TMS QUICK LINKS:
  • TMS ONLINE
  • MEMBERS ONLY
  • MEMBERSHIP INFO
  • MEETINGS CALENDAR
  • PUBLICATIONS

JOM QUICK LINKS:
  • JOM HOME PAGE
  • MORE HTML ARTICLES
  • COMPLETE ISSUES
  • BOOKS REVIEWS
  • SUBSCRIBE

Nanostructured materials may play a significant role in controlled release of pharmacologic agents for treatment of cancer. Many nanoporous polymer materials are inadequate for use in drug delivery. Nanoporous alumina provides several advantages over other materials for use in controlled drug delivery and other medical applications. Atomic layer deposition was used to coat all the surfaces of a nanoporous alumina membrane in order to reduce the pore size in a controlled manner. Neither the 20 nm nor the 100 nm TiO₂-coated nanoporous alumina membranes exhibited statistically lower viability compared to the uncoated nanoporous alumina membrane control materials. Nanostructured materials prepared using atomic layer deposition may be useful for delivering a pharmacologic agent at a precise rate to a specific location in the body. These materials may serve as the basis for “smart” drug delivery devices, orthopedic implants, or self-sterilizing medical devices.

INTRODUCTION

Recent industrial and academic research efforts have focused on the development of nanostructured materials for use in biomedical devices, including medical prostheses, implantable biosensors, and drug delivery devices. Nanostructured materials are defined as materials that contain clusters, crystallites, molecules, or other structural elements with dimensions in the 1 nm–100 nm range.¹ Recent advances in the use of nanostructured materials for medical applications have resulted from two motivations. First, there is a natural evolution to nanoscale materials as novel processing, characterization, and modeling techniques become available. Second, specific interactions between biological structures (e.g., enzymes and other proteins) and nanostructured materials may allow for devices with unusual functionalities to be developed.

For example, nanostructured materials may play a significant role in controlled release of pharmacologic agents for treatment of cancer.² Systemic administration (distribution throughout the entire body) of many common chemotherapeutic agents is associated with significant side effects. In the common chemotherapeutic agent doxorubicin hydrochloride, those side effects include myelodysplasic syndrome, congestive heart failure, and mucositis. In addition, many protein- and DNA-based treatments that are being developed for treatment of cancer have relatively short in vivo activities. These chemotherapeutic agents cannot be administered in oral form because they may be metabolized by the liver, intestine, kidneys, or lungs before reaching systemic circulation. Recent work has examined delivery of chemotherapeutic agents at the site where they are needed; this route avoids diffusional and enzymatic barriers and provides complete and instantaneous absorption. Nanostructured materials may provide constant delivery of a pharmacologic agent to the site in the body where it is needed, providing appropriate treatment over an extended time while minimizing damage to healthy tissue.

Nanoporous polymer materials are inadequate for use in drug delivery. Many porous polymer materials are created using solvent-casting techniques.³ These materials have poor mechanical properties and large pore size distributions; for example, pore size variation is as large as 30%. In addition, polymer membranes contain 100–200 μm tortuous pores. Ion-track etching has also been used to form membranes. This technique produces a much narrower pore size distribution than that observed in polymer membranes; for example, pore variation in membranes produced using ion-track etching is within 10%. However, ion-track etched membranes have low porosities; pore concentrations under 109 pores/cm² are commonly observed. Porous silicon is another material that has been considered for use in drug delivery.^4,5 This material may be produced by electrochemically corroding silicon in solutions containing hydrofluoric (HF) acid. The pores propagate in the <100> direction of silicon. The average diameter of the nanocrystalline porous silicon layers can be modified by altering the electrolyte composition, the electrochemical current, or the dopant characteristics. It should be noted that porous silicon films undergo degradation under physiological conditions. Although several investigators have examined the use of porous silicon for drug delivery, it is unclear how patient-to-patient differences in physiologic status may affect degradation of porous silicon or drug release rates.⁶

Nanoporous alumina provides several advantages over other materials for use in controlled drug delivery and other medical applications. Matsuda et al. demonstrated that anodization, stripping of the oxide, and re-anodization produces an unusual material with nanoscale pores.⁷ Nanoscale pores are randomly formed on the alumina surface at the beginning of the anodization process. These pores self-organize into a hexagonal arrangement during their growth into the bulk material. This first oxide layer is removed using an aqueous solution of 1.8 wt.% Cr (VI) oxide and 6 wt.% phosphoric acid. A second anodization process is carried out on this template. The resulting material, known as “alumite,” contains long, columnar, ordered nanopores. These nanopores demonstrate long-range order. The structure can be described as close-packed cells in a hexagonal arrangement, with pores at the center of each cell. The pore size can be modified by selection of appropriate processing temperature, electrical field strength, or electrolyte.

Nanoporous alumina provides several advantages over polymers for use in drug delivery. Alumina is a bioinert ceramic that is stable in physiologic solutions. Nanoporous alumina membranes can be processed with smaller pore sizes (20–100 nm range) and more uniform pore sizes than polymer membranes. Finally, the anodization process provides precise control over pore size and pore distribution. However, there is a significant disadvantage to the use of nanoporous alumina materials in medical applications. Although aluminum is a constituent of several medical alloys (e.g., Ti-6Al-4V, ASTM F136), it is currently unknown whether aluminum is a biocompatible material.⁸ In this paper, a nanoporous alumina membrane was coated with TiO₂ using atomic layer deposition.⁹ The biocompatibility and corrosion resistance of TiO₂ is well known.¹⁰ In fact, TiO₂ is commonly used as a passivation layer in dental, orthopedic, and cardiovascular implants.^11–13 Having a conformal coating of TiO₂ over the alumina membrane is important to minimize corrosion and improve cell compatibility. Previous work by A. Canabarro et al. has indicated that cells grown on TiO₂ surfaces contaminated by small amounts of alumina exhibit impaired activity; for example, contamination by alumina may lead to impaired mineralization of matrix by osteoblasts (bone cells).¹⁴

Atomic layer deposition was used to coat all the surfaces of the nanoporous alumina membrane in order to reduce the pore size in a controlled manner.⁹ Recent work by S. Kipke et al. on unmodified nanoporous alumina membranes has shown that release of biological molecules in nanoporous alumina membranes is determined by the diameters of the pores.¹⁵ The TiO₂ coating will reduce the pore size in a controlled manner, maintain a narrow pore size distribution, prevent aluminum ions from leaching into the surrounding tissues, and create a biocompatible pore/tissue interface.

Atomic layer deposition (also known as molecular layering or atomic layer epitaxy) is a thin film growth technique in which alternating chemical reactions occur between gaseous precursor molecules on a surface.^16,17 The self-terminating gas-solid reactions allow for material to be deposited in a layer-bylayer fashion. Individual reactions are separated by purge steps that involve saturation with an inert gas. By saturating the substrate at each individual reaction, all surfaces of a given substrate can receive a conformal coating of identical thickness. The atomic layer deposition cycles can be repeated to control precisely the coating thickness. The self-limiting nature of the reaction between these precursors and the surface ensure that all exposed regions of a substrate, including areas that are only accessible via long, tortuous pathways, are coated uniformly and precisely. It is this ability to produce conformal coatings on non-planar substrates that makes atomic layer deposition very useful for functionalizing nanoporous materials, including membranes and aerogels. The conformal capability of atomic layer deposition is quite different from that of physical vapor deposition technologies such as evaporation and sputtering, which are limited by line-of-site constraints and can only coat the outer surface of a porous substrate. As such, atomic layer deposition is uniquely suited for depositing a conformal nanometer-scale film with precise thickness onto the surface of a nanoporous membrane.

Atomic layer deposition technology has been developed over the last 30 years. This technology has been commercialized for a range of applications in microelectronics, including electroluminescent displays, logic chips, and disk drives. There is currently intense interest in developing atomic layer deposition methods to a range of new applications outside of the realm of microelectronics. Much of the ongoing work involves coating nanoporous or nanostructured templates to impart the templates with beneficial chemical or physical functionalities. For example, nanoporous alumina membranes are a convenient platform for synthesizing nanotubes and nanowires using atomic layer deposition-based templating methods.

SEM TESTING AND RESULTS

The nanoporous alumina membranes (Whatman Anodisc, Kent, U.K.) had a thickness of 60 μm and an average pore diameter of 200 nm for ~58 μm of the membrane thickness followed by a separation region of ~2 μm with either 20 nm or 100 nm diameter pores. The alumina membranes were coated by atomic layer deposition using alternating exposures to titanium tetrachloride (TiCl4) and water vapor at a growth temperature of 300°C along with Si(100) witness samples to facilitate thickness measurement by ellipsometry. Following the atomic layer deposition of the TiO₂ coating, ellipsometry of the Si(100) witness samples yielded a coating thickness in the range of 7.6–9.2 nm. These thicknesses yield an atomic layer deposition growth rate for TiO₂ of 0.86–1.0 Å/cycle, which is in agreement with previous reports. Scanning electron microscopy (SEM) was performed on TiO₂-coated nanoporous alumina membranes using an S4700 microscope (Hitachi, Tokyo, Japan) with field emission gun electron beam source. Scanning electron micrographs obtained from the large-pore side of the nanoporous alumina membranes (Figure 1a) show the openings of the pores with diameters of ~200 nm. A higher resolution scanning electron micrograph of the nanoporous alumina membrane surface (Figure 1b) reveals nanocrystals with lateral dimensions of ~20 nm. These nanocrystals were not observed on the uncoated nanoporous alumina membranes (SEM images not shown) and therefore are consistent with TiO₂ nanocrystals from the atomic layer deposition coating. In addition, cross-sectional scanning electron micrographs obtained after fracturing the membranes demonstrate that the TiO₂ nanocrystals extend to the middle of the 60 μm nanopores (Figure 1c). Raman spectroscopy was performed on the nanoporous alumina membranes before and after the atomic layer deposition of TiO₂ coatings using an RM2000 Raman microprobe spectrometer (Renishaw, Hoffman Estates, Illinois). Prior to the TiO₂ coating, the Raman spectrum is featureless. After atomic layer deposition of TiO₂ coating, the Raman spectra show distinct bands corresponding to the anatase TiO₂ peaks (Figure 2). In addition, powder x-ray diffraction measurements performed on the TiO₂-coated nanoporous alumina membranes (not shown) also confirm the anatase phase.

Annealed TiO₂-coated nanoporous alumina membranes and uncoated nanoporous alumina membranes with 20 and 100 nm pore size were sterilized using ultraviolet light. All membranes were exposed for three hours on each side and rotated 90° every 45 minutes to ensure that all sides were sterilized. Upon completion of ultraviolet sterilization, the membranes were placed in 24 well plates using Akwa Tears® (Akorn, Buffalo Grove, Illinois) to prevent floating of the membranes. Human epidermal keratinocytes (HEK) were seeded with 20,000 HEK in 1 mL of KGM-2 per well in 24 well plates. Media was changed after 24 h. Once human epidermal keratinocytes were 60% confluent (24 h after seeding) timed sampling began. Media was then harvested after 24 h and stored at –80°C and the human epidermal keratinocytes were assayed for viability by MTT (3-[4,5-dimethyl-2-thiazol]-2,5- diphenyl-2H-tetrazolium bromide).¹⁸ The membranes were moved to a new 24 well plate so human epidermal keratinocytes grown in the wells did not influence the data. Human epidermal keratinocytes were incubated under cell culture conditions in MTT medium (0.5 mg MTT per 1 mL KGM-2) for 3 h. The cells were rinsed in Hank’s Balanced Salt Solution and the tetrazolium metabolized in the mitochondria extracted with 70% isopropanol. The absorbance was quantitated at λ = 550 nm in a Multiskan RC plate reader (Labsystems, Helsinki, Finland). The data were normalized to the uncoated nanoporous alumina membrane and were expressed as percent viability. Each coating/treatment was run in triplicate. Human epidermal keratinocyte viability was statistically compared using ANOVA (SAS 9.1 for Windows, Cary, North Carolina). Comparisons were made between uncoated and coated membranes using the Student’s t-test at p < 0.05. Both the 20 nm and 100 nm TiO₂-coated nanoporous alumina membranes did not exhibit statistically lower viability compared to the uncoated nanoporous alumina membrane controls (Figure 3). It should be noted that the increase in viability for the TiO₂-coated nanoporous alumina membrane may be explained by the interaction between the TiO₂ and the MTT assay marker; recent work suggests that nanomaterial-dye physical interactions can affect results in dye-based assays.^19,20

Two bacteria cultures, Staphylococcus aureus ATCC 29213 and Escherichia coli ATCC 25922 (American Type Culture Collection, Manassas, Virginia), were used to examine the antimicrobial properties of TiO₂-coated and uncoated nanoporous alumina membranes. Both bacteria were cultured overnight to log phase in tryptic soy broth (Difco, Detroit, Michigan) prior to experimental procedures. Cultures were washed three times and the membranes were sterilized. Cell densities were brought to 107 cell/mL and examined using a spectrometer before testing. Coated and uncoated membranes were mounted (two each) on sterile microscope slides with autoclaved stainless steel clips. Slides were separately placed in sterile Petri dishes, and were covered with either 20 mL of bacteria in phosphate-buffered saline solution or uninoculated phosphate-buffered saline solution. The Petri dishes were placed on a shaker table and incubated for four hours at 25°C. The Petri dishes containing TiO₂-coated nanoporous alumina membranes were exposed to a Woods Light ultraviolet source. After 4 h the Petri dishes were removed from the incubator. Bacteria and/or phosphate-buffered saline solution were removed from the Petri dishes using pipettes. The Petri dishes were subsequently washed three times with phosphate-buffered saline solution. Slides were air dried in a sterile hood, stained with Acridine Orange for two minutes, and then washed three times with filter sterilized (2 μm) deionized water. Ten fields were counted on each filter using a Model 510 laser scanning confocal microscope (Zeiss, Thornwood, New York). Microbial results showed a significant decrease in the amount of S. aureus and E. coli attachment to 20 nm pore size TiO₂-coated nanoporous alumina membranes as compared to their uncoated counterparts (Table I). On the other hand, the larger (100 nm) pore size TiO₂-coated nanoporous alumina membranes did not show an antimicrobial effect. Figure 4 shows controls with phosphate-buffered saline solution, demonstrating as few as 1 cell/field. Figure 5 demonstrates S. aureus growth on a 100 nm pore size uncoated nanoporous alumina membrane. Small cell aggregates were also observed on the 20 nm pore size uncoated nanoporous alumina membrane (Figure 6). Recent work by M.S. Wong et al. demonstrated that nitrogen-doped TiO₂ has activity against pathogenic microorganisms, including Staphylococcus aureus, Streptococcus pyogenes, and other bacteria found in hospital settings.²¹

P.C. Maness et al. demonstrated that TiO₂ photocatalysis promotes peroxidation of E. coli phospholipid membranes, resulting in cell death.²² These findings could be attributed to lytic activity induced by the ultraviolet light-TiO₂ interaction. It is interesting to note that the 20 nm pore size TiO₂-coated nanoporous alumina membrane inhibited microbial adhesion while the 100 nm pore size TiO₂-coated nanoporous alumina membrane did not (Figure 7). Differences in photocatalytic activity or membrane-cell interaction with membrane surface area may account for this finding.

Nanostructured materials prepared using atomic layer deposition may be useful for delivering a pharmacologic agent at a precise rate to a specific location of the body. Ceramic materials are associated with less inflammation than polymeric materials that are currently used for local drug delivery. In addition, nanoporous ceramics may provide greater control over release rate than polymers, since there are often difficulties when pharmacologic agents and polymers are dissolved in a given solvent.²³ These materials may serve as the basis for a “smart” drug delivery device, which allows for controlled release of a pharmacologic agent in response to electric field, magnetic field, pH, temperature, or light intensity. A smart system could release a gene or drug at a precise rate to the location in the body where it is needed. Nanoporous titanium coatings may also be useful for improving bone synthesis in orthopedic implants or in preventing infection.^24,25

The authors would like to thank K. Evaul and A.O. Inman (Center for Chemical Toxicology Research and Pharmacokinetics, North Carolina State University) for their assistance with the MTT assays. This document was prepared in conjunction with work accomplished under Contract No. DEAC09- 08SR22470 with the U.S. Department of Energy. Argonne National Laboratory (ANL) is a U.S. Department of Energy Office of Science Laboratory operated under Contract No. DE-AC02-06CH11357 by UChicago Argonne, LLC. SEM was performed at the Argonne Electron Microscopy Center.

1. O.V. Salata, J. Nanobiotechnology, 2 (2004), p. 3.
2. G. Stylios et al., Injury, 36 (2005), pp. S6–S13.
3. N. Wisniewski et al., Colloids and Surfaces B: Biointerfaces, 18 (2000), pp. 197–219.
4. S.H.C. Anderson et al., Physica Status Solidi A: Applied Research, 197 (2003), pp. 331–335.
5. L.T. Canham et al., Thin Solid Films, 297 (1997), pp. 304–307.
6. M.P. Stewart et al., Advanced Materials, 12 (2000), pp. 859–869.
7. H. Matsuda et al., Science, 268 (1995), pp. 1466– 1468.
8. M.L. Hegde et al., J. Molecular Neuroscience, 22 (2003), pp. 19–31.
9. M.A. Cameron et al., Langmuir, 16 (2000), pp. 7435–7444.
10. V. Faust et al., Key Engineering Materials, 206 (2002), pp. 1547–1550.
11. A. Canabarro et al., J. Biomedical Materials Research, 87A (2008), pp. 588–597.
12. B.D. Ratner, J. Biomedical Materials Research, 27 (1993), pp. 837–850.
13. D. Stoeckel et al., European Radiology, 14 (2004), pp. 292–301.
14. A. Canabarro et al., J. Biomedical Materials Research, 87A (2008), pp. 588–597.
15. S. Kipke et al., Advanced Functional Materials, 14 (2004), pp. 1184–1188.
16. R.L. Puurunen et al., J. Applied Physics, 96 (2004), pp. 7686–7695.
17. H.M. Alsyouri et al., Langmuir, 19 (2003), pp. 7307–7314.
18. T. Mosmann, J. Immunological Methods, 65 (1983), pp. 55–63.
19. N.A. Monteiro-Riviere et al., Carbon, 44 (2006), pp. 1070–1078.
20. N.A. Monteiro-Riviere et al., Toxicology and Applied Pharmacology, 234 (2009), pp. 222–235.
21. M.S. Wong et al., Applied and Environmental Microbiology, 72 (2006), pp. 6111–6116.
22. P.C. Maness et al., Applied and Environmental Microbiology, 65 (1999), pp. 4094–4098.
23. D. Losic et al., Nanotechnology, 19 (2008), p. 24570.
24. Y.Z. Yang et al., Vacuum, 83 (2008), pp. 569–574.
25 S.R. Chae et al., J. Membrane Science, 329 (2009), pp. 68–74.

Roger J. Narayan and Nancy A. Monteiro-Riviere are with the Joint Department of Biomedical Engineering, University of North Carolina and North Carolina State University, 2147 Burlington Engineering Labs, Raleigh NC 27695-7115 USA; Monteiro- Riviere is also with the Center for Chemical Toxicology Research and Pharmacokinetics, North Carolina State University; Robin L. Brigmon is with Savannah River National Laboratory, Aiken, SC; Michael J. Pellin is with the Materials Science Division, and Jeffrey W. Elam is with the Energy Systems Division, Argonne National Laboratory, Argonne, IL 60439. Prof. Narayan can be reached at (916) 696-8488; e-mail roger_narayan@unc.edu.