Photofermentative Poly-3-Hydroxybutyrate Production by Rhodopseudomonas sp. S16-VOGS3 in a Novel Outdoor 70-L Photobioreactor

: In the present study, the performance of a 70 L photobioreactor, operating outdoors, was investigated using a purple bacterial strain as Rhodopseudomonas sp. S16-VOGS3 for producing poly-3-hydroxybutyrate (PHB). The novel photobioreactor was equipped with 5 rows L-shaped; the bottom of every row was placed in a stainless-steel tank containing water with controlled temperature. The photofermentation trials were carried out under fed-batch mode and under a semi-continuous regimen using lactic acid as the carbon source. The effect of the irradiance and the carbon/nitrogen ratio on the PHB accumulation was investigated, in order to evaluate the optimal bacterial growth. The results showed the feasibility of the prototype photobioreactor for the production of PHB by Rhodopseudomonas sp. S16-VOGS3 under the natural light/dark cycle. During the fed-batch growth (144 h long), the cumulative PHB increased quickly reaching a maximum value of 377 mg/L and decreased to 255 mg/L during the semi-continuous regimen (336 h long).


Introduction
Global petroleum-based plastic production passed 320 million tons (Mt) worldwide in 2016 and it is expected to double in the next 20 years [1]. The United Nations Environment Program (UNEP) showed that only the 54% of plastic waste went to landfills or recycled but the rest entered the ocean and accumulated as debris every year, representing a global problem for the marine ecosystems [2]. Globally, 5 to 13 Mt of plastics end up in the oceans every year [3]. Better recycling and the replacement of the petrol plastics with bioplastics (BPs), not only compostable but also biodegradable in different natural environments, could be a concrete answer to this serious environmental problem. The biodegradability of BPs in different natural environments, such as soil and marine/fresh water, is a key property that makes their life cycle more eco-sustainable compared to the conventional plastics. Despite the environmental benefits of BPs, their actual worldwide production is only about 1 Mt/y [1] but their demand is continuously growing and, in accordance with the last market data collected by European Bioplastics, the global BP production capacity is expected to increase from around 2 Mt/y in 2017 to around 2.4 Mt/y in 2022, in absence of government support measures, and to higher values with specific government incentives [4]. The main limitations that penalize the BPs market are the competition for the use of renewable materials for the food and energy sectors and their in batch mode and then semi-continuous mode, using lactic acid as carbon source. The aim of this study was to evaluate the performance of the novel photobioreactor in terms of biomass growth and PHB accumulation under the natural sunlight conditions and to investigate the effect of the irradiance (direct and diffuse) and carbon/nitrogen (C/N) ratio. The extracted PHB from Rhodopseudomonas lyophilized-biomass produced was characterized in terms of thermal properties and stability in view of its use as thermoplastic polymer.

Photobioreactor
The photo of the L-shaped row photobioreactor (L-SRP) used for the photofermentative PHB production is reported in Figure 1a. It is equipped with 5 parallel rows (Pyrex ® glass make) and the total working volume is 70 L. Every row is a L-shaped loop with 4-pipes laid horizontally in a stainless-steel tank that contains demineralized water at controlled temperature and 8-pipes placed one over the other and connected with U fittings as shown in the Figure 1a. The photobioreactor configuration with the 5 L-shaped loops permitted a good culture temperature control realized by the 4 pipes of each loop submerged in water, reducing the temperature excursion range experienced by bacterial cells. This solution avoided the use of an external refrigeration system. (direct and diffuse) and carbon/nitrogen (C/N) ratio. The extracted PHB from Rhodopseudomonas lyophilized-biomass produced was characterized in terms of thermal properties and stability in view of its use as thermoplastic polymer.

Photobioreactor
The photo of the L-shaped row photobioreactor (L-SRP) used for the photofermentative PHB production is reported in Figure 1a. It is equipped with 5 parallel rows (Pyrex ® glass make) and the total working volume is 70 L. Every row is a L-shaped loop with 4-pipes laid horizontally in a stainless-steel tank that contains demineralized water at controlled temperature and 8-pipes placed one over the other and connected with U fittings as shown in the Figure 1a. The photobioreactor configuration with the 5 L-shaped loops permitted a good culture temperature control realized by the 4 pipes of each loop submerged in water, reducing the temperature excursion range experienced by bacterial cells. This solution avoided the use of an external refrigeration system. The pipes and U-fittings have inner and outer diameter of 2.6 and 3.0 cm, respectively. All 5 rows are connected with 2 collectors: the first placed at the bottom and the second on the top of the culture system. The photobioreactor is joined with a vessel. A pump, provided with 3 flat blades  The pipes and U-fittings have inner and outer diameter of 2.6 and 3.0 cm, respectively. All 5 rows are connected with 2 collectors: the first placed at the bottom and the second on the top of the culture system. The photobioreactor is joined with a vessel. A pump, provided with 3 flat blades placed at 120 • to one another on the propeller shaft, is used to recirculate the culture in the system: the culture flows into the bottom collector which has 5 pots: one for each loop, then from the bottom up along each line until reaching the top collector. From the top collector, the culture returns to the vessel, where the pH is regulated at 7.0 ± 0.5 by adding 0.5mM HCl. The photobioreactor used in the present work derives from previous studies [16,17] on temperature controlled underwater tubular photobioreactor ( Figure 1b) operating in outdoor conditions to growth Rhodopseudomonas palustris 42OL.

Microorganism Strain and Culture Conditions
In the present study, Rhodopseudomonas sp. S16-VOGS3, belonging to the culture collection of the Istituto per lo Studio degli Ecosistemi (Consiglio Nazionale delle Ricerche (ISE, CNR), Florence, Italy), was employed. The 16S sequence of Rhodopseudomonas sp. S16-VOGS3 was deposited in GenBank under the following accession numbers: KU899101-KU899105.
Rhodopseudomonas sp. S16-VOGS3 cells were grown initially indoors and, then, outdoors in the 70-L photobioreactor. The bacterium was cultured in batch mode into 8-Sovirel bottles containing 100 mL medium at constant temperature of 30 ± 0.2 • C under continuous artificial irradiance of 75 W/m 2 in anaerobic conditions. After 7 days of growth, Rhodopseudomonas cells were harvested by centrifugation (9000 rpm) and transferred into two photobioreactors with 4 L working volume (not shown here) containing fresh medium. The growing conditions were the following: fed-batch mode, continuous irradiance of 150 W/m 2 , constant temperature (30 ± 0.2 • C) and anaerobic conditions obtained by fluxing argon (99.99% pure). A 150 W OSRAM power-star HQI-TS lamp was used to light the culture; the light intensity was measured using a Quantum/Radiometer/Photometer. Successively, cells of Rhodopseudomonas sp. S16-VOGS3 were used to inoculate the 70 L photobioreactor positioned outdoors containing fresh medium. A total of 1L of growth medium used for both indoor and outdoor bacterial growth contained 4.0 g lactic acid, 0.5 g NH 4 Cl, 1.0 g KH 2 PO 4 , 0.4 g NaCl, 0.05g CaCl 2 2H 2 O, 0.4 g MgSO 4 7H 2 O, g 0.005 ferric citrate and 0.1 mg para-aminobenzoic acid. Trace elements were provided by adding 10 mL (per 1 L) of a solution containing 30 mg H 3 BO 3 , 3.0 mg MnCl 2 4H 2 O, 500 mg Na 2 MoO 4 2H 2 O, 10 mg ZnSO 4 7H 2 O, 1.0 mg CuCl 2 2H 2 O, 2.0 mg NiCl 2 6H 2 O and 24.5 mg CoNO 3 6H 2 O. The pH of the medium was adjusted to 6.8 by using NaOH solution. During the outdoor bacterial growth, the pH was adjusted by adding HCl solution.

Outdoor Bacterial Growth
The L-SRP was inoculated in the evening by using the total volume of Rhodopseudomonas sp. S16-VOGS3 culture grown indoors. No argon was used for obtaining the anaerobic conditions since these conditions were reached two hours and half after the inoculation due to the dark natural conditions. The L-SRP was operated under fed-batch (FB) mode along the 144 h of biomass growth. A second experiment was carried out under a semi-continuous (SC) regimen, adopted to maintain the cell dry-weight concentration (CDWC) of about 4.0 g/L. Moreover, the medium replacement was accomplished every time (culture time of 72 h) adding to the culture an adequate amount of lactic acid to restore its initial concentration of 4.0 g/L. In this way, the unsuitable situation that could have stopped growth of Rhodopseudomonas sp. S16-VOGS3 due to lack of nutrients, such as the carbon source, is avoided. The culture samples were collected through a sampling point with respect to time to measure the CDWC, residual substrate (lactic acid) concentration and intracellular PHB concentrations in accordance with the analytic methods reported below. The investigation was carried out under the natural light/dark cycle in the summer period: 27 June-18 July 2017 at Sesto Fiorentino (Florence, Italy) (Latitude: 43 • 49 54 N; Longitude: 11 • 11 57 E; Altitude above sea level: 60 m).

PHB Recovery
Bacterial cells were separated from culture suspension by centrifugation at 9000 rpm and, then, the pellet was suspended in a physiological solution (9‰) to wash it and, then, the suspension was centrifuged at 9000 rpm. The washing operation was repeated twice before collecting the final biomass that was frozen and, then, lyophilized to be used for the subsequent PHB extraction. The extraction was carried out in accordance with Al-Kaddo et al. [18]: about 2 g of lyophilized cells were added to 200 mL of chloroform (CHCl 3 ) and the suspension was kept under stirring at room temperature for 6 days. Thereafter, the solution was filtered under vacuum utilizing a paper filter (Whatman No. 3) to remove cell debris and the filtrate was concentrated in a rotary evaporator to around 25 mL and PHB was precipitate by addition of 200 mL of cold methanol under vigorous stirring. Wet PHB was collected by centrifugation and then it was air dried, weighted and used for the thermal analyses.

Analytical Methods
Cell dry-weight concentration (CDWC) and bacteriochlorophyll (Bchl) concentration were determined in accordance with Carlozzi and Sacchi [16]. The PHB content accumulated in the bacterial cells was measured by HPLC (High-Performance Liquid Chromatography) using the Thermo Finnigan-Spectra System 6000 LP (Thermofinnigan, San José, CA, USA), following the procedure reported elsewhere [15]. The organic acid concentration in the bacteria cultures was determined as reported elsewhere [19] by using the HPLC (Thermo Finnigan-Spectra System 6000 LP) with a C18 analytical column (250 × 4.6 mm) operating at 25 • C. The cells were separated from the culture suspension by disposable syringe filter units (MFS-13 mm, pore size 0.45 µm) and the supernatant sample was injected into the C18 analytical column for the analysis of the lactic acid concentration. An aqueous solution 0.1 wt.% H 3 PO 4 was used as the mobile phase at a flow rate of 1.0 mL/min. All analyses were carried out in triplicate.

Thermal Analysis
Differential scanning calorimetry (DSC) and thermo-gravimetric analysis (TGA) were used to study the thermal properties of the recovered PHB in view of its use as thermoplastic polymer. DSC was performed using a DSC Q200 (TA Instruments, Waters LLC, New Castle, DE, USA) calibrated with standard indium. PHB samples (about 12 mg) were first heated from ambient temperature to 200 • C at a heating rate of 10 • C/min in order to erase any previous thermal history. After an isotherm at 200 • C for 5 min, the samples were cooled until ambient temperature and heated again to 200 • C with an increasing temperature rate of 10 • C/min. The glass transition temperature (T g ), melting temperature (T m ) and enthalpy (∆H m ) were determined from the second heating thermogram. The degree of crystallinity x c of PHB was calculated using the following equations: where ∆H m is the melting enthalpy of the sample (J/g); ∆H m,100 is the enthalpy value for a theoretically 100% crystalline PHB (146 J/g) [20].
The thermal stability of the polymeric samples produced by Rhodopseudomonas sp. S16-VOGS3 was determined by TGA using a TA Q-500 (TA Instruments, Waters LLC, New Castle, DE (USA)). About 10 mg of sample was loaded into an aluminum pan and heated from room temperature to 600 • C at a heating rate of 10 • C/min under nitrogen atmosphere. The thermal stability of the polymer was determined measuring the decomposition temperature (T d ), defined as a temperature corresponding to a sample weight loss of 5%.

Photobioreactor Performance
The improved geometry configuration of the novel L-SRP tested in the present study is evident and significant respect to some typical horizontal tubular photobioreactors. The verticalization of the cultural system allows the light to reach full pipe surface. In accordance with a previous investigation [21], spacing properly the rows of the 70-L photobioreactor, the light distribution improved inside Rhodopseudomonas culture causing a type of culture lamination that permitted the dilution of solar radiation (SR). In fact, the lamination of the culture can increase the irradiated surface (Si) and consequently the Si/V ratio of the photobioreactor and control the photic ratio by spacing or narrowing the parallel rows. Table 1 shows the internal diameter (di) and the Si/V ratio of several tubular photobioreactors proposed in literature [17,[21][22][23][24]: the lower the internal diameter the higher the Si/V ratio. With a di of 0.010 m, a Si/V ratio of 400 m -1 was reached [24]. A recent tubular photobioreactor with a Si/V ratio of 182 m -1 was identified as the best culture system among those tested by Palamae et al. [23]. They reported that the Si/V ratio affects the production of biomass and hydrogen differently: a high Si/V ratio favored the hydrogen production over biomass production when growing Rhodobacter sphaeroides S10.  Figure 2 shows the profiles of the daily maximum irradiance (Imax) and solar irradiation (SR) and Bchl, lactate and PHB concentrations in the bacterial culture during the cultivation of Rhodopseudomonas sp. S16-VOGS3 in the outdoor L-SRP under fed-batch mode (run1). After a lag-phase growth period, lasted about 48 h, during which Bchl concentration was stable and the irradiance had a fluctuating behavior (Figure 2a,b), it was observed a continuous increase of the Bchl amount up to double its concentration (3.8 g/L culture ) in the subsequent 96 h. During the fed-batch growth (lasted 144 h), lactic acid was consumed and, consequently, added twice (after 72 and 124 h) to restore its initial concentration to 4 g/L (Figure 2c). The lactate concentration restoring operations caused significant heightening of the C/N ratio in the medium since the concentration of ammonia nitrogen (NH 3 -N) in the culture broth decreased progressively over time (Figure 2c). It is known that stressed growth conditions caused by a high C/N ratio and/or nutrient limitation (e.g., Nitrogen and Phosphorus) in the culture medium promote the accumulation of intracellular PHB [25][26][27][28]. As expected, the PHB concentration in the culture grew over time and, more significantly, when the initial lactate concentration was restored (Figure 2d). In particular, PHB concentration augmented of 32% in the last period of bacterial growth when the amount of NH 3 -N was totally depleted and the growth occurred under nitrogen starvation. After 144 h of batch cultivation, a final amount of 377 mg(PHB)/L culture was reached corresponding to a PHB volumetric productivity of about 63 mg L −1 day −1 . Recently, a PHB production of 10.3 mg/L/day was reported by growing (indoors) Synechocystis PCC6803 [29]. In the present investigation, at the end of Rhodopseudomonas sp. S16-VOGS3 fed-batch growth (run 1), the CDWC was 2.92 g/L, which corresponded to the total dry In order to evaluate the effect of the solar light intensity on the productivity of PHB accumulated in the bacterial cells, a further investigation was carried out by covering about 80% of the photobioreactor with a dark cloth as shown in Figure 3a (run 2). In this way, the light intensity impinging on the L-SRP was significantly reduced and the culture was irradiated principally by diffuse irradiance, with the exception of the U-fittings. Figure 3b shows typical daily trends of the direct and diffuse irradiance on a horizontal plane measured at Sesto Fiorentino (Florence, Italy) in the period June-July 2017. The results obtained during the 336 h of Rhodopseudomonas growth under diffuse radiation (run 2) are shown in Figure 4a-d. During the first 192 h, the photobioreactor was operated in FB mode; when the Bchl concentration reached the value of 16.4 mg/Lculture and the CDWC 4.0 g/L, the regimen of growth was changed from FB to SC mode (Figure 4a). In order to maintain the CDWC ranging from 3.2 to 4.1 g/L, an appropriate culture volume was withdrawn from the reactor and replaced with an equal volume of fresh medium every 72 h. During the 336 h of cultivation, the maximum light intensity was quite constant and ranged from 900 to 956 W/m 2 ; on the contrary, the global solar radiation (SR) changed significantly from 22.3 to 29.8 MJ/m 2 day, as shown in Figure 4b. The lowest reached lactate content was about 1.0 g/L and its initial concentration (4.0 g/L) was restored when the fresh culture medium was added in accordance with the SC regimen of growth (Figure 4c). The NH3-N amount decreased progressively and reached zero after 164 h of cultivation. The accumulated PHB concentration decreased significantly during the first period of bacterial growth; subsequently, in the SC regimen the PHB concentration ranged from a minimum of 154 to a maximum of 286 mg/L (Figure 4d). This second outdoor investigation demonstrated that when Rhodopseudomonas sp. S16-VOGS3 was cultured under lower irradiance, the amount of PHB into bacterial cells decreased significantly, indicating the dominant role of irradiance on other parameters such as C/N ratio. In fact, as elsewhere demonstrated [27,28], stressed growth conditions caused by In order to evaluate the effect of the solar light intensity on the productivity of PHB accumulated in the bacterial cells, a further investigation was carried out by covering about 80% of the photobioreactor with a dark cloth as shown in Figure 3a (run 2). In this way, the light intensity impinging on the L-SRP was significantly reduced and the culture was irradiated principally by diffuse irradiance, with the exception of the U-fittings. Figure 3b shows typical daily trends of the direct and diffuse irradiance on a horizontal plane measured at Sesto Fiorentino (Florence, Italy) in the period June-July 2017. The results obtained during the 336 h of Rhodopseudomonas growth under diffuse radiation (run 2) are shown in Figure 4a-d. During the first 192 h, the photobioreactor was operated in FB mode; when the Bchl concentration reached the value of 16.4 mg/L culture and the CDWC 4.0 g/L, the regimen of growth was changed from FB to SC mode (Figure 4a). In order to maintain the CDWC ranging from 3.2 to 4.1 g/L, an appropriate culture volume was withdrawn from the reactor and replaced with an equal volume of fresh medium every 72 h. During the 336 h of cultivation, the maximum light intensity was quite constant and ranged from 900 to 956 W/m 2 ; on the contrary, the global solar radiation (SR) changed significantly from 22.3 to 29.8 MJ/m 2 day, as shown in Figure 4b. The lowest reached lactate content was about 1.0 g/L and its initial concentration (4.0 g/L) was restored when the fresh culture medium was added in accordance with the SC regimen of growth (Figure 4c). The NH 3 -N amount decreased progressively and reached zero after 164 h of cultivation. The accumulated PHB concentration decreased significantly during the first period of bacterial growth; subsequently, in the SC regimen the PHB concentration ranged from a minimum of 154 to a maximum of 286 mg/L (Figure 4d). This second outdoor investigation demonstrated that when Rhodopseudomonas sp. S16-VOGS3 was cultured under lower irradiance, the amount of PHB into bacterial cells decreased significantly, indicating the dominant role of irradiance on other parameters such as C/N ratio. In fact, as elsewhere demonstrated [27,28], stressed growth conditions caused by high C/N ratios should increase the amount of PHB into bacterial cells, but this was true as far as the results of our first investigation (run 1) is concerned, but not for those of the second one (run 2) ( Figure 5). In fact, despite the similar increases in the C/N ratio for both cases (Figure 5a), the PHB content into the cells had an opposite behavior (Figure 5b): during the run 1 (direct irradiance), the PHB content rapidly increased reaching the value of 19.6% of dry biomass weight; during the run 2 (diffuse irradiance), the PHB content decreased reaching a value of about 7% at the end of the investigation.

PHB Production
Recently, the same bacterial strain used in the present study (Rhodopseudomonas sp. S16-VOGS3) was identified as a potential candidate for employing crude glycerol coming from biodiesel industry [15]. The authors obtained (indoors) a PHB-rich biomass with a content of 18% on dry-weight biomass, demonstrating that a degradable plastic material as PHB could be efficiently produced by feeding Rhodopseudomonas sp. S16-VOGS3 with the byproduct coming from biodiesel industry, in order to obtain a significant reduction of PHB production-costs. A higher cost reduction for PHB production could be achieved when PHB biosynthesis is achieved by CO 2 bioconversion, which is a sustainable alternative to the non-renewable, petroleum-based polymer production [29]. The authors reported a cumulative PHB of about 290 mg/L, along 28 days of Synechocystis PCC6803 batch growth. In the present investigation, a similar cumulative PHB-value (286 mg/L) was obtained with Rhodopseudomonas sp. S16-VOGS3 grown outdoors during the semi-continuous regimen. An overview of the current carbon sources (e.g., industrial byproducts) used in the PHA production has recently reported by Jiang et al. [11]; they also reported the typical microorganisms that can use the various substrates. high C/N ratios should increase the amount of PHB into bacterial cells, but this was true as far as the results of our first investigation (run 1) is concerned, but not for those of the second one (run 2) ( Figure  5). In fact, despite the similar increases in the C/N ratio for both cases (Figure 5a), the PHB content into the cells had an opposite behavior (Figure 5b): during the run 1 (direct irradiance), the PHB content rapidly increased reaching the value of 19.6 % of dry biomass weight; during the run 2 (diffuse irradiance), the PHB content decreased reaching a value of about 7% at the end of the investigation. Recently, the same bacterial strain used in the present study (Rhodopseudomonas sp. S16-VOGS3) was identified as a potential candidate for employing crude glycerol coming from biodiesel industry 15]. The authors obtained (indoors) a PHB-rich biomass with a content of 18% on dry-weight biomass, demonstrating that a degradable plastic material as PHB could be efficiently produced by feeding Rhodopseudomonas sp. S16-VOGS3 with the byproduct coming from biodiesel industry, in order to obtain a significant reduction of PHB production-costs. A higher cost reduction for PHB production could be achieved when PHB biosynthesis is achieved by CO2 bioconversion, which is a sustainable alternative to the non-renewable, petroleum-based polymer production [29]. The authors reported a cumulative PHB of about 290 mg/L, along 28 days of Synechocystis PCC6803 batch growth. In the present investigation, a similar cumulative PHB-value (286 mg/L) was obtained with Rhodopseudomonas sp. S16-VOGS3 grown outdoors during the semi-continuous regimen. An overview of the current carbon sources (e.g., industrial byproducts) used in the PHA production has recently reported by Jiang et al. [11]; they also reported the typical microorganisms that can use the various substrates.     Figures 6a,b show the DSC and TG/DTG (thermogravimetric and derivate thermogravimetric) thermograms of the recovered PHB from Rhodopseudomonas sp. S16-VOGS3 cells, respectively. The DSC thermogram (second heating curve after cooling from the melt) shows an endothermal baseline shift associated with the glass transition (Tg = 4.2 °C) followed by a cold-crystallization exotherm (Tc = 52.8 °C) and, then, by a melting endotherm (Tm = 172 °C). On the basis of the melting enthalpy of the sample, a PHB crystallinity of 51% was evaluated. These results are in excellent accordance with    Figures 6a,b show the DSC and TG/DTG (thermogravimetric and derivate thermogravimetric) thermograms of the recovered PHB from Rhodopseudomonas sp. S16-VOGS3 cells, respectively. The DSC thermogram (second heating curve after cooling from the melt) shows an endothermal baseline shift associated with the glass transition (Tg = 4.2 °C) followed by a cold-crystallization exotherm (Tc = 52.8 °C) and, then, by a melting endotherm (Tm = 172 °C). On the basis of the melting enthalpy of the sample, a PHB crystallinity of 51% was evaluated. These results are in excellent accordance with  Figure 6a,b show the DSC and TG/DTG (thermogravimetric and derivate thermogravimetric) thermograms of the recovered PHB from Rhodopseudomonas sp. S16-VOGS3 cells, respectively. The DSC thermogram (second heating curve after cooling from the melt) shows an endothermal baseline shift associated with the glass transition (T g = 4.2 • C) followed by a cold-crystallization exotherm (T c = 52.8 • C) and, then, by a melting endotherm (T m = 172 • C). On the basis of the melting enthalpy of the sample, a PHB crystallinity of 51% was evaluated. These results are in excellent accordance with the results of Martino et al. [30] obtained for PHB produced by cultivation of Cupriavidus necator DSM 428, using cooking oil as carbon source, and recovered by an enzymatic and chemically assisted procedure and to those of PHB synthetized by another purple bacteria (Rhodococcus equi) investigated by Altaee et al. [31]. Good accordance was obtained also with the results obtained with Bacillus sphaericus NII 0838 by Sindhu et al. [32]. The TG and DTG curves of PHB produced with Rhodopseudomonas sp. S16-VOGS3 (Figure 6b) show a decomposition temperature (T d ) of 248 • C exhibiting a single weight loss with maximum degradation rate at Tmax of 262 • C. Similar results were reported by Martino et al. [30] for the PHB produced with Cupriavidus necator DSM 428. In Table 2, the thermal properties and stability of the PHB produced in the present study by Rhodopseudomonas sp. S16-VOGS3 are compared with those of PHB produced by another purple bacterial strain (Rhodococcus equi) [31]. As shown, the thermal properties are very similar, except a lower crystallinity (51% vs. 62%) and T d of PHB obtained in this study.

PHB Thermal Properties and Stability
the results of Martino et al. [30] obtained for PHB produced by cultivation of Cupriavidus necator DSM 428, using cooking oil as carbon source, and recovered by an enzymatic and chemically assisted procedure and to those of PHB synthetized by another purple bacteria (Rhodococcus equi) investigated by Altaee et al. [31]. Good accordance was obtained also with the results obtained with Bacillus sphaericus NII 0838 by Sindhu et al. [32]. The TG and DTG curves of PHB produced with Rhodopseudomonas sp. S16-VOGS3 (Figure 6b) show a decomposition temperature (Td) of 248°C exhibiting a single weight loss with maximum degradation rate at Tmax of 262 °C. Similar results were reported by Martino et al. [30] for the PHB produced with Cupriavidus necator DSM 428. In Table  2, the thermal properties and stability of the PHB produced in the present study by Rhodopseudomonas sp. S16-VOGS3 are compared with those of PHB produced by another purple bacterial strain (Rhodococcus equi) [31]. As shown, the thermal properties are very similar, except a lower crystallinity (51% vs. 62%) and Td of PHB obtained in this study.

Conclusions
The main aim of the present work was to evaluate the performance of the novel tubular photobioreactor for producing PHB by outdoor cultivation of Rhodopseudomonas sp. S16-VOGS3 using lactate as carbon source. The results showed that the peculiar configuration of the photobioreactor permitted an optimal temperature control and solar light capturing for producing PHB-rich biomass. The reduction of the light intensity reduced the cumulative PHB showing the dominant role of this parameter. The thermal characterization of the recovered PHB showed a semi-amorphous polymer (51% crystallinity) with good thermal properties and stability, similar to those of PHBs obtained in other studies.