As mentioned above, all metabolic pathways leading to H2 production in green algae are extremely sensitive to O2 due to fast and irreversible inhibition of [Fe—Fe]- hydrogenase enzyme(s) and competition from different respiratory processes for the reductants. Significant efforts to surmount the O2 sensitivity issue have been made, but still the algal strain with the O2-tolerant H2 photoproduction has not been generated (Ghirardi, 2006; Ghirardi and Mohanty, 2010). Therefore, in photosynthetically active algal cells under saturating light conditions the most efficient H2 photoproduction process (via direct biophotolysis) lasts for a few seconds. For the industrial system, however, H2 photoproduction should be extended from the scale of seconds to at least days. When optimized, such process may yield H2 at a cost of around $3/kg H2 for the upper bound performance (~9% STHE) and slightly above $8/kg H2 for the near term performance (~ 1.5—2% STHE) (Blake et al., 2008; James et al., 2009). Unfortunately, at the current state sustained H2 photoproduction in photosynthetically active green algae is only possible at the expense of efficiency. For example, long-term H2 production is usually observed in cultures under very low light intensities or even in the dark when O2 evolu­tion proceeds very slowly or does not proceed at all (Kondratieva and Gogotov, 1983; Aparicio et al., 1985). Interestingly, Batyrova et al. (2012) recently observed a stable but negligible rate of H2 production in very dense cultures placed under normal light conditions. The algal cells demonstrated a high hydrogenase activity, but efficient H2 photoproduction was not observed. Most probably, under these conditions H2 evolution was driven by the cells in the inner part of the photobioreac­tor, while H2 photoproduction in the illuminated algae was limited by the coevolved O2.

One of the possibilities of driving the process under normal photosynthetic conditions is to sparge cultures continuously with inert gas, such as argon or nitrogen, that removes rapidly the PSII-evolved O2 gas (Greenbaum, 1982; Greenbaum et al., 2001). Before the sulfur-deprivation protocol was developed, this was the only way for the long-term H2 generation in algal cultures. Using a confined bioreactor, Greenbaum et al. (2001) showed several cycles of simultaneous H2 and O2 photoproduction during 1 h intervals after extensive purging the cultures in the dark for 2 h by N2. The exper­iment lasted for over 1400 h (58 days) and required peri­odic additions of CO2 gas into the photobioreactor for restoration of photosynthetic activity and replenishment of carbohydrates in algal cells. The average stoichio­metric ratio of H2 to O2 was 2.8, indicating that reducing equivalents for H2 were derived from endogenous reductants, most likely starch, as well as water. There­fore, H2 photoproduction in this case was partly driven by the cells via indirect biophotolysis pathway. Never­theless, even under extensive purging of the cultures with N2 and good mixing conditions in the photobior­eactor, the H2 evolution rate was limited due to O2 buildup in the liquid phase (Greenbaum et al., 2001). These experiments were later repeated under sulfur — deprived conditions (but with some modifications) and demonstrated a simultaneous improvement in the rate of H2 photoproduction upon declining the O2-evolving activity in algal cultures (Ghirardi et al., 2000).

The prolonged H2 evolution by green algae can also be induced by a full or partial inhibition of the water­splitting activity of PSII in cells. The full inhibition can be achieved by applying DCMU (Gfeller and Gibbs, 1984; Fouchard et al., 2005). In contrast to the N2-sparging approach, H2 photoproduction in

DCMU-treated cultures depends totally on the amount of carbohydrates or other substrates stored by the cells during the growth period and, thus, the process is limited to only one cycle. The partial inhibition of the PSII activity occurs in sulfur-deprived algae (Melis et al., 2000) and in certain mutants with manipulated expression levels of the D1 reaction center protein (Surzycki et al., 2007) or in the mutant cells affected in the PSII water-splitting complex (Makarova et al., 2005,

2007) . For the establishment of an anaerobic environ­ment and H2 production in algal cultures, the partial inhibition of the PSII activity in cells should achieve the point when the O2 produced by the PSII centers is consumed sufficiently by respiration. However, H2 photoproduction under these conditions usually proceeds at lower rates as compared to the initial rates in dark-adapted algae exposed to the light. In contrast to the full inhibition approach, several cycles of H2 pro­duction are possible (Ghirardi et al., 2000), and thus the process can be driven continuously (Laurinavichene et al., 2006, 2008; Kim et al., 2010).