Team:Sheffield/modelling

Overview

This was the week where the initial concepts were conceived. There were 2 under-sink concepts and an In-sewer concept.

Under sink concept 1

The initial idea was to have a unit that would fit inside the existing pipework, or that would be used in place of a section of existing pipework. The key issues this design had to overcome were to maintain a constant environment for the bacteria to ensure they could survive and to ensure they were withheld in the unit to prevent release into the sewer systems (causing issues with antibiotic resistance in harmful bacterial strains). The limitations to this were boiling water and detergents being put down the drain.
The solutions were as follows;
A twin layer metal cooling fin arrangement was thought up to ensure that the heat from the boiling water could be dissipated quickly enough to ensure the survival of the bacterial culture.
A vertical Leaf filter arrangement was devised, where a frame would hold on each side two filter membrane layers, both of 180 kDa nominal pore size to prevent the passing of DNA out of the central volume. These filters would be separated by a small, sterile void to clearly separate the recombinant bacteria inside the central volume of the ‘leaf’ and the bacteria in the water surrounding the filter. These filters would be hydrophobic to allow water pass across them as quickly as possible and to also allow any lipids to more easily bind to the surface, hence remaining closer to the bacterial culture and likely to be exposed to greater concentrations of lipase, thus breaking down faster.
The filters were kept parallel to the flow so as to not allow large clumps of food to coalesce on the surface and hence block flow in the pipe, as this would be inherently counterproductive.

Under sink concept 2

The premise behind this design was to think of an alternative solution to the high temperature water passing over the filters. The concept was to use perforated plates with weir’s and down comers as seen in distillation columns. If designed correctly they would hold flow for long enough to ensure that the water would cool (by dissipating heat to the surrounding environment) sufficiently to ensure the survival of the culture in the leaf filters. N.B: Both heat dissipation designs would have induced significant turbulence across the filter membrane surfaces, which should have led to greater rates of diffusion of enzyme into the pipework.
N.B: Both units were designed with UV light systems to destroy all DNA that may have passed through the filter membranes.
No change was made to the filter arrangement.

Limitations of the under sink concepts

The main limitations were that the under sink units had no means of protecting the culture from detergents and other chemicals apart from allowing them to ‘hopefully’ pass by without causing too much damage.
Another limitation in hind sight was providing the culture with nutrients, however the bacterial culture were initially going to use the lipids they broke down as a source of fuel.

In-Sewer Concept

The in-sewer concept was a look at tackling the problem at a more industrial scale. After speaking to Dr Jensen and Prof Simon Tait, experts in sewers systems, it was clear that reducing or significantly restricting the flow in the sewers was a no go. As was increasing the rate of flow of water down the sewer systems as this would simply mean more raw sewage would be directly pumped into the sea.
As such the design was developed with limiting impact of throughput in the sewers as much as possible. Once again the filter leaf concept was used, where once again they were positioned parallel to the flow of water. This design also included ceiling mounted brackets to allow for access to the filters for replacement, cleaning and maintenance from above the pipework, so that it could be carried out faster, reducing down time.

Limitations of the in-sewer concept

The limitations of this concept were that irrespective of how streamlined the design may have been it was still going to impact on the flow of materials down the sewer.
Another issue with the design was that these membranes would allow other materials not passing through the sink units to coalesce on their surface, these most notably included wet-wipes. As such these filters could have simply provided a site for other materials to gather and hence block the sewage network.

Overview

Designs mentioned above were drawn-up on AutoCAD and discussed for potential improvement s and current limitations with the rest of the team. 2 Grease trap concepts were devised, after noticing that settling tanks are currently in use at industrial sites, such as local takeaways to collect fat prior to disposal via a specialise company. The grease traps were modified to see if we could come up with a design that enabled us to break down the fats during separation in these tanks.

Grease trap concept 1

The first concept was to house the bacteria in the solution found inside the tank itself. This would give a high degree of exposure of the fats to the lipase enzyme, thus facilitating rapid degradation of the lipids. The main risk with such a design is the fact the recombinant Bacteria are not well contained and more importantly separated from other bacteria. The design features to overcome this, were to introduce a 0.22 μm filter at the exit to ensure the bacteria were physically incapable of exiting the tank with the outflow. Such a filter perpendicular to flow will need to be periodically cleaned to allow for sufficient throughput, which is a limiting factor. Further along the process stream a white light was installed with the intention of having a ‘kill switch’ imbedded in the bacteria’s genetic sequence to kill the bacteria when exposed to white light. Along from this there will be a UVC light to destroy any remaining DNA to 100% check that the antibiotic resistance cannot be passed onto any other bacterial species.

Grease trap concept 2

The second concept was a mix of the under-sink concept and the previous grease trap concept. This would have a grease trap where the bacteria would be housed inside twin membrane vertical leaf filters to separate bacteria. This unit would only have the UVC light on the outflow to ensure all DNA is destroyed. The filter leaves, will have frames with surface mounted brackets in the ceiling of the tank, which would allow for access to the filters for faster, easier maintenance to reduce downtime of the unit.

Overview

This week initially saw a shift in the design of the unit, back to something that could fit under a sink unit, with the potential to operate at a higher level of efficiency by removing it from the pipe itself to facilitate a more controlled environment for the bacteria; something considered an issue in the previous designs.

Perfusion Bioreactor Design

The basis behind the shift towards a more conventional unit was to try an ensure the bacteria would remain alive, this is because irrespective of how well integrated design may be with the existing systems and piping networks in place, if the bacteria cannot survive or get outcompeted, then the exercise is pointless. Another point to add to this would be the batch nature required wrt using recombinant bacteria. Since the secretion of lipase does not benefit the bacteria to outcompete others in its area, it is wasted energy and therefore cannot reproduce as quickly as those without such a gene. Based on the rapid rates of reproduction of bacteria those that do not produce the lipase will soon out-compete the more ‘useful’ bacteria for our application. As such each unit will have a limited lifespan before the bacteria need to be replaced.
The perfusion Bioreactor consists of a bioreactor which has two perforated 316L SS plates that house a bed of fibrous disks in-between them. These disks bind the bacteria to the surface, immobilising them and thus limiting colony size, which helps to promote extended batch lifetimes. The immobilisation will also reduce the burden (hence increasing time between cleaning cycles) on the 0.22 μm filter at the outlet, since cake layer formation will occur at a much slower rate. The disks themselves are rotated within the tank as the metal plates are attached to the stirrer, which is propelled by water flowing down its central column and out of nozzles perpendicular to the axis of the column. Baffles are included to induce turbulence above and below the central area housing the fibrous disks. A ring sparger is situated at the base of the tank to provide sufficient oxygen diffusion to the tank.
Aside from the tank itself a flow system including a holding tank and a recycle system with tangential flow filtration would allow for the product stream to be more effectively controlled in terms of enzyme concentration and pH (due to buffer inclusion). This would allow water and any unused nutrients to be reused, improving the efficiency of the unit.

Limitations of perfusion Bioreactor Design

In terms of the operation of the system it would have been the most effective and efficient when in operation. However, this system would come at a great deal of expense, would be difficult to install and would require expertise for maintenance especially when failures occur in the numerous control systems that would be required to ensure near continuous operation.

Capsule Concept

This design was to be as simple as possible, whilst still providing an adequate environment for the bacteria to live. A T-junction connection would be fitted into the existing pipework under the water level, the new pipe would have a mesh to prevent food particles getting lodged in the pipe restricting flow. This pipe would also have a UVC LED to ensure all DNA is destroyed prior to entering the pipework. Cooling fins would run along the length of this pipe to ensure adequate heat diffusion could be achieved to the surrounding areas, so that any hot water could not affect the bacteria in the capsule. At the far end of the pipe there would be an auto-sealing valve (e.g. pneumatic) to close off flow when replacing the cartridge.
The cartridge itself is comprised of 2 chambers. One housing the nutrient feed and one housing the bacteria in solution. These would be separated by a ceramic filter, allowing nutrients to pass into the bacterial solution, by allowing water to enter the nutrient feed chamber and thus dissolve nutrient and then carry it back across. The Bacterial chamber is separated by another ceramic filter and another auto-sealing device to ensure leakage doesn’t occur during transportation and storage. This second filter would be 0.22 μm to ensure no bacteria escapes the capsule.

Limitations of the capsule concept

This design was deliberately made to be as simple as possible, as a base to build off of to enhance into the final product design. As such it is a little crude in nature and wouldn’t be very efficient e.g. a high level of nutrient loss would be recorded and a lack of effective control over its release would cause uncontrolled growth of the bacteria and hence excretion of the lipase, leading to unreliable performance.

Conversation with Dr. Paul Dobson, caused expected move for the design to regress somewhat to a slightly more traditional processing solution due to complications with the operation of the capsule concept. A meeting with an engineer for Anglian Water raised interesting points wrt the legislation regarding the unit.
The product itself would have to be something customers would purchase themselves and they would have no obligation to do so, since water company’s authority ends at the borders of the house and thus they are unable to enforce the installation of such units.
The idea of using a ‘soap on a rope’ style block of nutrients and bacteria for progressive release was brought forward due to the individual’s involvement in the innovation department. This idea was taken and adopted into a more complete design as per the conversation with Dr Dobson.

Vertical Stirred Bioreactor Concept

The concept consists of a twin chamber unit, where the upper chamber is a bioreactor housing a vertically mounted nutrient block, baffles on the walls, a water inlet pipe from the lid and a 0.22 μm ceramic filter as the base. A magnetic stirrer would sit on top of the filter to induce turbulence in the reactor and more significantly at the membrane surface, preventing cake layer formation and sedimentation. The magnetic stirrer would be powered by copper coils being situated on the outside of the tank, level with the stirrer in the x-axis. A ceramic filter would be necessary to ensure the level of erosion caused by the magnetic stirrer would be kept to a minimum, hence increasing product lifetime.
The second chamber would sit beneath the ceramic filter, this would obviously be gravity fed from above, with a drain in the base and a peristaltic pump to be used to control the outflow of enzyme solution into the connecting pipework. A sterile vent will be situated at the top of this chamber to prevent the unit collapsing due vacuum formation.

Limitations of vertical reactor concept

Adequate oxygen diffusion in the top chamber was the major drawback of this unit. Partially permeable membrane was considered however it would make the unit structurally unsound.

We considered developing multiple units to work in conjunction with each other across home, industrial and sewer environments as a complete network to effectively eradicate FOG formation. A Further meeting with Dr Falconer was used discuss the points raised by Paul and David. The idea of altering the aspect ratio of the product was in effect raised, as Rob suggested the use of a vertical filter to reduce the issues of potential caking caused by the filter location on the previous design. After further thought the final flattened cylinder style design was in effect born.