Embellishment of microfluidic devices via femtosecond laser micronanofabrication for chip functionalization.

Abstract

This paper demonstrates the embellishment of existing microfluidic devices with integrated three dimensional (3D) micronanostructures via femtosecond laser micronanofabrication, which, for the first time, proves two-photon photopolymerization (TPP) to be a powerful technology for chip functionalization. As representative examples, microsieves with various pore shape and adjustable pore size were successfully fabricated inside a conventional glass-based microfluidic channel prepared by wet etching for microparticle separation. Moreover, a fish scale like microfilter was also fabricated and appointed as a one-way valve, which showed excellent performance as we expected. These results indicate that such embellishment of microfluidic devices is simple, low cost, flexible and easy to access. We believe that, combined with TPP, the application of lab-on-chip devices would be further extended.

Full text

Page 1

Embellishment of microfluidic devices via femtosecond laser
micronanofabrication for chip functionalization†
Juan Wang,aYan He,aHong Xia,aLi-Gang Niu,aRan Zhang,aQi-Dai Chen,*aYong-Lai Zhang,a
Yan-Feng Li,bShao-Jiang Zeng,bJian-Hua Qin,bBing-Cheng Linband Hong-Bo Sun*ac
Received 16th February 2010, Accepted 27th April 2010
DOI: 10.1039/c003264f
This paper demonstrates the embellishment of existing microfluidic devices with integrated three
dimensional (3D) micronanostructures via femtosecond laser micronanofabrication, which, for the
first time, proves two-photon photopolymerization (TPP) to be a powerful technology for chip
functionalization. As representative examples, microsieves with various pore shape and adjustable
pore size were successfully fabricated inside a conventional glass-based microfluidic channel prepared
by wet etching for microparticle separation. Moreover, a fish scale like microfilter was also fabricated
and appointed as a one-way valve, which showed excellent performance as we expected. These results
indicate that such embellishment of microfluidic devices is simple, low cost, flexible and easy to access.
We believe that, combined with TPP, the application of lab-on-chip devices would be further
extended.
Introduction
In the past two decades, microfluidic systems have been inten-
sively investigated due to their broad applications in chemistry,
physics, biology and medicine. Comparing with macroscopic
settings, microfluidic devices have a series of distinct advan-
tages, such as low materials consumption, simplicity, high
safety, high sensitivity, high throughput and high integrity,1–9
which strongly stimulate the development of new techniques for
the fabrication of microfluidic devices. For example, ultraviolet
(UV),10e-beam,11X-ray lithographies,12softlithography,13,14
and nanoimprint15methodologies were successfully used to
create planar microfluidic channels. In order to further improve
the applications of microfluidic chips, efforts were also devoted
to fabrications of three-dimensional (3D) structured chips.
Through multi-exposure photolithography and soft lithog-
raphy, 3D microfluidic devices were manufactured and assigned
as advective micromixer,16immunoassay,17and neuron culture
platforms.18By utilizing replica remolding, complex micro-
fluidic channels were also developed for the use of valves or
pumps,9the study of cell–cell interactions,19computing,20and
DNA manipulation.21Besides, fabrication routes such as X-ray
ablation, layer-by-layer printing were also used for 3D micro-
structure prototyping.22Recently, a novel self-folding route was
reported for fabrication of real 3D microstructures from pre-
created 2D micropatterns.23–25Using self-assembly, 2D cruci-
form structures with porosity could be transformed into 3D
microcontainers with porous walls,26,27which shows great
potential for cell encapsulation and separation towards micro-
fluidic appliations.28All of these efforts have greatly advanced
the developments of functional microfluidic devices. However,
up to now, it is still difficult to make refined microfluidic chips
with complex 3D topological structures in a designable, flexible,
and controllable fashion. Moreover, it is still lacking in nano-
technology for local mending, modification, functionalization
and integration of existing microfluidic devices for special
applications.
Femtosecond laser direct writing by two-photon photo-
polymerization (TPP) of resins, due to its powerful capability in
3D fabrication and high spatial resolution, would be a promising
solution to this gap.29–35Previously, for the first time, we reported
that TPP prototyping of nanoshells by using negative tone resin
SU-8, shows great potential for fabrication of 3D microfluidic
devices.35Recently, we found that TPP prototyping was also
a powerful technology for embellishing existing microfluidic
devices towards advanced functions. Herein, we demonstrate
a novel post-embellishment strategy for appending extra func-
tions on traditional microfluidic channels via femtosecond laser
micronanofabrications. Through careful adjustment of the laser
power, transparent microchannels would be well integrated with
additional photopolymer functional parts without incompati-
bility. As a representative example, microsieves with different
pore sizes and shapes were successfully created inside the glass
microchannel prepared by wet etching for sieving particles with
different sizes. In addition, a fish scale-like structure was
designed as a one-way valve for microfluidic chips. It isbelievable
that,inthe near future,
nanofabrications would be widely used for embellishing micro-
chips and thus make microfluidic devices almighty in more
extensive applications.
femtosecondlasermicro-
aStateKey Laboratory onIntegratedOptoelectronics,CollegeofElectronic
Science and Engineering, Jilin University, 2699 Qianjin Street, Changchun,
130012, China. E-mail: hbsun@jlu.edu.cn; chenqd@jlu.edu.cn; Fax: +86
431 85168281; Tel: +86 431 85168281
bDalian Institute of Chemical Physics, 457 Zhongshan Street, Dalian,
116023, China
cCollege of Physics, Jilin University, 2699 Qianjin Street, Changchun,
130012, China
† Electronic supplementary information (ESI) available: Scheme for
‘‘wall effect’’; scheme for particle movements in one-way microvalve
test; SEM images of particles before and after sieving. See DOI:
10.1039/c003264f
This journal is ª The Royal Society of Chemistry 2010Lab Chip, 2010, 10, 1993–1996 | 1993
TECHNICAL NOTEwww.rsc.org/loc | Lab on a Chip
Downloaded by JILIN UNIVERSITY LIBRARY on 30 July 2010
Published on 27 May 2010 on http://pubs.rsc.org | doi:10.1039/C003264F
View Online

Page 2

Experimental section
1.Fabrication of microchannels
The microfluidic channel was fabricated by conventional wet
etching. Briefly, a glass slide was washed by acetone, alcohol and
deionized water, and then dried by nitrogen. After deposition of
Chrome (30 nm, adhesive layer) and gold (100 nm, sacrificial
layer) on the glass slide, a layer of photoresist was spin coated on
the sacrificial layer (SL). Required channel patterns could be
obtained by ultraviolet exposure under a mask and subsequent
development. After orderly removal of exposed SL and then
photoresist film, the glass slide was etched by hydrofluoric acid.
Finally,aglasschipwithchannelsimbeddedwasfinallyobtained.
In this work, the channel was 75 mm in width and 15 mm in depth.
2.Embellishment with inside 3D structures
The obtained microfluidic channel was firstly coated with nega-
tive resin SU-8 2075 (Microchem, US) on the channel side and
then scratched by straight-edged cover glass. Then the micro-
channels containing SU-8 were prebaked for 40 min at 95?C for
solidification, and cooled down to room temperature subse-
quently. The 3D structures were designed by 3D MAX, and
converted to operable computer data. The laser beam from
a femtosecond laser (80 MHz repetition rate, 120 fs pulse width,
800 nm central wavelength) was focused by a 60? oil immersion
objective lens (numerical aperture NA ¼ 1.35) to directly write
the desired microstructure. After that, the polymer embellished
glass chips were put in the oven for 15 min at 95?C for post-bake.
By rinsing the unphotopolymerized resin with SU-8 developer,
the designed microstructure was obtained in the microfluidic
channel. Finally, we covered the open channel with a cured
PDMS (Dow Corning, US) slab, and pressed it for adhesion.
Results and discussions
1. Embellishment of microfluidic channel with microsieves
For illustrating the feasibility of embellishment on existing
microfluidic chips, a conventional glass-based microfluidic
channel was used as a substrate for building inside 3D micro-
structures. As shown in Fig. 1, firstly, a ‘‘T’’ shaped microfluidic
channel was prepared by wet etching on a glass slide (75 mm in
width, and 15 mm in depth, see detailed procedures in the
Experimental section). Then the channel was covered with SU-8
2075 photoresist. It is worth pointing out that during TPP
fabrication, the 60? oil lens of the 1.35 numerical aperture (NA)
makes the work distance of the lens very short, so excess resin has
to be scraped off by straight-edged cover slide to protect the lens
and meanwhile let the laser focus at the bottom freely without
obstacle. After prebake of the resin, we use the femtosecond laser
to directly write in the microchannel according to the pre-
programmed microstructures. The entire process including the
positioning of sites to be addressed was monitored under
the Charge Coupled Device (CCD) set, because we had to induce
the laser light to focus36on the interior of the tiny channel but not
on the platform. After that the unphotopolymerized resin was
removed by SU-8 developer, and the embellished microstructure
was obtained in the microfluidic channel. For further micro-
separation tests, a PDMS slab was covered to make the channel
a closed system. At each end of the ‘‘T’’ shaped channel, there
was a hole used for the sample inlet, sample outlet, and waste
outlet respectively.
TPP prototyping inside the microchannel is quite different
from the processing on a smooth substrate. For example, when
we fabricated a simple sieve inside the microchannel under
conventional conditions, only collapsed structures were obtained
after development (ESI, Fig. S1a†). A possible reason might be
due to the interference of the channel walls, here called ‘‘wall
effect’’. When the laser wrote at the position near the wall of the
channel, the light passed through both the glass wall and
the resin. The different refractive index of SU-8 resin (1.58) and
Fig. 1
by femtosecond laser micronanofabrication. (i) Preparation of micro-
fluidic channel by conventional wet etching; (ii) surface coating with
photoresist of SU-8 and direct writing by femtosecond laser; (iii) devel-
oping; (iv) covering of the device with PDMS slab.
Schematic illustration of embellishment of a microfluidic system
Fig. 2
channel. (a) 3D structures designed by 3D max. The structures have
mainly two elements: a wall with high aspect ratio, and pores in it with
different shapes, such as round (b), square (c), round-end rectangle (d),
pentagrams (e), and triangle (f). The very high resolution and high aspect
ratio showed the powerful capability of fabricating complex 3D struc-
tures by TPP.
Microstructrues fabricated by direct writing of TPP inside the
1994 | Lab Chip, 2010, 10, 1993–1996This journal is ª The Royal Society of Chemistry 2010
Downloaded by JILIN UNIVERSITY LIBRARY on 30 July 2010
Published on 27 May 2010 on http://pubs.rsc.org | doi:10.1039/C003264F
View Online

Page 3

the glass (1.51)37defocused the incident beam (ESI, Fig. S1b
right†), and thus laser intensity was not strong enough to reach
the threshold of resin polymerization. As a result, the edge of the
sieve wafer did not stuck to the wall of the channel, and collapsed
easily after development. So the energy of the laser used for
fabricating microstructures inside the channel should be higher
than that on the flat surface because of the as-formed ‘‘wall
effect’’. Under this guidance, we successfully fabricated a well
stood sieve wafer in the microchannel (ESI, Fig. S1c†).
In order to show the variety of TPP in embellishing micro-
fluidic chips, microsieves with various high-resolution pores were
firstly fabricated inside the microfluidic channel. As shown in
Fig. 2, microsieve wafers 1 mm thick, 15 mm high with different
pore shapes including round, square, round-end rectangle,
pentagram, and equilateral triangle could be easily created
according to preprogrammed 3D structures (Fig. 2a). Although
the shapes seem to be simple, they include the basic building
elements of various complex structures, including straight
edge (edges of square and triangle), cambered edge (round),
acute angle (triangle), right angle (square), obtuse angle (ESI,
Fig.S1c†), and even angle with negative degree (pentagram). The
spatial resolution of the structures could achieve as high as
a submicrometer.
2. Microsieving tests
For microsieving tests, microsieves with round pores were
fabricated for the separation of microparticles with diameters in
the range of 2.5 to 5.5 mm. In our case, the pore size of micro-
sieves could be adjusted in a certain range. As representative
examples, a series of sieve wafers with 6, 5.5, 5, 4 and 3.5 mm
pores could be adopted for sieving particles with different sizes
(Fig. 3a–d). According to the diameter of the microparticles, we
chose the sieve with pores of 4.0 mm diameter for the test. As
shown in Fig. 3e, the ‘‘T’’ shape channel (75 mm in width, and
15 mm in depth) embellished with the above mentioned microf-
liter was coverd with PDMS slab to form close channels, and
there were three reservoirs at each end of the channel used for
sample inlet (A), sample outlet (B), and waste outlet (C)
respectively. By altering the depth of fluid in the reservoirs, we
could control the flow direction in the channels. Due to the good
wettability to both the PDMS and glass, herein, we used alcohol
as the media fluid to deliver the particles. It could be clearly
identified from the optical microscopy images that particles
smaller than the pores could pass the sieve (Fig. 3e), on the
contrary, bigger ones (Fig. 3f) could be headed off. If too many
blocked particles assembled at the sieve, we changed the flow
direction (A to B, A to C) to remove the particles by altering the
depth of the fluid in each reservoir. Statistic results of micro-
particles before and after sieving (Fig. 3g) show that particles
with diameter larger than 4.0 mm have been successfully headed
off, exhibiting its excellent separating capacity (ESI, Fig. S2†).
Fig. 3
fabricated sieves with pores of different diameters to sort particles. The
diameters of the pores are 5.5 mm (a), 5 mm (b), 4 mm (c), 3.5 mm (d)
respectively, and we used the 4.0 mm sieves for further testing (e). If the
particle size is smaller than the pores, it would pass the sieve, (f) but if the
particle is bigger than the pores, it would be blocked by the sieve. (g)
Statistic results of microparticles before and after sieving.
Microsieves and their working effects in the microchannel. We
Fig. 4
view (b, c, d) of the valve with fish scale like structures. If the particles
come from the non-eave direction, they would meet the eaves after they
passed the pores. If the particles come from the eaves direction, they meet
the eaves before the wall, and they lose their horizontal momentum when
they bang against the eaves, and would be blocked by the valve.
Fish scale like filters act as one way valves. Top view (a) and side
This journal is ª The Royal Society of Chemistry 2010Lab Chip, 2010, 10, 1993–1996 | 1995
Downloaded by JILIN UNIVERSITY LIBRARY on 30 July 2010
Published on 27 May 2010 on http://pubs.rsc.org | doi:10.1039/C003264F
View Online

Page 4

3.One-way microvalve
The functionalization of microfluidic chips by TPP prototyping
is not only limited to microsieves. For the second case, a fish scale
like microfilter was fabricated and designed as a one-way valve.
The unique filter wafer contained two major parts: a wall with
5.5 mm of round pores, and eaves half-covering the pores
(Fig. 4a–d). When the particles come from the eave direction,
they meet the eaves in advance, and lost their horizontal
momentums. As a result, they bang against the eaves due to the
flow force (ESI, Fig. S3†). On the contrary, if the particles come
from the opposite direction, they lost the horizontal momentums
after passing the pores, and then they would pass the valve with
flow of fluid (ESI, Fig. S3†). In the test, as shown in Fig. 4e, f, this
unique one-way microvalve was very employable, exhibiting
favourable performance as we expected.
Conclusions
In conclusion, we have found femtosecond laser micro-
nanofabrication a powerful technology for functionalizing
microfluidic devices with inside 3D microstructures. By using this
method, we have built a series of microsieves with round, square,
round-end rectangle, pentagram, equilateral triangle shaped
pores and adjustable pore sizes of 3.5–6.0 mm inside a vitreous
microfluidic channel, which, therefore, imparts the conventional
microchannel a separative function. Additionally, through
design and fabrication of a fish scale like microstructure (eaves
covered pores), the microchannel was functionalized with a one
way valve. The above results show that TPP prototyping has
already exhibited great potential for mending, modification,
functionalization and integration of general microfluidic chips,
and probably, it would be widely used in lab-on-chip systems in
the near future.
Acknowledgements
The authors acknowledge the financial support from NSFC
under grant nos. 60778004, 90923037, 60978062 and 60525412.
Notes and references
1 H. Craighead, Nature, 2006, 442, 387–393.
2 H. K. Wu, A. Wheeler and R. N. Zare, Proc. Natl. Acad. Sci. U. S. A.,
2004, 101, 12809–12813.
3 P. C. H. Li, Microfluidic Lab-on-a-Chip for Chemical and Biological
AnalysisandDiscovery,CRCpress,TaylorandFrancis,NewYork,2006.
4 H. Andersson, Sens. Actuators, B, 2003, 92, 315–325.
5 X. Gong, S. Peng, W. Wen, P. Sheng and W. Li, Adv. Funct. Mater.,
2009, 19, 292–297.
6 K. Sun, Z. X. Wang and X. Y. Jiang, Lab Chip, 2008, 8, 1536–1543.
7 M. Y. Zhang, X. Q. Gong and W. J. Wen, Electrophoresis, 2009, 30,
3116–3123.
8 M. Abdelgawad, M. W. L. Watson and A. R. Wheeler, Lab Chip,
2009, 9, 1046–1051.
9 M. Abdelgawad, M. W. L. Watson, E. W. K. Young, J. M. Mudrik,
M. D. Ungrin and A. R. Wheeler, Lab Chip, 2008, 8, 1379–1385.
10 B. F. Liu, M. Ozaki, H. Hisamoto, Y. Utsumi, T. Hattori and
S. Terabe, Lab Chip, 2004, 4, 368–371.
11 V. Kudryashov, X. C. Yuan, W. C. Cheonq and K. Radhakrishnan,
Microelectron. Eng., 2003, 67, 306–311.
12 F. Romanato, M. Tormen, L. Businaro, L. Vaccari, T. Stomeo,
A. Passaseo and E. Di Fabrizio, Microelectron. Eng., 2004, 73, 870–
875.
13 M. A. Unger, H. P. Chou, T. Thorsen, A. Scherer and S. R. Quake,
Science, 2000, 288, 113–116.
14 G. M. Whitesides, E. Ostuni, S. Takayama, X. Y. Jiang and
D. E. Ingber, Annu. Rev. Biomed. Eng., 2001, 3, 335–373.
15 X. G. Liang, K. J. Morton, R. H. Austin and S. Y. Chou, Nano Lett.,
2007, 7, 3774–3780.
16 A. D. Stroock, S. K. W. Dertinger, A. Ajdari, I. Mezic, H. A. Stone
and G. M. Whitesides, Science, 2002, 295, 647–651.
17 X. Y. Jiang, J. M. K. Ng, A. D. Stroock, S. K. W. Dertinger and
G. M. Whitesides, J. Am. Chem. Soc., 2003, 125, 5294–5295.
18 A. M. Taylor, M. Blurton-Jones, S. W. Rhee, D. H. Cribbs,
C. W. Cotman and N. L. Jeon, Nat. Methods, 2005, 2, 599–605.
19 D. T. Chiu, N. L. Jeon, S. Huang, R. S. Kane, C. J. Wargo, I. S. Choi,
D. E. Ingber and G. M. Whitesides, Proc. Natl. Acad. Sci. U. S. A.,
2000, 97, 2408–2413.
20 D. T. Chiu, E. Pezzoli, H. K. Wu, A. D. Stroock and
G. M. Whitesides, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 2961–
2966.
21 D. Huh, K. L. Mills, X. Y. Zhu, M. A. Burns, M. D. Thouless and
S. Takayama, Nat. Mater., 2007, 6, 424–428.
22 B. M. Weon, S. Chang, J. Yeom, S. K. Hahn, J. H. Je, Y. Hwu and
G. Margaritondo, J. Appl. Phys., 2009, 106, 153518.
23 J. S. Randhawa, M. D. Keung, P. Tyagi and D. H. Gracias, Adv.
Mater., 2010, 22, 407–410.
24 J. H. Cho, T. James and D. H. Gracias, Adv. Mater, 2010, DOI:
10.1002/adma.200904410.
25 T. Leong, A. Zarafshar and D. H. Gracias, Small, 2010, 6, 792–806.
26 T. G. Leong, C. L. Randall, B. R. Benson, A. M. Zarafshar and
D. H. Gracias, Lab Chip, 2008, 8, 1621–1624.
27 C. L. Randall, A. Gillespie, S. Singh, T. G. Leong and D. H. Gracias,
Anal. Bioanal. Chem., 2009, 393, 1217–1224.
28 J. Wang, M. Patel and D. H. Gracias, Nano, 2009, 4, 1–5.
29 H. B. Sun and S. Kawata, Adv. Polym. Sci., 2004, 170, 169.
30 S. Maruo, O. Nakamura and S. Kawata, Opt. Lett., 1997, 22, 132.
31 S. Kawata, H. B. Sun, T. Tanaka and K. Takada, Nature, 2001, 412,
697.
32 R. Guo, S. Xiao, X. Zhai, J. Li, A. Xia and W. Huang, Opt. Express,
2006, 14, 810–816.
33 S. Zhu, Y. F. Lu and M. H. Hong, Appl. Phys. Lett., 2001, 79, 1396–
1398.
34 T. A. Pham, D. P. Kim, T. W. Lim, S. H. Park, D. Y. Yang and
K. S. Lee, Opt. Lett., 2006, 16, 1235–1237.
35 D. Wu,Q. D. Chen,L. G. Niu,J. N. Wang,J. Wang,R.Wang,H. Xia
and H. B. Sun, Lab Chip, 2009, 9, 2391–2394.
36 M. Born and E. Wolf, Principles of Optics, Pergamon Press Ltd.,
London, 1959.
37 M. J. Weber, Handbook of Optical Materials, CRC Press, Boca
Raton, Florida, 2002.
1996 | Lab Chip, 2010, 10, 1993–1996This journal is ª The Royal Society of Chemistry 2010
Downloaded by JILIN UNIVERSITY LIBRARY on 30 July 2010
Published on 27 May 2010 on http://pubs.rsc.org | doi:10.1039/C003264F
View Online