Honeycomb-like polymeric films from dendritic polymers presenting reactive pendent moieties

Article abstract of DOI:10.1016/j.polymer.2014.01.032

In this paper we describe the fabrication of honeycomb-shaped polyurethane films from dendritic side-chain polymers presenting reactive pendent units. Two novel functional polyurethanes, poly(urethane-co-acylurea) (PU-PACY) and polyurethane-co-azetidine-2

Full text of DOI:10.1016/j.polymer.2014.01.032

Polymer 55 (2014) 1481e1490
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Honeycomb-like polymeric films from dendritic polymers presenting
reactive pendent moieties
,
b
d
Yu-An Su a, Wei-Fan Chen b, Tzong-Yuan Juang c , Wei-Ho Ting , Ting-Yu Liu ,
*
Chi-Fa Hsieh e, Shenghong A. Dai b, Ru-Jong Jeng a
,
**
a Institute of Polymer Science and Engineering, National Taiwan University, Taipei 106, Taiwan
b Department of Chemical Engineering, National Chung Hsing University, Taichung 402, Taiwan
c Department of Applied Chemistry, National Chiayi University, 300 Syuefu Road, Chiayi 60004, Taiwan
d Department of Materials Engineering, Ming Chi University of Technology, New Taipei City 243, Taiwan
e Chung-Shan Institute of Technology, Lungtan, Taoyuan 325, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
In this paper we describe the fabrication of honeycomb-shaped polyurethane films from dendritic side-
chain polymers presenting reactive pendent units. Two novel functional polyurethanes, poly(urethane-
co-acylurea) (PU-PACY) and polyurethane-co-azetidine-2,4-dione containing polyurethane (PU-PAZ),
were synthesized. The waxy dendrons featured focal urea/malonamide linkages favorable for hydrogen
bonding and peripheral alkyl chains. These intermolecular forces caused two functional PUs to undergo
phase separation and self-assembly. This resulted in honeycomb-like films with well-controlled surface
roughness characterized in terms of the ratio of rim width (W) to the pore size (D), i.e. W/D. The PU-PAZ
film had a higher contact angle (CA) and a lower value of W/D than did the PU-PACY analogue, due to the
presence of relatively more hydrophobic azetidine-2, 4-dione functionalities in the former film. Subse-
quent chemical modification of the PU-PAZ films through reaction with a hydrophobic poly(oxyalkylene)
amine enhanced the CA from 113 to 134ꢀ. Further physical modification through a peeling-off process
rendered the film surface with a three-dimensional (3D) rod-co-valley-structure having feature di-
mensions on a submicrometer scale. The 3D rod-co-valleyelike film exhibited superhydrophobicity with
a CA of 151ꢀ. The films also displayed excellent solvent-resistance after crosslinking with a diamine.
Through hydrophobic or hydrophilic chemical modification, we could readily manipulate the surface
properties of these honeycomb-like films with controllable surface roughnesses and reactive
functionalities.
Received 28 September 2013
Received in revised form
13 January 2014
Accepted 25 January 2014
Available online 11 February 2014
Keywords:
Dendrons
Honeycomb
Crosslinking
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
processes as etching, dip-coating, chemical vapor deposition, and
molecular self-assembly [12e24]. Molecular self-assembly is
Many superhydrophobic materials that mimic the water-
repelling surface of the lotus leaf have been reported recently [1e
11]. The preparation of these surfaces has relied mainly on two
co-operative effects: surface modification to impart low surface
energy and the introduction of particular surface morphologies.
These morphologies, either two-dimensional (2D) or three-
dimensional (3D), have been achieved by implementing such
particularly interesting because of the possibility of fabricating
highly regular structures, such as a honeycomb-like surfaces via a
breath-figure process (pioneered by Francois and co-workers)
[14,15]. In addition, variations in the chemical composition of the
films allow the rim width (W) and pore size (D) to be adjusted,
thereby manipulating the surface roughness [25e27].
Using this breath-figure technique, a variety of polymers
including functional linear polymers [28], star polymers [29e32],
amphiphilic copolymers [23,33e35], ionomer macromolecules
[36e38], polyion complexes [27,39,40], hyperbranched polymers
[41], and dendritic side-chain block copolymers [42e46], have been
employed to obtain honeycomb-like polymeric films. Apart from
that, Karthaus et al. reported that there were two methods to obtain
* Corresponding author. Tel.: þ886 5 271 7967; fax: þ886 5 271 7901.
** Corresponding author. Tel.: þ886 2 3366 5884; fax: þ886 2 3366 5237.
E-mail addresses: tyjuang@mail.ncyu.edu.tw (T.-Y. Juang), rujong@ntu.edu.tw
(R.-J. Jeng).
http://dx.doi.org/10.1016/j.polymer.2014.01.032
0032-3861/Ó 2014 Elsevier Ltd. All rights reserved.

1482
Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
the solvent-resistance honeycomb films [39], one was chemical
crosslinking via condensation of maleic anhydride and 1,8-
diaminooctane. The other was a photo-crosslinking system using
poly(vinyl cinnamate) [47]. Highly ordered honeycomb films were
prepared from photo-crosslinkable polymers. The 3-D honeycomb
structure was retained under UV irradiation. More recently, Wang
and co-workers reported another photo-crosslinked system [48].
Stable porous morphology was achieved by photo-induced cross-
linking during the breath-figure formation. The simple crosslinking
operation opens the door to facilely fabricate robust microporous
polymer films [49,50].
Our research group has taken an alternative approach to the
preparation of hydrophobic honeycomb-like films from dendron-
grafted polystyrenes and polyurethanes (PUs) through a combina-
tion of self-assembly and a breath-figure process [45,46]. These
waxy dendrons feature a focal part rich in hydrogen bonding groups
and a periphery rich in nonpolar units that undergo van der Waals
interactions. As part of our endeavor to enhance the surface prop-
erties of these films (i.e., tailored hydrophobicity/hydrophilicity and
solvent-resistance), in this study we developed two novel reactive
PU polymers: poly(urethane-co-acylurea) (PU-PACY) and poly-
urethane-co-azetidine-2,4-dione containing polyurethane (PU-
PAZ) [Scheme 1(a) and (b)]. PU-PACY presenting pendent aromatic
anhydride moieties was synthesized through carbodiimide (CDI)
chemistry [51e53], which is usedextensivelyinpeptide synthesis. In
esterification processes, the use of CDI chemistry is often hampered
by a side reaction that converts carboxylic acids to N-acylureas. On
the basis of this reaction, we developed a poly-CDI system and
subsequently added trimellitic anhydride (TMA) to obtain poly(N-
acylurea) as an intermediate for further chemical modification (e.g.,
to enhanced surface properties). In addition, we took further
advantage of the selectively reactive azetidine-2,4-dione moiety in
the generation-0.5 dendron (G-0.5) (Scheme 2); this unit reacts only
with aliphatic primary amino groups. We readily obtained the
azetidine-2.4-dioneecontaining PU-PAZ polymer through chain
extension of di-p-phenyl diisocyanate (MDI) with dendron-
containing diols such as [G-1.5]-C18-Diol and [G-0.5]-Diol
(Scheme 3). With their reactive pendent functionalities, these den-
dritic PU-PACYand PU-PAZ polymers possess side chains rich in not
only hydrogen bonding units but also pendent reactive moieties.
Thus, the surface properties of these honeycomb-like PU films could
be modified through reactions of these functionalities.
According to the literature [54e58], polymers featuring pendent
azetidine-2,4-dione or aromatic anhydride units can react rapidly
with aliphatic primary amino functional groups. Therefore, in this
study, we employed commercially available difunctional poly(-
oxyalkylene)amines (hydrophilic JeffamineÒM-1000 and hydro-
phobic JeffamineÒM-2005) as crosslinkers to impart enhanced
surface properties (tunable hydrophobicity up to the level of
superhydrophobicity and solvent-resistance) to our honeycomb-
like films. In addition, the values of W/D of the honeycomb-like
films could also be controlled through the self-assembly of these
various functional polymers. Accordingly, these materials have
potential applications in, for example, microanalysis, cell growth,
and stent tissue engineering.
2. Experimental
2.1. Materials
Dibutyltin dilaurate (DBTDL or T-12), N-(3-aminopropyl)dieth-
anolamine (APDEA), and 1,6-diaminohexane were purchased from
TCL. The poly(oxyalkylene)amines JeffamineÒM-1000 and
JeffamineÒM-2005 (abbreviated herein as M1000 and M2005,
respectively) were purchased from Huntsman. N,N-dime-
thylformamide (DMF), tetrahydrofuran (THF), dimethylsulfoxide
(DMSO), methylene di-p-phenyl diisocyanate (MDI), cyclohexane,
methanol, xylene, isobutyryl chloride, stearyl alcohol,
Scheme 1. Chemical structures of (a) PU-PACY and (b) PU-PAZ.

Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
1483
Scheme 2. Syntheses of the [G-0.5]-C18, [G-1.5]-C18, and [G-1.5]-C18-Diol dendrons.
triethylamine (TEA), 1-decanol, and diethylenetriamine (DETA),1,3-
dimethylphosphole 1-oxide (DMPO), and trimellitic anhydride
(TMA) were purchased from Aldrich and Acros. Scheme 2 displays
the chemical structures of the stearyl ester derivatives of 4-
isocyanato-4-(3,3-dimethyl-2,4-dioxo-acetidino)diphenylmethane
(IDD) grafted with dendrons of different generations. The building
block IDD and the dendrons (from [G-0.5]-C18 to [G-1.5]-C18) were
synthesized according to procedures in previous reports [55,59e
63].
64.93; H, 6.40; N, 9.88%. C23H27N3O5 requires C, 65.24; H, 6.60; N,
9.63%. FT-IR (nmax/cmꢁ1): 3305 (OH), 1857 (C]O), 1739 (C]O),
1638 [(NH)C]O(N)]. 1H NMR (400 MHz, DMSO-d6): dH 1.38 (3H,
s, CH3), 3.32 (4H, t, CH2), 3.55 (4H, t, CH2), 3.85 (2H, s, CH2), 5.06
(2H, s, OH), 7.02e7.62 (8H, m, PhH), 8.61 (1H, s, NH). FABMS: m/z
425 (Mþ)].
2.3. PU-PACY and PU-PAZ polymers [illustrated in Scheme 3(a)
and (b)]
2.2. [G-1.5]-C18-Diol dendron (Scheme 2)
(a) DBTDL (several drops) was added to a solution of [G-1.5]-
C18-Diol (0.884 g, 0.500 mmol) in dry DMF (5.00 mL).
This mixture was then added dropwise over 3 h to a so-
lution of MDI (0.250 g, 1.00 mmol) in DMF (5 mL) at 60 ꢀC.
After the addition of a catalytic amount of DMPO, heating
of the solution was continued for 8 h. At this point, TMA
(0.096 g, 0.500 mmol) was added to the mixture and then
the reaction was continued under a N2 atmosphere for
7 h. PU-PACY polymers were obtained after evaporation
of the solvent. FT-IR (nmax/cmꢁ1): 1852 (C]O), 1780 (C]
O).
(b) DBTDL (several drops) was added to a solution of [G-1.5]-
C18-Diol (0.884 g, 0.500 mmol) in dry DMF (5.00 mL). This
mixture was then added to a solution of MDI (0.250 g,
1.00 mmol) in DMF (5.00 mL). The solution was heated at
60 ꢀC for 3 h and then [G-0.5]-Diol was added under a N2
atmosphere over a period of 1 h. PU-PAZ polymers were
obtained after evaporation of the solvent. FT-IR (nmax/cmꢁ1):
1855 (C]O), 1740 (C]O).
APDEA (1.00 g, 6.16 mmol) was added dropwise to a solution of
[G-1.5]-C18 (2.47 g, 1.40 mmol) in dry THF (25.0 mL). The solution
was stirred at 75 ꢀC under a N2 atmosphere for 48 h. After evapo-
ration of the solvent, the product was purified through recrystal-
lization (MeOH) to give a light-yellow powder (2.23 g, 90.0 %)
[Found: C, 68.70; H, 8.22; N, 8.41%. C104H155N11O13 requires C,
70.67; H, 8.84; N, 8.72%. FT-IR (nmax/cmꢁ1): 3340 (OH), 1704 [(NH)
C]O(O)], 1654 [C]O(NH)]. 1H NMR (400 MHz, CDCl3): dH 0.83 (6H,
t, CH3), 1.35 [18H, C(CH3)2CONH], 3.78 (6H, PhCH2Ph), 4.0 (4H,
NHCOOeCH2), 4.29 (2H, OH), 7.03e7.68 (24H, PhH). FABMS: m/z
1768 (Mþ)].
2.2.1. [G-0.5]-Diol dendron (Scheme 3)
Diethanolamine (1.00 g, 9.50 mmol) was added to a solution of
IDD (3.80 g, 11.9 mmol) in dry THF (120 mL). The solution was
stirred at 0 ꢀC under a N2 atmosphere for 2 h. After evaporation of
the solvent, the product was purified through precipitation from
cyclohexane to give a white powder (3.50 g, 86.7%) [Found: C,

1484
Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
Scheme 3. Syntheses of (a) PU-PACY and (b) PU-PAZ.

Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
1485
2.4. All films prepared
2.5. Measurement
(a) Thin films prepared through spin-coating: z solution of PU-
PACY or PU-PAZ (20 mg/mL in CHCl3) was spin-coated
(2000 rpm) onto a silicon wafer.
1H NMR spectra of CDCl3 and DMSO-d6 solutions were recorded
using a Varian Gemini-400 FT-NMR spectrometer. IR spectra were
recorded using a PerkinElmer Spectrum One FT-IR spectrometer.
Thermal analysis was conducted under N2 using a TA Instrument
DSC2010 apparatus operated at a heating rate of 10 ꢀC/min. Ther-
mogravimetric analysis (TGA) was performed using a Seiko SSC-
5200 thermogravimetric analyzer operated at a heating rate of
10 ꢀC/min under N2. The thermal decomposition temperature (Td)
was taken to be the temperature corresponding to 5% weight loss of
the sample. Fast atom bombardment mass spectrometry (FABMS)
analysis was performed on a JEOL JMS SX/SX 102A mass spec-
trometer equipped with the standard FAB source, whose upper
limit for measuring molecular weight is 2000. Gel permeation
chromatography (GPC) was conducted using a Waters apparatus
equipped with Waters Styragel columns and a refractive index
detector (polystyrene calibration; mobile phase: THF). Scanning
electron microscopy (SEM) of the films was performed using a
field-emission scanning electron microscope (Hitachi S-5200) after
the films had been sputtered with a thin layer of gold/palladium
alloy. Average pore size (D) was measured from binary-contrasted
SEM images by using an imaging software (ImageJ [64]). The tiny
pores generated in the polymer frames were neglected. The contact
angles (CAs) of water droplets (5 mL) on the films were measured at
room temperature at 5-s intervals for 30 s using an optical contact
angle meter (Kyowa DropMaster); all the measured CAs were
constant during the 30-s period.
(b) Honeycomb-like polymer films prepared through breath-
figure processing:
A solution of PU-PACY or PU-PAZ
(10 mg/mL in chloroform) was cast onto a silicon wafer. The
solvent was evaporated at room temperature under a flow of
moist air (1 m/s). After complete evaporation of solvent, a
white polymeric film, featuring
a
honeycomb-like
morphology, was formed [illustrated in Scheme 4(a)].
(c) Hydrophobic or hydrophilic honeycomb-like PU films pre-
pared through chemical modification: A PU-PACY or PU-PAZ
honeycomb-like film was submerged for 24 h in a solution of
M2005, M1000, or 1,6-diaminohexane in MeOH at 60 ꢀC.
After complete evaporation of the solvent, a white PU film,
featuring a honeycomb-like morphology, was formed [illus-
trated in Scheme 4(b)].
(d) Rod-co-valleyelike surfaces prepared through a peeling-off
process: a rod-co-valley-like film was prepared by peeling
off the first layer of a honeycomb-like film with a sheet of
adhesive tape (Scotch Tape, 3M) [31] [illustrated in Scheme
4(c)].
3. Results and discussion
3.1. Synthesis and characterization of the dendritic polymers PU-
PACY and PU-PAZ
We used FT-IR spectroscopy to characterize the synthesized
polymers PU-PACY and PU-PAZ. Our synthesis of PU-PACY
involved the direct preparation of poly(urethane-co-carbodiimide)
(PU-CDI) from the aromatic diisocyanate MDI and [G-1.5]-C18-Diol
in the presence of DMPO catalyst and subsequent reaction of PU-
CDI with the carboxylic acid of TMA at room temperature
(Scheme 3a). After [G-1.5]-C18-Diol had been prepolymerized with
MDI, the characteristic absorption peak of the residual isocyanate
(NCO) units was evident at 2260 cmꢁ1 in the FT-IR spectrum (Fig.1).
The addition of catalytic DMPO converted the NCO groups to CDI
(N]C]N) units, with characteristic absorption peaks at 2106 and
2134 cmꢁ1. Subsequent reaction of the CDI units with the carboxylic
(COOH) functionality of TMA gave the polyacylurea, with charac-
teristic absorption peaks for the carbonyl (C]O) groups at 1715
cmꢁ1 and the anhydride (CeOeC) groups at 1778 and 1857 cmꢁ1
.
The signals for the N-acylurea and anhydride linkages in the FT-IR
spectra confirmed the successful preparation of PU-PACY based
on the CDI chemistry [51e53] in Fig. 1(a). We prepared the PU-PAZ
polymer by allowing the [G1.5]-C18-Diol-PU prepolymer to react
with G0.5-Diol. First, we reacted MDI with [G-1.5]-C18-Diol to form
the PU prepolymer (Scheme 3b); we subsequently reacted the
chain extender [G-0.5]-Diol featuring azetidine-2,4-diones termi-
nal groups with the PU prepolymer to form PU-PAZ. The disap-
pearance of the signal (
appearance of signals for the azetidine-2,4-dione units (
n
¼ 2260 cmꢁ1) for the NCO groups and the
n
¼ 1740
and 1855 cmꢁ1) in the FT-IR spectrum confirmed the successful
preparation of PU-PAZ [Fig. 1(b)]. We used GPC to determine
weight average molecular weights (Mw) for PU-PACY and PU-PAZ,
which were in the range of 17,470e42,310 g/mol with poly-
dispersities ranged from 3.74 to 3.92. Furthermore, thermal
decomposition temperatures (Td; 5% weight loss) of honeycomb-
Scheme 4. Schematic representations of (a) the formation of the surface pores of the
honeycomb-like films, (b) the chemical modification of the honeycomb-like films, and
(c) the honeycomb-like films obtained after applying the peeling-off method.

1486
Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
Fig. 2. SEM images demonstrating the pore distribution and uniformity on the sur-
faces of PU-PACY films prepared at polymer concentrations of 5, 10, and 15 mg/mL
under 80% RH and their respective W/D ratios.
interactions at high humidity. For this current study, we designed
two series of G1.5-C18 dendritic side-chainegrafted polymers with
reactive functional groups, thereby enabling us to control the sur-
face properties and manipulate the hydrophobicity of the
honeycomb-like films.
The surface properties of the films could be tailored by intro-
ducing pores at their surfaces. Previous reports have suggested that
the ratio between the values of W and D can be correlated to the CA;
specifically, a low W/D ratio generally indicates a high CA [25,26].
Accordingly, our first focus was to maximize the pore effect by
varying the concentration of the polymer solution used to form the
PU-PACY film. At 80% relative humidity (RH), we cast solutions of
PU-PACY 70/30 (where “70/30” represents the molar ratio of PU to
the dendritic polymer) of various polymer concentrations onto
silicon wafers. SEM revealed that a polymer concentration of 5 mg/
mL provided a film featuring broken, irregular pores (W/D ¼ 0.22)
distributed on its surface (Fig. 2), due to the low polymer concen-
tration failing to self-stabilize in a highly humid environment. In
comparison, at a higher concentration of 15 mg/mL, condensation
of water was enhanced, resulting in the formation of large pores
with a widened pore-to-pore distance and, thus, a high W/D ratio of
0.95 [38,39]. Thus, a low polymer concentration favored pore
Fig. 1. FT-IR spectra of (a) PU-PACY and (b) PU-PAZ.
like films were in the range of 208e238 ꢀC. The glass transition
temperatures (Tg) of PU-PAZ samples ranged from 48.2 to 71.3 ꢀC
(Table 1).
3.2. Factors influencing the surface properties of PU-PACY films
during film fabrication through the breath-figure process
Deepak et al. reported the preparation of honeycomb-like
polymeric films from a polymer presenting side chains having
high molecular weight and hydrogen bonding functionality [16].
The self-formation of the honeycomb-like structure arose from the
stabilization of water droplets through hydrogen bonding
Table 1
Physical properties of the PU-PACY and PU-PAZ honeycomb-like films.
a
b
Polymer
Mw (ꢂ03)
Mn (ꢂ103)
PDI
3.79
3.90
3.94
3.74
3.87
Td (ꢀC)
Tg (ꢀC)
PU-PACY 50/50
PU-PACY 60/40
PU-PACY 70/30
PU-PAZ 50/50
PU-PAZ 60/40
PU-PAZ 70/30
17.47
21.21
23.69
29.33
33.46
4231
4.61
5.44
6.01
7.84
8.65
208
209
227
233
235
238
n.d.c
n.d.
n.d.
71.3
60.6
482
1080
392
a
Temperature at which 5% weight loss occurred during TGA (N2; heating rate:
10 ꢀC/min).
Fig. 3. SEM images demonstrating the pore uniformity on the surfaces of PU-PACY
films prepared at a polymer concentration of 10 mg/mL under RHs of 50, 63, 80, and
93% and their respective W/D ratios.
b
c
Determined through DSC analysis under N2 (heating rate: 10 ꢀC/min).
Not detectable due to SSRR [51].

Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
1487
Fig. 4. (a) Inversely related CAs and molar percentages of the polymer PACY with respect to the W/D ratio. (b) SEM images revealing the pore uniformity on PU-PACY 50/50, 60/40,
and 70/30 films and corresponding CAs.
formation, whereas a high concentration favored the formation of a
honeycomb-like surface. Accordingly, when we tested an in-
between value of 10 mg/mL for the polymer concentration, we
successfully prepared e at 80% RH e a honeycomb-like film
featuring uniformly sized and evenly distributed pores. As ex-
pected, this film had a W/D ratio of 0.13, lower than those of 0.22
and 0.95 for the films prepared at polymer concentrations of 5 and
15 mg/mL, respectively. Thus, the W/D ratio serves as an index to
quantify the surface morphology of a honeycomb-like film, with a
low value representing a film surface presenting a uniform distri-
bution of evenly size pores.
With the PU-PACY 70/30 concentration optimized at 10 mg/mL,
we next examined the influence of the RH during film preparation on
the W/D ratio. As revealed in Fig. 3, the RH significantly affected the
pore-to-pore distance, pore size, and uniformity. For example, at a
low RH of 50%, the film surface featured scattered, nanoscale pores
with a W/D ratio of 0.58. We attribute this behavior tothe micro-sized
water droplets having difficultycondensingon the film surface under
such a comparatively low humidity. In contrast, under a relatively
moisture-saturated RH of 90%, the W/D ratio decreased to 0.43, due to
the inevitable collision and mutual condensation of water droplets
resulting in larger pores and a wider distance between them [42]. By
fine tuning the RH to 80%, we obtained a honeycomb-like film
featuring uniform pores and a W/D ratio of 0.18.
Having optimized the film forming conditions at a polymer
concentration of 10 mg/mL and an RH of 80%, we examined the
effect of the polymer composition on the pores and on the CAs of
the surfaces. Fig. 4(a) plots the CAs and percentages of PACY in the
overall composition with respect to the W/D ratios. The W/D ratios
and the CAs were inversely related over the entire range of polymer
compositions. For example, at 30 mol% of PACY, the W/D ratio of the
honeycomb-like film and the CA of the film surface were 0.96 and
Table 2
Molar compositions, W/D ratios, and CAs of the PU-PACY and PU-PAZ honeycomb-
like films.
Polymer
Composition (molar ratio)
Rim
width
Pore
size
W/D CA (ꢀ)
MDI [G-1.5]- [G-0.5]- TMA
C18-Diol Diol
W (
m
m)a D (
m
m)
PU-PACY 50/50
PU-PACY 60/40
PU-PACY 70/30
PU-PAZ 50/50
PU-PAZ 60/40
PU-PAZ 70/30
1
1
1
1
1
1
0.5
0.6
0.7
0.5
0.6
0.7
e
0.5 1.19
0.4 1.74
0.3 1.72
1.06
1.79
1.79
1.32
1.48
1.74
1.12 68
0.97 85
0.96 94
0.65 101
0.52 105
0.32 113
e
e
0.5
0.4
0.3
e
e
e
0.855
0.769
0.565
a
Rim width (W): measured along the hole center-hole center line.
Fig. 5. A comparison of the CAs of the PU-PACY and PU-PAZ systems.

1488
Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
Fig. 7. CAs of PU-PACY and PU-PAZ honeycomb-like films modified through reactions
with hydrophilic M1000 and hydrophobic M2005.
films [18]. The balance between the number of hydrophobic and
hydrophilic moieties significantly affects the surface tension be-
tween the polymer solution and water molecules. Thus, we pre-
pared the polymer PU-PAZ as a control for comparison of its
properties with those of PU-PACY. The structure of PU-PAZ features
azetidine-2,4-dione functional groups, making it relatively more
hydrophobic than PU-PACY. At the set conditions of a polymer
concentration of 10 mg/mL and an RH of 80%, we evaluated the CAs
of the respective polymer films (Table 2 and Fig. 5). In general, the
CAs of the PU-PAZ polymers were higher than those of their PU-
PACY analogues. For example, the film of PU-PACY 70/30 exhibited
a CA of approximately 94ꢀ, whereas its PAZ analogue had a higher
value of 113ꢀ. We suspect that the relatively more hydrophobic
azetidine-2,4-dione functionalities in the structure of PAZ
increased the degree of repulsion between the organic polymer
solution and the water molecules, thereby increasing the CAs.
Fig. 6. FT-IR spectra of amine-modified (a) PU-PACY, and (b) PU-PAZ honeycomb-like
films.
94ꢀ, respectively, while at 50 mol% the W/D ratio increased to 1.12,
whereas the CA decreased to 68ꢀ.
3.4. Surface modification of the honeycomb-like films through a
chemical approach
These results for the PU-PACY system are significantly different
from those we obtained in a previous study of a polystyrene (PS)-
type system [45]. During the film forming process, PS could not
effectively stabilize the water droplets, due to aggregation of the
micro-sized water droplets into large particles. In comparison, the
PU-type polymer has backbone functionalities that can partake in
additional hydrogen bonding interactions at the waterepolymer
solution interface to help stabilize the water droplets; these addi-
tional interactions enhanced the formation of small pores with a
wide pore-to-pore distance. Furthermore, the presence of the
dendron units also offered additional hydrogen bonding sites.
Because a decrease in the molar percentage of PACY implies an
increase in the content of dendron molecules, a polymer compo-
sition featuring a low molar percentage of PACY resulted in a low
W/D ratio and, inversely, a high CA. Our results are in good agree-
ment with those reported by Yabu, who demonstrated a general
trend of decreasing rim width (W) and increasing pore sizes (D)
correlating with higher CAs [25].
Tsujii et al. reported that a fractal structure generated from
hydrophobic wax crystals could strongly repel water [65]. Hosono
et al. also demonstrated that the water CA on inorganic nanopins
coated with hydrophobic surfactants could reach as high as 178ꢀ
[66]. Generally, the wettability of a surface can change drastically
depending on its chemical properties. Accordingly, we further
tailored the surface properties of our honeycomb-like PU-PACYand
PU-PAZ films through reactions with hydrophobic M2005, hydro-
philic M1000, and 1,6-diaminohexane. The FT-IR spectra in Fig. 6(a)
reveal that when the amines opened the rings of the cyclic anhy-
dride units in the dendritic PACY, the characteristic absorption
peaks of the anhydride groups originally at 1715, 1778, and
Table 3
Solvent-resistances of the crosslinked honeycomb-like films in various solvent
systems at 60 ꢀC for 24 h.
Polymer
THF
CHCl3
Acetone
DMF
PU-PACY 50/50
PU-PACY 60/40
PU-PACY 70/30
PU-PAZ 50/50
PU-PAZ 60/40
PU-PAZ 70/30
ꢁꢁ
ꢁꢁ
þꢁ
ꢁꢁ
ꢁꢁ
ꢁꢁ
ꢁꢁ
ꢁꢁ
þꢁ
ꢁꢁ
ꢁꢁ
þꢁ
ꢁꢁ
ꢁꢁ
þꢁ
ꢁꢁ
ꢁꢁ
ꢁꢁ
þꢁ
þꢁ
þþ
ꢁꢁ
þꢁ
þþ
3.3. Effect of hydrophobic/hydrophilic functionalities of PU
polymers on CAs
During the breath-figure process, the presence of an amphi-
philic copolymer can facilitate the formation of honeycomb-like
þþ: Soluble; þꢁ: partially soluble; ꢁꢁ: insoluble.

Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
1489
Fig. 8. SEM images of crosslinked PU-PAZ 50e50 before and after solvent treatment.
1850 cmꢁ1 disappeared [67]. Similarly, we used the same amines to
modify the PU-PAZ analogues. The characteristic absorption peak
of the azetidine-2,4-dione units originally at 1740 and 1855 cmꢁ1
also disappeared (Fig. 6(b)) [54]. For both PU-PACYand PU-PAZ, the
disappearance of these characteristic absorption peaks was evi-
dence for successful chemical modifications of their film surfaces.
We performed surface modifications through reactions with
M1000 and M2005 because of their similar propylene oxide/
ethylene oxide (PO/EO) backbones but different PO/EO ratios. Both
the PU-PACY and PU-PAZ films modified with the PO-rich M2005
had, in general, higher CAs than their pristine honeycomb-like films
(Fig. 7), presumably because the highly hydrophobic PO units
decreased the affinity of the film toward water. In comparison,
reactions with the EO-rich M1000 decreased in CAs because of
strong interactions between the hydrophilic EO units and water
droplets. Thus, we could readily manipulate the surface properties
of these honeycomb-like films through chemical modification with
hydrophobic/hydrophilic monoamines.
In contrast, modification with 1,6-diaminohexane did not
change the CAs of the films because of the absence of PO and EO
segments in the structure. Indeed, reactions with this difunctional
amine afforded honeycomb-like films with urea-crosslinked
structures favorable for solvent-resistance. We immersed the
crosslinked films in the polar solvents THF, CHCl3, acetone, and
DMF at 60 ꢀC for 24 h to test their solubilities; we found (Table 3)
that the film of PU-PAZ 50/50 exhibited dramatic improvements in
stability against these four solvents. SEM images of the urea-
crosslinked structure (PU-PAZ 50/50) revealed that the honey-
comb films were slightly swollen after THF and CHCl3 solvent
treatments (Fig. 8). However, the swollen effect was more pro-
nounced when treated by DMF and Acetone.
repellency of the hydrophobic PU-PAZ honeycomb-like films
could be enhanced further to superhydrophobicity through a
physical approach involving the creation of a rough, nanoscopic
architecture on the surface. This process involved peeling off the
first thin layer of the films to introduce a rod-co-valleyelike 3D
surface. The lengths and diameters of the rods ranged from 1 to 2
mm and from 250 to 300 nm, respectively. In this study, the com-
bination of rod-co-valleyelike 3D morphology and sub-
a
micrometer arrays increased the CA to 151ꢀ. Obviously, the
apparent CAs increase with surface roughness compared with a CA
value of 94ꢀ on the flat thin film of PU-PAZ. The relationship be-
tween contact angle and rough topography was analyzed via the
CassieeBaxter equation: cos q1 ¼ f1 cos q2 ꢁ f2 [68,69], where q1 and
q2 are the CAs on the flat thin film surface (spin-coating film) and
(including honeycomb-like film, hydrophobicity honeycomb-like
film or rod-co-valley-like film), respectively, f1 and f2 are the frac-
tions of liquid in respective contact with solid and air at the in-
terfaces (f1 þ f2 ¼1). After calculation, the f1,
and the f1,
honeycomb
were 0.655 and 0.328, respectively. In
hydrophobicity honeycomb
particular, the f1, rod-co-valley was dramatically decreasing to 0.135,
indicating that a much smaller portion of solid contacting with
water was found on the rod-co-valley surface.
3.5. Surface modification of the hydrophobic honeycomb-like films
through a physical approach
The pristine thin PU-PAZ film exhibited a low CA of less than
100ꢀ (Fig. 9). By employing the breath-figure process for film for-
mation, the resulting PU-PAZ honeycomb-like film featured a
porous surface exhibiting a CA of 113ꢀ. When we further chemically
modified the PU-PAZ honeycomb-like film through reaction with
the hydrophobic M2005 to decrease its surface energy, the CA
increased to 134ꢀ because of the presence of PO segments in the
backbone of the poly(oxyalkylene)amine M2005. The water-
Fig. 9. CAs and surface morphologies of chemically and physically modified PU-PAZ
honeycomb-like films.

1490
Y.-A. Su et al. / Polymer 55 (2014) 1481e1490
[23] Nurmawati MH, Renu R, Ajikumar PK, Sindhu S, Cheong FC, Sow CH, et al. Adv
4. Conclusion
Funct Mater 2006;16:2340e5.
[24] Zhao B, Zhang J, Wang X. J Mater Chem 2006;16:509e13.
[25] Yabu H, Takebayashi M, Tanaka M, Shimomura M. Langmuir 2005;21:3235e7.
[26] Wu X, Wang S. ACS Appl Mater Interfaces 2012;4:4966e75.
[27] Kon K, Brauer CN, Hidaka K, Löhmannsröben HG, Karthaus O. Langmuir
2010;26:12173e6.
[28] Wong KH, Hernandez-Guerrero M, Granville AM, Davis TP, Barner-Kowollik C,
Stenzel MH. J Porous Mater 2006;13:213e23.
[29] Connal LA, Vestberg R, Gurr PA, Hawker CJ, Qiao GG. Langmuir 2008;24:556e
62.
[30] Stenzel-Rosenbaum MH, Davis TP, Fane AG, Chen V. Angew Chem Int Ed
2001;40:3428e32.
[31] Barner-Kowollik C, Dalton H, Davis TP, Stenzel MH. Angew Chem Int Ed
2003;42:3664e8.
[32] Zhang Z, Hughes TC, Gurr PA, Blencowe A, Uddin H, Hao X, et al. Polymer
2013;54:4446e54.
[33] Yabu H, Hirai Y, Shimomura M. Langmuir 2006;22:9760e4.
[34] Hernández-Guerrero Maribel, Min E, Barner-Kowollik C, Müllera AHE,
Stenzel MH. J Mater Chem 2008;18:4718e30.
[35] Kim DG, Basavaraja C, Yamaguchi T, Huh DS. Polymer 2012;53:1347e54.
[36] Billon L, Manguian M, Pellerin V, Joubert M, Eterradossi O, Garay H. Macro-
molecules 2009;42:345e56.
[37] Yunus S, Delcorte A, Poleunis C, Bertrand P, Bolognesi A, Botta C. Adv Funct
Mater 2007;17:1079e84.
[38] Ghannam L, Manguian M, Francois J, Billon L. Soft Matter 2007;3:1492e9.
[39] Kabuto T, Hashimoto Y, Karthaus O. Adv Funct Mater 2007;17:3569e73.
[40] Yabu H, Tanaka M, Ijiro K, Shimomura M. Langmuir 2003;19:6297e300.
[41] Liu C, Gao C, Yan D. Angew Chem Int Ed 2007;46:4128e30.
[42] Cheng C, Tian Y, Shi Y, Tang R, Xi F. Macromol Rapid Commun 2005;26:1266e
72.
[43] Cheng CX, Tian Y, Shi YQ, Tang RP, Xi F. Langmuir 2005;21:6576e81.
[44] Hsieh SJ, Wang CC, Chen CY. Macromolecules 2009;42:4787e94.
[45] Ting WH, Chen CC, Dai SA, Suen SY, Yang IK, Liu YL, et al. J Mater Chem
2009;19:4819e41.
We have synthesized two functional PU polymers grafted with a
series of waxy dendrons, which feature a focal part rich in hydrogen
bonding groups and peripheral nonpolar units offering van der
Waals interactions. The PU-PACY and PU-PAZ polymers were
functionalized with pendent anhydride and azetidine-2,4-dione
units, respectively. Processing of the resulting polymers led to
their self-assembly into honeycomb-like films with controllable W/
D ratios. The presence of the azetidine-2,4-dione pendent units
caused the PU-PAZ species to be more hydrophobic than their
anhydride-presenting PU-PACY analogues. Thus, the PU-PAZ
honeycomb-like film exhibited a relatively high CA of 113ꢀ and a
low W/D ratio of 0.32. When we further modified the PU-PAZ film
with the hydrophobic poly(oxyalkylene)amine M2005, the CA
increased to 134ꢀ. A subsequent peeling-off process introduced a
submicrometer rod-co-valley-like 3D morphology onto the film
surface, further increasing the CA to 151ꢀ. We attribute the super-
hydrophobicity of this film to the combination of its hydrophobic
composition and the set of chemical and physical modifications.
Furthermore, we enhanced the solvent-resistance of the films by
crosslinking them through treatment with a diamine. Using this
approach, we could prepare superhydrophobic/superhydrophilic
honeycomb-like films with relative ease. We are currently
exploring the possible applications of these films in areas such as
microanalysis, cell growth, and stent tissue engineering.
[46] Chang CC, Juang TY, Ting WH, Lin MS, Yeh CM, Dai SA, et al. Mater Chem Phys
2011;128:157e65.
Acknowledgments
[47] Karthaus O, Hashimoto Y, Kon K, Tsuriga Y. Macromol Rapid Commun
2007;28:962e5.
We thank the Chung-Shan Institute of Technology and the Na-
tional Science Council (NSC) of Taiwan for financial support.
[48] Wang L, Maruf SH, Maniglio D, Ding Y. Polymer 2012;53:3749e55.
[49] Li L, Chen C, Li J, Zhang A, Liu X, Xu B, et al. J Mater Chem 2009;19:2789e96.
[50] Li L, Zhong Y, Li J, Chen C, Zhang A, Xu J, et al. J Mater Chem 2009;19:7222e7.
[51] Wei KL, Wu CH, Huang WH, Lin JJ, Dai SA. Macromolecules 2006;39:12e4.
[52] Chen AL, Wei KL, Jeng RJ, Lin JJ, Dai SA. Macromolecules 2011;44:46e59.
[53] Lin HL, Juang TY, Chan LH, Lee RH, Dai SA, Liu YL, et al. Polym Chem 2011;2:
685e93.
[54] Dai SA, Juang TY, Chen CP, Chang HY, Kuo WJ, Su WC, et al. J Appl Polym Sci
2007;103:3591e9.
[55] Chang HY, Chao TY, Yang CC, Dai SA, Jeng RJ. Eur Polym J 2007;43:3988e96.
[56] Kuo MC, Tung YC, Yeh CL, Chang HY, Jeng RJ, Dai SA. Macromolecules
2008;41:9637e42.
[57] Liu YL, Wu YH, Jeng RJ, Dai SA. J Colloid Interface Sci 2009;336:189e94.
[58] Shau SM, Chang CC, Lo CH, Chen YC, Juang TY, Dai SA, et al. ACS Appl Mater
Interfaces 2012;4:1897e908.
[59] Chen CP, Dai SA, Chang HL, Su WC, Jeng RJ. J Polym Sci Part A: Polym Chem
2005;43:682e8.
[60] Tsai CC, Juang TY, Dai SA, Wu TM, Su WC, Liu YL, et al. J Mater Chem 2006;16:
2056e63.
[61] Liu JK, Shau SM, Juang TY, Chang CC, Dai SA, Su WC, et al. J Appl Polym Sci
2011;120:2411e20.
[62] Juang TY, Tsai CC, Wu TM, Dai SA, Chen CP, Lin JJ, et al. Nanotechnology
2007;18:205606e13.
[63] Tsai CC, Chang CC, Yu CS, Dai SA, Wu TM, Su WC, et al. J Mater Chem 2009;19:
8484e94.
[64] Rasband WS. ImageJ. Bethesda, Maryland, USA: U. S. National Institutes of
Health; 1997e2013. http://rsb.info.nih.gov/ij/.
[65] (a) Shibuichi S, Onda T, Satoh N, Tsujii K. Langmuir 1996;12:2125e7;
(b) Yan H, Kurogi K, Mayama H, Tsujii K. Angew Chem Int Ed 2005;44:3453e6.
[66] Hosono E, Fujihara S, Honma I, Zhou H. J Am Chem Soc 2005;127:13458e9.
[67] Lin JJ, Hsu YC, Wei KL. Macromolecules 2007;40:1579e84.
[68] Zhao N, Xie Q, Kuang X, Wang S, Li Y, Lu X, et al. Adv Funct Mater 2007;17:
2739e45.
References
[1] Liu H, Szunerits S, Pisarek M, Xu W, Boukherroub R. ACS Appl Mater Interfaces
2009;1:2086e91.
[2] Guo Z, Liu W. Plant Sci 2007;172:1103e12.
[3] Gao L, McCarthy TJ. Langmuir 2006;22:2966e7.
[4] Feng L, Zhang Y, Xi J, Zhu Y, Wang N, Xia F, et al. Langmuir 2008;24:4114e9.
[5] Namavar F, Cheung CL, Sabirianov RF, Mei WN, Zeng XC, Wang G, et al. Nano
Lett 2008;8:988e96.
[6] He JH, Liu Y, Xu L, Yu JY, Sun GC. Solitons Fractals 2008;37:643e51.
[7] Geim AK, Dubonos SV, Grigorieva IV, Novoselov KS, Zhukov AA, Shapoval SY.
Nat Mater 2003;2:461e3.
[8] Yurdumakan B, Raravikar NR, Ajayanb PM, Dhinojwala A. Chem Commun;
2005. pp. 3799e801.
[9] Shi W, Zhang Y, Liu C, Wang Z, Zhang X. Langmuir 2008;24:1318e23.
[10] Brown PS, Talbot EL, Wood TJ, Bain CD, Badyal JPS. Langmuir 2012;28:13712e
9.
[11] Peng CW, Chang KC, Weng CJ, Lai MC, Hsu CH, Hsu SC, et al. Polym Chem
2013;4:926e32.
[12] Ishii D, Shimomura M. Chem Mater 2013;25:509e13.
[13] Yabu H, Hirai Y, Kojima M, Shimomura M. Chem Mater 2009;21:1787e9.
[14] Widawski G, Rawiso M, Francois B. Nature 1994;369:387e9.
[15] Pitois O, Fracois B. Colloid Polym Sci 1999;277:574e8.
[16] Deepak VD, Asha SK. J Phys Chem 2006;110:21450e9.
[17] Fracois B, Pitois O, Franqois J. Adv Mater 1995;7:1041e4.
[18] De León AS, del Campo A, Fernández-García M, Rodríguez-Hernández J,
Muñoz-Bonilla A. Langmuir 2012;28:9778e87.
[19] Wan LS, Lv J, Ke BB, Xu ZK. ACS Appl Mater Interfaces 2010;2:3759e65.
[20] Zhu Y, Sheng R, Luo T, Li H, Sun J, Chen S, et al. ACS Appl Mater Interfaces
2011;3:2487e95.
[21] Chen PC, Wan LS, Ke BB, Xu ZK. Langmuir 2011;27:12597e605.
[22] Ke BB, Wan LS, Xu ZK. Langmuir 2010;26:8946e52.
[69] Cassie A, Baxter S. Trans Faraday Soc 1944;40:546e51.