NBEXP0005

Aim
To determine an initial set of fabrication parameters that will serve as a starting point for optimisation of G/TiO2 solar cells.

Method

 * Step || Time (s) || RPM || Note ||
 * 1 || 5 || 400 || Sample deposition ||
 * 2 || 5 || 400 || Solution spreading ||
 * 3 || 20 || 3000 || Drying ||

Slides
Slides containing graphene:

Slides containing only TiO2:

UV-Vis
[|UV-Vis measurements]

==

I think I mis-labeled samples 5 & 10 (5 should be labeled 10 and vice versa). If this is true, then the film absorbance (transmission) increases (decreases) with increasing number of coats, as we'd expect. The increase is reasonably regular.

Also interesting is that the wavey absorbance of FTO in the transmission plot slowly diminishes as the number of coats increases. i.e. the uniform absorbance of graphene begins to dominate over the irregular absorbance of FTO.

Band gap determination
Band gaps of films calculated from UV-Vis absorbance data using equation for indirect band-gap semiconductor found here: http://www.mri.psu.edu/facilities/MCL/techniques/UV-Vis/UV-VisTheory.asp

[|NBEXP0005 - band gap data & calculations]

Slope used to determine x intercept was determined from values between 6.0E-19 and 6.7E-19. Measurements from coats 5 & 10 were swapped as samples were mislabelled.



The band gap values are a little high, but the trend is nice. However, in these samples the ratio of G to TiO2 (0.7mg G to 4mg TiO2) was not changed — only the number of coats. The substrate is FTO, not glass, so what we're probably seeing is a transition from the band gap of FTO to the band gap of the G/TiO2 composite.

UV-Vis reflectance measurements
[|NBEXP0005 — UV-vis reflectance measurements]

Reflectance measurements of the 9 films were made using the integrating sphere attachment. The diffuse reflectance measurements (R) were transformed using the Kubelka Munk function:

math f(R)=\frac{k}{s}=\frac{(1-R)^2}{2R} math

The transformed values (K) were converted into units of energy (eV) (as per this [|procedure]) using the following relation:

math \sqrt{K\times h\nu}=f(h\nu) math

The sections of minimum and maximum slopes were determined from visual inspection, lines of best fit made their equations solved simultaneously to find their intercept to find an estimated value for the band-gap of the material. For the pure TiO2 samples, the baseline values were extrapolated from measurements between 200 and 300nm, and line of greatest gradient from the values between 325 and 365nm. The equivalent values for the graphene/TiO2 samples were 200-290nm and 325-265nm respectively.

[|NBEXP0005 — UV-vis reflectance measurements]



The bad gap values for graphene-TiO2 obtained from this procedure are as follows:


 * No. coats || Band gap (eV) ||
 * 5 || 3.896 ||
 * 10 || 3.873 ||
 * 15 || 3.874 ||
 * 20 || 3.854 ||
 * 25 || 3.849 ||
 * 30 || 3.825 ||

And from the pure TiO2 films:


 * No. coats || Band gap (eV) ||
 * 5 || 3.916 ||
 * 10 || 3.846 ||
 * 23 || 3.827 ||



Photoelectochemical measurements

Photocurrent and IV measurements were carried out in 0.1M Na2SO4. Autolab xxx running Nova v1.5 software.

To calculate photocurrent increase, baseline was first corrected using fityk. The values were plotted and current measurements in the range 50-59, 110-119 and 170-179 seconds were extracted to use for calculations. This was done to capture the stable-state values after the peak had declined and avoid any off-cycle measurements.

To do
- Add TiO2 slide images - 4-point resistivity measurements - IV curves and photocurrent measurements (do the results correlate?)