Hybrid ceramic polymers: New, nonbiofouling, and optically transparent materials for microfluidics

Article abstract of DOI:10.1021/ac1004053

A new, commercial hybrid ceramic polymer, Ormocomp, was introduced to the fabrication of microfluidic separation chips using two independent techniques, UV lithography and UV embossing. Both fabrication methods provided Ormocomp chips with stable cathodic electroosmotic flow which enabled examination of the Ormocomp biocompatibility by means of microchip capillary electrophoresis (MCE) and (intrinsic) fluorescence detection. The hydrophobic/hydrophilic properties of Ormocomp were examined by screening its interactions with bovine serum albumin and selected amino acids of varying hydrophobicity. The results show that the ceramic, organic-inorganic polymer structure natively resists biofouling on microchannel walls even so that the Ormocomp microchips can be used in intact protein analysis without prior surface modification. With theoretical separation plates approaching 10⁴ m^-1 for intact proteins and 10⁶ m^-1 for amino acids and peptides, our results suggest that Ormocomp microchips hold record-breaking performance as microfluidic separation platforms. In addition, Ormocomp was shown to be suitable for optical fluorescence detection even at near-UV range (ex 355 nm) with detection limits at a nanomolar level (～200 nM) for selected inherently fluorescent pharmaceuticals.

Full text of DOI:10.1021/ac1004053

Anal. Chem. 2010, 82, 3874–3882
Hybrid Ceramic Polymers: New, Nonbiofouling, and
Optically Transparent Materials for Microfluidics
Tiina Sikanen,*,† Susanna Aura,‡ Liisa Heikkila¨ ,†,§ Tapio Kotiaho,†,§ Sami Franssila,| and
Risto Kostiainen†
Division of Pharmaceutical Chemistry, Faculty of Pharmacy, University of Helsinki, P.O. Box 56, FI-00014, Finland,
Department of Micro and Nanosciences, School of Science and Technology, Aalto University, P.O. Box 13500,
FI-00076, Finland, Laboratory of Analytical Chemistry, Department of Chemistry, University of Helsinki, P.O. Box 55,
FI-00014, Finland, and Department of Materials Science and Engineering, School of Science and Technology, Aalto
University, P.O. Box 16200, FI-00076, Finland
A new, commercial hybrid ceramic polymer, Ormocomp,
was introduced to the fabrication of microfluidic separa-
tion chips using two independent techniques, UV lithog-
raphy and UV embossing. Both fabrication methods
provided Ormocomp chips with stable cathodic electro-
osmotic flow which enabled examination of the Ormo-
comp biocompatibility by means of microchip capillary
electrophoresis (MCE) and (intrinsic) fluorescence detec-
tion. The hydrophobic/hydrophilic properties of Ormo-
comp were examined by screening its interactions with
bovine serum albumin and selected amino acids of varying
hydrophobicity. The results show that the ceramic,
organic-inorganic polymer structure natively resists bio-
fouling on microchannel walls even so that the Ormocomp
microchips can be used in intact protein analysis without
prior surface modification. With theoretical separation
plates approaching 104 m-1 for intact proteins and 106
m-1 for amino acids and peptides, our results suggest
that Ormocomp microchips hold record-breaking per-
formance as microfluidic separation platforms. In addi-
tion, Ormocomp was shown to be suitable for optical
fluorescence detection even at near-UV range (ex 355
nm) with detection limits at a nanomolar level (∼200 nM)
for selected inherently fluorescent pharmaceuticals.
(MCE), in combination with optical (fluorescence) detection, has
particularly found its place in bioanalysis.1,2
The first microchip-based capillary electrophoresis (CE)
separations3,4 were realized using glass or fused silica (FS) devices
which emulated the hydrophilic and relatively inert environment
of silica capillaries. Processing of glass or FS is, however,
somewhat laborious and does not exactly lend itself to rapid and
cost-effective mass production. During the past decade, there has
been a keen interest on finding more economic alternatives to
glass or FS as chip fabrication materials. This has led to extensive
research on less expensive polymer-based materials and the
progressive development of polymer microfabrication techniques.5,6
Today, there is a substantial number of different polymers that
have been utilized for the fabrication of microfluidic devices, plenty
of them presently in regular use and commercially available. There
are, however, certain requirements that govern the material choice
from the analytical point of view, namely, the materials properties
as well as the ease of mass production. The latter is an analytical
issue because it affects not only the cost but also the reproduc-
ibility of chip-based analyses. The materials properties, herein,
mainly include chemical stability, surface chemistry, and optical
properties. As most analytical operations are performed close to
room temperature, the differences in glass transition temperatures
or thermal expansion coefficients between materials can be
considered less relevant from the analytical point of view. Most
polymers also exhibit rather similar properties with respect to
thermal (approximately 0.1-0.2 WK-1m-1) and electrical con-
ductivity (resistivity on the order of 1012-1018 Ω cm-1 6).
Consequently, chemical stability and surface chemistry are
probably the most important materials properties as they
greatly define the applications in which the particular material
can, or cannot, be used.
Microfluidic devices are extensively applied to bioanalytical
applications in order to improve existing technology through
increased speed of analysis (higher throughput) and aggressive
integration of several unit operations on a single microchip (e.g.,
sample preconcentration, derivatization, separation, detection).
The development of bioseparation methods, many of which rely
on electrophoretic techniques, is of high importance in order to
explore the properties of amino acids, peptides, and proteins which
are the key elements in all living organisms. Since implementation
of electrophoretic separation techniques on microfluidic devices
is relatively straightforward, microchip capillary electrophoresis
In most cases, it is the high hydrophobicity of microfluidic
chips that complicate their use in bioanalyses, especially that of
(1) Pumera, M. Electrophoresis 2007, 28, 2113–2124
(2) Huikko, K.; Kostiainen, R.; Kotiaho, T. Eur. J. Pharm. Sci. 2003, 20, 149–
171
(3) Harrison, D. J.; Manz, A.; Fan, Z. H.; Ludi, H.; Widmer, H. M. Anal. Chem.
1992, 64, 1926–1932
(4) Manz, A.; Harrison, D. J.; Verpoorte, E. M. J.; Fettinger, J. C.; Paulus, A.;
Ludi, H.; Widmer, H. M. J. Chromatogr. 1992, 593, 253–258
(5) Shadpour, H.; Musyimi, H.; Chen, J. F.; Soper, S. A. J. Chromatogr., A 2006,
1111, 238–251
(6) Becker, H.; Gartner, C. Anal. Bioanal. Chem. 2008, 390, 89–111
.
.
* Corresponding author. E-mail: tiina.sikanen@helsinki.fi. Tel: 358-9-191 59169.
Fax: 358-9-191 59556.
.
† Division of Pharmaceutical Chemistry, University of Helsinki.
‡ Department of Micro and Nanosciences, Aalto University.
§ Department of Chemistry, University of Helsinki.
.
.
| Department of Materials Science and Engineering, Aalto University.
.
3874 Analytical Chemistry, Vol. 82, No. 9, May 1, 2010
10.1021/ac1004053  2010 American Chemical Society
Published on Web 04/15/2010

intact proteins. This is because high hydrophobicity often means
undesired nonspecific adsorption of the proteins onto the micro-
channel surface when their hydrophobic moieties (i.e., amino acid
residues on the protein structure) interact with the hydrophobic
surface functionalities resulting in loss of the active protein
conformation (protein fouling). Efforts have been made to
overcome such problems by various physical (prominently oxygen
plasma treatment) or chemical (prominently coating by polyacry-
lamide) surface modifications, but their effect is often temporary
or they require a lot of manual postprocessing, respectively.7,8
Often, changes in hydrophobicity also have implications on the
surface charge (zeta potential) and vice versa. Greater hydrophi-
licity typically means more positive or negative charges on the
surface (greater zeta potential) and, thus, greater electroosmotic
flow (EOF),9,10 though coatings resulting in nonionic surfaces also
exist.11,12 Nevertheless, any postprocessing treatment required
to taylor the surface chemistry toward better biocompatibility
increases manual work load and affects the analytical reproduc-
ibility. There is still a great demand on new, more economic
polymer materials that hold inert and stable surface chemistry
with inherent resistance to biofouling and that are, therefore, truly
applicable to mass production.
exploited to tissue engineering18,19 or laser desorption ionization-
mass spectrometry,20,21 ORMOCERs have not been used in
microfluidic applications. In our previous study, we used the
commercial Ormocomp in order to implement lidless microf-
luidic structures on silicon or glass substrates.22 MCE separa-
tion of fluorescein isothiocyanate (FITC)-labeled peptides and
pH-dependent EOF were also tentatively shown using the
lidless, lithographically defined Ormocomp microchannels
(floor and walls) with the roof made from poly(dimethyl
siloxane) (PDMS). However, prominent adsorption of the
peptides on PDMS roof was observed when such mixed
material channels were used. Here, for the first time, the
commercially available Ormocomp was used for fabrication of
fully enclosed microfluidic channels by UV-lithography or UV-
embossing techniques. An adhesive Ormocomp-Ormocomp
bonding method was utilized to implement microfluidic chan-
nels made merely from Ormocomp. With a view to bioanalytical
applications, the physicochemical properties of Ormocomp
chips, fabricated by either of the patterning methods, were
examined by means of standard microchip electrophoresis and
fluorescence detection at visible (ex 488 nm) or near-UV (ex
355 nm) range. Most importantly, the high optical transmission
of Ormocomp is a valued property in microfluidic applications.
Namely, apart from PDMS and some cycloolefin (co)polymers
(COP, COC), most polymer materials exhibit strong absor-
bance at UV and near-UV wavelengths5,23 which limits their use
in combination with optical detection techniques. Direct UV
(absorbance) detection, or intrinsic fluorescence detection exploit-
ing UV excitation, is practically impossible for most polymers.
Instead, bioanalytical processes need to be monitored on the basis
of fluorophore tagging of the amino acid residues so that both
fluorescence excitation and emission are in the visible range. Even
then, attention should be paid on the choice of the fluorophore
wavelengths so that detection sensitivity is not restricted by the
autofluorescence originating from the chip material.5,23 Typically,
it is the additives that give rise to autofluorescence, with the
PDMS and COC being rare exceptions. Thus, the ability to
perform efficient optical detection even at near-UV range, makes
Ormocomp an ideal material for microfluidics.
In this study, the physicochemical properties of a new,
commercial hybrid material, Ormocomp (from Microresist Tech-
nology GmbH), were examined with a view to microfluidic
bioanalytical applications. Ormocomp belongs to the family of
organically modified ceramics (ORMOCERs13,14), the properties
of which can be tailored to different applications, such as optical
devices15 or antistatic and antiadhesive coatings.16 ORMOCER
s
were first developed by Fraunhofer Institute in 1980s.17 The
manufacturing process includes sol-gel preparation of the
inorganic backbone (typically Si-O-Si) followed by cross-
linking of the organic side chains. By varying the amount of
inorganic and organic elements, the properties of ORMOCERs
can be adjusted in a wide range. The original idea was to
combine the properties of organic polymers (functionalization,
ease of processing at low temperatures, toughness) with those
of glass-like materials (hardness, chemical and thermal stability,
transparency).14 Following commercialization by Microresist
Technology GmbH, selected ORMOCERs have become avail-
able for common use, but so far, the number of papers making
use of ORMOCER-based microfluidics has been limited. Apart
from some micro- and nanostructured ORMOCER surfaces
EXPERIMENTAL SECTION
Materials and Reagents. Ammonium hydrogen carbonate,
monobasic sodium phosphate, disodium phosphate, and ꢀ-nico-
tinamide adenine dinucleotide 2′-phosphate reduced tetrasodium
salt (NADPH) were from Sigma-Aldrich (Steinheim, Germany).
Boric acid was from Riedel-de Hae¨n (Seelze, Germany), and
sodium hydroxide was from Eka Nobel (Bohus, Sweden). Sodium
hydrogen carbonate and magnesium chloride were from Merck
(7) Pallandre, A.; de Lambert, B.; Attia, R.; Jonas, A. M.; Viovy, J. L.
Electrophoresis 2006, 27, 584–610
(8) Makamba, H.; Kim, J. H.; Lim, K.; Park, N.; Hahn, J. H. Electrophoresis
2003, 24, 3607–3619
(9) Roman, G. T.; Hlaus, T.; Bass, K. J.; Seelhammer, T. G.; Culbertson, C. T.
Anal. Chem. 2005, 77, 1414–1422
(10) Kato, M.; Gyoten, Y.; Sakai-Kato, K.; Nakajima, T.; Toyo’oka, T. Electro-
phoresis 2005, 26, 3682–3688
(11) Wang, A. J.; Xu, J. J.; Chen, H. Y. Anal. Chim. Acta 2006, 569, 188–194
.
.
.
(18) Doraiswamy, A.; Jin, C.; Narayan, R. J.; Mageswaran, P.; Mente, P.; Modi,
R.; Auyeung, R.; Chrisey, D. B.; Ovsianikov, A.; Chichkov, B. Acta Biomater.
.
2006, 2, 267–275
(19) Dinca, V.; Ranella, A.; Popescu, A.; Dinescu, M.; Farsari, M.; Fotakis, C.
Appl. Surf. Sci. 2007, 254, 1164–1168
(20) Aura, S.; Jokinen, V.; Sainiemi, L.; Baumann, M.; Franssila, S. J. Nanosci.
Nanotechnol. 2009, 9, 6710–6715
(21) Jokinen, V.; Aura, S.; Luosujarvi, L.; Sainiemi, L.; Kotiaho, T.; Franssila, S.;
Baumann, M. J. Am. Soc. Mass Spectrom. 2009, 20, 1723–1730
(22) Aura, S.; Sikanen, T.; Kotiaho, T.; Franssila, S. Sens. Actuators, B 2008,
132, 397–403
(23) Piruska, A.; Nikcevic, I.; Lee, S. H.; Ahn, C.; Heineman, W. R.; Limbach,
P. A.; Seliskar, C. J. Lab Chip 2005, 5, 1348–1354
.
.
(12) Xiao, D.; Le, T. V.; Wirth, M. J. Anal. Chem. 2004, 76, 2055–2061
(13) http://www.ormocer.de/EN/.
.
.
(14) Haas, K. H. Adv. Eng. Mater. 2000, 2, 571–582
(15) Houbertz, R.; Domann, G.; Cronauer, C.; Schmitt, A.; Martin, H.; Park, J. U.;
.
.
Frohlich, L.; Buestrich, R.; Popall, M.; Streppel, U.; Dannberg, P.; Wachter,
.
C.; Brauer, A. Thin Solid Films 2003, 442, 194–200
(16) Haas, K. H.; Amberg-Schwab, S.; Rose, K. Thin Solid Films 1999, 351,
198–203
(17) Schmidt, H. J. Non-Cryst. Solids 1985, 73, 681–691
.
.
.
.
.
Analytical Chemistry, Vol. 82, No. 9, May 1, 2010 3875

(Darmstadt, Germany), and trifluoroacetic acid (TFA) was from
Fluka (Buchs, Switzerland). Acetone was from VWR International
(Stockholm, Sweden); methanol was from J.T. Baker (Deventer,
Holland), and acetonitrile and isopropanol were from Rathburn
(Walkerburn, Scotland). All buffer components and solvents were
of analytical or HPLC grade. All amino acids, angiotensin I
(g99.1%), angiotensin II (g96.2%), substance P (fragment 6-11,
>97%), fluorescein isothiocyanate (FITC, isomer I, >98.5%), FITC-
labeled BSA (analytical grade), coumarin (g99%), umbelliferone
(99%), and scopoletin (99%) were purchased from Sigma-Aldrich.
Human liver microsomes (pool of 20 donors) were from BD
Biosciences (Erembodegem, Belgium).
Figure 1. (a) Two layer, UV-lithography approach and (b) one layer,
UV-embossing approach with a PDMS stamp for Ormocomp micro-
channel patterning. In UV embossing (b), the thickness of the bottom
Ormocomp layer is defined by the residual layer under the stamp.
Sample Preparation. Stock solutions of the L-amino acids (20
mM), peptides (10 mg/mL), BSA (5 mg/mL), ꢀ-lactoglobuline
(5 mg/mL), and FITC-BSA (15 mg/mL) were prepared in
deionized Milli-Q water (Millipore, Bedford, MA) and diluted
accordingly with the specified buffer. Umbelliferone and scopoletin
were first dissolved in methanol-water (1:1) solution and then
by prebaking at 120 °C for at least 1 h to remove moisture
from the surface. Next, a layer of Ormocomp was spin-coated
(BLE delta 20 BM W8) on the wafer at 6000 rpm for 30 s, which
resulted in an Ormocomp thickness of approximately 18 µm.
The wafer was then prebaked on the hot plate at +80 °C for 2
min after which blank UV exposure (30 s) of the first
diluted with the buffer. Before analysis, all L-amino acids and
peptides were individually labeled overnight (16 h) at +37 °C with
a 2-fold molar excess of freshly prepared FITC solution in 50 mM
sodium hydrogen carbonate (pH 9.0)-acetone 80:20 (v/v). Before
use, all stock solutions and buffer solutions were filtered (0.2 µm)
and degassed by sonication for 10 min.
¨
Ormocomp layer was done with MA 6 mask aligner (SUSS
MicroTec Lithography GmbH, Garching, Germany) at a
wavelength of 365 nm (30 mW/cm2). The first layer was then
postbaked on the hot plate at +80 °C for 5 min, and a second
layer of Ormocomp was spin-coated on top of the first one at
4000 rpm for 30 s resulting in a thickness of approximately 21
µm. Since Ormocomp is liquid-like material before exposure,
the second Ormocomp layer was UV exposed (5 s) through a
photomask in proximity mode. Prebake and postexposure bake
were done in a manner similar to those of the first layer.
Development of the UV-patterned Ormocomp structures was
done with a mixture (1:1) of methylisobutylketone and isopro-
pyl alcohol (i.e., Ormodev from Microresist Technology GmbH)
for 3 min followed by a rinse with isopropanol.
Enzyme Incubation. Kinetic parameters of cytochrome P450
(CYP) 2A6 isoenzyme mediated coumarin 7-hydroxylation to
umbelliferone were determined by incubating coumarin with
human liver microsomes (HLM) under in-house optimized condi-
tions. In brief, coumarin was incubated at +37 °C in 50 mM
sodium phosphate (pH 7.4) buffer including 5 mM magnesium
chloride, 0.3 mg/mL human liver microsomes, and with 1 mM
NADPH as the cosubstrate. The incubations were performed in
duplicate at six different coumarin concentrations in a volume of
100 µL. After 30 min, the reactions were stopped by the addition
of 100 µL of ice-cold acetonitrile, and the proteins were precipitated
by centrifugation (5 min at 16 000g). The supernatant was then
spiked with the internal standard (5 µM scopoletin) and analyzed
using the Ormocomp microchips. The obtained activities of
coumarin 7-hydroxylation (in pmol/min/mg protein) were fitted
to the Michaelis-Menten equation.
Master Fabrication for UV Embossing. The master for the
fabrication of the PDMS stamp was made from SU-8 50 negative
photoresist (Microchem Corporation, Newton, MA). Single-side
polished silicon wafers with 〈100〉 orientation (L100 mm) were
used as substrates for the SU-8 patterns. Prior to processing,
silicon wafers were cleaned with standard RCA 1 cleaning (NH3
/
Microchip Fabrication. All microchips were fabricated from
a commercially available ORMOCER material, Ormocomp (Micro
Resist Technology GmbH, Darmstadt, Germany), which is a
solvent free, UV-curable inorganic-organic hybrid material. Or-
mocomp can be patterned using standard UV lithography or UV
embossing (Figure 1). Here, UV embossing was done using a
transparent PDMS stamp. In addition, double-side polished 7740
Pyrex wafers (L100 mm) were used as substrates for all Ormo-
comp structures. The microchip designs were based on an
accustomed geometry of standard capillary electrophoresis chip
incorporating a separation channel (55-85 mm in length) with
either double-T (100 µm) or simple injection cross. The micro-
channel width was typically 30 or 50 µm. The microchannel height
was defined by the layer thickness of the second Ormocomp layer
(UV-lithography approach) or by the PDMS stamp height in UV
embossing, and it was typically on the order of 20 µm.
H2O2/H2O, 1:1:5) followed by prebaking at 120 °C for at least
1 h in order to remove moisture on the surface. Next, SU-8
was spin-coated on the silicon wafer at 500 and 9000 rpm for
10 and 30 s, respectively. The resulting thickness of the SU-8
layer was typically 20 µm. The wafer was then soft-baked on a
hot plate at +65 °C for 5 min and then at +95 °C for 8 min
after which it was let to cool below +50 °C before UV exposure.
The exposure (15 s) was done through a photomask on a MA
6 mask aligner followed by a postexposure bake on the hot
plate at +95 °C for 8 min and slow cooling to room temperature.
The exposed SU-8 patterns were developed in propylene glycol
methyl ether acetate (PGMEA) solution for 10 min, rinsed with
isopropanol, and dried with nitrogen gas. Last, in order to
facilitate PDMS stamp detachment, Teflon-like fluoropolymer
was deposited on the wafer in a RIE reactor (Plasmalab 80+,
Oxford instruments plasma technology, UK) by applying 100
sccm CHF3 source gas flow, 50 W RF-power, and 250 mTorr
pressure for 5 min.
Ormocomp UV Lithography. The Pyrex wafers were cleaned
with standard RCA 1 cleaning (NH3/H2O2/H2O, 1:1:5) followed
3876 Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

separation voltages and to record current readings during experi-
ments. Before use, all microchannels were sequentially rinsed with
0.5 M sodium hydroxide, deionized water, and the specified
separation buffer for 2, 2, and 5 min, respectively, by applying
vacuum to a microchannel inlet. After rinsing, the microchannels
were filled with fresh buffer solution and quantities of 20, 20, and
10 µL of the buffer were applied to the buffer inlet, buffer waste,
and sample waste reservoirs, respectively. Last, an aliquot of 5-
10 µL of the sample solution was applied to the sample inlet
followed by immediate application of the injection voltages. Sample
introduction was performed in pinched injection mode at 1000
V/cm for 15-20 s. The separation conditions were optimized with
respect to the separation buffer composition and the electric field
strength. In order to prevent sample leakage, small push-back
voltages were typically applied to the sample inlet and sample
waste during separation.
EOF Determination. The EOF velocities were determined
on the basis of the current monitoring method.24 All measure-
ments were performed with the same instrumentation as in the
MCE experiments except that a 500 kΩ resistor was coupled in
series with the microchannel and the current changes over time,
calculated on the basis of the observed voltage drop across the
resistor, were monitored using the PicoScope 2203 AD converter.
Typically, a 5% decrease in the electrolyte concentration was utilized
in order to induce current drop in the circuit. The applied electric
field strength during measurements was between 300 and 700 V/cm.
Fluorescence Microscopy. Fluorescence detection of the
FITC-conjugates was accomplished with a Leica DMIL inverted
epifluorescence microscope (Leica Nilomark, Espoo, Finland)
equipped with an N PLAN 10 × /0.25 objective and a 75 W xenon
arc lamp with integrated lamp housing (Cairn Research, Faver-
sham, UK). The fluorescence excitation (450-490 nm) and
emission (515-600 nm) ranges were defined by appropriate filters.
Native fluorescence detection at 355 nm (excitation wave-
length) was accomplished with a Zeiss Axioscope A1 upright
epifluorescence microscope (Carl Zeiss Oy, Espoo, Finland)
equipped with a Plan-Neofluar 40 × /0.4 Corr objective and a
PowerChip-PNV diode-pumped UV355 nm laser (15 µJ at 1 kHz)
with custom-made laser housing (Cheos Oy, Espoo, Finland). The
fluorescence emission range (400-700 nm) was defined by
appropriate emission filters.
Figure 2. (a) A photograph of a full-wafer (diameter 100 mm) array
of enclosed Ormocomp separation microchips. (b) A scanning electron
micrograph of the cross-section of an Ormocomp separation channel
(on Pyrex substrate) demonstrating the good bonding quality between
the different Ormocomp layers.
Embossing Stamp Fabrication. PDMS stamp was fabricated
from Sylgard 184 silicone elastomer (Down Corning, Midland, MI)
by mixing the base elastomer and the curing agent in a ratio of
10:1 (w/w) and degassing the mixture in vacuum before being
poured on top of the SU-8 master. The PDMS stamp was cured
at 60 °C for 12 h and then peeled off the master.
Ormocomp UV Embossing. The Pyrex wafers were pre-
treated in a manner similar to that in the context of UV
lithography. Ormocomp was then spin-coated on the wafer at 500
and 2000 rpm for 10 and 30 s, respectively. The resulting thickness
of the Ormocomp layer was typically 70 µm. Embossing was done
with the PDMS stamp on the hot plate at +80 °C. Excess air
bubbles between the stamp and the Ormocomp layer were
removed manually with tweezers after which the exposure (60 s)
was done through the stamp. The exposure time as such was not
critical since the feature size was defined by the stamp. After
exposure, the wafer was transferred onto the hot plate at 80 °C for
a postexposure bake during which the stamp was easily peeled off.
Bonding of Ormocomp Microchannels. Enclosed Ormo-
comp microchannels, both lithographically defined and UV
embossed, were obtained by exploiting Ormocomp-to-Ormocomp
adhesive bonding (Figure 2). Such bonding is facilitated by a thin,
uncured Ormocomp film which remains on top of cured Ormo-
comp layers when exposure is done in the presence of oxygen.
Here, an Ormocomp bonding layer was spin-coated on top of a
standard overhead transparency cleaned with acetone and at-
tached onto a carrier wafer with double-sided tape. Spin-coating
was done at 500 and 4000 rpm for 10 and 30 s, respectively, after
which the Ormocomp layer was soft-baked at +80 °C for 2 min
and flood-exposed for 20 s. Next, the transparency was detached
from the carrier wafer and assembled on top of the previously
patterned Ormocomp wafer incorporating the microchannels. The
two Ormocomp layers were then manually pressed together with
tweezers, and a flood exposure of 30 s was done through the
transparency to finalize adhesion. Last, the transparency was
detached from the Ormocomp layers. The resulting thickness of
the top (bonding) Ormocomp layer was typically 18 µm.
Both microscopes were equipped with a similar Hamamatsu
R5929 photomultiplier tube with integrated housing and a signal
amplifier module (Cairn Research). Fluorescence emission was
collected perpendicular to the microchip by defining a detection
slit of 50 µm × (70-200) µm on top of the microchannel at a
predefined detection distance. A PicoScope 2203 AD converter
and PicoLog software (Pico Technology, St. Neots, UK) were used
for recording the signal acquired from the photomultiplier tube.
Instrumentation. The performance of the Ormocomp micro-
chips was characterized utilizing microchip capillary (zone)
electrophoresis (MCE), electroosmotic flow (EOF) measurements,
and fluorescence microscopy. Where applicable, the performance
of the Ormocomp microchips was examined against commercial
low fluorescence Schott Borofloat glass microchips (model MC-
BF4-TT100, Micralyne, Edmonton, Canada).
Microchip Capillary Electrophoresis. A computer-controlled
(Labview) high voltage power supply (Microfluidic Tool Kit,
Micralyne, Edmonton, Canada) was used for application of the
RESULTS AND DISCUSSION
In this study, for the first time, fully enclosed all-Ormocomp
microchannels were fabricated by either UV lithography or UV
embossing followed by adhesive bonding of the roof layer also
made from Ormocomp. The enclosed Ormocomp channels were
utilized in MCE separation of fluorescent labeled proteins, pep-
(24) Huang, X. H.; Gordon, M. J.; Zare, R. N. Anal. Chem. 1988, 60, 1837–
1838
.
Analytical Chemistry, Vol. 82, No. 9, May 1, 2010 3877

tides, and amino acids. The surface chemistry and biocompatibility
of native (untreated) Ormocomp microchannels, fabricated by
both lithographic and embossing techniques, were examined and
compared. Finally, the optical properties of Ormocomp microchips
were assessed against those of commercial, low-fluorescent
Borofloat glass microchips by determining the sensitivity of label-
free (intrinsic) fluorescence detection at near-UV (ex 355 nm)
range.
Surface Charge and Electroosmotic Flow. The surface
charge was first determined for both lithographically defined and
UV-embossed Ormocomp microchannels by following the EOF
velocity according to the current monitoring method.24 The
electroosmotic (EO) mobilities were measured under comparable
conditions for 18 mM sodium borate buffer (pH 10.0) and were
(3.3 0.3) × 10-8 m2V-1s-1 and (5.3 0.1) × 10-8 m2V-1s-1 for
lithographically defined and UV-embossed microchannels (n
) 6 runs), respectively. The corresponding zeta potential values
were calculated as per the Smoluchowski equation and were
-41.8 3.2 and -65.9 1.5 mV, respectively. The observed
differences in the surface charge were likely to result from the
different fabrication conditions between different methods and
also between different batches of microchips fabricated by the
same method. Previous studies support the fact that a relatively
large deviation (typically 15-30% RSD) can be attributed to
the batch-to-batch variation between both polymer and glass
Figure 3. MCE analyses of FITC-derivatized BSA (2.5 mg/mL) in
18 mM sodium borate (pH 10.0) using a lithographically defined
Ormocomp microchip (upper panel) or Borofloat glass microchip
(lower panel). The injections were performed in pinched injection
mode at 1200 Vcm-1 for 20 s, and the separations were performed
at 620 Vcm-1. The distance of detection was 4.0 cm (both microchips).
-27
microchannels fabricated by exactly the same method.25
Others have also shown that, besides material itself, EOF in
polymer microchannels may also depend on the fabrication
method. These kind of effects have been observed for instance
between hot embossed and laser ablated PMMA28 or COC29,30
microchannels. In this study, the measured batch-to-batch (n )
3) deviations were 19.0% and 12.9% (RSD) for lithographically
defined and UV-embossed microchannels, respectively, and thus,
in good accordance with the previous literature. In all cases,
however, substantial EOF with relatively low run-to-run repeat-
ability (typically 4-7%) was observed, indicating high surface
charge density and stable surface charge equilibrium on both
lithographically defined and UV-embossed Ormocomp surfaces.
Surface Interactions and Biofouling Resistance. The bio-
compatibility of Ormocomp was examined by screening the
surface interactions between Ormocomp and selected model
hydrophilic origin. In addition, high pH also assured maximum
fluorescence intensity from the FITC-derivatized biomolecules.31
BSA is a high molecular weight (66 kDa) protein which is
known to easily adhere to most plastic surfaces. Here, a com-
mercial fluorescein conjugate, FITC-BSA, was utilized to study
the biofouling resistance of Ormocomp by means of MCE and
fluorescence detection (ex 488 nm) of the conjugate. Repeated
injections of FITC-BSA were done under conditions described
in Figure 3 using both lithographically defined and UV-embossed
Ormocomp microchips. Borofloat glass microchips, known to be
less vulnerable to biofouling, were used for comparison. This
allowed us to compare the effects of nonspecific adsorption of
FITC-BSA on the analytical parameters between materials. As a
result, Ormocomp was found to effectively resist protein adsorp-
tion without any surface modification after microfabrication or
without any surface active ingredients applied prior to analysis.
The repeatabilities of the FITC-BSA migration time from run to
run were 0.9% and 1.9% (RSD, n ) 4) for untreated lithographically
defined and UV-embossed Ormocomp microchannels, respec-
tively. These values were well below the run-to-run variation
induced by EOF (i.e., 4-7% RSD in Ormocomp microchannels)
and were also consistent with the FITC-BSA migration time
repeatability measured for Borofloat glass microchips (1.1% RSD,
n ) 4). The number of theoretical plates was as high as (1.8
0.2) × 104 m-1 for FITC-BSA analysis in Ormocomp micro-
channels, against a plate number of (1.0 0.2) × 104 m-1 for
compounds, i.e., bovine serum albumin (BSA) and a set of L-amino
acids possessing different properties. The separation media (18
mM sodium borate (pH 10.0) buffer) was chosen on the basis of
two arguments. Namely, the relatively high ionic strength and
high pH (above the pI values of the model compounds) enabled
elimination of electrostatic interactions on the microchannel wall
and, thus, examination of interactions of purely hydrophobic-
(25) Locascio, L. E.; Perso, C. E.; Lee, C. S. J. Chromatogr., A 1999, 857, 275–
284
(26) Sikanen, T.; Tuomikoski, S.; Ketola, R. A.; Kostiainen, R.; Franssila, S.;
Kotiaho, T. Lab Chip 2005, 5, 888–896
(27) Sikanen, T.; Heikkila, L.; Tuomikoski, S.; Ketola, R. A.; Kostiainen, R.;
Franssila, S.; Kotiaho, T. Anal. Chem. 2007, 79, 6255–6263
(28) Johnson, T. J.; Waddell, E. A.; Kramer, G. W.; Locascio, L. E. Appl. Surf.
Sci. 2001, 181, 149–159
(29) Gaudioso, J.; Craighead, H. G. J. Chromatogr., A 2002, 971, 249–253
(30) Mela, P.; van den Berg, A.; Fintschenko, Y.; Cummings, E. B.; Simmons,
B. A.; Kirby, B. J. Electrophoresis 2005, 26, 1792–1799
.
.
.
.
.
.
(31) Lanz, E.; Gregor, M.; Slavík, J.; Kotyk, A. J. Fluoresc. 1997, 7, 317–319.
3878 Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

Figure 4. Apparent diffusion coefficients of (a) neutral, (b) hydrophobic, and (c) hydrophilic FITC-L-amino acids (50 µM) determined by plotting
the spatial peak variance as a function of migration time at different electric field strengths. The analyses were done in 18 mM sodium borate
(pH 10.0) using lithographically defined Ormocomp microchips. The injections were performed in pinched injection mode at 1200 Vcm-1 for
20 s. The distance of detection was constantly at 4.0 cm, and the electric field strengths were 360-450-630-720 Vcm-1 (Ala, Leu, and Glu),
360-450-540-630 Vcm-1 (Thr), or 450-540-630-720 Vcm-1 (Gly and Asp).
Borofloat glass microchannels. The peak profile (width, asym-
metry) was also perfectly comparable between runs made by
Ormocomp and Borofloat glass chips (Figure 3), suggesting
negligible interactions between the protein and the Ormocomp
surface.
tions. As can be seen from Figure 4, the spatial peak variances
of all amino acids followed linear regressions with correlation
coefficients between 0.98373 and 0.99997, confirming that band
broadening in Ormocomp microchannels was mainly diffusion
limited. Although the relatively large fluorescein label (FITC) is
likely to dominate the diffusion behavior, subtle differences were
observed between amino acids. For instance, the apparent diffu-
sion coefficients of hydrophobic and neutral amino acids were
found to decrease with increasing molecular volume (taking into
account molecular weight and hydrophobicity): DApp(Gly) >
DApp(Ala) > DApp(Leu) > DApp(Thr). Such behavior is similar to
that earlier observed in studies on the diffusion coefficients of
native (unlabeled) amino acids in aqueous solutions.34 How-
ever, according to Ma and co-workers,34 the diffusion coef-
ficient also decreases with increasing polarity, which was not
the case in our experiments. Instead, the apparent diffusion
coefficients of hydrophilic amino acids were somewhat higher
(2.19-4.45 × 10-6 cm2s-1) than those of hydrophobic or neutral
amino acids (1.31-1.63 × 10-6 cm2s-1), indicating that minor
band broadening may occur because of hydrophilic interactions
(e.g., hydrogen bonding) on Ormocomp surfaces. Nevertheless,
the obtained diffusion coefficients were generally in good
agreement with those earlier reported for both FITC-labeled
Next, selected L-amino acids were labeled in-house with FITC
and analyzed by MCE and fluorescence detection (ex 488 nm) in
a similar manner than FITC-BSA. Altogether six amino acids
were included, two of which were classified as hydrophilic (i.e.,
aspartic acid (Asp) and glutamic acid (Glu)), two as neutral
(i.e., glycine (Gly) and threonine (Thr)), and two as hydrophobic
(i.e., leucine (Leu) and alanine (Ala)).32 As a result, all amino acids,
regardless of hydrophobicity or charge state (pI), provided
symmetric peaks with plate numbers of (3.5-5.3) × 105 m-1
.
In order to confirm the absence of nonspecific interactions,
the apparent diffusion coefficients (DApp) were determined for
each amino acid selected through a series of runs performed
at a detection distance of 4.0 cm but at different electric field
strengths (according to Yao and Li33). By plotting spatial peak
variance (σ2) against migration time (t), a linear regression
results if band broadening is due to random processes only.
The slope of the regression line is twice the value of the
apparent diffusion coefficient (σ2 ) 2Dt), and the better the
correlation coefficient (R2), the less are the nonspecific interac-
(32) Kyte, J.; Doolittle, R. F. J. Mol. Biol. 1982, 157, 105–132
(33) Yao, Y. J.; Li, S. F. Y. J. Chromatogr. Sci. 1994, 32, 117–120
.
(34) Ma, Y.; Zhu, C.; Ma, P.; Yu, K. T. J. Chem. Eng. Data 2005, 50, 1192–
.
1196.
Analytical Chemistry, Vol. 82, No. 9, May 1, 2010 3879

amino acids ((3.0-3.9) × 10-6 cm2s-135) and fluorescein itself
3
((3.3-4.25) × 10-6 cm2s-1 ,36), although direct comparison is not
possible because of different experimental conditions. Altogether,
the results confirmed that both lithographically defined and UV-
embossed Ormocomp surfaces are natively resistant to biofouling.
This is also in line with the previous results on the good
biocompatibility of Ormocomp regarding cell viability and cell
growth on Ormocomp surfaces.18
Intrinsic Laser-Induced Fluorescence Detection. The
sensitivity of optical detection techniques does not directly benefit
from miniaturization. Nevertheless, fluorescence detection, laser-
induced fluorescence (LIF) in particular, remains as the most
sensitive technique available for microfluidic applications to such
a
degree that single-molecule detection has been accom-
plished.37,38 However, only a few compounds are inherently
fluorescent at visible range, and they must be derivatized to allow
detection through polymer-based substrates. On the other hand,
most compounds, such as many drugs and proteins, exhibit
intensive fluorescence at UV range, which is by default not
accessible through polymer substrates. This is why label-free
detection of underivatized compounds, so-called native fluores-
cence, is extremely rare among microfluidic applications.
In this study, a diode-pumped near-UV (355 nm, 15 mW) laser
was used for LIF detection of inherently fluorescent coumarin
dyes, umbelliferone and scopoletin, through Ormocomp micro-
structures. The MCE separation conditions were first optimized
with respect to buffer composition and electric field strength, and
a linear correlation of R2)0.9939 was established for umbellif-
erone over a concentration range of 500 nM to 2.5 µM, using
scopoletin (5 µM) as an internal standard. The limits of
detection (LOD) and quantitation (LOQ) of umbelliferone were
207 and 626 nM, respectively, as per ICH validation guidelines
(based on the standard deviation of the response and the slope
of a calibration curve). The repeatabilities for the relative peak
area (RSD, n ) 6) were also determined close to the lower
(750 nM) and upper (2.5 µM) level of quantitation and were
3.0 and 5.3%, respectively, suggesting good analytical reproduc-
ibility from run to run. The MCE separation of 5 µM scopoletin
(internal standard) and 250 nM umbelliferone (close to its
LOD) is presented in Figure 5a. For comparison, the corre-
sponding, in-house determined LOD and LOQ values through low-
fluorescent Borofloat glass were nearly the same, namely 163 nM
(LOD) and 494 nM (LOQ), suggesting equally low background
interference from Ormocomp or Borofloat glass. In addition, the
detection limits obtained were equal to those previously reported
for selected pharmaceuticals (81-475 nM) by label-free UV-LIF
detection through quartz substrates.39 Thus far, deep UV (266
nm) fluorescence detection has only been accomplished using
PDMS- or quartz-based microchips.40,41 Less UV-transparent glass
materials, such as Pyrex or Borofloat, have also been used when
Figure 5. (a) MCE separation of scopoletin (5 µM) and umbelliferone
(250 nM) in 18 mM sodium borate (pH 10.0) at 440 Vcm-1. The injection
was performed in pinched injection mode at 1000 Vcm-1 for 20 s. The
distance of detection was 3.5 cm. (b) Michaelis-Menten fit for coumarin
7-hydroxylation in human liver microsomes (HLM) as determined using
the Ormocomp chip. The separation conditions were as in (a).
excitation wavelengths have been at near-UV (300-385 nm)42 or
visible (488 nm) range43 or when two-photon excited fluorescence
has been utilized in order to overcome background interference.44
At best, detection limits between 81 and 475 nM have been
reported for small drug compounds and proteins by deep-UV (266
nm, 120 mW) excitation.39 A low micromolar detection limit was
also reported for Titan Yellow at near-UV (300-385 nm) excitation,
though a mercury lamp was used for excitation instead of a high-
power laser.42
Next, the Ormocomp chips were exploited in the analysis of
in vitro metabolism samples in order to examine the applicability
of intrinsic LIF detection through Ormocomp to real analytical
tasks. Kinetic parameters of cytochrome P450 (CYP) 2A6 isoen-
zyme mediated coumarin 7-hydroxylation to umbelliferone were
determined using off-chip incubated samples spiked with an
internal standard (5 µM scopoletin) and analyzed under separation
conditions described in Figure 5a. The obtained Michaelis-Menten
parameters for coumarin 7-hydroxylation were KM ) 6.0 1.2
µM and VMAX ) 597 40 pmol/min/mg protein (Figure 5b).
These values were in accordance with the previously reported
(35) Effenhauser, C. S.; Manz, A.; Widmer, H. M. Anal. Chem. 1993, 65, 2637–
2642
(36) Culbertson, C. T.; Jacobson, S. C.; Ramsey, J. M. Talanta 2002, 56, 365–
373
(37) Fister, J. C.; Jacobson, S. C.; Davis, L. M.; Ramsey, J. M. Anal. Chem. 1998,
70, 431–437
(38) Shelby, J. P.; Chiu, D. T. Anal. Chem. 2003, 75, 1387–1392
(39) Schulze, P.; Ludwig, M.; Belder, D. Electrophoresis 2008, 29, 4894–4899
(40) Schulze, P.; Ludwig, M.; Kohler, F.; Belder, D. Anal. Chem. 2005, 77, 1325–
1329
.
.
(41) Ros, A.; Hellmich, W.; Leffhalm, K.; Sischka, A.; Anselmetti, D. Mol. Cell.
Proteomics 2005, 4, S336–S336
(42) Yan, Q.; Chen, R. S.; Cheng, J. Anal. Chim. Acta 2006, 555, 246–249
(43) Liu, B. F.; Hisamoto, H.; Terabe, S. J. Chromatogr., A 2003, 1021, 201–
207
(44) Schulze, P.; Schuttpelz, M.; Sauer, M.; Belder, D. Lab Chip 2007, 7, 1841–
1844
.
.
.
.
.
.
.
.
3880 Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

Figure 6. (a) MCE separation of FITC-labeled substance P (6-11), angiotensin I, and angiotensin II (all 200 µM) in 18 mM sodium borate (pH
10.0) at 300 Vcm-1. The injection was performed in pinched injection mode at 1200 Vcm-1 for 15 s. The distance of detection was 4.0 cm. (b)
MCE separation of FITC-labeled L-amino acids (all 50 µM) in 36 mM sodium borate (pH 10.0) at 720 Vcm-1. The injection was performed in
pinched injection mode at 1200 Vcm-1 for 20 s. The distance of detection was 4.5 cm.
parameters of coumarin 7-hydroxylation in human liver mi-
L-amino acids within 60 s (Figure 6b). The peak widths were on
the order of 0.4 s for the peptides and between 0.4 and 0.8 s for
the amino acids, and the migration time repeatabilities (RSD,
n ) 4) with and without internal standard were 0.05-0.53% (amino
acids) and 1.9-2.2% (peptides), respectively. These values are in
general better than those previously reported for microchip-based
separations of fluorescent labeled peptides12,27 or amino acids,9,10,48
though direct comparison is hardly possible due to differences in
the experimental conditions and fluorescent labels.
The results also show that, under extreme conditions, the
absolute EOF velocity may have a decisive role in defining the
final separation efficiency. In general, EOF velocity attributed to
polymer-based microchips is lower than that of glass-based
microchips.5 This allows extension in the separation time and,
thus, better resolving power if nonspecific interactions of analytes
can be eliminated. In our work, the effect of EOF velocity was
found especially crucial on the separation of the coumarin
derivatives presented in Figure 5a. Umbelliferone and scopoletin
are weakly acidic in nature (pKa ∼ 9.0-9.5) and differ from one
another only by the neutral methoxy group. Despite their very
similar electrophoretic mobilities, scopoletin and umbelliferone
were resolved (RS ) 1.1) within 15 s and had a separation path
of 3.5 cm in Ormocomp microchannels. Meanwhile, the same
resolution in Borofloat glass microchannels was not possible
unless the separation length was extended to 6.0 cm and 5%
isopropanol was added to the separation buffer.
-47
crosomes, i.e., KM values between 0.2 and 2.3 µM.45
Clearly, the results suggest that Ormocomp structures can be
utilized as more economic alternatives to highly expensive and
rare quartz microchips in optical detection even at UV range.
Although deep-UV range (<300 nm) can hardly be reached, LIF
detection at near-UV range enables numerous possibilities for
label-free analysis. Often, the deep-UV range is also inaccessible
to standard microscope optics that are commonly made from
borosilicate and show high transparency only at excitation
wavelengths above 300 nm (and autofluoresce when excited by
deep UV). Nevertheless, near-UV range between 330 and 380 nm
forms an absorption zone characteristic for many pharmaceuticals
(e.g., hallucinogens, diuretics, nonsteroidal antiinflammatories)
and Ormocomp enables their label-free detection by LIF.
Separation Performance. Materials properties, both surface
and bulk, have their implications on the resulting performance of
the separation system in question. Obviously, all nonspecific
surface interactions need to be eliminated in order to achieve
maximum performance through narrower peak width. In addition,
the better the detection sensitivity, the smaller is the risk of
contamination since lower concentrations can be used. It is also
clear that the stability of the surface charge and EOF greatly affect
the analytical reproducibility. According to our results, Ormocomp
appears as an ideal chip fabrication material for high-efficiency
bioseparations because of its favorable and inert surface chemistry
and good optical transparency. The results suggest that analytical
performance similar to that of glass-based devices is easily
obtained using Ormocomp microchips. As an example, fast (<60
s) and highly efficient separation of FITC-labeled peptides, with
plate numbers reaching record-breaking million plates per meter
(i.e., (2.6-3.2) × 104 plates per separation length of 4.0 cm),
was demonstrated using Ormocomp microchannels (Figure
CONCLUSIONS
A new, nonbiofouling material, Ormocomp, was introduced for
the fabrication of microfluidic separation chips. Ormocomp was
patterned by two independent techniques, i.e., photolithography
and UV embossing, resulting in similar biocompatibility and
surface properties of the end product. Enclosed microstructures
with all inner surfaces of the same material were obtained through
adhesive bonding of Ormocomp layers. The physicochemical
properties of Ormocomp were characterized by means of micro-
chip capillary electrophoresis (MCE) and fluorescence detection
of a variety of biologically active compounds, i.e., amino acids,
6
a). Similar separation efficiency, i.e., (2.5-3.8) × 104 plates per
separation length of 4.5 cm, was also obtained for FITC-labeled
(45) Pelkonen, O.; Sotaniemi, E. A.; Ahokas, J. T. Br. J. Clin. Pharmacol. 1985,
19, 59–66
(46) Pearce, R.; Greenway, D.; Parkinson, A. Arch. Biochem. Biophys. 1992,
298, 211–225
(47) Draper, A. J.; Madan, A.; Parkinson, A. Arch. Biochem. Biophys. 1997, 341,
47–61
.
.
(48) Culbertson, C. T.; Jacobson, S. C.; Ramsey, J. M. Anal. Chem. 2000, 72,
.
5814–5819.
Analytical Chemistry, Vol. 82, No. 9, May 1, 2010 3881

peptides, and intact proteins. It was shown that regarding
separation efficiency and detection sensitivity, in particular,
Ormocomp holds a record-breaking performance over most other
polymers so far introduced to microfluidics. This is because
Ormocomp was found to be natively resistant to biofouling and
to maintain stable cathodic EOF on the order of (3-5) × 10-8
m2V-1s-1 (in 18 mM sodium borate at pH 10). As a result,
theoretical separation plates of up to 106 per meter were easily
obtained for fluorescently labeled amino acids and peptides by
MCE in Ormocomp microchannels. Good separation efficiency
(∼104 plates per meter) was also obtained in intact protein
analysis without any chemical or physical modification of the
native Ormocomp surfaces. In addition, Ormocomp enabled
highly sensitive intrinsic fluorescence detection even at near-
UV range (ex 355 nm) so that detection limits at nanomolar
level (∼200 nM) were reached for inherently fluorescent
pharmaceuticals. According to our results, Ormocomp appears
as an ideal chip fabrication material for microfluidic bioanalysis,
especially when material biocompatibility is of paramount
concern.
ACKNOWLEDGMENT
The work was financially supported by the Academy of Finland
(121882), the Finnish Funding Agency for Technology and
Innovation (40380/06), the University of Helsinki Research Funds,
and the Graduate School of Electrical and Communication
Engineering. Ms. Elisa Moilanen is gratefully acknowledged for
her help with the metabolism assays.
Received for review February 13, 2010. Accepted April 6,
2010.
AC1004053
3882 Analytical Chemistry, Vol. 82, No. 9, May 1, 2010