The transmission spectra of coated and uncoated glass are comparatively presented in Fig. 4. It can be observed that the light transmittance of the coated glass is significantly enhanced in comparison to the uncoated glass granting antireflective properties of the coated surface. The effect is especially prominent in the visible light range, which is important for better utilization of the incident solar irradiation by the PV units.
Download high-res image (49KB)
Download full-size image
Fig. 4. Transmission spectra of coated and uncoated glass.
The measured water drop contact angle θ on the uncoated and the coated glass surfaces is evident in Fig. 5a and b, correspondingly. The θ value was reduced from 39° on the uncoated glass to 6° on the coated surface demonstrating superhydrophilic property of the latter. This property allows the chemisorbed H2O layer on the TiO2 to attract water molecules through van der Waals forces and hydrogen bonds obstructing thus the contact between the glass surface and the adsorbed contaminants. The impurities deposited on the coated surface can be easily removed by the spreading action of water and consequently, the coated TiO2 glass surface exhibits a self-cleaning effect (Ganesh et al., 2012).
Download high-res image (37KB)
Download full-size image
Fig. 5. Contact angle between water droplet and (a) uncoated glass and (b) coated glass.
The photocatalytic degradation of Methylene Blue on the glass surfaces under UV illumination is shown in Fig. 6. It is important to note that the concentration and the amount of MB deposited on the uncoated (a) and the coated glass (c) is the same. The lighter color on the coated surface (a) is caused by the wider spreading of the liquid due to the superhydrophilic property of the coating. After 30 min irradiation, the color intensity on the coated surface was significantly decreased in comparison to the uncoated one (d). The outcome is attributed to the photocatalytic activity of the nano-sized TiO2 component of the coating. According to the well-established photocatalytic mechanism (Fujishima et al., 1999) the photo-generated electrons and holes reach the surface of the TiO2 nanoparticles and participate in the formation of highly reactive radicals which attack and decompose the organic compound to inorganic CO2 and H2O. The photocatalytic activity tests indicate that the SurfaShield G coated PV panels are expected to exhibit self-cleaning activity towards other organic contaminants reaching the glass surface and consequently increased efficiency in solar light utilization.
Download high-res image (147KB)
Download full-size image
Fig. 6. Images of Methylene Blue on: coated glass before (a) and after (b) irradiation; uncoated glass before (c) and after (d) irradiation. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
3.3. Outdoor power generation performance of PV panels
The power per minute difference (ΔPm) and the intensity of the incident solar irradiation (R) recorded during a sunny day right after coating in Attica Greece (Fig. 9a) revealed that ΔPmbetween the coated and uncoated panels during the day is constant at ∼6%, while the ΔPmreaches an increase up to 30% during the morning and the evening hours. This outcome can be attributed to the antireflective properties of the coated glass surface that utilized the sunlight irradiation with high incident angle (Fig. 9b). Since one of the components of the SSG coatingTiO2 is known to have high refractive index, the increase of the light transmittance can be explained by the relation between the refraction indexes of the two media, i.e. air and glass surface (3):(3)n1sinθ1=n2sinθ2
where n1 and n2 are the refraction indexes of the air and the glass respectively; θ1 is the angle of incidence and θ2 is the angle of refraction.
Download high-res image (319KB)
Download full-size image
Fig. 9. Outdoor performance of PV panels: (a) power per minute difference measured during the first day after coating the panel; (b) schematic presentation of light transmition through non-coated and coated glass; (c) power per minute difference measured in a cloudy day; (d) power per minute difference measured 16 days and 86 days after coating the panel.
It is known that the refractive index of the air (n1) is 1.00 and the refractive index of uncoated soda lime glass (n2) is 1.51. The measured refractive index of the coating (n2′) was 1.47 which is lower than that of the uncoated glass. Consequently, during the morning and evening, when the incident sunlight angle obtained high values, the smaller θ2′caused by lower index of the coating could ensure higher light transmition leading to increased power output.
The antireflective behavior of the coating exhibited during a sunny day in the morning and evening when sunlight is limited, can be noted during a cloudy day as well (Fig. 9c). The explanation is based on the altering of the surface roughness. The uncoated PV panel glass exhibited sub-micron roughness i.e. between 200 and 300 nm, while the roughness of the coated surface was between 40 nm and 60 nm. After application, the nanoscaled particles fill up the gaps on the glass creating nano-roughness and pores which help the panel to trap more scattered and diffused light. Therefore, for the cloudy days with higher component of scattered and diffused light, the transmition of light through the glass is higher.
The effect of dust accumulation on the power generation of the PV panels can be assessed by comparing two sunny days with time difference ∼70 days. The selected days were in June and August that is day 16th and day 83rd after coating application, respectively. The type of the ΔPm curves (Fig. 9d) is similar revealing large ΔPm values for the morning and evening hours and lower ΔPm values stable during the rest of the day. It is evident that the average 5% power difference recorded at 17/06 (day 16th) increased to 9% at 26/08 (day 83th). It must be underlined that there was no rain between the two dates which practically means that the larger difference in power generation between the coated and uncoated panels is caused by dust accumulation on the uncoated panel.
The power difference ΔPd between the two PV panels for each day of the monitored period (from June to December) is presented in Fig. 10. It can be observed that from the very first day of application (1stof June) there is an increase of 2.43%, which is undoubtedly due to the antireflective property of the coating as both of the panels were clean.
Download high-res image (206KB)
Download full-size image
Fig. 10. Outdoor performance of PV panels in Greece: (a) power difference per day for the examined period of time; (inset) picture of the uncoated and coated PV panels after rain; (b) power difference during the day where the maximum value was measured.
It is important to evaluate the change in power ΔPd with respect to the weather conditions. At the beginning of the monitored period (during the summer), the ΔPd values revealed a moderate power increase of ∼5%. Dust accumulation was observed on the uncoated panel only demonstrating the antistatic properties of the coating since no rains occurred meanwhile. Higher power increase was recorded in the beginning of September when the first rains appeared. Rain converted dust into muddy stains on the uncoated panel while it was more efficient in washing off dust on the coated panel due to the hydrophilic nature of the coating (inset in Fig. 10a). During the next months, a satisfactory gain in energy production was recorded. In December, the gained power difference was attributed to the antireflective property of the coating as it was raining continuously and both of the panels were clean.
On the 4th of October which was a cloudy day with sunlight intervals, a spike with high power difference of 19.81% was recorded (Fig. 10b). A closer look at the instantaneous power difference values during this day revealed that the phenomenon can be related to the antireflective property of the coating. Specifically, when the light intensity was low (cloudy intervals), the coated panel produced more energy reaching a maximum of 50%. On the other hand, when the light intensity was high (sunlight intervals) the power difference reached minimum evidencing that the coating transmits better the diffuse radiation.
The monitored period of time covered parts of summer, autumn and winter seasons in Greece and the recorded phenomena are expected to be observed during the rest of the year. An average of 5% power gain due to the coating was calculated for the entire period of time.
The performance of the PV arrays in Qinghai province (China) was monitored for the time period from November to March. The experimental data on the power output difference between uncoated and coated PV panels is presented in Fig. 11a. The average power difference per day (ΔPd) exhibited increasing trend for the entire period. The highest increase of the ΔPd was recorded in January that was attributed to the sandstorm occurred in the PV plant location on January 21st. The ΔPd value reached 12.43% and kept growing to 13.17% for 6 days due to the dust accumulated on the uncoated array. Although the rain on January 29th ended the continuous increase, the muddy stains observed on the untreated panels after the rain (Fig. 11b) made the power difference go up again. The consequent drop of the ΔPd to approximately 2% was associated with cleansing performed at January 31st. It should be mentioned that regular cleansing procedure (every two months) is practiced for simulating the real maintaining conditions in most of the PV plants in western China. Overall, the data collected revealed an average 6% power increase during the entire exposure period.
Fig. 11. Outdoor performance of PV panels in China: (a) power difference per day for the examined period of time; (b) picture of the uncoated and coated PV panels after rain (finger scrubbing marks clearly show the dust accumulated).