Micropatterning of Porous

Structures from Co-Continuous

Polymer Blends
WEI ZHANG
School of Polymer, Textile and Fiber Engineering, Georgia Institute of Technology, Atlanta, GA 30332

MIN LI, CHAOSHENG WANG

School of Polymer, Textile and Fiber Engineering, Georgia Institute of Technology, Atlanta, GA 30332
School of Materials Science and Engineering, Donghua University, Shanghai, China

JACK G. ZHOU
Department of Mechanical Engineering and Mechanics, Drexel University, PA 19104

DONGGANG YAO
School of Polymer, Textile and Fiber Engineering, Georgia Institute of Technology, Atlanta, GA
Received: June 2, 2011
Accepted: December 11, 2011

ABSTRACT: In contrast to the immense literature in porous material

generation, micropatterning of porous devices remains a technical challenge. In
this study, a new process for micropatterning of porous structures with a
controllable morphology was developed and investigated. This process combines
polymer melt blending, hot embossing, and in-mold annealing for geometrical
pattern transfer and simultaneous morphological control. A special effort was
made to generate a microgroove pattern with an open pore structure. Parametric
experimental studies were conducted on stamps with different feature sizes

Correspondence to: Donggang Yao; e-mail: yao@gatech.edu.

Contract grant sponsor: National Science Foundation.
Contract grant numbers: CMMI-0800016 and CMMI-0800735.

Advances in Polymer Technology, Vol. 00, No. 0, 1–14 (2012)


C 2012 Wiley Periodicals, Inc.
MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS

under different processing conditions. The results demonstrated the feasibility

and the versatility of the proposed technique in fabricating micropatterned
porous structures. By varying the geometrical and boundary conditions during
in-mold annealing, micropatterns with graded porous structures were
demonstrated.  C 2012 Wiley Periodicals, Inc. Adv Polym Techn 00: 1–14, 2012;

View this article online at wileyonlinelibrary.com. DOI 10.1002/adv.21260

KEY WORDS: Annealing, Blends, Hot embossing, Molding, Porous structures

modynamically unstable process are hard to control.

Introduction The resulting devices also tend to have a dense skin
formed, which is undesired in filtration and scaffold-

D ue to their large internal surface area, porous

materials have been widely used in chemi-
cal and biomedical applications, where high surface
ing applications.
A desired process for micropatterning porous
devices would be able to in situ control the gen-
eration of the porous structure while achieving
activity is desired. Examples are extracellular scaf-
high geometrical patterning accuracy. One versa-
folds for tissue engineering,1,2 porous substrates for
tile technique for controllable generation of a co-
catalytic reaction,3,4 and permeable media for mem-
continuous phase morphology is immiscible poly-
brane filtration.5,6 In addition to the internal porous
mer blending;28−34 a porous structure can be pro-
structure, some emerging applications, such as cell
duced by extraction of the sacrificial polymer. By
aligning scaffolds,7−12 microfluidic devices,13−17 mi-
tuning the material and process parameters, one
cropatterned porous electrodes,18,19 and fuel cell
can generate a desired porous structure with well-
membranes,20−25 further require these materials to
defined pore size and distribution. Particularly, the
be shaped into specific geometries, including mi-
authors35,36 have developed techniques for control-
cropatterns.
lable growth of gradient porous structures by reg-
In contrast to the immense literature in porous
ulating the boundary conditions during an in-mold
material generation, micropatterning of porous de-
annealing process. In the meantime, there has also
vices remains a technical challenge. Most well-
been extensive work on precision micropattern-
known techniques for processing porous materials,
ing of polymers.37,38 It would therefore be inter-
such as gas foaming, powder sintering, and solid
esting to explore the feasibility of combining co-
freeform fabrication, have limited capability due to
continuous polymer blending with precision mi-
their relatively poor morphological control and/or
cropatterning for creating micropatterned porous
low geometrical resolution at smaller sizes. Recently,
devices.
Wessling and coworkers26,27 developed a phase sep-
In this study, co-continuous polymer blending
aration micromolding process for generating mi-
was combined with hot embossing, a widely used
cropatterns with internal pores down to submicron
micropatterning technique, for the fabrication of mi-
sizes. They cast a concentrated polymer solution into
cropatterned porous devices. A salient feature of
a micropatterned mold and employed freeze drying
this new process is that a controllable in-mold an-
or coagulation for extracting the solvent and form-
nealing stage is applied to the embossed blend to
ing the desired porous structure. This process has
regulate its structural evaluation. A special effort
been used for generating micropatterned devices,
was made to generate an open pore structure at
such as microsieves5,6 and porous scaffolds,7 for fil-
the mold contact surface. Parametric experimental
tration and biomedical applications. However, there
studies were conducted on stamps with different
are some fundamental limitations with this process.
feature sizes under different processing conditions.
The use of a large amount of solvent leads to a slow
The results demonstrated the feasibility and the
and environmentally unfriendly process. More criti-
versatility of the proposed technique in fabricating
cally, the pore size and its distribution from this ther-

2 Advances in Polymer Technology DOI 10.1002/adv

 MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS
micropatterned porous structure with controllable
morphology.
Fabrication Process
The process sequence used in the new process
for fabricating micropatterned porous materials is
shown in Fig. 1. Initially, two immiscible polymers
are melt–mixed under suitable conditions to form
a co-continuous blend with phase size in the mi-
crometer range. The quenched blend is then re-
heated above the softening temperatures (i.e., glass
transition temperature for amorphous polymers or
melting temperature for semicrystalline polymers)
of both polymer components and hot embossed (or
compression molded) in a heated mold. After the
embossing stage, the embossed blend further under-
goes an in-mold annealing stage, where controlled
phase evolution occurs for generating the desired
morphology. Finally, after cooling and demolding
of the embossed article, one polymer component is
selectively extracted, leaving behind a porous mi-
cropattern.
The critical step in this special hot embossing pro-
cess is in-mold annealing. In contrast with the stan-
dard hot embossing process for purely geometrical
patterning purposes, the new process is able to con-
trol the evolution of the internal phase structure.
First, the initial phase structure will be coarsened
from thermal annealing effects. Due to the interfa-
cial tension, the interfacial area between the polymer
components will decrease, resulting in a larger phase
size. Second, the phase structure can further be regu-
lated by applying proper boundary conditions dur-
ing in-mold annealing. Particularly, by employing a
FIGURE 1. Process sequence of porous micropatterning: (a) polymer blending to form a co-continuous blend; (b) hot
embossing with controllable thermomechanical history; (c) sacrificial extraction and formation of porous patterns.
Advances in Polymer Technology DOI 10.1002/adv
temperature difference at the boundary, a gradient
phase structure can be generated, as demonstrated
in the authors’ previous work.35 Furthermore, var-
ied kinematical and dynamical conditions at the
mold surface can be utilized for fine controlling the
phase structure near the surface. It has been found
that a different chemical interaction at the mold
surface can yield a different surface morphology.36
By modifying the mold surface properties, different
morphology, including a segregated layer and co-
continuous dual phases, can be generated right at
the surface. Last but not least, geometrical confine-
ment can also be used in the annealing step to influ-
ence the coarsening kinetics and produce globally or
locally aligned microporous structures.
Experimental Procedures
MATERIALS
The two polymers, poly(lactic acid) (PLA) and
polystyrene (PS), used in this study were obtained
from commercial sources. The PLA was a semicrys-
talline polymer produced by NatureWorks LLC un-
der the product name of INGEOTM 6202D. It has
a Tg (glass transition temperature) of 55◦ C and a
Tm (melting temperature) of 165◦ C, as measured
by a differential scanning calorimeter, model TA-
Q200 from TA Instruments (New Castle, DE), with
a temperature rate of 10◦ C/min. The PS was an
amorphous polymer, provided by Dow Chemical
(Midland, MI) under the product name of Styron
675. The Tg for this polymer was determined to
be 95◦ C and the rubbery plateau termination (or
flow) temperature is around 130◦ C. The thermal sta-
bility of these resins was examined using thermal
3
MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS

gravimetric analyzer, model TA-Q5000IR from TA
Instruments, with a heating rate of 10◦ C/min in a ni-
trogen environment. The thermal degradation tem-
peratures of PLA and PS were determined to be 270
and 380◦ C, respectively. To minimize the hydrolysis
of the PLA, both polymers were dried inside a vac-
uum oven at 90◦ C for 12 h prior to melt processing.
BLENDING
The dried PLA and PS at a 50/50 weight ra-
tio were melt mixed in a recirculating batch mixer
(RCPBM-2000, Randcastle Extrusion Systems, Inc.,
Cedar Grove, NJ) at 180◦ C and 30 rpm with nitrogen
purging. The mixing operation was stopped when
an equilibrium torque was achieved. Although still
molten, the mixture was scraped out and quenched
into tap water to preserve the blend morphology.
The solidified blend was dried in a vacuum oven
for 6 h at 90◦ C and stored in a desiccator before hot
embossing.
HOT EMBOSSING
The experimental setup for hot embossing,
retrofitted onto a hydraulic press with upper and
lower platens, is illustrated in Fig. 2a. Silicon stamps
with square microchannels having widths of 10, 50,
and 250 μm, as shown in Figs. 2c and 2d, were used
as the embossing mold. Samples with dimensions of
1 × 0.5 × 0.5 cm (length by width by thickness) were
cut from the as-mixed blend. The sample surface was

FIGURE 2. Experimental setup for embossing and annealing: (a) hot embossing setup; (b) in-mold annealing setup; (c)
top view of the silicon stamp; (d) side view of the silicon stamp.

4 Advances in Polymer Technology DOI 10.1002/adv

 MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS
polished with a fine sandpaper. The polished sam-
ple was preheated in an oven set at the same tem-
perature as the target embossing temperature. Once
reaching the target temperature, the sample was im-
mediately removed from the oven and hot embossed
against the silicon stamp placed on the lower heated
platen of the hot press. The combined embossing
and holding stages were carried out under constant
pressure control. A load cell (LCM305-2KN, Omega
Engineering, Inc., Stamford, CT) installed on the up-
per platen was used for pressure monitoring. A rub-
ber pad, approximately 10 mm thick, inserted be-
tween the load cell and the upper platen was used
to facilitate uniform transfer of the embossing pres-
sure. The total embossing and holding time was set
to be 1 min, of which the initial pressure loading
step (i.e., the embossing step) took about 5 s and
the holding stage used the remaining 55 s. The pri-
mary experimental parameters investigated in this
study included the embossing force (10, 20, 50, and
80 N), the embossing temperature (100, 120, 140, 160,
and 180◦ C), and the microchannel size (10, 50, and
250 μm) of the stamp.
IN-MOLD ANNEALING
The tendency for a co-continuous blend struc-
ture to coarsen during thermal annealing offers the
opportunity to control internal morphology with-
out using complex setups. To determine a suitable
scheme for morphological control in micropattern-
ing, three different annealing protocols/sequences
were investigated.
In the first approach, the as-mixed blend was first
hot embossed using the procedure described in the
previous section. At the end of the holding stage,
the embossed polymer blend and the silicon stamp
(without ejection) were placed inside an oven at
180◦ C for thermal annealing, with constant nitrogen
purging. After annealing, the molten blend together
with the mold was quenched into tap water.
In the second approach, the sequence of hot em-
bossing and annealing was inverted. The anneal-
ing step was first conducted to grow the blend
phase structure to the desired size; then, the an-
nealed blend was hot embossed against the silicon
stamp.
The third approach is a further modification of the
first one. Specifically, the annealing step was modi-
fied to include a gradient thermal field. Instead of a
convective oven, a hot press with a differential tem-
perature setup, as shown in Fig. 2b, was employed.
The bottom platen was set at 200◦ C, whereas the up-
Advances in Polymer Technology DOI 10.1002/adv
per platen was kept at room temperature. By doing
this, a temperature gradient was generated in the di-
rection perpendicular to the platen surface. During
the entire in-mold annealing process, a small force
was applied to maintain the contact but not to de-
form the material.
EXTRACTION OF SACRIFICIAL POLYMER
Samples with desired phase structures and mi-
cropatterns were submerged into cyclohexane at
75◦ C for 24 h to selectively remove the PS phase and
form a PLA mircoporous structure. This dissolution
process was monitored by measuring the weight loss
of the sample. The dissolved samples were collected
and dried thoroughly under vacuum at 80◦ C for at
least 4 h to remove the residual solvent.
CHARACTERIZATION
Scanning electron microscopy (SEM) was used
for observing the pattern geometry and the micro-
porous structure. To prepare the SEM sample, the
dried porous patterns were brittle fractured in liq-
uid nitrogen and sputter coated with gold.
Results and Discussion
EFFECTS OF EMBOSSING
Prior to embossing, a co-continuous, intercon-
nected 50/50 wt% PLA/PS blend was prepared
first. The processing strategies discussed in the pre-
vious work35,36 were strictly followed to generate
the required phase structure. The co-continuity or
connectivity of the phase structure was confirmed
by weight loss measurement and SEM observation
on samples selectively dissolved by cyclohexane
(which is a solvent to PS but not to PLA). From the
weight loss test, nearly all PS was extracted from
the blend after dissolved in cyclohexane, indicating
a nearly 100% co-continuity in the PLA/PS blend.
Figure 3 shows an SEM of the porous structure af-
ter PS extraction. From this micrograph, the average
phase size was estimated to be 4 μm based on image
analysis. Specifically, the pore size, S, is defined as
follows:
4A
S= (1)
L
5
MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS

FIGURE 3. Phase morphology of as-mixed PLA/PS

50/50 wt% blend.

where A and L are the area and the perimeter of a

phase droplet/island/striation in the 2-D image.
During embossing, the blend is subjected to a de-
formation field. The effect of this deformation pro-
cess on the structural changes was studied first. The
silicon stamps with three different feature sizes (10,
50, and 250 μm) were used in this study. Despite
the difference on feature size, all stamps have as-
pect ratio of 1. Five temperature levels (100, 120,
140, 160, and 180◦ C) and four compression force lev-
els (10, 20, 50, and 80 N) were applied to cover a
broad range of processing conditions. Note that the
temperatures employed in the experiment were all
above the Tg of PS to ensure the deformability of
the blend. Likewise, the low limit of the emboss-
ing force, 10 N, was set to achieve a full replica-
tion of the groove. Compression force higher than
80 N was not applied as this tended to cause the
breakage of the silicon stamp. The total embossing
stage lasted for 1 min, and after embossing, the sam-
ple was immediately quenched into tap water. For
all samples, the PS phase was extracted before SEM
imaging.
Figure 4 shows the replication and morphology
of samples prepared at 100◦ C and 80 N with three
different sized stamps. Under these embossing con-
ditions, the silicon grooves are reasonably replicated,
except local sharp edges. For all three cases, an ori-
entated structure was observed at the entrance of
the microribs, where primary deformation occurred.
One can find that the microstructures were actu-
ally stretched and oriented into the cavities on the
stamp. At this low embossing temperature, close to
Tg of PS and well below the PLA melting tempera- FIGURE 4. Samples produced at 100◦ C and 80 N with
ture, structural recovery or relaxation during defor- stamps of different groove sizes: (a) 10 μm; (b) 50 μm;
mation is anticipated to be insignificant. Therefore, (c) 250 μm.

6 Advances in Polymer Technology DOI 10.1002/adv

 MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS
the deformation-caused structural orientation can be
retained. The degree of orientation appeared to be
affected by the micropattern dimension. The orien-
tation is minimal for the 10 μm groove pattern but
became more obvious at an increased groove dimen-
sion. The different orientation at different pattern
sizes may be explained by considering the differ-
ent levels of deformation involved in embossing the
three different microgrooves. As the groove size in-
creased, more materials would be transported into
or from the substrate, leading to increased orien-
tation in the substrate. Furthermore, different rates
of relaxation at different sizes may also contribute
to the orientation effect. The structural relaxation is
primarily driven by the interfacial tension. One can
define a relaxation time, λ, from a dimensional argu-
ment as follows:
ηD
λ= (2)
 evated temperature. Based on these observations, it
where  is interfacial tension, η is equivalent vis-
cosity, and D is dimension. Therefore, the structural
relaxation time is expected to be smaller at a reduced
size scale for deformation, resulting in reduced ori-
entation effect. At first glance of the micrographs, the
porosity also appeared to be affected by the pattern
size. However, this is only a visual illusion. Note that
the initial blends used for embossing the three differ-
ent microgrooves all had the same blend structure,
with a phase size around 3 μm. For larger grooves, a
larger scale bar is needed in the micrograph. At the
larger scale bar, the micropores are more difficult to
observe. This produced a falsely denser structure.
In fact, weight measurement of the embossed pat-
terns before and after extraction of the PS phase all
showed that a 50% weight fraction remained in the
porous material.
The replication and morphology of samples pre-
pared at increased temperatures but still below the
PLA melting temperature are shown in Figs. 5–
7, with both side and top views of the embossed
structures. From the side view, it can be seen that
the morphological orientation was reduced at these
higher temperatures. At 160◦ C, close to the PLA
melting temperature, a nearly isotropic porous struc-
ture was found even in those regions near the mi-
croribs, where large deformation occurred (Fig. 7).
The embossed structure after PS extraction also ex-
hibited an open pore structure, even at the out-
side surfaces. This can be clearly seen from the
top views in Figs. 5–7. At these increased temper-
atures, the required force for complete filling of the
Advances in Polymer Technology DOI 10.1002/adv
silicon groove was significantly reduced. Although
a force of 80 N was needed to fill the grooves at
100◦ C, only 20 N was needed to fill the grooves
at 140◦ C.
Embossing experiments at temperatures above
the PLA melting temperature were also conducted.
As an example, Fig. 8 shows the replication and
morphology of patterns embossed at 180◦ C, approx-
imately 15◦ C above the PLA melting temperature.
At this elevated temperature, only 10 N compres-
sion force was needed to achieve complete filling
of the microcavities. This reduced embossing force
is understandable due to the highly increased flu-
idity of the blend at a temperature above the Tg
or Tm of both components. Under these embossing
conditions, faithful pattern duplication was realized,
even at the sharp edges of the microgrooves. From
the side view in Fig. 8, it can be seen that a highly
isotropic porous structure was produced at this el-
seems that embossing a completely melted blend
is similar to embossing a semimolten blend at re-
duced temperatures (Figs. 5–7). However, at 180◦ C,
a dense skin was found on the external surfaces of
the embossed pattern. In fact, close examination of
replicated porous patterns indicated an earlier ini-
tiation of this skin formation phenomenon even at
a temperature below Tm of PLA. If comparing the
surface appearances shown in Figs. 6–8, one can eas-
ily see a gradual decrease of the surface openness
when the embossing temperature increased from
140 to 180◦ C. When high permeability is a priori,
the formation of a denser skin is undesired. On the
other hand, the modification of the surface openness
can render a change in permeability of the struc-
ture. One may therefore utilize this phenomenon
for controlling the permeability of the resulting
device.2
EFFECTS OF ANNEALING
For a co-continuous polymer blend, its internal
structure tends to coarsen when the processing tem-
perature is higher than the flow temperatures (Tm
for PLA and the rubber plateau temperature for
PS) of the involving components. Through quies-
cent annealing at different temperatures and time,
material structures with different phase sizes can
be generated. In the current study, annealing and
hot embossing were combined to create micropat-
terned porous structures with different pore sizes.
Two procedures were tested for this fabricating
7
MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS

FIGURE 5. Samples produced at 120◦ C and 20 N compression force with stamps of different groove sizes: (a1) and
(a2) 10 μm; (b1) and (b2) 50 μm; (c1) and (c2) 250 μm.

8 Advances in Polymer Technology DOI 10.1002/adv

 MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS

FIGURE 6. Samples produced at 140◦ C and 20 N compression force with stamps of different groove sizes: (a) 10 μm;
(b) 50 μm; (c) 250 μm; (d) 250 μm, top view.

concept and the influence of the embossing/
annealing sequence on the final sample structures
was investigated. In the first procedure, anneal-
ing was conducted prior to embossing or mold-
ing, whereas the second one reversed the sequence.
These two fabricating procedures were referred to as
Annealing-Molding (AM) and Molding-Annealing
(MA), respectively, for further reference.
The experimental results from the AM protocol
are given in Fig. 9. These samples were first an-
nealed inside an oven at 180◦ C for 10 min. The an-
nealed samples were cooled to room temperature to
freeze the phase morphology before being reheated
to 120◦ C and embossed with 20 N compression force.
In Figs. 9a and 9c, most of the surface pores were
closed after hot embossing for samples with both
50 and 250 μm features. The cross sections of these
samples (Figs. 9b and 9d) demonstrated an oriented
structure especially around the corners. Such a high
level of orientation was not observed in the em-
bossing experiments without the preannealing step.
This can be seen by comparing with the structure
in Fig. 5. It can also be seen that the pore sizes are
very different in these two cases; with the anneal-
ing step, the pore size was increased by at least five
times. From these contrasted observations, it may
be reasonable to conjecture that the initial phase size
plays an important role in affecting the embossed

Advances in Polymer Technology DOI 10.1002/adv 9

MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS

FIGURE 7. Samples produced at 160◦ C and 20 N compression force with stamps of different groove sizes: (a) 10 μm;
(b) 50 μm; (c) 250 μm; (d) 250 μm, top view.

structure; the large phase structure appeared to be
more affected by the deformation. The reduced ori-
entation at smaller pore sizes may be attributed to
the increased elastic effects caused by the increased
interfacial area.
The experimental results from the MA protocol
are given in Fig. 10. The annealing step was con-
ducted after embossing, but the embossing step was
performed under the same condition as in the AM
protocol. Compared to Fig. 9, one can see that the
phase structures in Fig. 10 are much more uniform.
Furthermore, open pore surfaces were obtained with
this processing sequence. The difference may be ex-
plained as follows. As mentioned earlier, smaller
pore size helps suppress orientations from defor-
mation due to the increased elastic effect. Likewise,
the annealing step facilitates additional elastic re-
covery/relaxation, thus further reducing the orien-
tation. Therefore, the MA protocol is a better one for
generating porous structures with a uniform mor-
phology. Figure 10 also indicates that the pore size
in the embossed pattern may be easily controlled by
the annealing time. With an annealing time of 3 min,
the pore size grew from the original size of about 4
μm to about 12 μm and further grew to about 40 μm
after 5 min.

10 Advances in Polymer Technology DOI 10.1002/adv

 MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS

FIGURE 8. Samples produced at 180◦ C and 10 N compression force with a stamp of 250 μm groove size: (a) side view;
(b) top view.

FIGURE 9. Samples prepared by annealing at 180◦ C for 5 min followed by hot embossing at 120◦ C and 20 N
compression force, with stamps of different groove sizes: (a) 50 μm, top view; (b) 50 μm, cross section; (c) 250 μm, top
view, (d) 250 μm, cross section.

Advances in Polymer Technology DOI 10.1002/adv 11

MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS

FIGURE 10. Samples embossed at 120◦ C and 20 N compression force before annealed in oven at 180◦ C for different
times: (a) 50 μm and 3 min, cross section; (b) 50 μm and 3 min, top view; (c) 250 μm and 5 min, top view; (d) 250 μm
and 5 min, cross section.

MICROPATTERNING WITH GRADED crogroove on the stamp were used to impose a

POROUS STRUCTURES geometric confinement locally onto the blend mi-
crostructure. When the phase structure grew to a
The MA method can be further developed for cre-
level comparable to the size of the microgroove, the
ating porous micropatterns with a graded porous
existence of the groove walls could strongly impinge
structure. The coarsening rate during the annealing
the structure evolution of the blend. Instead of grow-
stage may be controlled spatially by varying the ge-
ing equally in all directions, the structure tends to
ometrical and/or boundary conditions applied on
grow along the less obstructed directions. As a re-
the blend. Two examples are given below to demon-
sult, the microstructure grew faster along the groove
strate this valuable modification.
direction, which caused the formation of an orien-
In the first example, geometrical confinement was
tated morphology. Similar geometrical confinement
utilized to generate orientated structure. Figure 11
effects were found previously by the authors during
shows a phase morphology highly modified by the
compression molding of a planar geometry.36
geometrical conferment. Here, the sidewalls of mi-

12 Advances in Polymer Technology DOI 10.1002/adv

 MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS
FIGURE 11. Surface-aligned structure formed by embossing at 120◦ C and 20 N compression force and
consequentially annealing inside oven at 180◦ C for 10 min: (a) top view; (b) cross section. The groove size is 50 μm.
FIGURE 12. Micropatterned graded porous structure
formed by embossing at 120◦ C and 20 N compression
force and consequentially annealing between two plates
with the upper plate at 20◦ C and the bottom plate at
200◦ C: (a) annealing unit design; (b) SEM image of the
micropatterned graded porous structure.
Advances in Polymer Technology DOI 10.1002/adv
In the second example, a varied thermal boundary
condition was applied to generate a gradient thermal
field and consequently a gradient coarsening rate.
This resulted in the generation of a gradient phase
structure in the micropattern, as shown in Fig. 12.
Instead of annealing in a uniform temperature field
(as used for generating the morphology in Fig. 10),
a temperature difference was imposed on the lower
and upper plates during annealing (Fig. 12a). With
this setup, the material in contact with the silicon
stamp had a higher temperature than the remain-
ing. As a result, the phase size on the patterned side
was much larger. The increased coarsening rate is
caused by the decreased viscosity at increased tem-
perature; this mechanism is well understood.35 Us-
ing the same principle, one can easily generate mi-
cropatterns with other types of graded structures.
Conclusions
In this study, a new process for micropatterning
of porous structures with a controllable morphol-
ogy was developed and investigated. Specifically,
a co-continuous polymer blend was hot embossed
against a micropatterned mold and then annealed
inside the mold to obtain the desired internal mor-
phology. It was found that embossing at a low tem-
perature below the polymer flow temperature led
to a significantly deformed internal structure. This
orientation effect was largely suppressed by hot
13
MICROPATTERNING OF POROUS STRUCTURES FROM CO-CONTINUOUS POLYMER BLENDS
embossing the blend above the flow temperature,
especially when an in-mold annealing stage was ap-
plied after embossing. In-mold annealing was found
to be a versatile technique for controlling the inter-
nal structure of the embossed pattern; not only the
overall phase size, but also its gradient and distri-
bution can be controlled by varying the boundary
conditions during in-mold annealing. This in-mold
annealing process is superior to preembossing an-
nealing, because the latter in combination with hot
embossing led to the formation of a closed-pore sur-
face and undesired structural deformation.
References
1. Virgilio, N.; Sarazin, P.; Favis, B. D. Biomaterials 2010, 31,
5719–5728.
2. Xiang, Z. Y.; Sarazin, P.; Favis, B. D. Biomacromolecules 2009,
10, 2053–2066.
3. Sun, C. Y.; Liu, S. X.; Liang, D. D.; Shao, K. Z.; Ren, Y. H.; Su,
Z. M. J Am Chem Soc 2009, 31(5), 1883–1888.
4. Selvam, P.; Mohapatra, S. K. Microporous Mesoporous Mater
2004, 7(3), 137–149.
5. Girones, M.; Akbarsyah, I. J.; Nijdam, W.; Van Rijn, C. J. M.;
Jansen, H. V.; Lammertink, R. G. H.; Wessling, M. J Membrane
Sci 2006, 283, 411–424.
6. Bikel, M.; Culfaz, P. Z.; Bolhuis-Versteeg, L. A. M.; Perez, J.
G.; Lammertink, R. G. H.; Wessling, M. J Membrane Sci 2010,
347, 93–100.
7. Papenburg, B. J.; Schuller-Ravoo, S.; Bolhuis-Versteeg, L. A.
M.; Hartsuiker, L.; Grijpma, D. W.; Feijen, J.; Wessling, M.;
Stamatialis, D. Acta Biometerialia 2009, 5, 3281–3294.
8. Lam, M. T.; Clem, W. C.; Takayama, S. C. Biomaterials 2008,
29, 1705–1712.
9. Isenberg, B. C.; Tsuda, Y.; Williams, C.; Shimizu, T.;
Yamato, M.; Okano, T.; Wong, J. Y. Biomaterials 2008, 29, 2565–
2572.
10. Yucel, D.; Kose, G. T.; Hasirci, V. Biomaterials 2010, 31,1596–
1603.
11. Berry, C. C.; Campbell, G.; Spadiccino, A.; Robertson, M.;
Curtis. Biomaterials 2004, 25, 5781–5788.
12. Sarkar, S.; Lee, G. Y.; Wong, J. Y.; Desai, T. A. Biomaterials
2006, 27, 4775–4782.
13. Jang, H. S.; Cho, M. W.; Park, D. S. Sensors 2008, 8, 700–710.
14
14. Song, Y. A.; Batista, C.; Sarpeshkar, R.; Han, J. J Power Sources
2008, 183, 674–677.
15. Mekaru, H.; Fukushima, E.; Hiyama, Y.; Takahashi, M. J Vac
Sci Technol B 2009, 27(6), 2814–2819.
16. Choi, H. W.; Bong, S.; Farson, D. F.; Lu, C. M.; Lee, L. J. J Laser
Appl 2009, 21, 196–204.
17. Pemg, B. Y.; Wu, C. W.; Shen, Y. K.; Lin, Y. Polym Adv Technol
2010, 21, 457–466.
18. Kang, M. G.; Guo, L. J. Adv Mater 2007, 1391–1396.
19. Rao, J. L.; Keeney, R. W. Opt Commun 2010, 283, 2446–2450.
20. Zhang, Y.; Lu, J.; Shimano, S.; Zhou, H.; Maeda, R. Elec-
trochem Commun 2007, 9, 1365–1368.
21. Lozano, A.; Barreras, F.; Valino, L.; Marin, C. Exp Fluids 2007,
42, 301–310.
22. Zhang, Y.; Lu, J.; Zhou, H. S.; Itoh, T.; Maeda, R. J Miroelec-
tromech Syst 2008, 17(4), 1020–1028.
23. Ishizuka, M.; Houjou, H.; Motokawa, S.; Mizuno, J.; Momma,
T.; Osaka, T.; Shoji, S. Jpn J Appl Phys 2006, 45(10A), 7944–
7948.
24. Taylor, A. D.; Lucas, B. D.; Guo, L. J.; Thompson, L. T. J Power
Sources 2007, 171, 218–223.
25. Kim, H. T.; Reshentenko, T. V.; Kweon, H. J. J Electrochem
Soc 2007, 154(10), B1034–B1040.
26. Vogelaar, L.; Barsema, J. N.; Van Rijn, C. J. M.; Nijdam, W.;
Wessling, M. Adv Mater 2003, 15(16), 1385–1389.
27. Vogelaar, L.; Lammertink, R. G. H.; Barsema, J. N.; Nijdam,
W.; Bolhuis-Versteeg, L. A. M.; Van Rijn, C. J. M.; Wessling,
M. Small 2005, 1(6), 645–655.
28. Wu, G. Z.; Xu, H. B.; Zhou, T. /Polymer 2010, 51, 3560–
3567.
29. Yu, W.; Zhou, W.; Zhou, C. X. Polymer 2010, 51, 2091–2098.
30. Pyun, A.; Bell, J. R.; Won, K. H.; Weon, B. M.; Seol, S. K.; Je, J.
H.; Macosko, C. W. Macromolecules 2007, 40, 2029–2035.
31. Galloway, J. A.; Koester, K. J.; Paasch, B. J.; Macosko, C. W.
Polymer 2004, 45, 423–428.
32. Zhang, W.; Yao, D. G.; Zhang, Q. W.; Zhou, J. G.; Leikes, P. I.
Biofabrication 2010, 2, 035006.
33. Steinmann, S.; Gronski, W.; Friedrich, C. Rheol Acta 2002, 41,
77–86.
34. Lopez-Barron, C. R.; Macosko, C. W. Langmuir 2009, 25(16),
9392–9404.
35. Yao, D. G.; Zhang, W.; Zhou, J. G. Biomacromolecules 2009,
10, 1282–1286.
36. Zhang, W.; Deodhar, S.; Yao, D. G. Ann Biomed Eng 2010,
38(6), 1954–1964.
37. Heckele, M.; Schomburg, W. K. J Micromechanics Microeng
2004, 14, R1–R14.
38. Becker, H.; Gartner, C. Electrophoresis 2000, 21, 12–26.
Advances in Polymer Technology DOI 10.1002/adv