Nickel Oxide (NiO) nanoparticles (particle size ~20 nm, 99.9%, Informant Advanced Materials) were ball-milled in ethanol with few droplets of acetate. The mixture of above colloidal solution, ethyl cellulose (Aldrich) and terpinol anhydrous (≥99.5%, Fluka) were sonicated and stirred alternatively to obtain a fine dispersion. A paste was made by evaporating the ethanol from the mixture on a rotary evaporator. FTO glass (Nippon sheet glass, resistance 13 Ω/square) were coated with nickel acetate (+98%, Alfa Aesar) ethanol (≥99.7%, Merck) solution (0.05M) by dip coating and subsequently dried before screen print [30] [31].

SECM experiments were performed on cell contained a Pt wire counter electrode and a Pt wire quasi-reference electrode. Positioning was performed with an x-y-z stepper motor system. A Pt wire radius 12.5 µm was sealed into a 5 cm Pyrex glass capillary under vacuum. The UME was polished and shaped conically by a wheel with 180-grid Carbimet paper disks and micro polishing cloths with 0.3 μm for 3 minuties. The UME was sharpened to RG ~ 10, where RG is the ratio of the diameters of the glass sheath and the Pt wire. Before each experiment, the UME was polished with 0.3 μm powder, rinsed with water and ethanol. The sensitized NiO electrode was placed at the bottom of a small volume electrochemical cell and short circuited by a Pt wire to the electrolyte. The emission spectra of the LEDs compared with the absorption spectrum of NiO/C343 and NiO/CW1 film. The LEDs were placed close to the cell and focused on the dye-sensitized film by an objective lens (so that the photo-illuminated spot had a diameter of about (0.0785 cm²) [32] [33] [34] [35].

According to various applications SECM has been a prevailing technique for probing interface kinetics [36] [37]. SECM feedback mod based on an ultra microelectrode (UME) to substrate and subsequent its current response as a function of the distance from the surface provide dye regeneration kinetics in sensitized solar cells [19] [38] [39] [40] [41]. In feedback mode scanning probe based on the motion of ultra microelectrode (tip) close to the surface of conductive or non-conducting (insulting) substrate. For the UME, tip reaction for charge transfer reaction and the study state current of given by Equation (1a) and Equation (1b) respectively

O + n e − → R (1a)

I T ∞ = 4 n F ( k eff κ ) C r T 2 (1b)

where, F: Faraday’s constant, n is number of electrons transferred in reaction Equation (1b), k_eff reaction rate constant, C is concentration of electrolyte, r_T is UME radius and κ, normalized rate constant.

The photon induces electrochemical reaction on the NiO/dye surface a change of the tip current as the UME approaches the interface. In order to scrutinize the kinetics of dye regeneration we measured SECM current-distance curves of dye sensitized NiO film with blue illumination in different wavelengths. The effect of illumination intensity on the kinetics of dye regeneration was studied by measuring approach curves at different J_hv, its value increases clearly as the J_hv is increased. The kinetics of regeneration by the electrolyte was studied using SECM feedback mode approach on the NiO films with two dyes of C343 and CW1 different intensity as documented in Table 1 and Table 2.

Figure 2(a) shows the normalized curve at UME to CW1/NiO under blue LED illumination photon flux increased from 2.2 × 10⁻⁹ mol·cm⁻²·s⁻¹ to 22.4 × 10⁻⁹ mol·cm⁻²·s⁻¹. The rate constant k_eff increased from 1.45 × 10⁻³ to 8.18 × 10⁻³ cm·s⁻¹ in blue LED for CW1/NiO. Figure 2(b) shows the normalized curve at NiO/C343 under illumination with blue LED at different intensities. The rate constant k_eff of NiO/C343, increased from 0.92 × 10⁻³ to 4.93 × 10⁻³ cm·s⁻¹ in blue LED as documented in Table 1.

Figure 3(a) shows the normalized approach curve on C343/NiO under red illumination flux density increased from 2.12 × 10⁻⁹ mol·cm⁻²·s⁻¹ to 14.7 × 10⁻⁹ mol·cm⁻²·s⁻¹ its rate constant k_eff increased from 2.43 × 10⁻³ to 7.39 × 10⁻³ cm·s⁻¹. Figure 3(b) shows the approach curve on NiO/CW1 under illumination with red illumination at different intensities rate constant increased from 1.82 × 10⁻³ to 5.64 × 10⁻³ cm·s⁻¹ in red LED as documented in Table 2.

Generally, according to several studies illuminated dye excited (D^*) injects a hole into the valence band (VB) of the P-type semiconductor succeed to the reduction of the dye (D⁻). There are a number of reactions mechanism of dye regeneration in P-type DSSCmost likely by 2a-e [42] [43]. The regeneration of the dye at the dye-sensitized electrode-electrolyte interface. Therefore, at the illuminated D-sensitized NiO electrode-electrolyte-UME probe [44] [45]:

NiO / D + hv → D * / NiO : excitation (2a)

NiO / D * → hv D − → h + / NiO : chargrseparation (2b)

NiO ( h + ) + D − → NiO / D : germinaterecombination (2c)

NiO ( h + ) / D − + I 3 − → D / NiO ( + ) + I 2 − + I − : regeneration (2d)

3 I − − e − → I 3 − : regenerationontheUME (2e)

when illuminated dye is excited there is charge separation and a hole injection into the NiO valence band, and the dye is reduced. Then dye will react with the oxidized species of the electrolyte ( I 3 − ) and regenerate to its ground state. Under illumination excited dye injects holes to valance band NiO oxide (charge separation) represented by Equation (2b) heterogeneous electron transfer (regeneration) I 3 − / D − Equation (2d). In order to rationalize influence of light on dye regeneration we performed measurement at two different dyes in blue LED in different intensity as shown in Figure 2 & Figure 3.

Figure 4 shows the plot k_eff vs J_hν for CW1 and C343 sensitized (a) in blue illumination and (b) in red illumination. An experiment values of k_red = 6.95 × 10⁵ mol⁻¹·cm³·s⁻¹ and Φ_hv(λ) = 3.16 × 10⁶ cm²·mol⁻¹ for C343 blue, LED, respectively. An experiment values of k_red = 7.95 × 10⁵ mol⁻¹·cm³·s⁻¹ and Φ_hv(λ) = 3.32 × 10⁶ cm²·mol⁻¹ for CW1 in the blue, LED, respectively Table 3. When the sensitized NiO film back-illuminated, the ground state dye is denoted by D, photo-reduced dyes D⁻ and photo excited dye molecules D^*. The mathematical expression for excited dye D^* can be derivative from the mass conservation and the steady-state approximations for surface concentrations of the photo excited dye represented by Equation (3a) the reduced dye Equation (3b)

∂ Γ D * ∂ t = ϕ hv J hv Γ D − k inj Γ D * (3a)

∂ Γ D − ∂ t = k red Γ D − [ I 3 − ] S 1 / 2 + k inj Γ D * (3b)

Thus, the rate constants k_red for the regeneration of the reduced dye after injection the hole into the valance band of P-type semiconductor by the reduced state of redox at incident light of given by Equation (4). A kinetic analysis rate constant k_eff, for regeneration processes expressed in terms of reduction rate constant k_red, absorption cross-section of dye ϕ_hv, thickness of sample L, dye concentration on NiO film D^o and electrolyte concentration [I⁻] as

1 k red = 2 L D o ( [ I − ] ϕ hv J hv + 1 k eff ) (4)

where, k_eff represents the regeneration rate, and Φ_hv absorption rate.

Due to dependence of regeneration on illumination wave length, result in different values of k_eff of the different wavelengths. Under illumination of the thin film with blue LED incident photon fluxes the rate constants extracted by fitting k_red of 7.95 × 10⁵ mol⁻¹·cm³·s⁻¹ for CW1 Φ_hv of 3.3 × 10⁶ cm²·mol⁻¹ and 6.73 × 10⁵ mol⁻¹·cm³·s⁻¹ for C343, Φ_hv of 2.3 × 10⁶ cm²·mol⁻¹.

A number of studies on film thickness state to macroscopic characteristics of dye sensitized solar cells parameters depend on the rate of charge transfer reactions. SECM feedback analysis permitted to investigate dye regeneration kinetics at a microscopic sample [46] [47] [48] [49]. The film thickness has implicitly affected on the performance of dye sensitized solar cells which increase the accumulation probability, which promotes more light absorption [19] [20]. In order to investigate the effect of dye thickness absorption spectra of C343 adsorbed on the porous NiO films of varied thickness as shown in (Table 4). In this work, we used six samples with t thickness (2.8 μm, 34 μm, 3.8 µm, 4.4 µm 5.4 μm, and 5.8 μm respectively). Figure 5(a) shows absorption spectra of C343 adsorbed on the porous NiO films in different thicknesses. The spectra clearly show broad absorption bands peaking at approximately 490 nm, superimposed to a background signal due to scattering of light by the nickel oxide film. This proves the successful saturation of the films with the dye of the pores for all films. The film thickness is in line with the trend of peak heights in the solid-state absorption on the surface NiO spectra, pick highest for 5.8 μm and minimum spectra pick for 2.8 μm.

Figure 5(b) shows the normalized SECM approach curves recorded with Pt UME (r_T = 12.5 mm) approaching to C343/NiO films of thickness of 38 μm illuminated at different J_hν. The k_eff value obtained at the illuminated C343/NiO film, increases as the J_hv increased. Increasing J_hv increased hence k_eff, as shown in Table 3. As J_hv increased from 2.21 × 10⁻⁹ mol·cm⁻²·s⁻¹ to 22.4 × 10⁻⁹ mol·cm⁻²·s⁻¹, the cross ponding increased from 1.87 × 10⁻³ cm·s⁻¹ to 7.73 × 10⁻³ cm·s⁻¹. The result revealed that the normalized approach curves depend on the thickness of the film. SECM approach curves of the other films recorded on supporting information. Figure 5(d) shows the L (10⁻⁶ m) vs k_red,Φ_hv for six samples as shown Table 5. As film thickness increases the absorption crossection increases because when thickness of sample increases the dye deposition increase, it provides probability of much light absorption as shown Figure 5(d). In contrary as film

thickness increases the regeneration rate constant decrease due to, the dye concentration increases on surface of film, the reaction rate on the sample surface is faster than the UME reaction.