(181i) Investigation of Fluid Flow Maldistribution in Microchannel-Based Hemodialysis Using Residence Time Distribution Analysis

Coblyn, M. Y., Oregon State University
Sharp, K., Oregon State University
Wagner, R., Oregon State University

More than 1.2 million people worldwide require regular hemodialysis therapy as a result of end-stage renal disease (ESRD) and that number is growing at a rate of 6-7% annually.  The current method of hemodialysis treatment requires a patient to visit a dialysis center approximately 3 times a week for a 4-5 hour treatment session. The current filtration method utilizes a hollow-fiber dialysis unit. This method inherently requires high dialysate volumes and high blood/dialysate flowrates to achieve sufficient removal of blood waste products.

A microchannel-based hemodialyzer provides improved efficiency, lower dialysate volumes, lower flow rates in a more compact system. This opens up opportunities in frequent long-duration dialysis (6-8 hour sessions during sleep) which better simulate natural kidney function and provide less physical stress to the ESRD patient.

Previous research has gone into the development of a microchannel-based hemodialysis device and although we have found very promising results in terms of performance, size, and fabrication, there are significant challenges to address before a device and fabrication process could be reasonably proposed for commercial production and use.  The issues stands with fluid flow maldistribution within the microchannel unit.  Flow maldistribution is due to header/channel design, fabrication tolerance, and the presence of gas bubbles in the device.  This flow maldistribution poses a significant problem to device efficiency and blood compatibility (blood shearing and coagulation effects). Channels in the array may receive uneven flow from the header and partially occluded channels act as a source of high shear stress on blood cells passing though.  In our current NIH funded research we are investigating the effect of all these factors and looking to provide a solution.

In order to investigate this issue, we require an accurate and repeatable methods/system for sensitive, high resolution investigation of: lamina design and architectural features, fluid dynamics, bubble behavior, tracer impulse-response, mass transfer efficiency, blood compatibility. We’ve propose the development of a test loop system that would allow for inline absorbance measurement and  high speed video capture of fluid flow, along capabilities for mass transfer characterization and hemolysis testing. The tracer impulse-response studies would allow for the construction of residence time distribution curves for our prototype devices and allow us to determine mean residence time and time dispersion.

We are currently assembling and benchmarking the entire system. In the late spring we will begin investigating microchannel geometry and their effects on fluid flow distribution in the dialyzer systems, along with comparison to results from a computational simulation package that is being developed in parallel with the test loop. The ultimate goal is to provide a computational tool for those working with microchannel dialyzers but for now, it allows us to compare experimental results with computational predictions and add a level of confidence to both systems.

This test loop will also allow for investigation of gas bubble/liquid interactions in microchannel systems. This bubble problem can be divided into three categories: macro-, meso-, and microbubbles. Macrobubbles exist in the header region and block off whole sections of channels in the array. Mesobubbles exist in a channel and completely block flow through that channel. Microbubbles also exist in a channel but only partially occlude the channel, adhering strongly to the channel wall. Both macrobubbles and mesobubbles are a significant problem to address but their solution lies in better assembly techniques and system priming methods. Microbubbles on the other hand will enter your system regardless of how tightly sealed the laminae are or how well the unit is primed with fluid before use.  Instead of avoid microbubbles, the proper solution is to allow them to easily pass through the system. The solution lies in developing a coating that would mask the energetic properties of the channel wall and therefore decrease the interfacial attraction between bubble and wall. Particular attention is being given to the design and fabrication of the Microbubble Generator, which would be a unique and innovative subsystem component. We are currently working on the fabrication of a device that would utilize sonic flow of gas through a micro-nozzle. Such a device could provide significant control over size of microbubbles formed by adjusted the fluid flowrate surrounding the micro-nozzle.

Our current projection is to complete header/microchannel geometry investigation midsummer 2011 and begin investigation of microbubble presence on fluid flow during the latter half of that summer.