asymmetry arises from differences in elastic modulus, partial molar volume, and chemical strain between the positive electrode, electrolyte, and negative electrode. Therefore, pressure design for a solid state cell cannot be based only on the average contact factor. The weaker contact side and the higher-stress side must both be considered.

The thickness variation result is also informative. The maximum thickness variation increases at low pressure because the accessible SOC range expands. At higher pressure, the variation flattens. Its nanometer-scale amplitude is small relative to the total thickness. Thus, thickness breathing is not the main reason to limit pressure in my model. Interfacial stress difference and stack preload cost are more restrictive. This distinction is useful when translating a solid state cell pressure window into a module or pack design.

My model has several limitations. I treat the solid state cell as an ideal layered structure with uniform properties and isothermal operation. I do not explicitly simulate interfacial damage, debonding, void growth, or fracture. I use a pressure-dependent contact factor rather than a full rough-surface contact simulation. The model also does not include end-plate stiffness, module mass, thermal gradients, or pressure non-uniformity. These factors can shift the optimal pressure window. Future work should include interface strength, fracture energy, allowable stress thresholds, temperature effects, high-rate operation, overcharge and overdischarge, and abnormal pressure conditions. Experimental validation of interfacial mechanical response in a solid state cell would also help refine the model.

Conclusions

I developed a coupled electrochemical-mechanical model for a solid state cell that includes axial pressure, pressure-dependent interfacial contact, stress-coupled transport, Butler-Volmer interfacial kinetics, and interfacial stress mismatch. I used the model to analyze how axial pressure affects contact, electrochemical output, internal stress, interfacial stress difference, and thickness variation. I then formulated a multi-criteria optimization problem with usable areal capacity, minimum interfacial contact factor, maximum interfacial stress difference, and pressurization cost.

My main conclusions are as follows. First, axial pressure improves electrochemical performance mainly by increasing real interfacial contact and reducing interfacial overpotential. The capacity and terminal voltage improve as pressure rises, but the gains diminish after contact approaches saturation. Second, the maximum interfacial stress difference increases with pressure and is larger at the positive electrode-electrolyte interface. This makes interfacial stress mismatch a key mechanical cost of high pressure. Third, the maximum thickness variation increases at low pressure and then flattens. Its absolute magnitude is small, so it is not the dominant pressure-limiting factor in my model. Fourth, under the specified material parameters, ideal layered structure, and isothermal condition, the trade-off axial pressure window is \(10.0\) to \(15.8\ \mathrm{MPa}\). The balanced optimum is \(12.2\ \mathrm{MPa}\). This window balances capacity utilization, interfacial contact recovery, interfacial stress difference, and pressurization cost for the solid state cell.

These results provide a modeling and optimization framework for pressure regulation and interfacial reliability analysis of solid state cells. They also show that pressure selection should not aim only at maximizing electrochemical performance. A proper pressure window must consider solid-solid contact, electrochemical kinetics, interfacial mechanics, and engineering preload cost together. The same framework can be extended to multilayer solid state cell stacks, real rough interfaces, damage evolution, thermal effects, and module-level structural constraints.

Scroll to Top