to make an answer this question easier, could you please specify:
1) what you mean by "silicon thin film solar cell"? I.e. amorphous Si, or crystalline Si? The efficiency depends on the band gap, which in turn depends on its crystallinity in the case of Si.
2) what you mean by maximum efficiency? The theoretical limit, or the current record held by actual cells?
The efficiency of a single-junction solar cell is limited. Shockley and Queisser theoretically calculated the highest possible efficiency to be 44% (which differs if calculated using different incident photon fluxes), a limit dependent on the band gap of the active-layer semiconductor. Based on the quantum characteristics of light absorption.To circumvent the Shockley-Queisser limit and absorb a larger portion of the sunlight, a photon up conversion process can be used.
Commercially available cells span from about 7%, for low-quality a-Si cells, to a realistic 12-14% for polycrystalline Si cells. To get a somewhat better efficiency you should move to mono_sI (that is not thin film) cells, bur the cost is high...
Or you may abandon Silicon and get higher-efficiency thin film CIGS objects...
Some years ago, there was an Australian/German company named "CSG Solar" (where CSG is short for "Crystalline Silicon on Glass"; not to be confused with the Chinese "CSG PVTech"!) that fabricated highly advanced Si thin-film modules (see, e.g., www.researchgate.net/publication/224686138_CSG-1_Manufacturing_a_New_Polycrystalline_Silicon_PV_Technology or www.researchgate.net/publication/222537756_Crystalline_silicon_on_glass_CSG_thin-film_solar_cell_modules) at the "Solar Valley" production site in Germany.
The most impressive feature of these modules was the comparably high short-circuit current of (if I remember it right) nearly 30 mA/cm2; this is so remarkable because it was reached despite the very low light absorption in thin-film silicon (below 2 micron!) due to an excellent light trapping. On the other hand, the open-circuit voltage was very poor; this was due to high internal recombination losses at the laser-cut grooves separating the series-connected active cell stripes (see Fig. 1 of www.researchgate.net/publication/257975255_Qualitative_and_quantitative_evaluation_of_thin-film_solar_cells_using_solar_cell_local_characterization for a cross section sketch).
Taking this example as a reference, one can say that thin-film Si solar modules are possible, but to make them work well isn't as easy as it seems -- let alone to produce well-working modules at a competitive price.
Idris Bouchama -- unfortunately, your answer goes in the wrong direction, because the paper that you refer to is about wafer-based solar cells, whereas the question of Deepak Suthar is about thin-film silicon devices.
Thanks a lot for this recent chart. Zooming in on the right side one can see that the curve for thin-film Si (symbol: open downward blue triangle) has come to an end in the year 2014, peaking at 21.2 %, which was reached by the company Solexel. Searching for more information about this on the web I found the following article: "Revisiting thin silicon for photovoltaics: a technoeconomic perspective" (Energy Environ. Sci. 2020, 13, 12–23; https://pubs.rsc.org/en/content/articlehtml/2020/ee/c9ee02452b).
There it says that "One notable result is that Solexel achieved an efficiency of 21.2% for full-size solar cells with 35 μm thick kerfless wafers."