Goh Dirong the diffusion coefficient estimated by the Randles-Sevcik equation depends on the electroactive area, I do not think it is correct to do it with the geometrical area. There is a way to determine the electrode area without depending on the value of the diffusion coefficient. This is done by determining the capacitance of the double layer in cyclic voltammetry experiments. You should use the current values at different scan rates in a region of the voltammogram where there is no faradaic current contribution (dropout region before the current starts to increase in the voltammogram, see attached links). In this region the current measured corresponds mostly to the charging current of the electric double layer. You will obtain different values of current versus scan rate. You should plot I vs. v, if you get a line, the slope will be the capacitance (in Farads, F). Then you should look for the specific capacitance of your electrode material, which should be reported (F/cm2). By dividing the capacitance by the specific capacitance, you will get the electroactive area of your electrode. The drawback of this method is to find the standardized specific capacitance value of the electrode material. There are many ways to obtain the area of an electrode, but you can also look for other options. Here are some links where the way I propose is discussed; the first paper explain it better:

https://www.degruyter.com/document/doi/10.1515/zpch-2019-1514/html

Article Physical and Electrochemical Area Determination of Electrode...

https://osf.io/gyexd/download