Project Number: P13373
Nano-Fluidic Characterization
David WestMechanical Engineering / Justin Davis
Mechanical Engineering
David Sharp
Mechanical Engineering (Leader)
Abstract:
The main objective for this project is to build, test, and validate an apparatus to test various Nano-porous membranes and write a program to collect the characteristics of the flow across these membranes. Such characteristics include the pressure drop between both sides of the membrane, fluid flow rate through the system, and a measurement of the fluid temperature. The system itself consists of various fittings to ensure the same size diameter piping is used throughout the system, a membrane holder to house and protect the membrane, pressure transducers to observe the pressure difference, a flow sensor, a thermocouple, and a syringe pump to inject the fluid into the piping. The entire system will be kept secured on a ridged frame that will keep all parts connected and in place, preventing them from moving during the procedure. As fluid is injected through the pipe using a syringe pump, data for each of the fluid characteristics will be collected using various sensors. A fitting for secondary fluid injection can also be used for a secondary testing procedure, in which electrolytes and food coloring can be injected into the system during testing. The data collected during the experiment will then be exported onto LabVIEW, where the data can be observed and edited for further analysis.
NOMENCLATURE:
Nanoporous Membranes- A Small disk like material with microscopic pores throughout its surface area.
Carbon Infused Membranes- Membranes with carbon lining through each of its pores
NBIL- Nano-bio Interface Lab – Dr. Michael Schrlau’s lab at Rochester institute of technology that analyses nano-materials.
CNT- Carbon Nanotubes, membranes lined with carbon throughout their pores.
DAQ- Data Acquisition Device used to collect data from the pressure transducers.
bACKGROUND
Nano-porous membranes are used for a multitude of uses. They are primarily used for filtration uses in the medical industry where quick and precise filtration is necessary when injecting fluids into the body. These membranes primary usage in the NBIL will be as a template for building CNT’s. Currently however, there are no definitive ways to characterize the flow across these nanoporous membranes accurately for usage at the NBIL.
The current testing system uses a syringe pump to inject fluid through the membrane, and then have it immediately exit into a beaker so that an average flow rate can be determined. Since the exit of the system goes directly outside the system, the pressure is assumed to be atmospheric, so only the pressure just before the membrane would need to be measured. The new system design will ensure that all values obtained will be directly from the sensors themselves, eliminating any measurement errors during the analysis. Also, it is desired that the flow going through the membrane should be visibly seen. Below are the main components that the system must accommodate throughout the design process.
· Quantify expected flow across membrane through calculations and assumptions
· Reduce the cost of measurement devices and system parts
· Physically observe the flow from outside the system
· Collect and store all data from within the system
· Change the flow rate of the system
· Control the sampling rate of data collection
Design Process
Customer Needs:
Spec # / Specification (metric) / Unit of Measure / Marginal Value / Expected ValuesS 1 / Test rig size limit (l x w x h) / mm / 200x120x100 / 510X115X40
S 2 / Membrane diameter / mm / 13 - 25 / 13
S 3 / Development cost / $ / < 2,500 / < 2,500
S 4 / Measuring pressure range / kPa / < 500 / <690
S 5 / Measuring pressure accuracy / Pa / 0.1 / 1725
S 6 / Measuring flow rate range / mL/min / 0 - 3 / 0-5
S 7 / Measuring flow rate accuracy / microliter/s / 0.001 / 2.5
S 8 / Measuring temperature range / ᵒC / -20 to 100 / -20 to 100
S 9 / Measuring temperature accuracy / ᵒC / 0.01 / 1
S 10 / Fine sampling rate / per second / 100 / 100
S 11 / Coarse sampling rate / per minute / 1 / 1
In order to accurately design a system that meets the customer specifications, each need and specification was analyzed and changed based on how the proposed design would meet each specification. Shown in table 1, each of the main specification for the system to meet was listed along with the desired value for it to achieve once completed. The column to the right shows the proposed design’s specification values, allowing the two values to be compared and analyzed in order to come to a consensus as to which specifications are able to be obtained and which ones would need to be sacrificed in order to obtain the most accurate results possible. This process was repeated until the final design of the system was able to meet customer satisfaction.
Before the final design, an initial analysis of the fluid flow through the system was conducted so that a baseline of expected values can be obtained. This step in the design procedure is critical since the values that the system obtains will need to be cross-referenced with values from both the previous system as well as calculated values obtained prior to testing. That way, the validity of the measurements found from the constructed system is tested and secured. These measurements incorporate the fluid flow from the entrance of the system where the syringe pump will inject fluid, to the end of the system where the flow sensor is located. The calculations themselves will incorporate measuring the velocity of the flow throughout the system as well as the pressure difference found between both sides of the membrane.
To begin calculations, various assumptions of the flow across the system need to be made. Firstly, it is assumed that flow going through the pipe is fully developed laminar flow, since the flow will have enough time to become fully developed once it has reached the membrane. It will also be assumed that the fluid itself will remain incompressible and at a standard temperature and pressure. Since it is hard to predict exactly how many nano-pores are within each membrane, it is assumed that each membrane has a constant pore density across its surface area. This assumption allows for the calculation of pore number for the effective surface area of that membrane. Lastly, the effects of head loss throughout the system can be considered negligible, since previous data calculations have shown that the head loss of the system did not greatly impact the values found.
Once the assumptions are obtained, equations for the mass flow rate and the differential pressure in the system are derived based on standard fluid mechanic laws. These simplified equations are shown below:
Equation 1: Flow Velocity Equation 2: Differential Pressure:
The first equation shows the velocity of the fluid going through the system and the flow going through each the membrane’s pores. In this case, the area of the pipe refers to the area of the membrane pore. In the pressure change equation, the diameter of the membrane is pore and the thickness of the membrane is used along with the fluid flow going across the membrane. This calculation is valid since the pressure difference of both ends of the tube represents the pressure difference between both sides of the entire membrane. These equations were then input into MATLAB so that the effects of the velocity of the fluid and the exposed surface area of the membrane as well as the membrane diameter can be compared to see its effect on the pressure difference in the system.
After performing the calculations in MATLAB, the results from figure 1 are obtained. This shows that as the pore diameter of the membrane decreases, the flow rate of the fluid increases for the same pressure drop. The current standard membrane pore diameter is 147nm with carbon injected within the pores; this value is able to range between 200nm with no carbon inside to enough to reduce the tube to 100nm.
Software System Architecture:
In order to be able to collect information from each of the sensors in the system, a program was created in LabVIEW so that the data can be observed throughout the experiment and then exported to Excel for further analysis. In this system, there are the two pressure transducers, the thermocouple, and the flow sensor each taking separate measurements at the same instance. These sensors are all connected a computer via USB.
The two pressure transducers are connected to a data acquisition device connected to the computer and powered with a 19V power source. The DAQ device will serve as a way to convert the voltage values from the pressure transducers into data LabVIEW understands, which can then be edited so that a pressure value can be seen for each sensor. In order to adjust the voltage to display the correct pressure value, each transducer needs to be calibrated to see the relation between the voltage signal read and the actual pressure being given. That way, a simple equation can be written in LabVIEW to change the voltage signal into pressure values.
The temperature sensor is a T type thermocouple, which signifies the two types of metal used to measure change in resistance. It is first put into a National Instruments Thermocouple reader, which is then plugged into the computer for LabVIEW to detect. The block diagram for the thermocouple is similar to the one used for the pressure transducers; however the set-up for the temperature sensor is far more simple as it does not require calibration due to the sensor specific reader purchased.
The flow sensor comes with its own cable and program that can be used to read the flow values from it. However, to ensure that the values obtained for the flow rate matches the values for the pressure and temperature at the same instance in time, the measurements will be taken through LabVIEW using the flow sensor’s driver software. This driver software is integrated into the VI written for the other sensors, that way all values obtained will be able to be assured of their timing. Once all the sensors are connected, the block diagram should look similar to the one shown above. Note: Sensirion driver software is not shown, as it is extremely complex and would be hard to depict in Figure 3.
Figure 4 shows the LabVIEW front panel of the block diagram depicted in Figure 3. There are two graphs, thermometer, COM port protocols, and time controls. The first graph shows the pressure difference between the transducers while the next shows the flow rate of the fluid in the system (recorded as a function of time). The thermometer shows the current temperature of the flow. In order to set a time step, a “Time Step” field is available to the user that can be changed from any maximum level down to the minimum processing time of the flow sensor communication with the USB port. This value changes from computer to computer, dependent and the version of USB used as well as the allotted ram allocation of the computer used. If a lower value is entered than this minimum, the times step will bottom out at this value. A time step of “0” seconds will result in this minimum.
System Design:
The system itself consists of various couplings and fittings so that each of the sensors can be fastened securely to the system. First, a straight compression fitting is used so that the pipe from the syringe pump and the system piping can be connected. Next, a T type compression fitting is placed so that the first pressure sensor can be connected and take readings before the membrane holder. Another type of T-fitting (Luer Lock T) is put in place after to serve as both a pressure release point as well as a secondary fluid injection point just before them membrane holder. A custom made fitting is then used in order to attach the holder to the Luer T. Clear piping is pressed onto the holder exit so that the flow leaving the membrane can be seen. This leads to another compression fitting so a cross-type NPT coupling (where the thermocouple and the second pressure transducer are connected) can be used. Lastly, the flow sensor is attached after the cross fitting so that the flow rate of the fluid can be verified. The entire system is supported by two extruded aluminum ASS railings to ensure the sensors remains stable and components can be easily rearranged.
Each of the sensors (save the thermocouple) also has an extra support part designed to make sure the system does not bend due to their weight. The transducer supports consists of an L shape metal piece in which the transducers can secured with a set screw to keep the transducer from moving. The flow sensor is supported with a flat plate connected to either rail and screwed to the base of the flow sensor itself. The frame is supported with the option of two ¾ inch aluminum pieces each with four rubber bumpers to ensure added stability in the system, or the 4 detachable legs used to raise the system.
The system in Figure 5 shows a computerized design of the system made in SolidWorks. This system was used as a reference model so that the true system could be constructed. The picture in Figure 6 shows the fully constructed system used for the experimentation of the nano-porous membranes. Also, the membrane holder can be switched between a plastic or metallic version depending on user requirements. The plastic holder lets the user observe flow around the membrane during experimentation, but cannot survive high pressure, whereas the metallic holder cannot be seen through, but can survive higher amounts of pressure.