Perhaps the duration of pulses (10min) is a little short, and 30 min may be suitable. Morever, GITT test is better to carry out after the first cycle. You can refer this paper "Enhanced Electrochemical Performance of Fe0.74Sn5@Reduced Graphene Oxide Nanocomposite Anodes for Both Li-Ion and Na-Ion Batteries". Best wishes!

1. Refer manuals for your instruments

2. Rest time depends on a gradient

https://www.researchgate.net/publication/237896336_Phase_formation_and_electrochemical_properties_of_cryochemically_processed_Li_1_x_V_3O_8_materials

Article Phase formation and electrochemical properties of cryochemic...

Hi,

I'm not sure about your GITT parameters, but looking at your curve, you seem to be close to equilibrium, and very close to previously published voltage curves for LiV3O8 until end of discharge just after 400 on your x-axis. The remaining part of the curve is electrolyte decomposition or another irreversible reaction.

As you are close to equilibrium, and see no relaxation when the load is stopped, seem to me to indicate, that you should use a much higher load (e.g. C1).

For previously published emf ve x curves for LiV3O8 see e.g.:

https://www.researchgate.net/publication/255036331_Comparison_of_LiV3_O_8_Cathode_Materials_Prepared_by_Different_Methods

Article Comparison of LiV3 O 8 Cathode Materials Prepared by Different Methods

Thank you all for your contributions!

Keld, I followed your advice and changed the current to 1C. Please find attached the obtained curve. I am still not fully satisfied with the result as, especially when approaching 2.5V, the curve starts to show sharp needle-like voltage drops but during relaxation the potential almost reaches its original value. Are there any further improvements which I can make?

 Yes, the 1C load is clearly to much, and in the recovery after the last current spikes, you see your next problem: As the emf-curve is not smoothly sloping, but consists of a series of steps, where at least the last step (around 2.3 V vs Li) is a genuine two-phase transition, your recovery will also have inflexions, and be difficult to interpret.

Maybe your first attempt to characterise the kinetics should be a series of discharges at different loads, plot the capacity vs 1/(the load) and se at which load you reach the full capacity. This can be refined by doing it for each plateau, starting with well equilibrated electrodes. The kinetics at two-phase transitions do not follow the diffusion kinetics normally assumed in the analysis of GITT experiments.

Unfortunately this is time consuming, and require that your electrodes preparation is reproducible.

Good luck

Dear Sebastian Maletti,

in your case, we may consider an alternative galvanostatic intermittent titration technique program as a proposal. It is a "signal-based condition" schedule for both the 1)current and 2)relaxation modes, if  your software support conditions.

1) Set the end of current mode (not based on a preset time condition), but on a "dV-signal based condition" .e.g. when Vt-Vo is not very high (~0,1V) to avoid (common error of an early) pre-triggering "future phases" on the interface with electrolyte; Note : Vo is the "OCV" of the previous step.  

2) Also, set the end of relaxation mode (not at a preset time), but when a "V-Rate-signal trigger condition" .e.g. when abs(dVt/dt) < Rv,  where Rv~1mV/s. The step time will be much shorter (for the case of a possible two-phase transition) triggered on time,  i.e. on certain charge Q2p.

Thanks again!

Unfortunately, our testing software does not allow your proposed changes, Ioannis. So I have to figure out other solutions.

I am also starting to think that, as Keld put it, it is maybe not possible to record a proper GITT curve at a stretch, maybe I will try recording "small" GITT curves starting from different (equilibrated) potentials. I am aware that in the two phase region I can only obtain something like an effective diffusion coefficient, because there is, strictly spoken, no such thing as a defined diffusion coefficient. For that region I could try to check against values obtained from CV measurements.

Dear Sebastian, I'm new to electrochemical experiments. can you please suggest me the document/manual/paper with step by step procedure for GITT.

Thanks

Dear Mangesh, the GITT technique has been developed by Weppner and Huggins. I recommend to read their article as it gives you much information about the theory behind GITT as well as about its practical application:

http://jes.ecsdl.org/content/124/10/1569.short

Thanks Sebastian.