Microbial fuel cell arrangement and method for operating it

Abstract

The invention relates to a microbial fuel cell arrangement comprising a cell reactor. The cell reactor comprises a membrane, which has an active surface and a support surface, as well as a pore size of 10 nm and/or a divalent ion rejection of 50%; an anode and a cathode, which are connected with each other through an external electrical circuit; an influent inlet for liquid medium arranged at the active surface side of the membrane and at least one permeate outlet arranged at the support surface side of the membrane; an influent line connected to the influent inlet; a concentrate outlet, arranged at the active surface side of the membrane and connected to a concentrate line; and pressurisation means for creating pressure difference between the active surface side and support surface side of the membrane. The invention relates also to a method for operating a microbial fuel cell.

Claims

1. A microbial fuel cell arrangement, comprising a cell reactor, which comprises: a membrane, which has an active surface and a support surface, as well as a pore size of 10 nm and/or a divalent ion rejection of 50%, an anode and a cathode, which are connected with each other through an external electrical circuit, an influent inlet for liquid medium arranged at the active surface side of the membrane and at least one permeate outlet arranged at the support surface side of the membrane, an influent line connected to the influent inlet, a concentrate outlet, arranged at the active surface side of the membrane and connected to a concentrate line, a recirculation line, which is arranged to connect the concentrate line and the influent line, pressurisation means for creating pressure difference between the active surface side and support surface side of the membrane, and at least one sensor means for measuring the quality of the liquid medium, a permeate and/or a concentrate, arranged in connection with the influent line, a permeate line and/or the concentrate line.

2. The arrangement according to claim 1, wherein the anode is arranged at the active surface side of the membrane and the cathode is arranged at the support surface side of the membrane.

3. The arrangement according to claim 1, wherein the pressurisation means are arranged in connection with the influent line for creating pressure difference between the active surface side and the support surface side of the membrane.

4. The arrangement according to claim 3, wherein the pressurisation means are arranged to create a pressure in the range of 0.5-5 bar on the active surface side of the membrane.

5. The arrangement according to claim 1, wherein the membrane is a semipermeable reverse osmosis membrane.

6. The arrangement according to claim 1, wherein the membrane is a nanofiltration membrane, which has divalent ion rejection50%.

7. The arrangement according to claim 1, wherein the cell reactor is an elongated tubular reactor which is divided in its axial direction by the membrane.

8. The arrangement according to claim 1, wherein the recirculation comprises pH regulating means.

9. The arrangement according to claim 1, wherein a hydrolysis unit is arranged in connection with the influent line before the influent inlet.

10. The arrangement according to claim 1, wherein the cathode is air cathode.

11. The arrangement according to claim 1, wherein the anode and the cathode are arranged at the active surface side of the membrane.

12. A method for operating microbial fuel cell, which comprises a cell reactor, which comprises an anode and a cathode, which are connected with each other through an external electrical circuit, as well as a membrane having an active surface and a support surface, the method comprising feeding liquid medium comprising organic substance(s) through an influent inlet at the active surface side of the membrane, creating a pressure difference between the active surface side and the support surface side of the membrane, allowing a part of liquid medium to permeate through the membrane to the support surface side of the membrane and to form a permeate, removing the permeate from the support surface side of the membrane through a permeate outlet, removing a part of the liquid medium as a concentrate through a concentrate outlet at the active surface side of the membrane, recirculating at least a part of the concentrate back to an influent line, and measuring the quality of the liquid medium, the permeate and/or the concentrate and adjusting the volume of the recirculated concentrate on basis of the measurement.

13. The method according to claim 12, wherein the method comprises adjusting pH of the concentrate before it is recirculated back to the influent line, and/or hydrolysing at least part of the liquid medium before feeding to the cell reactor.

14. The method according to claim 12, wherein the ratio of the permeate to the liquid medium fed through the influent inlet is of 10-99%.

15. The method according to claim 4, wherein the applied pressure on the active surface side of the membrane is in the range of 0.5-5 bar and the pressure level on the support surface side of the membrane is the atmospheric pressure.

16. The method according to claim 12, wherein the liquid medium, which is used as influent, is selected from effluents from pulp and paper industry process, from oil and gas industry process, or from mining process, or the liquid medium originates from food or beverage industry, municipal or agricultural waste water.

Description

Brief Description of the Drawings

(1) In the following, the invention will be described in more detail with reference to the appended schematic drawing, in which

(2) FIG. 1 shows an exemplary arrangement according to the invention.

(3) FIG. 2 shows the daily average of power production, expressed in W/m.sup.3, based on the anode chamber volume.

(4) FIG. 3 shows the coulombic efficiency of the microbial fuel cell of Example 1,described in further detail below.

(5) FIG. 4 shows the daily average of power production, expressed in W/m3, based on the anode chamber volume.

(6) FIG. 5 shows the reduction of soluble COD values, given in percentages, for microbial fuel cells of Examples 1 and 2, which Examples are described in further detail below.

Detailed Description of the Invention

(7) FIG. 1 shows a microbial fuel cell arrangement 1. The arrangement comprises a cell reactor 2, which is divided by a membrane 3. In the embodiment shown in FIG. 1 the membrane 3 is a reverse osmosis membrane. At the active surface side of the membrane 3 is arranged an anode 4 and at the support surface side of the membrane is arranged a cathode 5. Anode 4 and cathode 5 are connected with each other through an external electrical circuit 6.

(8) In the first end 2 of the cell reactor 2, at the active surface side of the membrane 3, is arranged an influent inlet 7 and in the second end 2 of the cell reactor 2, at the active surface side of the membrane 3, is arranged a concentrate outlet 8. Liquid medium that enters the cell reactor 2 through the influent inlet 7 is partly filtrated through the membrane 3 by the pressure difference between the active surface side and the support surface side of the membrane 3. The positive pressure on the active surface side is created by the pressurisation means 9, such as pressure pump, arranged in connection with the influent line 10 leading to the influent inlet 7. The part of the liquid medium, which does not permeate through the membrane 3 is exited from the cell reactor 2 through the concentrate outlet 8. The liquid medium that permeates the membrane 3 to the support surface side of the membrane 3 is exited through a permeate outlet 11.

(9) Air may be fed to the support surface side of the membrane through an air inlet 12. Air may exit the support surface side through the permeate outlet 11 or through a separate air outlet (not shown).

(10) A part of the concentrate flow can be recirculated from the concentrate line 13, which connected to the concentrate outlet 8, back to the influent line 10 via a recirculation line 14. Recirculation line 14 may comprise a recirculation pump 15, as well as pH regulating means for optimising the pH of the concentrate before it is combined with the liquid medium used as an influent. A flow 16 of suitable chemical, such as base or buffer may be added to the recirculation line. A part of the concentrate flow can be exited the arrangement 1 as an excess flow 18.

(11) The arrangement 1 may also comprise a hydrolysis unit 17. A part of the liquid medium can be treated in the hydrolysis unit 17 before it is fed to the cell reactor 2. Alternatively, or in addition a part of the concentrate flow can be directed to the hydrolysis unit 17 for hydrolysis of slowly biodegradable organic substances. After hydrolysis this hydrolysed concentrate may be introduced to the liquid medium before it is fed to the cell reactor 2.

EXPERIMENTAL

(12) Some embodiments of the invention are described in the following non-limiting examples.

(13) Construction of the Microbial Fuel Cell Arrangement Used in the Experiments

(14) A microbial fuel cell reactor, which comprised an anode chamber and a cathode chamber, similar to that shown in FIG. 1 was assembled in laboratory. The cell reactor was operated positioned in upright. The anode side comprised an influent inlet and a concentrate outlet and the cathode side comprised a permeate outlet. The casing of the cell reactor was made of plastic (acetal). A single piece of rectangular membrane was arranged in the cell. The membrane was supported by a porous stainless steel plate that also functioned as a permeate carrier. The permeate flow was collected in manifold before exiting through the permeate outlet.

(15) Through the casing of cell reactor two metal alloy screws were bored. The first screw penetrated the casing into the anode chamber and the second screw penetrated the casing on the permeate side. In Examples 1 and 2 a folded metal alloy strip was soldered to the tip of the screw on the anode side.

(16) Two additional flow connections were made to the cathode side of the cell reactor, in order to allow air to the cathode side and upright position of membrane. In Examples 1 and 2 the cell reactor configuration allowed simultaneous flow of air through the cathode side and outflow of the permeate from the cathode side.

(17) Anode comprised stainless steel meshes, which were placed on top of each other in the anode chamber. The area of one layer was 34 cm.sup.2. In Examples 1 and 2 one of the stainless steel meshes of the anode was in contact with the folded alloy metal strip, which thus connected the anode to the external circuit. In Example 2 a carbon cloth was placed between the meshes and the membrane. Anode chamber volume was 7.5 ml.

(18) An air cathode was used as cathode. Cathode comprised a carbon cloth, which was placed between the membrane and a metal plate. The carbon cloth contained a gas diffusion layer with 2 mg/cm.sup.2 platinum as catalyst. Active cathode area was 34 cm.sup.2.

(19) The influent feed was pumped from the feed tank to the influent inlet of the cell reactor. The feed tank volume was ca. 0.6 l in Examples 1 and 2 and ca. 2.2 l in Example 3. The influent inlet was arranged on the lower part of the anode chamber. The concentrate outlet was arranged on the upper part of the anode chamber. Anode side was pressurized by feeding pressurised nitrogen gas to the anode side. A portion of the liquid medium permeated through the membrane to the cathode side and flowed through the permeate carrier. In Examples 1 and 2 humified air was pumped on the cathode side through two upper inlet connections. Permeate flowed out with air through the permeate outlet connection into a permeate collection vessel.

(20) The concentrate stream, which contains the material rejected by the membrane, exited the cell reactor through the concentrate outlet and was directed back into the feed tank.

(21) Operation of the Microbial Fuel Cell Arrangement

(22) A variable external resistor was connected between the anode and the cathode electrodes. Anode potential against cathode potential was measured at 10 minute intervals. The cell voltage and external resistor value were used to calculate power and current. All power production (W/m.sup.3) results are expressed in relation to anode chamber volume.

(23) The arrangement was fed with a liquid medium comprising either brewery wastewater or a mixture of effluents from other microbial fuel cells which had been fed with pre-fermented brewery wastewater.

(24) The pressure applied on the anode side was ca. 3.5 bar, and the cell reactor operated at ca. 30 C. temperature. Liquid medium was circulated in the arrangement at a flow rate of ca. 20 l/h. Air was pumped through the cathode at a flow rate of ca. 2 l/min.

(25) The arrangement was operated in batch mode. An influent batch was recirculated in the arrangement until the concentrate was removed and new influent batch was put to the feed tank. At the same time the permeate vessel was also changed.

(26) Analysis of soluble COD was performed from the liquid medium at the start of each batch and from permeate and concentrate after finishing each batch. Conductivity was also measured. Influent, permeate and concentrate were weighed for each batch.

Example 1

(27) The reactor of the microbial fuel cell comprised an anode electrode made of 3 layers of stainless steel mesh, a membrane which was a polyamide reverse osmosis membrane and a cathode electrode which was a carbon cloth with catalyst.

(28) First 3 batches were run using effluents from other microbial fuel cells as influent. Then 3 consecutive batches were run using brewery wastewater as influent. Then the pH of the concentrate went below 7, and the following batches were run using again effluents from other microbial fuel cells as influent. As dissolved solids removal and permeate flow decreased over time, the membrane was cleaned by performing an air backflush on day 23 between batches 9 and 10. A total of 13 batches were run. The duration of each batch varied between 1-7 days.

(29) FIG. 2 shows the daily average of power production, expressed in W/m.sup.3, based on the anode chamber volume. The data points for batches using brewery wastewater as influent are indicated with crosses, the data points for batches using microbial fuel cell effluents as influent are indicated with black squares.

(30) Power generation started after 2 days of operation. It can be seen from FIG. 2 that when the influent was changed to untreated brewery wastewater, power decreased although soluble COD concentration of the influent changed from 1100 mg/l to 3500 mg/l. It was assumed that the untreated wastewater started to ferment within the microbial fuel cell arrangement, suppressing exoelectrogenic activity and lowering the pH of the concentrate.

(31) The coulombic efficiency (CE %) was calculated for each batch using the ratio of total coulombs obtained, C.sub.out to the theoretical amount, C.sub.in, available from complete influent oxidation:
CE %=C.sub.out/C.sub.in100%=(lt)/((FnCOD)/M)
where
l is the average current (A), calculated from cell voltage and resistor value;
t is the time interval;
M is the molecular weight of oxygen;
F is the Faraday constant;
n is the number of electrons exchanged per mole of oxygen;
COD is the weight of removed amount in soluble COD.

(32) FIG. 3 shows the coulombic efficiency of the microbial fuel cell of Example 1. The data points for batches using brewery wastewater as influent are indicated with crosses, the data points for batches using microbial fuel cell effluents as influent are indicated with black squares.

(33) Open circuit voltage of the microbial fuel cell arrangement was in the beginning exceptionally high, 910 mV. However, in later measurements the open circuit voltage was between 780-830 mV. An oxide layer accumulated on the metal alloy surface over time thus increasing resistance within the electrical circuit.

(34) Volatile Fatty Acid (VFA) analysis was performed for influent, concentrate and permeate of batches 6 and 7 of the Example 1. The results of VFA analysis are shown in Table 2.

Example 2

(35) The reactor of the microbial fuel cell comprised an anode electrode made of 2 layers of stainless steel mesh and a carbon cloth, a membrane which was a polyamide reverse osmosis membrane, and a cathode electrode, which was a carbon cloth with catalyst.

(36) First 7 batches were run using effluents from other microbial fuel cells as influent, and the last 2 batches were run using brewery wastewater as influent. The duration of each batch varied between 1-5 days.

(37) During the first 3 batches cell voltage was very low. The cell was opened, and connection to the external circuit on cathode side was fixed on day 7.

(38) FIG. 4 shows the daily average of power production, expressed in W/m.sup.3, based on the anode chamber volume. The data points for batches using brewery wastewater as influent are indicated with crosses, the data points for batches using microbial fuel cell effluents as influent are indicated with black squares.

(39) The batch number 6 of Example 2 was run until power production declined rapidly on day 19. The soluble COD value of the concentrate was low, below 300 mg/l, as expected. The batch number 7 of Example 2 had very low soluble COD value already in the start of the batch. This is apparent also from power production in FIG. 4.

(40) It can be seen from FIGS. 2 and 4 that the highest power production is at the similar level for Examples 1 and 2 despite the different anode configurations. The highest open circuit voltage, OCV, was 850 mV in Example 2. Even though carbon cloth has lower conductivity than the metal mesh, it provides more surface area for biofilm to grow.

(41) The anode configuration of Example 2 removed more soluble COD than the anode configuration of Example 1, which is seen from Table 1. The anode configuration of Example 2 also tolerated the influent comprising untreated wastewater better, as the concentrate pH did not go below 7 during those batches.

Example 3 (Reference)

(42) The cell reactor comprised an anode electrode made of 5 layers of stainless steel mesh and a carbon cloth, a membrane which was a polyamide reverse osmosis membrane and a cathode electrode, which was a carbon cloth with catalyst.

(43) In Example 3 there was no folded metal strip on the anode side touching the stainless steel meshes of the anode, and therefore the anode was not properly connected to the external electrical circuit. Consequently there was not a working microbial fuel cell arrangement. This was reflected in cell voltage which was at highest 1 mV at 500 Ohm and 7 mV at ca. 10 kOhm. The highest open circuit voltage with this configuration was 480 mV. The permeate flow pattern on the cathode side also caused cathode flooding. The conditions did not favour exoelectrogenic activity.

(44) First batch was run using a mixture of microbial fuel cell effluents and brewery wastewater as influent. Following 6 consecutive batches were run using brewery wastewater as influent. The duration of each batch varied between 2-5 days.

Example 4 (Reference)

(45) A filtration unit was used to filtrate brewery wastewater. Plain pressure filtration at 3.5 bar and with the same type of reverse osmosis membrane as used in the microbial fuel cell examples was run for 20 hours. The VFA analysis of this filtration is included in table 2.

Comparison of Examples 1-4

(46) In Table 1 the unfavourable conditions of Example 3 are compared against key results from working microbial fuel cells of Examples 1 and 2 having a pressure applied on the anode side. All the given results are average values. In terms of the organic load, the permeate quality of the results for Example 1 and 2 are significantly better, even when untreated brewery wastewater was used as influent. The reduction in soluble COD concentration from influent to permeate is 95% in Example 1, 97% in Example 2 and only 83% in Example 3.

(47) In Table 2 VFA composition of influent, concentrate and permeate are compared for selected batches of Example 1 and Example 4.

(48) Batch 6 of Example 1 used untreated brewery wastewater as influent and batch 7 of Example 1 used effluent of other microbial fuel cells as influent. Example 4 used untreated brewery wastewater as influent.

(49) It can be observed that the VFA concentration of microbial fuel cell permeate is slightly lower than that of conventional reverse osmosis filtration.

(50) TABLE-US-00001 TABLE 1 Conductivity, soluble COD and pH for examples 1, 2 and 3. Example 3 Example 1 Example 2 (reference) Feed conductivity (mS/cm) 2.1 2.3 1.9 Concentrate conductivity (mS/cm) 3.5 3.3 2.6 Permeate conductivity (mS/cm) 1.5 1.4 0.6 Feed CODsol (mg/L) 1900 1500 3200 Concentrate CODsol (mg/L) 2200 600 3600 Permeate CODsol (mg/L) 100 40 560 Concentrate pH 7.5 7.6 6.1 Permeate pH 9.2 9.1 7.6

(51) TABLE-US-00002 TABLE 2 VFA composition for Examples 1 and 4. VFA Acetic Propionic Butyric Valeric concen- acid acid acid acid tration (ppm) (ppm) (ppm) (ppm) (ppm) Exam- influent 739 323 541 29 1631 ple 1, concentrate 2 473 400 226 1100 batch 6 permeate 38 5 1 1 45 Exam- influent 586 219 41 48 894 ple 1, concentrate 689 379 151 120 1339 batch 7 permeate 25 2 1 1 28 Exam- influent 670 260 72 76 1078 ple 4 permeate 43 12 3 1 59 (refer- ence)

(52) FIG. 5 shows the reduction of soluble COD values, given in percentages, for microbial fuel cells of Examples 1 and 2. The remaining COD in permeate and concentrate is compared to the COD of the influent for each batch. It is seen that Example 2 removes more soluble COD than Example 1. Conventional pressure filtration does not reduce COD value, so FIG. 5 shows the benefit of combining microbial fuel cell and pressure filtration within the same arrangement.

(53) Cell resistances were evaluated using electrochemical impedance spectroscopy, EIS. The scans were run in two electrode mode, using cathode as working electrode. EIS was run for Examples 1 and 2 at 0.7 V and for Example 3 at 0.3 V. Scan results were evaluated using equivalent circuit fitting to Randles circuit with Warburg element. However, results of Example 2 required two charge transfer (R.sub.ct) and two capacitance elements within the circuit to fit. In Table 3 cell resistances are compared for Examples 1, 2 and 3. Example 3 results clearly show the connection problem within the external electrical circuit.

(54) TABLE-US-00003 TABLE 3 Cell resistances evaluated using electrochemical impedance spectroscopy. R.sub.s () R.sub.ct () W (1//sqrt(Hz) Example 1 24 42 17 Example 2 15 19 20 Example 3 18 000 4 000 74 000

(55) Even if the invention was described with reference to what at present seems to be the most practical and preferred embodiments, it is appreciated that the invention shall not be limited to the embodiments described above, but the invention is intended to cover also different modifications and equivalent technical solutions within the scope of the enclosed claims.