The
use of MEMS devices in biotechnology requres accurate control of very small
flows. For example, peptide drugs need to be administered at rates on the order
of 1-10 uL/hour, and uTAS systems require strict control of fluid delivery in
order to maintain the continuity of reactions. The regulation of such flows
becomes increasingly complex as the flow rates decrease. In many situations,
accuracy must be maintained at a minimum of ten percent. Improved flow
regulation to meet such a requirement will come from physically plausible
modeling and prediction. However, one of the main problems with such micro-scale
devices is the prediction of flows in micro-scale channels – especially when
the molecular size of the fluid or particles is a significant percentage of the
diameter of the channel. Large molecular length-to-channel diameter ratios are
especially apparent with proteins, polysaccharides, DNA, RNA, synthetic
polymers, and other macromolecules. These macromolecular flows are found in a
wide range of integrated biological microfluidic devices that are currently
under development. A few of the potential applications include analysis and
diagnostics of biological fluids, sample extraction, purification and separation
of proteins, cell culture media delivery, protein flows over surfaces with
adhered substrates, and drug delivery. It is important to characterize such
macromolecular microflows to simplify and accelerate the design of MEMS devices
in biomedical applications, and to develop means for improved flow resolution
and control. Once these types of flows are understood, microdevices can be
designed so that they are better suited to specific applications.
In the medical industry, the localized delivery of macromolecules such as
large peptides, enzymes, and DNA to
specific sites in the human body is becoming increasingly important. Such localized
delivery often requires very small, highly concentrated doses of these
molecules, which, for synthetic peptides,
requires improved methods of delivery for clinical applications. This is because
traditional bolus injections, intravenous infusion, and oral administration are
not viable options for peptides, which are quickly and easily degraded in the
digestive system or by proteases (non-specific enzymes that break down peptides)
present throughout the body. These three delivery methods rely upon the
circulatory system to distribute the drug to its intended site, however the
circulatory system also distributes the drug to most of the body. Such
non-specific delivery allows the drug to affect many other sites and may cause
undesirable systemic effects. The dosage required for effective treatment is
also very high. Because the drug is diluted as it is spread by the circulatory
system, the concentration of drug decreases as the drug is advected and diffused
throughout the body. Thus, sufficient drugs must be administered to compensate
for the large-scale dilution that takes place throughout the volume of the body.
Peptides, in particular, must be administered at or very close to the site of
intended affect to minimize degradation before reaching the desired target.
While this type of delivery is more difficult, it has very distinct advantages.
Site-specific delivery not only minimizes the required dose (little or no
dilution before reaching the intended site), it almost eliminates undesirable
side effects in the rest of the body. Any drug that escapes the site is then
either quickly degraded or not concentrated enough to have a noticeable effect.
Investigations underway at the
University
of
Utah
are aimed at exploring peptide flows in
micro-scale passages to develop improved engineering models for the prediction
of these flows. In the past, modeling of peptide flows has been limited to
complicated molecular dynamics (MD) simulations that require specialized
computers that are not available to most engineers. Development of a method for
accurate finite element modeling (FEM) of these flows would make quick,
efficient peptide suspension modeling available to all engineers without the
large computational requirements of MD simulations. The development of FEM
methods requires inclusion of phenomena that are negligible or non-existent in
most single-phase flows, but have a significant effect on the characteristics of
microscale suspension flows of large molecules. In particular, slip phenomena
have large effects on flow characteristics. Cohesive slip between macromolecules
adhered to the wall and that in bulk is believed to be the main mechanism of
slip in macromolecular suspensions. This is illustrated schematically in Figure
1. Such cohesive slip is affected by a number of factors including the size,
make-up, geometry, and charge of the molecules as well as the make-up, charge,
and surface roughness of the channel walls.
Figure
1. Slip velocity V is caused by disentanglement of peptide in the bulk from adsorbed
peptides. This process leads to an “apparent” slip very close to the
surface. The slip length b is
determined by extrapolating the velocity profile to zero.
To our knowledge, characterization
of peptide or protein flows in microchannels is not available in the open
literature, with very little information on the behavior of slip flows in
suspension flows, in general. Most related past studies considered easily
attainable synthetic polymers, with simple molecular make-ups, dissolved in
water. In contrast, proteins have a much more complicated molecular composition
because they consist of a chain or chains of amino acids linked through peptide
bonds. The amino acid sequence is not ordered like the synthetic polymers that
have been investigated in the past. Twenty standard amino acids are arranged in
a different combination for each protein. Each amino acid residue in the protein
has a different side chain with different characteristics. Combined interactions
of polar, non-polar, and reactive side-chains with the solvent and other
side-chains determine the configuration of the protein. Such complicated protein
structure makes cohesive slip much different than that produced by simple
polymers. Other types of slip may also play a larger role in slip phenomena that
may be unique to protein suspensions. For example, the complicated structure of
proteins may have an effect on adherence to the wall or coherence to other
proteins in the bulk. They may repel each other and create a thin protein-free
layer near the wall that increases apparent slip, they may adhere to other
proteins so strongly that cohesive slip is greatly affected, or they may be
repelled from the wall such that true molecular slip occurs along the surface.
The goals of present research efforts at the University of Utah includes
the characterization of microchannel flows of peptide suspensions, and the
development of numerical models that accurately predict the behavior of these
types of flows in fundamental and practical applications. Specifically,
theoretical models will be developed to simulate the slip, viscous dissipation,
shear thinning, and molecular distribution phenomena present in
bio-macromolecular suspensions. These findings will then be used in the design
and validation of microdevices used with protein and peptide suspensions.
State of Utah Center of Excellence for Biomedical Microfluidics 
