ultrahydrophobic textiles using nanoparticles: lotus · pdf filetraditional textile wet...

Ultrahydrophobic Textiles Using Nanoparticles: Lotus Approach

Karthik Ramaratnam, Ph.D.1, Swaminatha K. Iyer, Ph.D.1, Mark K. Kinnan2, George Chumanov, Ph.D.2,

Phillip J. Brown, Ph.D.1, Igor Luzinov, Ph.D.1

1School of Materials Science and Engineering, Clemson University, Clemson, South Carolina, USA

2Department of Chemistry, Clemson University, Clemson, South Carolina, USA

Correspondence to: Phillip J. Brown, Ph.D. email: [email protected] Igor Luzinov, Ph.D. email: [email protected]

ABSTRACT It is well established that the water wettability of ma-terials is governed by both the chemical composition and the geometrical microstructure of the surface.1 Traditional textile wet processing treatments do in-deed rely fundamentally upon complete wetting out of a textile structure to achieve satisfactory perform-ance.2 However, the complexities introduced through the heterogeneous nature of the fiber surfaces, the nature of the fiber composition and the actual con-struction of the textile material create difficulties in attempting to predict the exact wettability of a par-ticular textile material. For many applications the ability of a finished fabric to exhibit water repellency (in other words low wettability) is essential2 and po-tential applications of highly water repellent textile materials include rainwear, upholstery, protective clothing, sportswear, and automobile interior fabrics. Recent research indicates that such applications may benefit from a new generation of water repellent ma-terials that make use of the “lotus effect” to provide ultrahydrophobic textile materials.3,4 Ultrahydropho-bic surfaces are typically termed as the surfaces that show a water contact angle greater than 150°C with very low contact angle hysteresis.4 In the case of tex-tile materials, the level of hydrophobicity is often determined by measuring the static water contact angle only, since it is difficult to measure the contact angle hysteresis on a textile fabric because of the high levels of roughness inherent in textile structures. INTRODUCTION Barthlott and Neinhuis revealed the interdependence between surface roughness, reduced particle adhesion and water repellency based on experimental data ob-tained on microscopically smooth and rough plants.4 It was also demonstrated that hydrophobic substrates that are rough on a micro and nanoscale tend to be more hydrophobic than smooth ones because of the

reduced contact area between the substrate and the liquid and vice versa for hydrophilic substrates. It is clear that wettability strongly depends on two properties the surface free energy and the surface roughness.6-9 Surface free energy is an intrinsic prop-erty of the material that can be controlled by chemi-cal modification, such as fluorination9 and other hy-drophobic coatings.10,11 In particular a variety of fluo-rine-based polymers have been used because of their high water and oil resistances, organic solvent resis-tance, and lubricity.12 Fluoro-polymers/coatings have been commonly used for hydrophobic applications because of their lowest surface free energy property. It has been reported that the surface free energy de-creased in the order –CH2 > -CH3 > -CF2 > -CF2H > -CF3, which predicts that the closest hexagonal pack-ing of -CF3 groups gives the lowest surface free en-ergy of the materials.13 Hence, ultrahydrophobic sur-faces have been generally prepared by modifying the surfaces with various fluorinated polymers, such as PTFE coatings,9,14 fluoroalkylsilanes,15,16 and per-fluorinated polymer monolayers.17 The focus of our research was to explore the possibility of achieving the ultrahydrophobic effect on fabrics using nanopar-ticles as the roughness initiating component and a non-fluorinated hydrophobic polymer (polystyrene) as the low surface free energy component. The effect of fabric roughness due to the different fabric con-structions on achieving the lotus effect was also stud-ied. In our fundamental approach we have employed dif-ferent methods to fabricate ultrahydrophobic textile materials. By using combinations of polystyrene grafted layers (the low surface free energy compo-nent) and silver/silica/calcium carbonate nanoparti-cles (roughness initiation component) we have cre-ated textile materials that demonstrate ultrahydro-

Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org Volume 3, Issue 4 - 2008

1

mailto:[email protected]

mailto:[email protected]

phobicity.18 The authors recognize that silver has long played a role in antimicrobial finishing products. However in this paper we did not investigate the an-timicrobial nature of the treated fabrics and the silver nanoparticles were synthesized to have the desired range of particle sizes for this studies primary con-cern, i.e., ultrahydrophobicity. However, the cost of silver is not unsubstantial and therefore cheaper par-ticle technologies were also investigated including silica and calcium carbonate. The multi-dimensional nature of roughness arises in these fabrics due to sur-face modification with nanoparticles along with fiber size and the macroscopic fabric structure. The hydro-phobic polymer then provides “the icing on the cake” to give the ultrahydrophobic fabric response. Rough-ness gives rise to the phenomena of limited impreg-nation (water penetration into the rough structures) and is greatly responsible for obtaining higher levels of repellency. Several thermodynamic models have been proposed to study the depth of water penetration onto such rough surfaces and predict the degree of repellency. De Gennes et al. explained that a liquid film deposited on a porous substrate recedes via two separate mechanisms: suction and dewetting.19 The detailed behavior of the roughness levels and the wet-ting behavior are not discussed here. To initiate this study, first the roughness of the lotus leaf (nelumbo nucifera) was studied using scanning probe microscopy. Unfortunately the SPM did not lead to consistent measurements and the study showed that the leaf surface is “very rough” in fact too rough in our case for the SPM tip to track the surface profile. The high level of roughness (mi-crons) and the non-homogeneous distribution of the bumps on these leaves meant that good imaging us-ing AFM was difficult. Despite such difficulties the RMS roughness (root mean square roughness value) could still be approximated and was found to vary considerably in different areas of the leaves (mi-crons). We considered that micron level modifica-tions may lead to deterioration of the properties of the textile fibers and fabrics and in addition mimicking nature’s non-homogenous rough surface seemed overly complex. Thus in this study attempts were made to reproduce artificial lotus surfaces by making use of submicron particles and we restricted our-selves to preparing homogeneous rough surfaces in the order of nanometer/submicron levels using poly-mer thin films and a variety of particles. MATERIALS Polyester (PET) fabrics style # 777H (Fabric 1), #703 (Fabric 2) and a PET microfiber fabric (Fabric 3) were obtained from Testfabrics Inc. Polystyrene (PS) (45900 g/mol) with monofunctional carboxy termi-

nated end groups was obtained from Polymer Source Inc. Glycidyl methacrylate (GMA) obtained from Aldrich was polymerized using free radical polym-erization to obtain poly (glycidyl methacrylate) (PGMA) (Mn- 24 000 g/mol, PDI 1.7 (obtained from GPC)). Poly (2-vinylpyridine) (PVP) was obtained from Aldrich (Mn- 37500 g/mol). Poly (styrene-b-(ethylene-co-butylene)-b-styrene) (SEBS, Kraton FG1901X) triblock copolymer was obtained from Kraton Polymers US LLC. The polymer contained ~29 wt% of styrene and 1.4 wt% of reactive maleic anhydride (MA) groups. The block copolymer was reported to have Mn=41,000 g/mol, Mw/Mn=1.16, and Rg = 6.3 nm, where Rg is the radius of gyration of SEBS macromolecules.20 The polymer was analyzed with gel permeation chromatography employing polystyrene standards. Silver nanoparticles of various sizes 52, 84, 96 and 105 nm were synthesized by procedure described by Evanoff et al.21 Silica nanoparticles of sizes about 50 and 100 nm were obtained from Nissan Chemicals. Silica Microspheres of uniform size of 150nm were obtained from Polysciences Inc. Calcium carbonate nanoparticles of different shape and size were do-nated by Solvay chemicals. ACS grade toluene, tetra-hydrofuran (THF) and methyl ethyl ketone (MEK) were obtained from VWR and used as received. Highly polished single-crystal silicon wafers of {100} orientation (Semiconductor Processing Co.) were used as a model substrate for flat surfaces. INSTRUMENTATION The polyester fabrics were initially functionalized using either a Harrick plasma cleaner/ sterilizer PDC-32G or a Sicatech corona generator (Unisystems LFI). SEM studies were performed using a FESEM- Hitachi S4800. All of the polymer coatings were per-formed using a dip coater (Mayer Fientechnik D-3400). Ellipsometry measurements were done to measure the thickness of the polymer films after coat-ing on model silicon wafer substrates using a COMPEL discrete polarization modulation automatic ellipsometer (InOmTech Inc.) at an incidence angle of 70°. Scanning probe microscopy (SPM) was used to analyze the morphologies obtained on silicon wa-fer using Dimension 3100 microscope (Digital In-struments, Inc.). Silicon tips with spring constants of 50 N/m were used to scan the surfaces. All of the imaging was done at a scan rate of 1 Hz employing tapping mode. Water contact angle (WCA) was measured using a goniometer (Kruss, Model DSA10) to determine the extent of water repellency. The fab-ric wetting properties after the surface modification were studied using a Textest FX3300 Air permeabil-ity tester, Textest FX3000 Hydrostatic pressure


2

tester, and the tensile properties using Instron (Model 1125). The determination of the fabric stiffness was performed using a cantilever-bending tester. EXPERIMENTAL Model Substrates Silicon wafers were used as model (flat) substrates to optimize the experimental conditions required to cre-ate surfaces with high repellency. These wafers were chosen as model substrates to study polymer grafting thickness and nanoparticle adsorption. After optimiz-ing treatment conditions on the model silicon wafer substrates, the same treatments were then imple-mented on three different polyester fabrics, each hav-ing an inherently different fabric roughness. Silicon Wafer Preparation Silicon wafer substrates before applying the surface modification processes were initially cleaned with de-ionized water in an ultrasonic bath for 30 minutes. The wafer was then placed in piranha solution (3:1 concentrated sulfuric acid/30% hydrogen peroxide) for approximately one hour, and then rinsed several times with de-ionized water. After rinsing, the sub-strates were dried under a stream of nitrogen in clean room 100 conditions.

PET Fabric Preparation The three polyester fabrics were initially rinsed in the following solvents (water, acetone, toluene, and etha-nol) to remove surface contaminants. Fabrics were thoroughly rinsed in de-ionized water for about 1 hour to remove water soluble residuals and dried in an oven at 80°C until they achieved constant weight. Fabrics were later subjected to an air plasma dis-charge (frequency 13.56MHz) 6.8 W power for 2 minutes at a pressure of about 200 torr. For larger sample sizes, corona treatments were used as an al-ternative for the plasma discharge due to the limited capacity of the plasma equipment. The corona treat-ment was performed at standard conditions (70°F/21°C temperature and 65% relative humidity) with the fabric swatch placed exactly 1 cm below the corona treatment head. The plasma/corona treated fabrics were then rinsed in THF to remove any low molecular weight remnants formed due to the chain scission process during the treatment. Both plasma and corona treatments add surface functionalities that increase the reactivity of the PET fiber/fabric sur-faces and they were performed to ensure that any subsequent grafting processes were successful.

SILVER NANOPARTICLES APPROACH Preparation of hydrophobic substrates using sil-ver nanoparticles approach The following procedure illustrated below (Figure 1) was performed on clean silicon wafers to optimize the experimental procedure and later the optimized procedure was applied to the PET fabrics. Step 1: The cleaned substrate was dip coated using different blend compositions of the PGMA/PVP solu-tion prepared in MEK (0.2 wt/vol %) and then an-nealed at 110°C for 10 minutes to aid self cross-linking of the epoxy groups of PGMA. Crosslinking of PGMA stabilizes the microstructure of the blend. Step 2: The annealed substrate was treated with etha-nol (good solvent for PVP) to ensure the presence of PVP on the surface. PVP is responsible for the ad-sorption of the nanoparticles providing a metal-ligand complex22 between the silver nanoparticle and PVP. The PGMA/PVP morphology formed on the silicon substrate was then studied using SPM.

P E T

P G M A /P V P

S ilve r N a n o p a rtic le

G ra fte d P S B ru s h

P G M A

P E T

P G M A /P V P

S ilve r N a n o p a rtic le

G ra fte d P S B ru s h

P G M A

FIGURE 1. Schematic representation of the lotus approach using silver nanoparticles to achieve lotus effect. Step 3: The substrate was then exposed to an aqueous suspension of silver nanoparticles under constant ultrasonication for about 3 hours. Here the effect of nanoparticle size on the wettability was studied by exposing four model substrates (simultaneously pre-pared) with four aqueous suspensions of silver nanoparticles that differed in the nanoparticle size. Each of the four aqueous suspensions contained ap-proximately 1010 particles/ml with particles sizes of 52 nm, 84 nm, 96 nm and particles greater than 105 nm respectively (Table I). Step 4: The substrate was then dip coated with a sec-ond layer of PGMA (0.1 wt. % PGMA in MEK).


3

This second layer traps silver particles in a cage be-tween the first PGMA/PVP layer and PGMA layer. The formation of a “nanoparticle sandwich” ensures nanoparticles entrapment and it is assumed that this arrangement is very robust and durable.


4

Step 5: Carboxy terminated PS (using 0.5 wt.% PS solution in toluene) was then grafted to the uncross-linked epoxy functionality of the top layer at 150ºC for approximately 4 hours via reaction of the PS car-boxylic groups with the epoxy groups of PGMA. Unreacted PS was later removed by multiple rinses in toluene. Scanning probe microscopy (SPM) analysis was per-formed on the resultant substrates after each step. The wettability characteristics of these substrates after the final step of PS grafting were studied using the static contact angle measurements. Based on the results

obtained on the samples using the above procedure the experimental conditions were optimized. Preparation of hydrophobic and ultrahydropho-bic PET fabrics using silver nanoparticles SEM images of the three different PET fabrics used in this study are shown in Figure 2. Fabric 1 and Fab-ric 2 had different fabric densities at 335 g/m2 and 122 g/m2 respectively. The third fabric (Fabric 3) is composed of PET microfibers. Four samples of each of the 3 fabrics were prepared to study the adsorption effect of each of the 4 different sized nanoparticle suspensions (as shown in Table II). Scanning electron microscopy (SEM) was used to verify nanoparticle coverage. The hydrophobicity of these fabrics after PS grafting was studied using WCA measurements.

TABLE I. Listing of the Experimental Design on Silicon Wafer Substrates

Name Step 1 Step 2 Step 3 Step 4 Step 5

Substrate 1 PGMA/PVP dip-coating Ethanol treatment 52 nm silver nanoparti-

cle suspension PGMA dipcoat-ing PS grafting





Substrate 4 PGMA/PVP dip-coating Ethanol treatment > 105 nm silver

nanoparticle suspension PGMA dipcoat-ing PS grafting

TABLE II. Listing of the Experimental Design on PET Fabrics

Name Step 1 Step 2 Step 3 Step 4 Step 5

Fabric 1 (70: 30)

PGMA/PVP dip-coating

Ethanol treat-ment

52 nm/ 84 nm/ 96 nm/ > 105 nm silver nanoparticle suspension

PGMA dipcoat-ing PS grafting

Fabric 2 (70: 30)


Ethanol treat-ment



Fabric 3 (70: 30)


Ethanol treat-ment



1 mm

(a)

500 µm

(c)

1 mm

(b)

FIGURE 2. Polyester Fabrics of different constructions. (a) Fabric 1 with 335 g/m2 density, (b) Fabric 2 with 122 g/m2 density and (c) Fabric 3 made from Polyester microfibers.


5

Dip coating in SEBS polymer solution con-taining CaCO3 nanoparticles

SILICA NANOPARTICLES APPROACH The application procedure described for the silver nanoparticles approach above was also applied for silica nanoparticles. This approach was performed to verify if the creation of ultrahydrophobic textiles us-ing nanoparticles could be prepared from other nanoparticle materials. Optimization of surface roughness on silicon wa-fer via silica nanoparticles approach Silicon wafer substrates were treated using silica mi-crospheres of size 150 nm ± 30 nm (Polysciences Inc). Different concentrations of nanoparticles sus-pensions were prepared by diluting the initial 5.2 wt % aqueous suspension as follows:- (10x, 20x, 30x, 50x, 300x, and also 1800x). This empirical optimiza-tion approach was used to evaluate which nanoparti-cle concentration would give the highest contact an-gle on the silicon wafer. This nanoparticle concentra-tion would then be applied to the three different PET fabrics. CALCIUM CARBONATE (CaCO3) NANOPARTICLES APPROACH Synthesis of ultrahydrophobic silicon wafer sub-strates/polyester fabrics using CaCO3 nanoparti-cles In an attempt to reduce the number of steps to pre-pare ultrahydrophobic fabrics as compared to the multi-step approach using silver/silica nanoparticles the following procedure was used. The substrates (silicon wafers/fabrics after initial sample prepara-tions) were dip coated under constant ultrasonication in PGMA solutions in MEK (0.2 wt/vol %) for 3 minutes, to create the macromolecular anchoring layer. The PGMA coated fabrics were annealed at 110°C for 10 minutes. The fabrics were then exposed to a suspension of CaCO3 nanoparticles (~ 100 nm to 200 nm) dispersed in 2% SEBS polymer solution prepared from toluene, under constant ultrasonication for 3 minutes. The amount of CaCO3 nanoparticles added to the SEBS solution was varied at different % concentrations by volume (30, 40, 50, 60 and 70) with respect to the SEBS polymer concentration. Variations in the % composition of CaCO3 nanoparti-cles to SEBS helped empirically optimize the nanoparticle content and identify the best combina-tion of the nanoparticle and the SEBS polymer. The presence of CaCO3 nanoparticles along with the SEBS polymer on the silicon wafer/fabric provides a rough/bumpy profile as illustrated in Figure 3. Tak-ing one further step, the substrates (Si wafter/fabrics) were rinsed in 10 wt % acetic acid (AA) to dissolve the calcium carbonate nanoparticles. This in-turn left

FIGURE 3. Schematic of the lotus approach using calcium car-bonate nanoparticles (represented by blue spheres) and SEBS polymer micro-pores (in the form of dimples) in place of the nanoparticles that had previously made the surface bumpy/rough. The porous surface was then studied as a reverse lotus template and was examined in order to verify if dimpled surface profiles are sufficient to increase water contact angle. In concept if this is found to work well, then the nanoparticles can be removed from the substrates after initial modification and as a result the may improve the handle and com-fort properties of the fabrics. Scanning probe micros-copy (SPM), scanning electron microscopy (SEM) analysis and contact angle measurements were per-formed to analyze the nanoparticle distribution and the wetting behavior of the substrates after the modi-fication. Effect of the shape of CaCO3 nanoparticles on ul-trahydrophobicity The effects of cube-like and needle-like CaCO3 nanoparticles23 as shown in Figure 4 (of different sizes) were used for the study. The nanoparticle ap-plication procedure (as explained above) was also used for CaCO3 nanoparticles. The CaCO3 nanoparti-cles were applied to silicon wafer substrates but stud-ies were not extended to fabrics. This study was only intended to investigate the effect of nanoparticle shape on surface properties. The compositions of both the cube-like and needle-like nanoparticles were varied with respect to the hydrophobic SEBS poly-mer concentration. The substrates after the modifica-tion were analyzed using SPM to study the nanopar-ticle distribution. The wettability of the substrates

Ss

Silicon Wafer

Silicon WaferSEBS Polym er

was verified using the water contact angle measure-ments.

(a) (b)

FIGURE 4. (a): cube like CaCO3 nanoparticles of size 40-70 nm and (b) needle like CaCO3 nanoparticles of size 200-350 nm. EVALUATION OF THE FABRIC PROPERTIES AFTER NANO-PARTICLE MODIFICATION It could be expected that adsorption of the nanoparti-cles onto fiber surfaces could adversely change other fabric properties. In order to estimate the deteriora-tion of some other fabric properties the following test methods were employed: air permeability test, hydro-static pressure test, strip test, stiffness test and wash tests. All tests were performed on fabrics before and after the modification processes. The initial fabrics samples prepared using the above mentioned ap-proaches were of smaller size (~ 2 x 5 cm). Clearly larger sample sizes were needed to test these bulk fabric properties. Thus new samples of size 25 x 25 cm were prepared for performing these tests. PET fabric 2 was chosen for the study and was divided into 4 samples such as, the fabric “as is” (Control 1), fabric after cleaning to remove fabric finishes before the modification (Control 2), cleaned PET fabric with polymer (SEBS) coating only (Control 3), and the PET fabric after modifying with nanoparticles and polymer coating (Nanocoated fabric). Air Permeability Test ASTM test method D-737 was used for this study at a test pressure of 125 pascal. This test was used as an objective measure of the air permeability of the fab-rics before and after the nanoparticle modification using the four samples of fabric 2 described above (control 1, 2, 3 and nanocoated fabric). The results are reported in terms of cubic feet per minute (CFM). Hydrostatic Pressure Test AATCC test method 127-1998 was used for this study. All of the four samples were tested at a gradi-ent pressure of 10 mbar/min. This dynamic test

measures the resistance of the fabrics to water pene-tration. Strip Test ASTM test method D-5035 was used for this study to analyze the change in fabric strength before and after the modification. Samples of 1” x 6” were cut from each fabric i.e., controls 1, 2, 3 and the nanocoated fabric. Tests were performed in both the warp and weft direction. Stiffness Test ASTM test method D-1388 utilizing a cantilever bending tester was used for this study to test each of the fabrics, i.e., controls 1, 2, 3 and the nanocoated fabric. Wash Test A slightly modified version of AATCC wash test method 124-1996 was used to estimate the durability of the nanoparticle treatment. AATCC standard ref-erence detergent was used as the surfactant. The modified fabrics were washed once in 0.1% detergent solution at 60°C for 2 hours. After the test, the fabrics were thoroughly washed with distilled water (numer-ous rinses) for several hours to remove all the surfac-tant and thereafter the WCA was measured on these fabrics. RESULTS AND DISCUSSION Silver Nanoparticles Approach Characterization of the hydrophobic silicon wafer substrates The thickness of the PGMA/PVP polymer layer (after step 1) was 4.0 ± 1 nm as measured by Ellipsometry24 on the model silicon wafer substrate. The second layer of PGMA had a thickness of around 2 nm and the final polystyrene layer thickness ranged between 5 to 15 nm. Figure 5 shows the SPM images after step 1 (Figure 5a), step 2 (Figure 5b) and step 3 (Figure 5c). The PGMA/PVP blend morphology of the substrates after step 2 (ethanol treatment) showed dispersed morphology due to the immiscibility of the polymers. Different blend percentages of the PGMA/PVP system were also experimented with. It was found that the silver nanoparticle adsorption was dependent on the size of the PVP domains formed in the PGMA matrix. Varying the amount of PVP in the PGMA/PVP blend also regulated the density of silver nanoparticles adsorbed. The ideal PVP inclusion size was expected to be at least 100 nm (as it approxi-mately matched the size of the largest nanoparticles used) and such domains should ensure a good cover-


6

age of the surface with the silver nanoparticles. The 70/30 blend ratio of the PGMA/PVP system was found to form PVP inclusion size of approximately 100 to 150 nm (measured from the SPM morphology in Figure 5a). Other blend ratio’s showed varied sizes and inconsistent phase separations. Hence it was decided to use this particular composition of 70/30 blend ratio for further studies. The PVP islands successfully immobilized silver particles (after over-night exposure to a suspension of silver nanoparticles in deionized water) due to the affinity of pyridyl groups to silver through metal ligand interactions of the nitrogen atoms.22 The adsorption of the silver nanoparticles (post Figure 5a) and thus coverage on the 70/30 composition can be seen in the SPM mor-phologies Figure 5b. From Figure 5c, it was also observed that the nanoparticles stayed intact after the PS grafting and after several rinses of the sample in toluene to remove the ungrafted PS polymer.


7

(a) (b) (c)

FIGURE 5. 1 X 1 µm (a) and 3 X 3 µm (b, c) SPM topography images. Silicon wafer modified with: (a) PGMA/PVP (70/30), (b): silver nanoparticles sandwiched between PVP/PGMA and PGMA layers, (c): PS grafted on the top PGMA layer. Vertical scale (a): 10 nm, (b, c): 200 nm. Polyester (PET) lotus fabrics obtained from silver nanoparticle approach Although silver is a well known antimicrobial agent the primary objective for using silver nanoparticles was to obtain the appropriate surface roughness.

Though the fabrics covered with the silver nanoparti-cles might indeed render some anti-microbial proper-ties, these studies were not undertaken. Figure 6 shows the SEM micrographs of polyester fabric 1 samples (after step 3) obtained using aqueous nanoparticle suspensions of size 52 nm, 84 nm and 96 nm. It can be observed from Figure 6, that not all 3 different sizes of the nanoparticles show complete and even surface coverage. Figure 7 shows the SEM micrographs of fabric 1 (after step 3) using nanopar-ticles with a size greater than 105 nm. From Figure 7, it can be observed that this particle size gave more even nanoparticle coverage on the fiber surface than the smaller sized particles. The distribution density of nanoparticles (size > 105 nm) was manually calculated from the SEM micro-graph in Figure 7c and was found to be in the range of 20 ± 5 nanoparticles/µm2. This calculation was made only on this fiber surface and is at best an indi-cator of particle distribution. We do not claim this distribution is the same for all fibers in the fabric as it is expected that varied nanoparticle adsorption will occur on the individual fibers and yarns that make up the fabric. Also observed from the Figures 6, 7 was that the density of silver nanoparticles adsorbed on fiber surfaces increases as the nanoparticle size in-creases and was a maximum for the nanoparticles with size greater than 105 nm. Here it should be noted that the PVP inclusion size ranged from 100 to 150 nm (as measured on the silicon wafer substrates). Similar adsorption behaviors were observed for the other fabrics i.e., Fabrics 2 and 3. However the water contact angles measured after the final modification step (step 5) was different for all of the 3 fabrics. Fabric 1 showed water contact angles (WCA) of about 150° ± 5°, whereas fabric 2 showed WCA of 160° ± 8° and fabric 3 showed WCA of about 138° ± 3°.

5 µm 5 µm 5 µm

FIGURE 6. Silver nanoparticles of different size adsorbed on PET fabric (1) at 10k magnification. (a): 52 nm, (b) 84 nm, (c) 96 nm

(c) (b) (a)


8

FIGURE 7. Silver nanoparticles of different size adsorbed on PET fabric (1) at 10k magnification. (a): 52 nm, (b) 84 mn, (c) 96 nm

The differences in the WCA values could be attrib-uted to the difference in the fiber and fabric struc-tures. On comparing SEM’s of fabrics 1 and 2 it can be seen that fabric 2 is a more open fabric (Figure 2). The openness of the yarns (in both warp and weft directions) leads to larger inter-yarn spaces. In this case it appears that a combination of ultrahydropho-bic treatments on fabrics of sufficient inter-yarn spac-ing leads either to increased treatment efficacy or a greater entrapment of air within the fabric structure (or both) to apparently increase levels of hydropho-bicity as measured by water contact angle measure-ments. From the WCA measurements, fabric 2 can be considered to show ultrahydrophobic properties as the contact angle is above the 150° mark. It is ac-knowledged that the error bars are high for these fab-rics, but we consider that such assumptions are in any case reasonable due to the fact that water contact an-gle measurements are not accurate for any textile surface that has a high degree of roughness. Such roughness is present in the form of protruding fibers and also due to the weave pattern. Fabric 1 was ob-served to be in the border line of ultrahydrophobicity while the PET microfiber fabric (3) is well below the ultrahydrophobic border. As an aside, the multilayered PS/PGMA/SILVER/ PVP/PGMA system on the PET fabric is appears to have reasonable robustness since the particles did not detach at high temperature (during PS grafting) or in toluene under ultrasonic treatments. A static contact angle analysis (Figure 8) was per-formed on the PET fabric having the PS/silver multi-layer treatment and on a PET fabric modified with only PS (no silver). The contact angle of the fabric increased from 113° ± 4° (Figure 8a) for the PET fabric with PS only to 157° ± 3° for PS/silver multi-layer system (Figure 8b). Thus the reason for the dramatic increase in the WCA of the fabric (making it to be classified as ultrahydrophobic) must be at-tributed due to the synergistic effect of the hydropho-

bicity of PS polymer and the roughness caused by the silver nanoparticles.

Silica Nanoparticles Approach Optimization of surface roughness on silicon wafer A variety of rough surfaces were created on the model silicon wafer substrate using different dilution rates of the silica nanoparticles of approximately 150 nm size. After the samples were prepared SPM was used to analyze the RMS roughness values of the surface. Figure 9 shows the SPM topography images after the nanoparticle modification at different dilu-tion rates as described in the experimental section. Using the SPM software, the RMS values of the samples were measured. The WCA was measured for all of the samples and the results were plotted against the RMS values. The data shown in Figure 10 identi-fied the RMS values that give the highest WCA. For nano-level roughness, this RMS value was chosen as the empirical optimum value. As observed in Figure 10, the optimum RMS value was found to be around 40 to 50 nm and corresponds to the 30-50 times dilu-tion range.

10 µm 5 µm 2 µm

FIGURE 8. Static water contact angle on (a): PET fabric grafted with PS only (no silver), (b): ultrahydrophobic fabric.

(a) (b)


9

PET lotus fabrics obtained from silica nanoparticle approach The three PET fabric types (335, 122 gsm fabric and also PET microfiber fabric) were then treated using the optimized procedure described above and the apparent WCA measurements after the modification are shown in Figure 11. The 335 gsm fabrics showed a WCA of 150° ± 5° suggesting that the fabric is of borderline ultrahydrophobicity, the 122 gsm fabrics

showed ultrahydrophobicity with contact angles well above 150° with a WCA at 165o. In the case of the PET microfiber fabric, the WCA was observed to be ca. 135° ± 3°. This implies that the roughness of the fabric 3 (microfiber fabric) did not in combination with the nano-rough surface modified fiber/fabric contribute sufficiently to achieve ultrahydrophobic-ity. Figure 12, shows the adsorption of the silica nanoparticles on the PET fiber surfaces on fabric 1.

FIGURE 9. SPM topographies of homogeneous rough surfaces created via silica nanoparticles approach at various dilution rates. (a) 30 x dilution (b) 50 x dilution and (c) 300 x dilution. Vertical scales are (a, b) 300 nm and (c) 100 nm

100

105

110

115

120

125

130

135

30 35 40 45 50 55

Roughness, nm

Con

tact

Ang

le, d

egre

es

FIGURE 10. WCA (Water contact angle) Vs. RMS roughness plot of silicon wafer modified with silica nanoparti-cles and SEBS polymer.


10

FIGURE 11. Water contact angles on the 3 different PET fabrics after silica nanoparticle modification(a): Fabric 1 with WCA ~ 150°, (b): Fabric 2 with WCA ~ 165° and (c) Fabric 3 with WCA ~ 135°.

FIGURE 12. Silica nanoparticles (100 nm) adsorbed on the PET fiber surface of fabric 1. (a) 5k and (b) 10k magnifications.

(b)

(c)

(a)

(b) (a)

FIGURE 13. WCA (Water contact angle) Vs. % CaCO3 composition plot on silicon wafer treated substrates.

30 40 50 60 70 80 90

110

115

120

125

130

135

140

145

150

155

Con

tact

Ang

le, d

egre

e

% C A C O 3 (V o l. % )

B e fo re A A R IN S E A te r A A R IN S E

10 µm 5 µm

Calcium Carbonate Nano-Particles Approach The CaCO3 nanoparticles approach (as described in the experimental section) uses less application steps in the fabrication process compared to our previous approaches. The thickness of the PGMA layer was found to be 2.0 ± 1 nm and the final SEBS polymer layer thickness was measured to be between 100 and 200 nm. Though this approach showed promising results, the nanoparticles did tend to form aggregates and this could then lead to undesired variability in coating consistency. Hydrophobic and Ultrahydrophobic silicon wafer substrate obtained from CaCO3 nanoparticle ap-proach Figure 13 shows the effect of % composition of the CaCO3 nanoparticles on the WCA. CaCO3 nanoparti-cles with % compositions of 60 and 70 exhibit ultra-hydrophobic water contact angles < 150° even on the model silicon wafer substrates. But after the nanopar-ticles were dissolved using 10 % acetic acid (AA), the WCA drastically dropped as observed from Fig-ure 13. This shows that the high level of roughness obtained in the form of bumps (with the presence of nanoparticles) are not replicated in the form of pores (i.e., a reverse lotus effect after dissolving the nanoparticles). It is thought that this might be due to depth distribution issues. For example if the nanopar-ticles rest mainly in the topmost part of the film and are just touching the SEBS hydrophobic layer instead of the nanoparticles being half immersed in the SEBS layer the roughness levels will not be maintained after the removal of the nanoparticles. Figure 14 shows the ultrahydrophobic WCA obtained on the silicon wafer substrate before dissolving the nanopar-ticle.

It is also important to note that these high contact angles are achieved on the modification of a virtually

perfectly flat surface, i.e., silicon wafer without using any fluorinated polymers. Obtaining ultrahydropho-bicity on a flat surface without fluoropolymers opens way for other applications such as automobile exte-rior surfaces, buildings, housing, walls, airplane sur-faces hornings and many more.

10 µm 500 µm

FIGURE 15. CaCO3 nanoparticles (~100 to 200nm) ad-sorbed on the PET fiber surface. (a) 100 and (b) 5k magni-fications.

As seen in Figure 15, though aggregation was a prob-lem using this approach the WCA obtained on all of the substrates such as wafer and also both fabrics (1 and 2) showed ultrahydrophobic properties. With aggregation of the nanoparticles on both the silicon wafer and fabric substrates, the RMS roughness val-ues increases abruptly to a sub-micron or micron level roughness, and hence the substrates show the “lotus effect.”

Effect of CaCO3 nanoparticles shape on obtaining ultrahydrophobic effect

FIGURE 16. Water droplet suspended on PET fabric of different fabric structures with CaCO3/SEBS coating. See magnifications.

The effect of shape was studied on the silicon wafer substrate. Figure 17 illustrates the SPM analysis of the model silicon substrate covered with CaCO3 nanoparticles dispersed in SEBS matrix. Formation of CaCO3 nanoparticle aggregates was observed on silicon wafer substrates. Different % compositions by volume of CaCO3 nanoparticles to SEBS were pre-pared to create numerous surface topographies. It was observed that certain volume 70% concentrations of CaCO3 nanoparticles gave increased hydrophobic

FIGURE 14. Water droplet suspended (158°) on a CaCO3 modified silicon wafer substrate.


11


12

properties relative to other concentrations applied. The cube shaped nanoparticles were, in general, su-perior to the needle shaped particles as shown in Fig-ure 18. From the results, it was observed that though the needle shaped nanoparticles have sharp asperities than the cube shaped particles they did not exhibit high WCA for any of the CaCO3 % compositions. This study limits only to CaCO3 nanoparticles and hence the shape effect does not apply to other nanoparticles like silica or silver.

FIGURE 17. SPM images of silicon wafer substrates coated with shaped CaCO3 nanoparticles dispersed in an SEBS matrix. (a) (10x10 µm) cube and (b) (5x5 µm) needle shaped particles.

Evaluation of fabric properties after the nanoparti-cle modification The main focus of this study has been to examine a number of fabric treatment methods to obtain ultra-hydrophobic textile materials. Whilst ultrahydropho-

bicity is desirable for many applications it seemed likely that other fabric properties may have changed during the treatments. The silica nanoparticles (100nm) approach was used in conjunction with fabric 2 to evaluate some basic fabric properties. In order to estimate the deteriora-tion of the fabric properties ASTM standard textile methods were used namely; air permeability test, water penetration through the fabric using the hydro-static head test, and the tensile properties of the fabric before and after fabric modification using the strip test. Air permeability test ASTM test method D-737 was used for this study. Control PET fabrics “as is” (Control 1), PET fabric after cleaning to remove fabric finishes before the modification (Control 2), cleaned PET fabric with polymer (SEBS) coating only (Control 3), and the PET fabric after modifying with nanoparticles and polymer coating (Nanocoated fabric) were tested. It was found that all the above fabrics recorded an air permeability reading of 15 cfm with no significant changes in the reading when tested repeatedly. Hydrostatic pressure test AATCC test method 127-1998 was used for this study. It was found that the fabric coated with hydro-phobic polymer only (Control 3) and the fabric after the modification with nanoparticle and polymer coat-ing (Nanocoated fabric) showed increased resistance to water penetration as compared to the other fabric samples (see Figure 19). It was interesting to find that the PET fabric with only hydrophobic polymer coating (Control 3 has a WCA of about 130°) showed increase in resistance to that of the fabric modified with both nanoparticle and hydrophobic polymer coating (nanocoated fabric has a WCA greater than 150°). This very interesting result suggests that the increases in nano-roughness (due to the presence of nanoparticles and the consequential increases in fab-ric hydrophobicity) do not necessarily reduce the resistance to pressurized fluid flow through the fab-ric. In other words, this type of ultrahydrophobic ma-terial relatively promotes the flow of water through the fabric structure compared to the simple hydro-phobic material, which showed increased resistance to the flow of water through the fabric.

(a) (b)

30 40 50 60 70 80 90110

120

130

140

150

cubic needle

Con

tact

Ang

le, d

egre

e

CaCO3 content, %

FIGURE 18. Effect of CaCO3 nanoparticle shape on the WCA of the silicon substrate


13

Strip test ASTM test method D-5035 was used for this study. Fabric strength was analyzed using the strip test. 1” x 6” samples were cut for each of the fabrics described above (control 1, 2, 3 and nanocoated fabric). Both the warp and weft direction were tested for all of the above fabric samples. There were no significant changes in the breaking strength (measure in lbf) and elongation % for both the warp and weft direction measured for all the above fabric samples. This shows that the fabric strength did not deteriorate after the nanoparticle modification. Wash test AATCC test method 124-1996 was used for this study. The wash test provided further information on the durability of the nanoparticle treatment on the PET fabric. It was found that the average contact angle reduced by about 5°. The WCA before washing was found to be around 150 to 155°, whereas after washing it varied between 145 to 150°. This showed that the nanocoating along with the hydrophobic polymer is quite robust and at least under these wash-ing conditions the fabric did not significantly change it hydrophobic nature. CONCLUSIONS For a fabric to show ultrahydrophobic properties, it is necessary to have the correct roughness component and hydrophobic components. It may be expected that by having the appropriate fabric roughness and the appropriate hydrophobic component coating a

lotus fabric with ultrahydrophobicty may be created without the need of nanoparticles. Nevertheless the addition of nano-roughness using nanoparticles gives the experimentalist another op-tion for fabric tunability. We have described a num-ber of methods to create surfaces that lead to ultrahy-drophobicity using hydrophobic polymer/nano-particle systems. In order to create a lotus fabric “the finisher” should also consider the type of nanoparticles to be used. Typically this selection process would be based on base material costs, ease of application and fabric property enhancement. Regarding the systems used in this study the silver nanoparticles approach was found to have little or no aggregation problems. The results with both the sil-ver and silica nanoparticles independently showed that fabric construction is an important parameter to consider when attempting to create lotus type fabrics. While the silica and calcium carbonate nanoparticles are relatively cheap (compared to silver) they also showed some aggregation problems. In the case of the CaCO3 nanoparticles the effect of shape and size of these CaCO3 nanoparticles was found to have an influence on hydrophobcity. From the results it ap-pears that certain shapes and sizes gave increased hydrophobicity relative to other shapes and this is an area for future work. Ideally one would like a perfect dispersion of particles, however, even though CaCO3 nanoparticles had some aggregation problems they still displayed ultrahydrophobic properties when ap-plied to the flat model substrate (silicon wafer). The implication of this is very encouraging since the sili-con wafer is virtually perfectly flat it clearly opens up application areas for ultrahydrophobic coatings out-side those of textile materials. ACKNOWLEDGEMENT The authors wish to thank the Department of Com-merce (National Textile Center) for financial support of this work. Additionally we would like to acknowl-edge support from (CAEFF), the Center for Ad-vanced Engineering Fibers and Films and COMSET the Center for Optical Materials Science and Engi-neering at Clemson University. The authors would like to thank Dr. V. Klep for the synthesis of PGMA and Solvay Chemicals for donat-ing calcium carbonate nanoparticles of different shapes and sizes.

FIGURE 19 . Resistance to water penetration vs. fabric sample (hydrostatic pressure test).

Control 1 Control 2 Control 3 Nanocoated Fabric

8

9

10

11

12

13

Res

ista

nce

to w

ater

pen

etra

tion,

mba

r

Fabric Sample


14

REFERENCES

[1] Jiang, L., R. Wang, B. Yang, T. J. Li, D. A. Tryk, A. Fujishima, K. Hashimoto, and D. B. Zhu, Pure and Applied Chemistry, Vol. 72, No. 1-2, January 2000, pp73-81.

[2] Heywood, D., Textile Finishing, Society of Dyers and Colourists, 2003, pp187.

[3] Project C04-CL06 National Textile Center Report, http://www.ntcresearch.org/pdf-rpts/Bref0605/C04-CL06-05e.pdf, June 2005.

[4] Wier, K.A., and T. J. McCarthy, Langmuir, Vol. 22, No. 6, February 2006, pp2433-2436.

[5] Feng, L., S. Li, Y. Li, H. Li, L. Zhang, J. Zhai, Y. Song, B. Liu, L. Jiang, and D. Zhu, Ad-vanced Materials, Vol. 14, No. 24, December 2002, pp1857-1860.

[6] He, B., N. A. Patankar, and J. Lee, Langmuir, Vol. 19, No. 12, May 2003, pp4999-5003.

[7] Wenzel, R.N; Industrial and Engineering Chemistry, Vol. 28, No. 8, August 1936, 988-994.

[8] Cassie, A.B.D; Discussions of the Faraday Society, No. 3, January 1948, pp11-16.

[9] McCarthy, J. T., and D. Oner, Langmuir, Vol. 16, No. 20, August 2000, pp7777-7782.

[10] Tadanaga, K., N. Katata, and T. Minami, Jour-nal of the American Ceramic Society,Vol. 80, No. 12, April 1997, pp1040-1042.

[11] Tadanaga, K., N. Katata, and T. Minami, Jour-nal of the American Ceramic Society, Vol. 80, No. 12, December 1997, pp3213-3216.

[12] Nishino, T., M. Meguro, K. Nakamae, M. Mat-sushita, and Y. Ueda, Langmuir, Vol. 15, No. 13, May 1999, pp4321-4323.

[13] Hare, E.F., E. G. Shafrin, and W.A. Zisman, Journal of Physical Chemistry, Vol. 58, No. 3, March 1954, pp236-239.

[14] Minko, S., M. Müller, M. Motornov, M. Nitschke, K. Grundke, and M. Stamm, Journal of American Chemical Society, Vol. 125, 2003, pp3896-3900.

[15] Schondelmaier, D.; S. Cramm, R. Klingeler, J. Morenzin, C. Zilkens, and W. Eberhardt, Langmuir, Vol. 18, No. 16, July 2002, pp6242-6245.

[16] Nakajima, A.; K. Abe, K. Hashimoto, and T. Watanabe, Thin Solid Films, Vol. 376, No. 1-2, November 2000, pp140-143.

[17] Jung, D-H.; I. J. Park, Y. K. Choi, S-B. Lee, H. S. Park, and J. Rühe, Langmuir, Vol. 18, No. 16, July 2002, pp6133-6139.

[18] Ramaratnam, K., K. S. Iyer, M. K. Kinnan, G. Chumanov, P. Brown, I. Luzinov, Polymer Preprints, Vol. 47, No. 1, March 2006, pp576-577.

[19] Pierre, G.G, B. W. Françoise, and D. Quéré, Capillarity and Wetting Pheno-mena, Springer, 2003, pp215.

[20] Luzinov, I., D. Julthongpiput, and V. V. Tsukruk, Macromolecules, Vol. 33, No. 20, September 2000, pp7629-7638.

[21] Evanoff, D.D., and G. Chumanov, Journal of Physical Chemistry B, Vol. 108, No. 37, Au-gust 2004, pp13948-13956.

[22] Malynych, S., I. Luzinov, and G. Chumanov, Journal of Physical Chemistry B, Vol. 106, No. 6, January 2002, pp1280-1285.

[23] http://www.solvaypcc.com “Solvay Preci-pitated Calcium Carbonate 28th July 2008

[24] Iyer, K.S. and I. Luzinov, Macromolecules, Vol. 37, No. 25, November 2004, pp 9538-9545.

AUTHORS’ ADDRESSES Karthik Ramaratnam, Ph.D.; Swaminatha K. Iyer, Ph.D; Phillip J. Brown, Ph.D.; Igor Luzinov, Ph.D. Clemson University School of Materials Science and Engineering Clemson, SC 29634 USA Mark K. Kinnan; George Chumanov, Ph.D. Clemson University Department of Chemistry Clemson, SC 29634 USA

http://www.ntcresearch.org/pdf-rpts/Bref0605/C04-CL06-05e.pdf

http://www.ntcresearch.org/pdf-rpts/Bref0605/C04-CL06-05e.pdf

http://www.solvaypcc.com/

ultrahydrophobic textiles using nanoparticles: lotus · pdf filetraditional textile wet...

Documents