A calculated capacitive contribution of 99% at low scan rates almost certainly stems from flaws in the fitting procedure or data selection, rather than the material truly exhibiting near-pure capacitive behavior. The Dunn method is extremely sensitive to data quality at low scan rates, and a common systematic error can artificially inflate the result to nearly 100%.
Most probable cause: fitting using a single point or an inappropriate potential window
The standard workflow of the Dunn method is to select multiple potential points on the CV curve and perform a linear fit of \(i(V)/\nu^{1/2}\) against \(\nu^{1/2}\) for each potential. The slope yields \(k_1\) and the intercept yields \(k_2\). If only one point near the current peak is selected, or a potential where the current approaches zero (such as near the endpoints of the CV curve), the fitted \(k_1\) and \(k_2\) will be severely distorted. At low scan rates, the absolute current magnitude is small and the signal-to-noise ratio is poor, which amplifies the error from single-point fitting and may produce an anomalous capacitive contribution close to 100%.
Verification: Ensure you select a sufficient number of points uniformly across the potential window (typically dozens or more), and that each point is fitted using data from all scan rates. Results are unreliable if only one or two points are used, or if fitting is performed solely with data from low scan rates.
Second possible cause: an overly narrow scan rate range or incorrect entry of scan rate values
The Dunn method requires good linearity of \(i(V)/\nu^{1/2}\) versus \(\nu^{1/2}\). If the scan rate range used for fitting spans only 5–30 mV/s, this narrow interval makes the fitted slope \(k_1\) and intercept \(k_2\) highly sensitive to data fluctuations, which tends to artificially overestimate the fraction of total current attributed to \(k_1\nu\). In the literature, the Dunn method commonly covers a broad scan rate range from 0.1–1 mV/s up to 100 mV/s or higher. Fitting outcomes become unreliable when low-scan-rate data points are scarce or noisy.
Verification: Confirm consistent units for scan rate (mV/s instead of V/s), and check whether raw \(\nu\) values or unit-converted values are substituted into the fitting formula. A frequent trivial mistake is substituting 0.005 for 5 mV/s, which completely breaks the magnitude balance between \(k_1\nu\) and \(k_2\nu^{1/2}\).
Third possible cause: breakdown of the underlying formula assumptions for low-scan-rate CV curves
The Dunn equation \(i(V)=k_1\nu+k_2\nu^{1/2}\) assumes the total current is a linear superposition of surface-controlled current (proportional to \(\nu\)) and diffusion-controlled current (proportional to \(\nu^{1/2}\)). Significant ohmic drop, leakage current or side reactions at low scan rates violate this superposition assumption and spuriously raise the fitted \(k_1\). Longer charging durations at low scan rates amplify the cumulative effects of side reactions and leakage current.
Verification: Inspect the shape of low-scan-rate CV traces. Distortions such as pronounced bulges, aggravated asymmetry, or current values that fail to decrease regularly with falling scan rate indicate that the raw data is unsuitable for direct Dunn-method analysis.
Rapid troubleshooting checklist
1. Check the number of potential points: fit separately for at least 20 potential points; avoid relying only on peak positions.
2. Validate scan rate units: substitute scan rates as numerical values in mV/s (e.g., 5, 10, 20, 30), not converted values in V/s.
3. Examine fitting \(R^2\): the coefficient of determination \(R^2\) for the linear fit at each potential should be close to 1. Points with low \(R^2\) violate the model assumptions and should be discarded or noted.
4. Cross-check using the b-value: calculate the slope b from a plot of \(\log i\) against \(\log \nu\). A b value near 1 confirms dominant capacitive response; for b between 0.5 and 0.8, a 99% capacitive contribution at low scan rates is physically implausible and points to computational error.
If none of the above issues apply, plot the raw scatter data of \(i(V)/\nu^{1/2}\) versus \(\nu^{1/2}\) for the low scan rates. If the points deviate markedly from a straight line, the Dunn method is not valid for this scan rate range, and the 99% value carries no physical meaning.