In medical device applications, the development of anodic aluminum oxide membranes has been concentrated on biological fluid separation.
Nanoporous anodic aluminum oxide (AAO)
structures have received considerable attention in the research community for a
range of uses. The self-ordering and straight channel porous structure has been
exploited primarily as a template structure for producing nanorods/nanofibers
of Au, Pt, Cu and Ag, as well as several polymers.1-4
More recently, nanoporous AAO has also been used to develop nanoporous
membranes for a variety of medical devices.5-7 Here, the
development of AAO membranes has been concentrated on applications for
biological fluid separation. Proteins, toxins and viruses on the molecular
level are capable of being separated from human blood using synthetic polymer
membranes.
Hemodialysis, one of the applications that
uses polymer membranes, refers to the extracorporeal filtering of blood through
an artificial kidney for the purpose of maintaining or supplementing kidney
function. In this process, a blood circuit is formed when blood is drawn from a
patient and circulated continuously past a synthetic polymer membrane to remove
small- and middle-molecular-weight (MW) solutes normally passed by the kidney.
As the primary treatment modality for chronic kidney disease (CKD) and
end-stage renal disease (ESRD) patients, hemodialysis is the limiting step in
determining the performance of a dialysis session.
A uremic toxin is characterized by different molecular weight, degree of
protein binding, volume of distribution and charge. Low-MW nitrogenous waste
products are the most extensively studied group of these
toxins.8 These toxins feature a high diffusivity of mass
transfer characteristics due to their low molecular weight and lack of protein
binding.9
Improving the efficiency of maintenance dialysis can be accomplished in part
through the development of new dialysis membranes, whereby the survival rate of
the patient may be improved. The non-uniformity of pore distribution,
irregularities in pore shapes and size, and the limited reusability of current
dialysis membranes has led to extensive research in the area of membranes for
hemodialysis.
Nanoporous AAO membranes have a self-ordering pore arrangement with high
pore densities that are essential to maximize permeation and molecular flux
across the membrane in a fluid separation application (see Figure 1). AAO
membranes with this highly uniform and self-organized nanoporous structure are
an ideal choice over the contemporary cellulose-based and synthetic polymer
membranes in the hemodialysis application. AAO membrane advantages include high
porosity while maintaining uniform pore size between 5 and 300 nm, high
hydraulic conductivity (water permeability), uniform distribution of pores,
straight pore structure, and high resistance to chemical and temperature
degradation (sterilization).
Nanoporous AAO is created as a thin-film oxide during the electrochemical
process of anodization of an aluminum substrate. The aluminum substrate is
placed in an acid electrolyte as the anode in an electrochemical cell setup. As
the electrical potential is raised, a nanoporous array of aluminum oxide is
grown at a specified rate. Membrane characteristics, such as pore size,
interpore spacing and thickness, are highly dependent on the anodization
parameters, the details of which are shown
elsewhere.10-12 AAO membranes have been successfully
manufactured in both sheet and tube form.
High Porosity and Uniform Pore Size
Figure 2. A state-of-the-art PES membrane (surface view, a; cross-section, b).
High-resolution scanning electron microscope (SEM) images of the
morphology of an AAO membrane are shown in Figure 1. In comparison, a
polyethersulfone (PES) membrane currently used as a dialysis membrane is shown
in Figure 2. The pore sizes on the surface of the PES dialysis membrane are not
uniform, and the regularity of pore shapes is also unsatisfactory. Some pores
appear oval in shape while some appear as slits. In contrast, the pore shape on
the AAO membrane surface is uniformly circular. In addition, the cross-section
of the AAO membrane reveals straight channels of nanometer scale, whereas the
polymer membrane exhibits a tortuous structure.
The primary determinants of convective solute removal are the
sieving properties of the membrane used and the
ultrafiltration rate.13 The mechanism by which
convection occurs is termed solvent drag. If the molecular dimensions of a
solute are such that transmembrane passage occurs, the solute is swept
(“dragged”) across the membrane in association with ultrafiltered plasma water.
Thus, the rate of convective solute removal can be modified by changing either
the rate of solvent (plasma water) flow or the mean effective pore size of the
membrane. If a straight cylindrical pore model is considered (such as the case
in this ceramic membrane), the fluid flow along the length of a cylinder is
governed by the Hagen-Poiseuille equation:14 where DP is
the pressure gradient across the membrane (transmembrane pressure), Q is the
flow rate or ultrafiltration rate across the membrane, L is the length of pore
channel, and r is the radius of pore. So, the rate of ultrafiltration is
directly related to the fourth-power of the pore radius at a constant
transmembrane pressure. In other words, the convective transfer of solute is
determined by the fourth-power of the pore radius.
The diffusive properties of a dialysis membrane are determined mainly by the
porosity and pore size.15 Based on a cylindrical pore
model,16 membrane porosity is directly proportional to
both the number of pores and the fourth power of the pore radius (r4).
Therefore, diffusive permeability is strongly dependent on pore size. Studies
over the past 15 years suggest a direct relationship between delivered
urea-based hemodialysis doses and patient outcome.17-21
Since the elimination of low-MW nitrogenous waste products is mainly obtained
by diffusion through a dialysis membrane, higher porosity will achieve better
elimination of these uremic toxins. In other words, AAO membranes with higher
porosity can deliver a higher urea-based hemodialysis dose than polymer
membranes, given the same timeframe.
High Hydraulic Conductivity
A previous study by Huang et al. showed that hydraulic
conductivity (water permeability) of a sheet AAO membrane was approximately
twice that of a PES dialysis membrane.22 As stated
previously, PES dialysis membranes have an irregular pore structure (both inner
and outer surfaces) and a wide pore size distribution, whereas AAO membranes
have a highly monodisperse pore size distribution. The more uniform the pore
size of the membrane, the higher its hydraulic conductivity. Enhanced
convective transfer of middle- and large-molecular-weight solutes can therefore
be achieved by using nanoporous AAO membranes with these properties.
When comparing the cross-section of the two
membrane types, large differences in their structures are evident (see
Figures 1 and 2). The channel or path from the interior to the exterior
surface of the PES dialysis membrane is not straight. It is unclear if there
are true channels or paths from one side to the other. Instead, the
cross-section of the polymer membrane resembles a sponge-like material.
Accordingly, in practice, some blood or blood fragments might be left inside
this structure even after a thorough cleaning with chemical reagents. In
contrast, the channel or path connecting both surfaces of an AAO membrane is
straight and smooth. In addition to the above structure limitation, heat
disinfecting methods cannot be used on cellulose or synthetic polymer dialyzers
due to their inadequate temperature resistance. The high temperature resistance
of AAO membranes makes them better for heat disinfecting processes, thereby
significantly increasing the reusability.
Uniform Pore Size Distribution
Figure 3. Equivalent diffusion coefficient vs. molecular weight of AAO and PES membranes.
Hemodialysis membranes currently used in
dialysis equipment do not have a cylindrical pore structure with monodispersed
pore sizes; rather, they exhibit a wide pore size distribution and a tortuous
structure.
Experimental studies have been conducted on
mini-module hemodialyzers to measure the hydraulic conductivity, diffusive
permeability, rejection coefficient/sieving coefficient and clearance of
solutes (urea, creatinine, vancomycin and inulin) of nanoporous AAO tube
membranes.23-25 For the nanoporous alumina membrane, the
measured hydraulic permeability was: Lp = 8.7 × 10-9
m•s-1•Pa-1 (K = 30.3 × 10-15
m2•s-1•Pa-1),
while for a PES membrane, it has been reported as: Lp =
5.02 × 10-15 m•s-1•Pa-1 (K =
15.06 × 10-15
M2•s-1•Pa-1).26
The hydraulic conductivity of the AAO tube membrane is therefore approximately twice
that of the PES membrane.
The solutes’ diffusive permeability results are shown in Figure 3. The
diffusive permeability for three investigated solutes passing through the
alumina membrane is approximately 140% higher than for the PES
membrane.26 The difference is most likely due to the
pore structure that allows easier passage of solutes through the straight
channels on the AAO membrane.
Future Work
While
the methodology of manufacturing nanoporous AAO templates and membranes has
been studied and the effect of processing parameters on membrane properties has
been established, the extent
of biocompatibility (cytotoxicity, genetotoxicity, biofouling, complement
activation, etc.) still needs to be more closely scrutinized before it sees
widespread use in biological environments. In addition, the incorporation of
ion-sensing coatings or a drug could still be explored to further add value to
AAO structures in a very competitive medical device market.
For more information regarding AAO membranes,
contact EMV Technologies, LLC, 205 Webster St., Bethlehem, PA 18025; (610)
419-4952; fax (610) 419-2568; e-mail info@emvtechnologies.com; or visit www.emvtechnologies.com.
William H.Van Geertruyden William H. Van Geertruyden is a general manager at EMV Technologies, LLC, Bethlehem, Pa.
ZhongpingHuang Zhongping Huang is an assistant professor at Widener University, Chester, Pa.
References 1. C.R. Martin, Science 266, (1994) 1961. 2. C.A. Foss, Jr., G.L. Hornyak, J.A. Stockert, Journal of Physical Chemistry,
98 (1994) 2963. 3. C.K.
Preston, M. Moskovits, Journal
of Physical Chemistry, 97 (1993) 8495. 4. S.L. Pan, D.D. Zeng, H.L. Zhang, H.L. Li, Journal of Applied Physics, A
70, 637–640 (2000) 1.D. 5. E. Briggs, A. Walpole, P. Wilshaw, M. Karlsson,
E. Palsgard, Journal of Materials
Science: Materials in Medicine, 15 (2004) 1021-1029. 6. P.A.M. Darder, M. Hernandez-Velez, E. Manova, E.
Ruiz-Hitzky, Thin Solid Films,
495 (2005) 321-326. 7. G. Dawai, V. Yadavalli, M. Paulose, M. Pishko, C.
Grimes, “Therapeutic Micro and Nanotechnology,” Biomedical Microdevices, 5 (2003) 75-80. 8. W. R. Clark, J. H. Shinaberger, Blood Purification, 18 (2000)
260-263. 9. W. R. Clark, L.W. Henderson, Kidney International, 59
(Suppl 78) (2001): S298-303. 10. H. Masuda, K. Fukuda, Science, 268 (1995) 1466.
11. H.
Masuda, M. Satoh, Japanese
Journal of Applied Physics, 35, (1996) L126. 12. H. Masuda, F. Hasegwa, Journal of Electrochemical Society,
Vol. 144, No. 5, (1997) L127-L130. 13. L. Henderson, In Replacement of Renal Function by Dialysis (4th ed). Edited
by Jacobs C. Dortdrecht, Kluwer Academic Publishers (1996) 114-118. 14. R. Bird, W. Stewart, Lightfoot EN, in Bird RB, Stewart
WE, Lightfoot EN (eds), Transport
Phenomena, 1st edition, New
York, John Wiley and Sons (1960) 34-70. 15. W. Clark, C. Ronco, Nephrology Dialysis Transplantation,
16 (suppl. 5) (2001) 56-60. 16. W.
Clark, D. Gao, Contrib Nephrol,
Vol. 137 (2002) 70-77. 17. F. Gotch, J. Sargent, Kidney International, Vol. 28
(1985) 526-534. 18. W.
Owen, N. Lew, Y. Liu, E. Lowrie, J. Lazarus, New England Journal of Medicine, Vol. 329 (1993)
1001-1006. 19. T. Parker, L. Husni, W. Huang, N. Lew, E.
Lowrie, American Journal of
Kidney Disease, Vol. 23 (1994) 670-680. 20. A. Collins, J. Ma, A. Umen, P. Keshaviah, American Journal of Kidney Disease,
Vol. 23 (1994) 272-282. 21. R. Hakim, J. Breyer, N. Ismail, G. Schulman, American Journal of Kidney Disease,
Vol. 23 (1994) 661-669. 22. Z.
Huang, W. Zhang, S. Tang, et al., ASME IMECE
2004 Paper #59404. 23. A.
Attaluri, Z. Huang, A. Belwalkar, W. Van Geertruyden, D. Gao, W. Misiolek,
ASAIO Journal (January
2009) in press. 24. Z. Huang, A. Attaluri, A. Belwalkar, W. Van
Geertruyden, D. Gao, W. Misiolek, Proceedings of the ASME 2008 Summer Bioengineering
Conference (SBC2008), Marco Island, Florida, USA. 25. A.
Belwalkar, E. Grasing, W. Van Geertruyden, Z. Huang, W.Z. Misiolek, Journal of Membrane Science, Vol.
319 (2008) 192–198. 26. Z. Liao, E. Klein, C. Poh, Z. Huang, J. Lu, P.
Hardy, D. Gao., Journal of
MembraneScience, Vol. 256 (2005) 176-183.