Team:Brasil-USP/Project/Reactor

Project

Bioreactors and Reactor


As said previously, we intend to implement our process in scaled up bioreactors. The whole process would be divided in 3 stages, as proposed in figure 1.

Figure 1- Scheme of the whole process for the rubber degradation showing the 3 foreseen bioreactors.

The first 2 bioreactors are supposed to contain microorganisms while the last one is a chemical reactor, commonly found in polymeric industry.

Bioreactor 1 - Devulcanization

The process that takes place in this first bioreactor is the devulcanization of tire particulates. Tires are vulcanized, a process required to confer hardness and resistance to natural rubber forming sulfur bonds, that cross-link the polyisoprene chains of the natural rubber. This process allows tires to be used in the tire manufacturing procedure but, at the same time, makes it more difficult for the tire to be reused/recycled, turning the devulcanization into a very important step in our project, even if at this moment we are not genetically engineering it but only optimizing its cultivation in rubber presence.

This bioreactor will contain a special strain of Acidithiobacillus ferrooxidans, which were kindly provided by Professor Denise Bevilaqua, from UNESP (Estadual University of São Paulo). This microorganism is a wild bacteria that has the gene for tetrathionate hydrolase (TetH), which is able to reduce inorganic sulfur compounds [1]. TetH has its maximal activity in a pH range from 3 to 4, adding to that, the bacterial ideal growth is in pH 4, what let us think that the ideal bioreactor pH would be 4 [1]. Though Professor Denise Bevilaqua, who provided us the Acidithiobacillus ferrooxidans strain, recommends that this bacteria should be cultivated at 30°C, while TetH has its best activity at 25°C [2]. To achieve the best relation between bacterial growth and tire devulcanization some tests with different temperatures and pHs should be performed. The ideal stirring in cultivation is 150 rpm, to scale up maybe the rotation will also change.

Acidithiobacillus ferrooxidans is being maintained in a simple culture medium that Professor Denise Bevilaqua provided us; its composition is as follows (g/L): (NH4)2SO4 0.5; MgSO4.7H2O 0.5; K2HPO4 0.5 and FeSO4 33.3. To activate the microorganism sulfur metabolism, it is necessary to use a different culture medium with sulfur instead of iron (g/L): (NH4)2SO4 0.5; MgSO4.7H2O 0.5; K2HPO4 0.5 and S 10. This medium composition were also informed by Professor Denise Bevilaqua. Once the sulfur metabolism is activated, the bacteria can be incubated with the tire scrapes or powder, in order to initiate the devulcanization process as shown in Figure 2. The medium required for this process is very similar (g/L): (NH4)2SO4 1.5; MgSO4.7H2O 0.5; K2HPO4 0.05; Ca(NO3)2 0.01; KCl 0.05 and FeSO4.7H2O 4.43.

Figure 2- In the first flask Acidithiobacillus ferrooxidans is incubated in the medium to activate sulfur metabolism, while in the last two flasks the bacteria is in the media for devulcanization with tire powder.

As it is possible to see, the tire powder floats, suggesting that an intense stirring will be needed in the bioreactor. Feng et al. described a 7 L batch using Acidithiobacillus ferrooxidans, they used a 3 step batch with different pHs, 400 rpm and 30°C in a 40-day process [5].

We would also pretend to make the devulcanization in a fed-batch process in a stirred tank, for initial parameters we could start with a two pH step batch, starting with 4, which is the one that there are more bacterial growth and after achieving a high cell concentration we could decrease the pH to get the best TetH active.
After the devulcanization is ready, as soon as the particulate floats, we think it will be possible to separate it from the media as in a clarification process, wash it and introduce it into the second bioreactor.

Bioreactor 2 - Degradation

The second bioreactor is the core of this project, it contains our modified Escherichia coli. The process in this bioreactor is the degradation of the rubber to ODTD. We chose E. coli because it has its genetics well known and can grow fast in high cellular density cultivation with low cost substrates, besides there are already many established process using this bacteria in bioreactors, what makes the process easier to be defined [3]. High density cultivation can reduce the effluents and also reduce the production costs [3]. The strain with our circuit is BL21 and the best conditions would be a pH between 5.5 and 8.5 and temperatures from 35 until 40°C, although they can grow from 8 until 48°C [3, 4]. Using E. coli to express heterologous proteins is possible to achieve productions of 0.5 - 0.8 g/L, furthermore there are reports saying 5 - 10 g/L for some therapeutic proteins [3]. The enzymes we want to express are: RoxA (best activity in pH 7 and 40°C) and Lcp (best activity in pH 7 and 30°C). Thus, we think the better conditions to the bioreactor would be pH 7 and that tests trying to find the best temperature for the higher efficiency in cleavage of the rubber should be performed.

As exist many kinds of bioreactors we had help from Professor Teresa Cristina Zangirolami from Federal University of São Carlos. At first we thought to chose the one that causes less environmental impact, the solid state bioreactor, but, as our circuit needs to be induced, it would not be possible in this system. The second option was the fed-batch in a stirred tank in two steps: growth and induction. The problem here was that our circuit should always be induced or there would be HokD production and it would cause cell death. The solution was a fed-batch with constant induction, but with lower temperature around 25°C. The lower temperature makes the process slower, but it is easier to supply O2 in the right demand and avoid fermentative pathways. The fed-batch cultivation allows us to control the substrate supply, thus the specific grown velocity can also be controlled [3]. The addition of other nutrients is also an important step in the cultivation, which can affect the maximal cell concentration, productivity and also product formation [3].

Thinking about these fermentative pathways, glycerol instead of glucose as carbon supply would avoid acetate production that would reduce the production of our recombinant proteins (overflow metabolism) [3].

A proposed media for fed batch would be the one used by Sargo 2011 without antibiotics and with our inducer [3]:

Table 1- Fed-batch cultivation media.

Feeding media (per L)
Glycerol
800.00 g
KH2PO4
21.28 g
(NH4)2HPO4
6.40 g
Citric acid
2.27 g
Fe (III) citrate
40.00 mg
CoCl2.6H2O
4.00 mg
MnCl2.4H2O
23.50 mg
CuCl2.2H2O
2.50 mg
Zn(CH3COO)2.2H2O
16.00 mg
Na2MoO4.2H2O
4.00 mg
H3BO3
5.00 mg
EDTA
13.00 mg
MgSO4.7H2O
20.00 mg
Thiamine
4.50 – 45.00 mg
Antifoam PPG 30%
1 mL

Although the airlift equipment is cheaper to acquire, it is not well recomended to high density cultures, it is more used for fungus and in our particular case, the rotor of the stirring bioreactor would help to maintain polyisoprene in suspension.

The idea of working with batches and not in a continuous process is to avoid mutations and loss of productivity.

Chemical Reactor

Finally, at the reactor 3, the transformation of the main final product of RoxA and Lcp enzymes, the ODTD molecule, into hydrocarbon chain with fuel properties will take place. Besides granting the final product considerable economic interest and generate a high-energy molecule directly from a renewable source (namely, natural rubber), resulting in a more sustainable end for this material, it would also decrease the final pollutant emission from the burning process that usually is a solution for unusable tires.

Before properly transforming ODTD, it should be purified from Bioreactor 2. A simple flowchart was designed by us, Figure 3.

Figure 3 - Flowchart for the purification process of ODTD


The first step for the purification of ODTD consists in the microfiltration to separate through a semipermeable membrane the RoxA and Lcp catalyzed cleavage products, that may be larger than triisoprene units, from the tire particles and the bacterial cells. After that, these products should be solvent extracted with ethyl acetate, dissolved in methanol, and then separated by HPLC on a reversed phase C8 column, as described in scientific papers [6, 7]. The specific retention time, interval taken for a component to travel along a chromatography column and be detected, of the ODTD peak must be previously determined by using the HPLC column coupled with a mass spectrometry unit using the electrospray ionization (ESI)-MS technique.

As reported previously, since the ODTD molecule has a ketone and an aldehyde groups in the C1 and C15 positions (Figure 4), in order to transform this molecule into a hydrocarbon fuel, two different steps should be performed: the reduction of the aldehyde and ketone groups into alcohol groups; and then the dehydration of the alcohol groups (Figure 5).


Figure 4 - ODTD 2D structure and molecular weight value


Figure 5 - Reduction and dehydration of ODTD molecule to obtain a hydrocarbon fuel.


Instead of using a strong reducing agent, such as sodium borohydride (NaBH4), and then introducing a strong acid catalyst to dehydrate the alcohol groups, economically and industrially talking, catalytic hydrogenation is the best option to accomplish this transformation. Catalytic hydrogenation consists of a chemical reaction between molecular hydrogen (H2) and another compound or element, usually in the presence of a catalyst. The process is commonly employed to reduce or saturate organic compounds and has wide industrial application since the reactions are highly selective and the catalyst can often be recovered and recycled. Furthermore, hydrogen gas is considerably cheaper than hydrogen deriving from other sources, as hydrides. Therefore, it generally is an economically viable process. [8] In this procedure, some parameters should be previously determined, as the best catalyst and solvent for the reaction, the specific conditions of pressure, temperature, concentration, time of hydrogenation, as well the hydrogen donor source.

There are two families of catalyst: homogeneous and heterogeneous catalysts. The homogeneous catalysts are soluble in the reactional mix, while heterogeneous catalysts are insoluble. Industrially, heterogeneous catalysts are more utilized (80%), being the metals palladium, platinum, rhodium, nickel, cobalt and ruthenium the most frequently ones. [9] The properties requested for a catalyst are high activity, high selectivity, fast filtration rate, and recycle capability. Depending on the choice of catalyst, the product may vary. One of the facts that make catalytic hydrogenation an economically viable process is that catalyst can be refining and remanufacturing after use. It allows that the intrinsic value of the metal may be recovered. It should be taken into account, the efficiency of metal recovery varies; on Table 2 we can see some examples. [8]

Table 1- Fed-batch cultivation media.

References

references

Back to top

Feeding media (per L)
Glycerol
800.00 g
KH2PO4
21.28 g
(NH4)2HPO4
6.40 g
Citric acid
2.27 g