Nanofabrication
with biomolecules
by Dominic C. Chow1,3, Matthew S. Johannes2,3, Woo-Kyung Lee2,3, Robert L. Clark2,3, Stefan Zauscher2,3, and Ashutosh Chilkoti1,3*
The advances in biotechnology and nanotechnology
are spawning a new and exciting manufacturing tool
– bionanofabrication – which enables revolutionary
ways of building complex bionanostructures on
surfaces with nanometer precision. Here, we
highlight the opportunities and challenges in the
development of this emerging technology and discuss
its use in genomics, proteomics, nanostructures,
nanomaterials, drug discovery, and synthetic biology.
To provide useful insights into the future directions of
bionanofabrication, current developments in
characterization techniques, fabrication methods, and
their integration and control in manufacturing
processes are discussed.
1Department of Biomedical Engineering,
2Department of Mechanical Engineering and Materials Science,
3Center for Biologically Inspired Materials and Material Systems,
Duke University,
Durham, North Carolina 27708-0281, USA
*E-mail: chilkoti@duke.edu
30 December 2005
The excitement generated by nanotechnology
derives from the promise of manipulating matter
atom-by-atom and molecule-by-molecule to create
devices with performances and functionalites that are
orders-of-magnitude better than those provided by
current manufacturing technologies. The rapid
development of nanoscale patterning and
lithography technologies plays a substantial role in
realizing this dream. The invention of synthetic
photosensitive compounds, e.g. poly(vinyl cinnamate)-
based negative photoresists1 and diazoquinone-based
positive photoresist2, in the 1930s laid the
foundation for many modern-day lithography and
etching techniques used in microfabrication. For
years, devices such as microprocessors3 and
microelectromechanical systems (MEMS)4,5 have
been machined out of crystalline Si by lithography
(e.g. photolithography, X-ray lithography, and
electron- and ion-beam lithography), as well as by
bulk micromachining processes such as wet and
dry chemical etching, SiO2 growth, and vapor
deposition6.
While engineers and scientists race to shrink the size of
transistors and MEMS components through nanofabrication
to create the next generation of high-performance electronic
devices, biologists and life scientists have just begun to
employ micropatterning and, to a more limited extent,
nanopatterning techniques to build high-throughput
detection systems for genomic and proteomic studies7-9.
ISSN:1369 7021 © Elsevier Ltd 2005

Bionanofabrication lies at the intersection of nanotechnology
and biotechnology, and in many cases can trace its origins
back to technologies that are at least a century old. Many
strategies employed in bionanofabrication, especially those
for nanolithography and nanopatterning of biomolecules
(top-down nanofabrication), can trace their origins to
analogous strategies in conventional micro- or
nanofabrication. For example, soft lithography, used to
deposit biomolecules over a large area in a parallel fashion10,
is an analog of photolithography used to create transistors on
a Si substrate, and arguably owes much to methods
developed in the printing industry. Similarly, dip-pen
nanolithography (DPN) directly places biomolecules on
surfaces at the nanoscale using micromachined cantilevers of
an atomic force microscope (AFM)11 and is the nanoscale
analog of writing with quills.
The early development of bionanofabrication has largely
focused on the creation of biologically active structures for
biosensing and diagnostics by fairly straightforward
modifications and, in some cases, the extension of
conventional nanofabrication techniques. More recently,
there is a growing interest in using biomolecules as active
components in nanofabrication processes and the
development of new biomolecule-based materials systems.
For example, enzymes can be deposited to synthesize and
excavate nanostructures of nucleic acids and proteins at the
nanoscale12,13. In addition, the use of biomolecules for
nanoscale assembly has become a promising and effective
solution to difficult positioning problems in the fabrication of
nanostructures (bottom-up nanofabrication).
The capability of bionanofabrication to manipulate
biomolecules with nanometer precision on surfaces in a
massively parallel fashion is potentially revolutionary
because it would enable the creation of novel devices with
performance metrics that go beyond those currently
envisioned. However, the direct application of
nanofabrication techniques and processes to realize the
potential of bionanofabrication remains a difficult, though, in
our opinion, not insurmountable task. In this review, we
highlight the opportunities and challenges that researchers
and scientists face in developing bionanofabrication
processes. We provide specific examples of the uses of
bionanofabrication in the areas of biosensing, nanomaterials,
drug discovery, and synthetic biology. Next, we discuss
current developments in bionanofabrication techniques and
RESEARCH REPORT
processes. Finally, we outline potential future directions, with
particular emphasis on fabrication processes and their
control, integration, and characterization.
Opportunities and challenges
The incorporation of ‘soft-wet’ biological components into
conventional nanofabrication platforms designed and built for
‘hard-dry’ semiconductors, conductors, and dielectrics brings
both new opportunities and challenges since biomolecules
possess some unique properties:
• A wide range of biomolecules, including nucleic acids,
proteins, lipids, and oligosaccharides, react with other
biological components by molecular recognition14,15,
which is important for bottom-up nanofabrication based
on self-assembly;
• Enzymes can catalyze the synthesis and removal of both
biological and synthetic molecules16,17;
• Biomolecular reactions (biotransformations) are highly
selective and site-specific17. There are many enzymes that
cleave DNA at particular sites (restriction enzymes) and
link two pieces of DNA together (ligases)17. Proteases that
digest proteins at specific sites16,17 and enzymes that add
functional motifs to proteins18 are just a few examples of
protein-modifying enzymes;
• Biomolecular reactions are often highly efficient under
physiological conditions, so the yields of
biotransformations are substantially higher than those by
chemical syntheses19,20; and
• The use of biomolecules and biological processes is often
environmentally friendly, so treatment of the waste
products can be minimal and the by-products pose little
health risk compared to the reagents used in
semiconductor processing.
On the other hand, there are also significant constraints
associated with the use of biomolecules in
bionanofabrication. These are:
• The ultrahigh-vacuum (UHV) conditions used in
conventional nanofabrication approaches are incompatible
with biomolecules, whose function and structural integrity
in most cases are destroyed in a high-vacuum
environment;
• Even in an aqueous environment, many proteins can easily
lose their native, active conformation after deposition
because they can unfold and bind nonspecifically to a
surface; and
December 2005 31
 RESEARCH REPORT
• Biological reactions, with some notable exceptions21,22,
must take place in an aqueous solution or buffer, which
limits their uses in many electronic applications.
Current applications and potential uses
The unique challenges posed by the use of biomolecules in
nanofabrication are substantial, but surmountable. To
fabricate efficiently with biomolecules and make use of their
unique properties, a radical shift away from the high-energy,
high-vacuum processing paradigm of conventional micro- and
nanofabrication is needed. Although the field of
bionanofabrication is still in its infancy, its impact and
promise go far beyond biological applications where creative
uses of biomolecules remotely resemble those envisioned by
biologists who originally studied and isolated them.
Genomics and proteomics
Early implementation of bionanofabrication targets genomics,
proteomics, and high-throughput biosensing and diagnostic
applications. Bionanofabrication strategies are used to reduce
the spot size of the deposited biomolecules, to increase the
spot density of arrays, and to improve the throughput of
deposition processes23-25. Through miniaturization of
diagnostic devices and equipment, the cost efficiency, speed,
and accuracy of micro- or nanoarray-based detection
technology can be improved substantially by reducing the
amounts of reagents required, the reaction volumes, and
detection errors26.
Nanostructures and nanomaterials
As bionanofabrication becomes increasingly mature and
sophisticated, more and more research efforts will turn to
Fig. 1 DNA-based bionanofabrication by self-assembly of DNA molecules. (A) DNA strands, which have complementary sticky-end overhangs, self-assemble into a branched junction.
(B) These branched junctions can further self-assemble into DNA nanogrids owing to the orientation of the complementary sticky ends, as shown in the AFM images of DNA lattices.
(Reprinted with permission from102. © 2003 AAAS.)
32 December 2005
exploit the intrinsic functionality of biomolecules to create
new nanostructures, processes, and nanomaterials systems.
For example, bacterial S-layer proteins27,28 and DNA tiles
(Fig. 1)29 self-assemble into crystalline two-dimensional
templates with highly controllable and predictable periodic
features for patterning metal nanostructures. The assemblies
have potential applications in DNA computing and
nanoelectronics30-32.
Recently, the direct deposition of biomolecules has gained
considerable attention, because it allows the precise
placement of biomolecules at the nanoscale. A variety of
biomolecules such as collagen33, oligonucleotides34, biotin-
streptavidin nanostructures35, elastin-like polypeptides
(ELPs)36, and DNase I12 can be deposited to create
bionanostructures by DPN for biosensing and nanoscale
surface modifications. In addition, the roles of novel
biomolecules in miniaturized fluidic systems, and potentially
nanofluidics, are exemplified by the recent use of thermally
or pH-responsive proteins as biological actuators in
microfluidic systems37.
Drug discovery and synthetic biology
More recently, new and exciting opportunities for
bionanofabrication have emerged in the fields of synthetic
biology and recombinant protein engineering. The firm
integration of recombinant protein engineering techniques
such as phage display38, directed evolution39,40, and
combinatorial libraries41,42, as well as the discovery and
isolation of novel proteins from extremophiles (microbes that
thrive in extreme environments)43,44, would further
strengthen bionanofabrication approaches by providing

Fig. 2 Micropatterned DNA arrays used as templates for high-throughput gene synthesis
and protein expression. (A) Single-strand DNA immobilized on a microchip surface was
used as a template for generating double-strand DNA by polymerase chain reaction
(PCR). (B) Synthesis (left) and cleavage (right) of oligonucleotides from a microfluidic
microchip, visualized by fluorescent scanning microscopy. Details of microfluidic
chambers and connecting channels are shown in the inset. (Reprinted with permission
from45. © 2004 Nature Publishing Group.)
proteins with improved thermal stability, binding specificity,
and solvent compatibility.
This integrated approach, albeit at the microscale, is
exemplified by a recent report that micropatterned DNA
arrays could be used as templates for high-throughput gene
synthesis and protein expression (Fig. 2)45. In this specific
example, the use of patterned oligonucleotide sequences is
extended to drug discovery and lead validation, also providing
a valuable tool for systems biology. This technology can
easily be applied and adapted to DNA nanoarrays,
significantly improving their throughput and processing
capability.
Another emerging trend is the use of recombinant
nanostructured biocompartments such as vesicles, viruses,
and phages, which allow predefined functional biological
components to self-assemble into a sophisticated self-
contained system. These biocompartments are ideal for
bionanofabrication. They not only serve as nanoscale reaction
chambers for mineralization and metallization46,47, but also
provide a means to introduce docking sites to spatially direct
the assembly of biological, metallic, and semiconducting
nanostructures (Fig. 3)48.
Finally, another major opportunity for bionanofabrication
lies in harnessing the amazing biotransformation capabilities
of enzymes to potentially synthesize semiconducting
materials. For example, silicatein filaments and subunits have
been isolated in a marine sponge and were shown to catalyze
RESEARCH REPORT
Fig. 3 Nanostructured biocompartment M13 bacteriophage for nanowire synthesis.
(A) Visualization of different processes in nanowire synthesis for the nucleation, ordering,
and annealing of virus-particle assemblies. (B) Nucleated particles self-assemble
symmetrically on the virus for aggregation-based annealing. (C) The rigidity and packing
of the expressed peptides promote high-order self-assembly of M13 bacteriophage along
the preferred orientation in nucleated particles. (D) The bacteriophage contains
genetically modified capsid and ends that are encoded for in the enclosed phagemid DNA.
(Reprinted with permission from48. © 2004 AAAS.)
polymerization of silica and Si at low-temperatures19,20. This
discovery heralds the future development of Si deposition at
both micro- and nanoscales with novel enzymes under
ambient conditions, and blurs the distinction between
biological and nonbiological components used in
bionanofabrication.
Current developments in techniques
and processes
Taking soft lithography down to the nanoscale
Soft lithography was first developed in the Whitesides
laboratory at Harvard in 199349. Initially, the process was
termed microcontact printing (µCP), in which
polydimethylsiloxane (PDMS) stamps with micron features
were cast against negative masters generated by
photolithographic processes50. These stamps were coated in
alkanethiol inks and used to create self-assembled
monolayers (SAMs) on substrates through conformal contact
in a massively parallel fashion. The technique has spawned
many other related techniques and variations10,51.
Most notably, in the context of this review, soft
lithography was first used to pattern proteins in a direct print
fashion in 199852; chicken immunoglobulins (IgGs) were
directly patterned onto glass and polystyrene surfaces at the
December 2005 33
 RESEARCH REPORT
nanoscale for protein recognition52, and on polymers using
the biotin-streptavidin adapter system53,54. In addition to
PDMS, other materials are being studied for their applicability
to pattern biomolecules. By increasing the strength of the
stamps used for printing, feature sizes of µCP can be reduced
to less than 100 nm (nanocontact printing, nCP) by avoiding
the complication of stamp collapse. For example, stamps
made out of polyolefin plastomers (POPs) exhibit better
mechanical stability than PDMS and are able to print
proteins, particularly fibrinogen, down to 100 nm feature
sizes55. The development of protein patterning was further
extended to the printing of multiple proteins using
microchannels56 and single proteins (various antibodies) on a
surface (line widths <100 nm) (Fig. 4)57. These studies
highlight the potential impact and current limits of soft
lithography for construction with proteins at the nanoscale.
Soft lithography has also been used to directly pattern
DNA. For example, Xiao and coworkers used µCP with precise
alignment to fabricate and synthesize oligonucleotide arrays
on surfaces in a step-wise, single-nucleotide fashion for DNA
hybridization studies58. Another study demonstrated the
direct printing of DNA-surfactant molecules on the
microscale59. More recently, patterning of 20 bp
oligonucleotides and 500 bp and 1600 bp PCR fragments was
accomplished with a feature size of about 800 nm60. These
Fig. 4 Nanocontact printing (nCP) of proteins. (A) Steps highlighting the nCP process.
Typically a polymeric stamp is cast against a master, coated with an inking solution, and
then brought into conformal contact with the target substrate, which transfers target
molecules to the substrate. (B) AFM tapping mode image of rabbit antibodies patterned
on a glass substrate using nCP. Patterned line widths are around 100 nm. (Reprinted with
permission from57. © 2003 American Chemistry Society.)
34 December 2005
studies show how soft lithography can be used to fabricate
DNA arrays and point toward the simultaneous printing of
multiple oligonucleotides.
Scanning probe lithography for bionanofabrication
DPN is a versatile scanning probe-based technique for
fabricating both organic and biomolecular nanostructures. In
general, DPN involves the coating of an atomic force
microscope (AFM) cantilever tip with the desired molecule to
be transferred. Once the tip is brought into contact with the
target substrate, the molecules assemble on the surface
because of their specific or nonspecific affinity. When the tip
is translated across the surface, the molecules are deposited
in the wake. Originally, this technique was used to directly
deposit organic molecules on surfaces at the nanoscale,
e.g. an alkanethiol onto Au61, an organosilane onto Si62, and
a positive onto a negative polyelectrolyte, or vice visa63. This
approach was made more generic in later work by
nanopatterning a SAM of 16-mercaptohexadecanoic acid
(MHA) on Au surfaces to serve as a template for
nanopatterning biomolecules. The advantage of this approach
is that MHA can be reproducibly nanopatterned by DPN and
is easily functionalized with biomolecules by reaction with
amine groups, therefore the DPN process does not have to be
optimized empirically for each combination of substrate and
ink. For example, immunoglobulin G64, lysozyme64, biotin-
streptavidin35, ELP (Fig. 5)36, alkylamine-modified DNA65, and
cowpea mosaic virus66 could be nanopatterned with feature
sizes on the order of 100 nm by this approach.
Direct-write DPN, though more difficult, has also been
developed for the deposition of biomolecules onto surfaces.
For example, a hexanethiol-modified oligonucleotide ‘ink’ was
deposited on Au and oxidized Si substrates, with feature sizes
from 200 nm to microns34. Similarly, this strategy was
applied to DNA labeled with fluorescent molecules of
different colors for direct optical detection67. This direct-
write approach is also applicable to proteins, demonstrated
by the successful deposition of thiolated collagen and
collagen-like peptides onto Au substrates, generating 30 nm
wide line patterns for cell binding assays33. This approach has
been extended to catalyze biotransformations with nanoscale
spatial precision; in the first example of this approach,
DNase I molecules were deposited locally by DPN to digest
an oligonucleotide SAM with nanoscale precision (Fig. 6)12.
Conversely, this strategy should also be applicable to create
DNA nanostructures enzymatically, for example, by terminal

Fig. 5 Fabrication of a nanoarray of ELP. (A) A template of MHA was created by DPN,
followed by backfilling with SH-(CH2)11(OCH2CH2)3-OH (EG3-OH), activation of MHA
with N-ethyl-N’-(dimethylaminopropyl) carbodiimide/N-hydroxy-succinimide
(EDC/NHS) and end-grafting of ELP. (B) AFM tapping-mode image of a 10 x 9 ELP dot
array in PBS buffer at room temperature, with an enlarged view of an area indicated and a
representative cross section, showing a typical feature height of 5-6 nm and a lateral
feature size of ~200 nm. (Reprinted with permission from36. © 2004 American
Chemistry Society.)
DNA
Fig. 6 Scheme for procedures of enzyme nanolithography. (A) An oligonucleotide SAM
was digested with DNase I deposited by DPN after activation with magnesium Mg2+.
(B) Tapping-mode AFM image and line profile across the image of the oligonucleotide
SAM after digestion. The decrease in height in the trenches is ~3 nm. (Reprinted with
permission from12. © 2004 American Chemistry Society.)
RESEARCH REPORT
deoxynucleotidyl transferase, which repeatedly adds
mononucleotides to the 3’ end of oligonucleotide initiators
(Fig. 7)68.
In a complementary bionanofabrication approach that was
demonstrated on peptides, a Staphylococcal serine V8
protease-functionalized AFM probe was used to digest
carboxylic acid groups of peptides coated on a surface
substrate69. In another study, streptavidin-conjugated
enzymes were attached to the tip of an AFM cantilever
coated with biotinylated groups, by scanning on surface
substrates coated with an enzyme of choice. Using this
technique, a probe functionalized with alkaline phosphatase
was prepared and used to locally generate nanostructures of
a water-insoluble complex in the presence of its substrate
and cofactor (Fig. 8)13. These prototypical bionanofabrication
approaches demonstrate the capability of DPN for both
creating and excavating bionanostructures.
One of the limitations of DPN is the requirement of
reinking for patterning biomolecules on a large area.
A technique called anodization lithography, in which an
electric potential is applied between a conducting substrate
and an AFM cantilever to oxidize the area beneath the tip70,
has been developed. An attractive feature of anodization
lithography is the elimination of an ink from the process, so
that nanopatterned templates can be generated with high-
spatial resolution in a single uninterrupted scan. This
approach is particularly powerful for bionanofabrication if the
anodization reaction is carried out on a protein-resistant
material. It generates nanoscale patterns of reactive groups
DNA
Fig. 7 (A) Surface-initiated growth of DNA by terminal deoxynucleotidyl transferase
(TdTase) from Au substrates that present an SAM of an oligonucleotide-terminated thiol
5'-SH-(CH2)6-T25. TdTase repeatedly adds mononucleotides in 5' to 3' direction to the
oligonucleotide initiator. (B) Tapping-mode AFM image in air for Au arrays of 100 nm with
the DNA SAM after a 2 hr incubation with TdTase. Inset: image at a higher magnification
(1 µm x 1 µm) and the line profile of the nanostructures below the AFM image. The dotted
line represents the average height of Au arrays and immobilized DNA SAM. (Reprinted
with permission from68. © 2005 American Chemistry Society.)
December 2005 35
 RESEARCH REPORT
Fig. 8 Enzyme-assisted nanolithography. (A) An enzyme alkaline phosphatase was
immobilized on an AFM tip. In the presence of an enzyme substrate in the surrounding
medium, alkaline phosphatase generates a water-insoluble complex, which precipitates
on the surface. Surface modification was performed by two procedures: (B) the tip was
rested at a particular position for 20 s, then moved rapidly to a different position to induce
a localized enzymatic reaction. (C) Alternatively, the tip was slowly moved with a velocity
of 10 nm/s across the surface to form a continuous line. The surfaces were then imaged
with the same tip in tapping mode AFM after enzymatic reaction. (Reprinted with
permission from13. © 2005 American Chemistry Society.)
to which protein molecules can be conjugated, while
ensuring minimal nonspecific protein adsorption in the
background. Bovine serum albumin (BSA) dot patterns with a
feature size as small as 26 nm have been fabricated with this
strategy71.
Along a similar line, nanopatterns of biomolecular
nanostructures can be generated on resist-thiol-precoated Au
substrates by mechanical removal of resist molecules
(nanoshaving) followed by subsequent chemical
modifications, or replacement of resist molecules by
biomolecules in the solution in which nanoshaving is
performed (nanografting). The nanografting strategy is
applicable to a variety of proteins as demonstrated by the
fabrication of 40 × 40 nm2 to 200 × 250 nm2 square
patterns of BSA, lysozyme, and rabbit IgG using this
technique72. In addition, nanografting of DNA-derivatized
Au nanoparticles on Au surfaces at below 100 nm lateral
resolution has also been demonstrated73. However,
nanografting is often limited by its slow patterning speed.
This problem can be overcome by maintaining a small droplet
of the desired solution between an AFM tip and a hydrophilic
surface (instead of submerging the AFM tip in solution) while
36 December 2005
nanopatterning is performed. This technique is known as
meniscus force nanografting, which is capable of generating
nanopatterns with a lateral resolution as low as 60 nm on
Au surfaces74.
Nanopipettes for bionanofabrication
Nanopipettes are micromachined pipettes that typically have
inner diameters down to 3 nm75. Pioneering work with
nanopipettes as force sensors in scanning probe microscopy76
led to their use to transfer a liquid to a surface in a controlled
fashion75. For example, this technique was used to deliver an
etchant onto a Cr layer to produce trenches with line-widths
down to 100 nm. Bruckbauer and coworkers were the first to
report the deposition of biotinylated single-stranded DNA
onto a streptavidin-functionalized surface77 and of multiple
biological entities onto unmodified surfaces78 by applying a
voltage bias between the surrounding aqueous medium and
the solution inside the nanopipette, based on scanning ion
conductance microscopy (SICM)79.
In later work, nanopipettes were used in an ambient
environment for the direct printing of protein G at a
linewidth of 500 nm on an aldehyde-functionalized surface
and green fluorescent protein (GFP) onto a BSA-coated
surface at ~250 nm80. In a different study, a protease,
trypsin, was delivered through a nanopipette to locally digest
adsorbed BSA, resulting in holes that were 340 nm in width
and 660 nm in depth (Fig. 9)81. To better control the channel
size of nanopipettes, a focused ion beam was used to create a
nanochannel inside an AFM tip to serve as a reservoir that
acted like a nanopipette82, as demonstrated by the
deposition of polystyrene nanoparticles in attoliter volumes
to both hydrophobic and hydrophilic surfaces. This approach
should also be applicable to the deposition of biomolecules
on a surface.
Future directions
Although the field of bionanofabrication has undergone
impressive advances, it is still in its infancy and requires
significant research and development to fully realize its
potential. Below, we outline what we believe to be key
areas that will see significant advances over the next few
years.
Fabrication processes
Current research initiatives are focused primarily on adapting
existing nanolithography techniques, which are often slow
and serial, to deposit biomolecules. The majority of the

Fig. 9 (A) Micrograph of a nanopipette with a 200 nm aperture for dispensing biomolecules.
(B) Nanochannels were enzymatically created by dispensing a protease (trypsin) from the
nanopipette onto a BSA layer on glass. Line one was created first at a constant scanning
velocity, followed by line two. At the intersection, the depression is the sum of the two
lines. (Reprinted with permission from103. © 2005 American Chemical Society.)
research efforts are engaged in prototyping the deposition
and excavation of biomolecules through single-step
biotransformation processes, while little attention is currently
paid to parallel, multiple-step processing. The next challenge
in bionanomanufacturing will be the integration of these
individual processes into an overall process strategy that
allows the fabrication of devices and systems designed for
real-world applications.
For soft-lithography and scanning-probe-based
bionanofabrication techniques, the improvement of feature
resolution is still one of the most important challenges, as it
limits the patterning of biomolecules with nanometer
resolution83. Empirical optimization of these methods, while
useful, must be supplemented by fundamental studies that
combine the physical chemistry and mechanics of these
processes. For scanning-probe-based approaches, transport-
related factors such as surface diffusion, scan speed, probe-
ink interaction, substrate-ink interaction, and ambient
humidity must be quantified and controlled to maximize
resolution, fidelity, and process reliability84-86. For soft-
lithography approaches, the aspect ratio of the stamp
features, printing contact time, and mechanical properties of
RESEARCH REPORT
the stamp and substrate need to be evaluated as they are the
predominant factors controlling resolution87-90.
Process control
Another major issue in bionanofabrication is the need to
radically increase process throughput. While scanning-probe-
based approaches are inherently serial, great strides are being
made toward the implementation of massively parallel
cantilever arrays that work either passively or actively in
modifying surfaces to improve throughput91-93. However,
substantial challenges remain in controlling individual
cantilevers in these arrays. These massively parallel
approaches also require more sophisticated process control
systems that aid in locating and aligning nanostructures
repeatedly and reliably. Since positioning accuracy depends
largely on the performance of sensing elements, there is an
urgent need for better nanoscale metrology technologies94,95.
Bionanofabrication approaches that are based on the
sequential deposition and/or removal of biomolecules require
continuous position confirmation and realignment. One
approach to achieving this is by fast AFM imaging (<1 s),
which allows for a quick alignment registration between
fabrication steps. Alternatively, AFM can be used in
conjunction with confocal microscopy to locate patterned
areas optically96. For the registration of multiple-step
stamping processes with µCP or nCP, an innovative solution
employs a lock-and-key type alignment after each
fabrication step97.
An area that deserves further research is the repeated
inking of the tip or stamp with multiple biomolecules, which
is crucial to increasing process throughput. One approach to
deal with this problem is to create a PDMS stamp with
positive features with differential heights, so that various
biomolecules can be patterned when the applied force is
varied98. Another approach focuses on the development of
instruments and techniques that use micro- or nanofluidic
networks to deliver multiple inks in the proximity of the
patterned area.
Process integration
A significant bottleneck in bionanofabrication is the lack of
seamless integration among fabrication and assembly steps
that operate at length scales ranging from the molecular to
the millimeter scale. Research on bionanofabrication remains
fragmented with most endeavors focused on individual
processes, although integration among different
manufacturing platforms and control systems is much
December 2005 37
 RESEARCH REPORT

needed. A vision of a hierarchical, multiscale

biomanufacturing process involving several
(bio)nanofabrication steps is shown in Fig. 10. One can
imagine that bionanostructures could be created through
step-wise biotransformations using fast scanning or optical
detection for alignment, together with sophisticated control
algorithms. These nanoscale transformations could be
observed and controlled by relaying optical, chemical, or
electrical signals to and from the macroscopic world.
However, the fusion of the nanoscale world with the
macroscale world is far from trivial, and remains a grand
challenge in bionanomanufacturing.
Characterization and analysis
Current implementations of bionanofabrication processes are
severely limited by the lack of ability to thoroughly
characterize and analyze the bionanostructures that are
created by these processes. The structural complexity of
biomolecules, their fragility, and their need to remain bathed
in water to retain their function often preclude the use of
techniques routinely employed in the characterization of
metallic and semiconducting materials. The small amounts of
analytes and their confinement to substrate surfaces further
restrict the choice of analytical techniques and render many
commonly used molecular biology techniques impractical or
useless. Therefore, much of the current characterization for
bionanostructures is based on physical attributes such as
height, width, friction, or mechanical strength. The advent of Fig. 10 Integration of various components and processes in bionanofabrication using a
bottom-up or top-down approach. A bioinspired manufacturing platform that enables
in situ analytical techniques such as environmental scanning parallel assembly of devices requiring integration of biological (protein or DNA) and
electron microscopy (eSEM)99, cryo-transmission electron nonbiological components (Au nanoparticle (Au NP) or quantum dot (QD)) across length
scales ranging from the nanometer to the millimeter. A multifunctional wafer-scale device
microscopy (cryo-TEM)100, and near-field scanning optical can be built upon a multi-layer microelectronic chip foundation that provides the optical,
electrical, and chemical inputs (through waveguides) and outputs (light scattered by
microscopy (NSOM)-based techniques101, which do not surface plasmon resonance (SPR) and light couple out by grating) to simultaneously build
require elaborate sample preparations and modifications, ‘on-chip’ multiple copies of functional nanoscale devices in an automated, parallel
assembly process. (Copyright 2005 Duke University.)
would potentially allow more meaningful characterization of
bionanostructures. structures and devices. Attempting these tasks, however, is
clearly worthwhile because of the enormous benefits that can
Summary be gained by harnessing the unique properties biomolecules
Building on the technological foundation of nano- and offer, enabling the creation of nanotools and devices with
microfabrication, bionanofabrication offers us enabling functionalities and properties that rival those built with their
technologies that will undoubtedly transform the current nonbiological counterparts. NT
state of biosensing, nanomaterial development, drug
discovery, and synthetic biology. To this end, we must tackle Acknowledgments
The authors would like to acknowledge financial support from the National Science
the daunting task of overcoming all the challenges in dealing Foundation through a US National Science Foundation NIRT grant EEC-0210590.
We thank Jingdong Tian (Biomedical Engineering, Duke University) for Fig. 2, and
with biomolecules, as well as in resolving the problems that Scott Kennedy and Martin Brooke (Electrical and Computer Engineering, Duke University)
impede fabrication processes, their process control and for assembling Fig. 10 based on illustrations provided by Thomas LaBean (Computer
Science, Duke University), and L. Jay Guo (Electrical and Computer Engineering,
integration, and characterization of the resulting nanoscale University of Michigan).

38 December 2005

REFERENCES
1. Minsk, L., et al., Appl. Polym. Sci. (1959) 2 (6), 302
2. Thompson, L., et al., Introduction to Microlithography, 2nd edn., American
Chemical Society, Washington, DC, (1994)
3. Thompson, L., and Kerwin, R., Annu. Rev. Mater. Sci. (1976) 6, 267
4. Judy, J. W., Smart Mater. Struct. (2001) 10 (6), 1115
5. Bustillo, J. M., et al., Proc. IEEE (1998) 86 (8), 1552
6. Madou, M. J., Fundamentals of Microfabrication: the Science of
Miniaturization, 2nd edn., CRC Press, Boca Raton, (2002)
7. Hong, J., and Quake, S., Nat. Biotechnol. (2003) 21 (10), 1179
8. Zhu, H., and Snyder, M., Curr. Opin. Chem. Biol. (2003) 7 (1), 55
9. Lipshutz, R. J., et al., Nat. Genet. (1999) 21 (1s) 20
10. Xia, Y. N., and Whitesides, G. M., Annu. Rev. Mater. Sci. (1998) 28, 153
11. Binnig, G., et al., Phys. Rev. Lett. (1986) 56 (9), 930
12. Hyun, J., et al., J. Am. Chem. Soc. (2004) 126 (15), 4770
13. Riemenschneider, L., et al., Nano Lett. (2005) 5 (9), 1643
14. Philp, D., and Stoddart, J. F., Angew. Chem. Int. Ed. (1996) 35 (11), 1155
15. Lehn, J. M., Angew. Chem. Int. Ed. (1990) 29 (11), 1304
16. Alberts, B., et al., Molecular Biology of the Cell, Garland, New York, (1994)
17. Stryer, L., Biochemistry, 4th edn. W. H. Freeman, New York, (1995).
18. Chapman-Smith, A., and Cronan, J. E., Trends Biochem. Sci. (1999) 24 (9), 359
19. Morse, D. E., Trends Biotechnol. (1999) 17 (6), 230
20. Cha, J. N., et al., Proc. Natl. Acad. Sci. USA (1999) 96 (2), 361
21. Klibanov, A. M., Adv. Appl. Microbiol. (1983) 29, 1
22. Klibanov, A. M., Trends Biochem. Sci. (1989) 14 (4), 141
23. Braun, E., et al., Nature (1998) 391, 775
24. Chan, W. C. W., and Nie, S. M., Science (1998) 281, 2016
25. Chen, R. J., et al., Proc. Natl. Acad. Sci. USA (2003) 100 (9), 4984
26. Polla, D., et al., Annu. Rev. Biomed. Eng. (2000) 2., 551
27. Douglas, K., et al., Science (1992) 257, 642
28. Sleytr, U. B., et al., FEMS Microbiol. Rev. (1997) 20 (1-2), 151
29. Winfree, E., et al., Nature (1998) 394, 539
30. Adleman, L. M., Sci. Am. (1998) 279 (2), 54
31. Adleman, L. M., Science (1994) 266, 1021
32. Mao, C., et al., Nature (2000) 407, 493
33. Wilson, D. L., et al., Proc. Natl. Acad. Sci. USA (2001) 98 (24), 13660
34. Demers, L. M., et al., Science (2002) 296, 1836
35. Hyun, J., et al., Nano Lett. (2002) 2 (11), 1203
36. Hyun, J., et al., J. Am. Chem. Soc. (2004) 126 (23), 7330
37. Lopez, G. P., et al., US Patent 6,615,855, (2000)
38. Clackson, T., et al., Nature (1991) 352, 624
39. Crameri, A., et al., Nature (1998) 391, 288
40. Beaudry, A. A., and Joyce, G. F., Science (1992) 257, 635
41. Xiang, X.-D., et al., Science (1995) 268, 1738
42. Houghten, R. A., et al., Nature (1991) 354, 84
43. Jaenicke, R., and Bohm, G., Curr. Opin. Struct. Biol. (1998) 8 (6), 738
44. Ventosa, A., and Nieto, J. J., World J. Microbiol. Biotechnol. (1995) 11 (1), 85
45. Tian, J., et al., Nature (2004) 432, 1050
46. Bazylinski, D. A., et al., Appl. Environ. Microbiol. (1995) 61 (9), 3232
47. Dujardin, E., et al., Nano Lett. (2003) 3 (3), 413
48. Mao, C., et al., Science (2004) 303, 213
49. Kumar, A., and Whitesides, G. M., Appl. Phys. Lett. (1993) 63 (14), 2002
50. Xia, Y. N., et al., Langmuir (1996) 12 (16), 4033
RESEARCH REPORT
51. Whitesides, G. M., et al., Annu. Rev. Biomed. Eng. (2001) 3, 335
52. Bernard, A., et al., Langmuir (1998) 14 (9), 2225
53. Hyun, J., et al., Langmuir (2001) 17 (20), 6358
54. Yang, Z. P., and Chilkoti, A., Adv. Mater. (2000) 12 (6), 413
55. Csucs, G., et al., Langmuir (2003) 19 (15), 6104
56. Bernard, A., et al., Adv. Mater. (2000) 12 (14), 1067
57. Renault, J. P., et al., J. Phys. Chem. B (2003) 107 (3), 703
58. Xiao, P. F., et al., Nanotechnology (2002) 13 (6), 756
59. Xu, C., et al., J. Mater. Chem. (2003) 13 (12), 3044
60. Lange, S. A., et al., Anal. Chem. (2004) 76 (6), 1641
61. Richard, D., and Piner, C. A. M., Science (1999) 283, 661
62. Jung, H., et al., J. Am. Chem. Soc. (2003) 125 (40), 12096
63. Lim, J. H., and Mirkin, C. A., Adv. Mater. (2002) 14 (20), 1474
64. Lee, K. B., et al., Science (2002) 295, 1702
65. Demers, L. M., et al., Angew. Chem. Int. Ed. (2001) 40 (16), 3071
66. Smith, J. C., et al., Nano Lett. (2003) 3 (7), 883
67. Su, M., and Dravid, V. P., Appl. Phys. Lett. (2002) 80 (23), 4434
68. Chow, D. C., et al., J. Am. Chem. Soc. (2005) 127 (41), 14122
69. Takeda, S., et al., Nano Lett. (2003) 3 (11), 1471
70. Gu, J. H., et al., J. Am. Chem. Soc. (2004) 126 (26), 8098
71. Yoshinobu, T., et al., Electrochim. Acta (2003) 48, 3131
72. Liu, G. Y., and Amro, N. A., Proc. Natl. Acad. Sci. USA (2002) 99 (8), 5165
73. Schwartz, P. V., Langmuir (2001) 17 (19), 5971
74. Case, M. A., et al., Nano Lett. (2003) 3 (4), 425
75. Lewis, A., et al., Appl. Phys. Lett. (1999) 75 (17), 2689
76. Shalom, S., et al., Rev. Sci. Instrum. (1992) 63 (9), 4061
77. Bruckbauer, A., et al., J. Am. Chem. Soc. (2002) 124 (30), 8810
78. Bruckbauer, A., et al., J. Am. Chem. Soc. (2003) 125 (32), 9834
79. Hansma, P. K., et al., Science (1989) 243 (4891), 641
80. Taha, H., et al., Appl. Phys. Lett. (2003) 83 (5), 1041
81. Ionescu, R. E., et al., Nano Lett. (2003) 3 (12), 1639
82. Meister, A., et al., Appl. Phys. Lett. (2004) 85 (25), 6260
83. Quate, C. F., Surf. Sci. (1997) 386 (1-3), 259
84. Rozhok, S., et al., J. Phys. Chem. B (2003) 107 (3), 751
85. Peterson, E. J., et al., J. Phys. Chem. B (2004) 108 (39), 15206
86. Jang, J. Y., et al., J. Chem. Phys. (2001) 115 (6), 2721
87. Li, H. W., et al., Langmuir (2003) 19 (6), 1963
88. Rogers, J. A., et al., J. Vac. Sci. Technol. (1998) 16 (1), 88
89. Michel, B., et al., Chimia (2001) 56 (10), 527
90. Delamarche, E., et al., J. Phys. Chem. B (1998) 102 (18), 3324
91. Bullen, D., et al., Appl. Phys. Lett. (2004) 84 (5), 789
92. Hong, S. H., et al., Science (1999) 286, 523
93. Zhang, M., et al., Nanotechnology (2002) 13 (2), 212
94. Abdelhady, H. G., et al., Nanotechnology (2005) 16 (6), 966
95. Doumanidis, H., Nanotechnology (2002) 13 (3), 248
96. Noy, A., et al., Nano Lett. (2002) 2 (2), 109
97. Biebuyck, H. A., et al., J. Res. Dev. (1997) 41 (1-2), 159
98. Tien, J., et al., Proc. Natl. Acad. Sci. USA. (2002) 99 (4), 1758
99. Donald, A. M., Curr. Opin. Colloid Interface Sci. (1998) 3 (2), 143
100. Gustafsson, J., et al., Biochim. Biophys. Acta Biomembr. (1995) 1235 (2), 305
101. Hoppener, C., et al., Biophys. J. (2005) 88 (5), 3681
102. Yan, H., et al., Science (2003) 301 (5641), 1882
103. Ionescu, R. E., et al., Nano Lett. (2005) 5 (5), 821
December 2005 39