Development of a Semicontinuous Spray Process for the Production of

Article pubs.acs.org/IECR

Development of a Semicontinuous Spray Process for the Production of Superhydrophobic Coatings from Supercritical Carbon Dioxide Solutions Pontus Olin,*,† Caroline Hyll,‡,§ Louise Ovaskainen,† Marcus Ruda,∥ Oskar Schmidt,∥ Charlotta Turner,⊥ and Lars Wågberg*,†,# †

Fibre and Polymer Technology, §Industrial Metrology and Optics, Production Engineering, and #Wallenberg Wood Science Centre, KTH Royal Institute of Technology, Teknikringen 56, SE-100 44 Stockholm, Sweden ‡ Innventia AB, Drottning Kristinas väg 61, SE-114 28 Stockholm, Sweden ∥ Cellutech AB, GreenHouse Laboratories, Teknikringen 38, SE-114 28 Stockholm, Sweden ⊥ Centre for Analysis and Synthesis, Department of Chemistry, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden S Supporting Information *

ABSTRACT: Superhydrophobic surfaces have been fabricated in a continuous spray process, where an alkyl ketene dimer (AKD) wax is dissolved in supercritical carbon dioxide (scCO2) and sprayed onto the substrate. The mass of extracted AKD from scCO2 has been investigated as well as the pressure, temperature, and flow of CO2 at the steady-state spray conditions. Several different substrates such as glass, aluminum, paper, poly(ethylene terephthalate) (PET), and polytetrafluoroethylene (PTFE) have been successfully coated, and the superhydrophobic properties have been evaluated by measurement of water contact angle, water drop friction, scanning electron microscopy (SEM), and surface topography. The most efficient spray process, considering surface properties and mass of extracted AKD, is obtained at the lowest temperature investigated, 67 °C, and the highest pressure evaluated in this study, 25 MPa. We also show that the influence of preexpansion conditions (p, T) on the surface temperature at the selected spray distance (3 cm) is negligible by measurement with an infrared camera during spraying. sticky superhydrophobicity.15 In this case, the area wetted by the drop is larger, leading to higher adhesion,16,17 and the selfcleaning properties are lost. Superhydrophobic coatings and surfaces can be fabricated in several different ways, such as lithography,18 templating,19 plasma treatment,20 chemical vapor deposition,21 layer-by-layer deposition,22 colloidal self-assembly,23 sol−gel processes,24 electrospinning,25 electrospraying,26 wet chemical reaction,27 and hydrothermal reactions.28 Many of these methods involve multistep procedures and harsh conditions, require specialized reagents, are expensive, or are restricted to small, flat surfaces or specific materials.29 As a result, industrial scale fabrication of superhydrophobic coatings has not been fully realized. A method showing good potential is liquid flame spray (LFS), where TiO2 nanoparticles are deposited on the surface of paper board in a continuous roll-to-roll process.30 A drawback associated with this technique is that the substrate is subjected to a high temperature jet that can be detrimental to delicate surfaces. Supercritical carbon dioxide (scCO2) is a commonly used solvent in extraction of oils31,32 and algae,33 in separation and fractionation of paraffin waxes,34,35 as well as in cleaning processes36 due to its low critical temperature, Tc = 31.1 °C,

1. INTRODUCTION Wetting of surfaces by water drops is important in many natural processes as well as industrial applications. The phenomenon of wetting has gained renewed interest over the past decade, in large part due to what has become known as superhydrophobicity,1 a field which was to a great extent inspired by the self-cleaning properties of the lotus leaf.2 The criterion for superhydrophobic behavior is a high water contact angle,3 usually above 150°.4 Some superhydrophobic surfaces can cause water and even oil5 to roll off leaving little or no residue and carry away surface contamination,6 a self-cleaning property referred to as the lotus effect.2 These surfaces are associated with low contact angle hysteresis7 (CAH) and low roll-off angle due to low adhesion between the surface, water, and contaminants. In order to achieve sufficiently low adhesion, a surface with low surface energy and a suitable roughness structure is required8 which leads to the formation of air pockets beneath the water drop.9 This wetting mode was first described by Cassie and Baxter,10 who also formulated a simple model describing the wetting of this type of surface. Their model has been improved recently.11,12 A hierarchical roughness is also believed to improve the stability of this superhydrophobic wetting mode.13 Another group of selfcleaning surfaces are so-called super slippery surfaces,14 where contaminants are carried away on a low surface energy liquid impregnated in a porous hydrophobic substrate. Surfaces with a water contact angle greater than 150° but with high CAH and roll-off angle are referred to as possessing © 2015 American Chemical Society

Received: Revised: Accepted: Published: 1059

September 26, 2014 January 5, 2015 January 5, 2015 January 5, 2015 DOI: 10.1021/ie503798k Ind. Eng. Chem. Res. 2015, 54, 1059−1067

Article

Industrial & Engineering Chemistry Research

Figure 1. Diagram of semicontinuous RESS equipment used for spraying AKD from scCO2.

and low critical pressure, pc = 7.39 MPa. If compounds with sufficiently low surface energy are dissolved in supercritical carbon dioxide, superhydrophobic coatings can be formed. In this process, the mixture is expanded through a nozzle onto a surface via the so-called rapid expansion of supercritical solutions (RESS) technique. The rapid drop in pressure leads to supersaturation of the solution and subsequent nucleation and formation of small particles during gas expansion.37 Previously this technique has been operated in a batchwise procedure.38 In order to be able to coat larger surfaces faster and more economically, the process needs to be upscaled. There are difficulties involved in upscaling this batchwise procedure that could be mitigated by instead employing a semicontinuous spray process. Superhydrophobic paper surfaces have been fabricated using alkyl ketene dimer (AKD) in the RESS process.38 AKD is a very efficient material for forming superhydrophobic coatings using the RESS process, since it is a crystalline, hydrophobic wax showing complete miscibility with scCO2 above pressures of 20 MPa and temperatures of 50 °C39 and that can form fractal, superhydrophobic structures upon crystallization from the melt.40 In this work, we show that simultaneous extraction and spraying from a supercritical solution of AKD in scCO2 is a viable method for the continuous fabrication of superhydrophobic coatings. The effect of temperature and pressure in the spraying vessel on the superhydrophobicity and structure of the sprayed surfaces have been investigated. Such coatings can be applied to several different types of surfaces and sensitive materials.

(Ljungby, Sweden) was used as received. Substrates used for coating were silica wafers p-doped with boron (Si-mat, Kaufering, Germany), microscope glass slides (VWR International, Radnor, PA, USA), aluminum foil with a thickness of 15 μm (VWR International, Radnor, PA, USA), poly(ethylene terephthalate) (PET) film with a thickness of 100 μm (Bolloré Plastic Films Division, Quimper, France), polytetrafluoroethylene (PTFE) film with a thickness of 100 μm (Fluortek AB, Knivsta, Sweden), and grade 00H quantitative filter paper (Binzer & Munktell Filter GmbH, Battenberg, Germany). The substrates were cut into approximately 10 × 50 mm pieces and subsequently rinsed with the following sequence of solvents prior to spray coating: Milli-Q water, ethanol, and Milli-Q water. The filter paper was coated as received. 2.2. Equipment and Experimental Procedure. 2.2.1. Semicontinuous RESS Equipment. The setup for the semicontinuous rapid expansion of supercritical solutions (RESS) is schematically illustrated in Figure 1. Specially designed equipment with a variable volume cell with a capacity of 20 mL was built and used for the spraying experiments. The vessel was heated by a heating jacket (IHP International Heating Products AB, Sävedalen, Sweden), and a thermocouple was placed inside the vessel in order to monitor the internal temperature. The vessel temperature was controlled by a cascade controller (Model FGH 2000, Eurotherm, Malmö, Sweden) with a precision of ±0.1 °C. A pressure gauge (Model PGI-63B-BG400-LAQX, Swagelok, Stockholm, Sweden) with a precision of ±0.2 MPa was used to monitor the pressure inside the vessel during the steady-state experiments. CO2 was pressurized and delivered to the vessel by a syringe pump (Model 260D, ISCO, Teledyne, Lincoln, USA) which was cooled to approximately 5 °C by a water bath. The pump had a capacity of 103 mL of CO2 at 5 °C and 6 MPa. Approximately 5 m of stainless steel tubing was coiled and placed in an oil bath in order to heat the CO2 before it entered the vessel. A piece of stainless steel tubing (Scantec Lab, Sävedalen, Sweden) with a length of 5 cm and an inner diameter of 0.12 mm was used as a nozzle. The nozzle was heated with a custom-built heater that consists of a cylindrical piece of steel with a bore through hole to fit the nozzle and additional holes to fit a thermocouple and three heat cartridges. The heating of the nozzle was regulated by a heat controller (Model ATR 243, PIXSYS Electronics, Venice, Italy). The outlet tubing from the vessel to the valve

2. MATERIALS AND METHODS 2.1. Materials. The wax alkyl ketene dimer (AKD) was supplied by EKA Chemicals (Bohus, Sweden) in the form of pellets with a mean particle diameter of approximately 4.5 mm. Since the wax is used in the paper industry as a sizing agent, the wax contains some unknown additives, and the AKD was therefore purified prior to use by dissolving the pellets in warm ethanol. The solution contained a yellow emulsion that was manually removed. The AKD was recrystallized at room temperature and the crystals were filtered and washed with ethanol. This cleaning step was repeated twice, and the white crystals were dried in a vacuum overnight. Ultrapure carbon dioxide (≥99.995%) purchased from Strandmö llen AB 1060

DOI: 10.1021/ie503798k Ind. Eng. Chem. Res. 2015, 54, 1059−1067

Article


coverage was determined gravimetrically for one surface of each substrate by weighing on a Sartorius BP301 analytical balance with a resolution of 0.1 mg. 2.2.5. Examination of Surface Temperature during Spraying. In order to measure the temperature of the substrate during spraying, a high resolution infrared camera, FLIR SC6000 MWIR (FLIR Systems Inc., Wilsonville, OR, USA), was used. A filter paper was placed 3 cm from the spray nozzle, and the camera was mounted a distance of 30 cm from the rear side of the filter paper. The temperature of a 7 cm2 circular area centered at the nozzle exit of the sample was measured during 60 s of spraying. The same spray parameters were used as for the solubility measurements: 65 and 85 °C at pressures of 20 and 25 MPa. 2.3. Characterization Methods. 2.3.1. Differential Scanning Calorimetry (DSC). A differential scanning calorimeter (DSC 820, Mettler-Toledo) was used to determine the melt temperature of AKD and to establish that most of the impurities had been removed by the cleaning step in ethanol. The experiments were run at 10 °C/min, and a N2 flow rate of 50 mL/min was used. The experiments were performed at temperature ranges between −70 and 100 °C. Three temperature cycles were run, and the melt temperature was determined in the third cycle. Approximately 5 mg of unpurified and purified AKD was used. 2.3.2. Contact Angle Measurements (CAM). Static water contact angles (θ*) were determined for uncoated and coated samples using a CAM200 contact angle meter (KSV instruments Ltd., Helsinki, Finland) with an automatic dispenser. A droplet of Milli-Q deionized water with a volume of 5 μL was placed on the sample, and the static contact angle was determined. Ten images were recorded by the CCD camera of the instrument. The images were processed using the CAM200 software, applying the Young−Laplace fitting method to calculate the mean contact angle from 10 images. Contact angles were determined at five positions on each surface, and an average value was calculated. 2.3.3. Water Drop Friction Measurements. The superhydrophobic sliding resistance was measured on coated silica surfaces according to a previously developed method.41 In brief, water drops with volumes ranging from 3 to 40 μL were allowed to slide down coated surfaces with an inclination of 10°. The acceleration was measured using high-speed video, and the superhydrophobic sliding resistance, bsh, was calculated as a function of the square root of the Bond number according to

before the nozzle had the same inner diameter as the nozzle (0.12 mm) to maintain a constant pressure inside the vessel during spraying. All tubing before the vessel had an inner diameter of 0.5 mm. 2.2.2. Extraction of AKD with the Semicontinuous RESS Setup. Approximately 5 g of AKD was placed in the vessel, and the vessel was heated to the desired temperature. The vessel was placed in an upright position so that the bottom of the vessel contained melted AKD and the outlet for the nozzle was at the top of the vessel. The oil bath was heated to a temperature approximately 15 °C above the cell temperature to ensure efficient heat transfer. Pressurized CO2 was subsequently delivered to the vessel and bubbled up through the melted AKD. All the valves in the setup were opened so that the CO2 could pass directly through the entire system. A plastic bottle filled with glass wool was placed directly in front of the nozzle to collect the AKD. The spraying continued until the pump with CO2 was empty, and the amount of sprayed AKD was measured gravimetrically by weighing on a Sartorius BP301 analytical balance with a resolution of 0.1 mg. The pump was filled again and pressurized with CO2, and the spraying of AKD continued until the pump was emptied. This cycle was repeated at least five times at each temperature and pressure. The mass flow rate was obtained by dividing the collected mass of AKD by the spraying time and the volume flow of CO2 from the pump. We chose to normalize with the flow from the pump since the accuracy of the measured flow was higher compared to the measurements of the pressure and temperature in the vessel. The temperature and pressures investigated ranged from 66 to 87 °C and from 20 to 25 MPa, respectively. These conditions were chosen in order to compare the results with a previous study.39 2.2.3. Examination of Steady-State Parameters for the Semicontinuous RESS Setup. The time required to reach a steady state for the semicontinuous process was investigated in order to be able to spray at constant temperatures and pressures. It is only the nozzle that restricts the flow of scCO2 since all the valves in the system are opened during spraying. There is also a pressure drop when the CO2 enters the vessel because of the void volume in the vessel. The same procedure as for the solubility experiments was used, except that only the highest and lowest cell temperatures were examined, the AKD was not collected during spraying, and at least three spray cycles were performed. The pressure during spraying was monitored both for the pump and from a pressure gauge connected to the vessel. The cell temperature and the flow of CO2 from the pump were also monitored. 2.2.4. Spraying of AKD with the Semicontinuous RESS Setup. The same experimental procedure was used as for the examination of the steady-state parameters. The silica surfaces were sprayed using vessel temperatures of 66 °C (low temperature) and 82 °C (high temperature) and pressures of 20 MPa (low pressure) and 25 MPa (high pressure). The aluminum foil, filter paper, PET film, and PTFE film were all sprayed at 66 °C and 25 MPa since the extracted amount of AKD was expected to be highest at these conditions. The distance between the nozzle and the substrate surface was held constant at 3 cm. The spray coating started when the pressure was constant as determined from the steady-state experiments, and the spraying continued for approximately 2 min. The samples were simultaneously moved in a direction perpendicular to the nozzle at a speed of approximately 8 cm/s. The samples were allowed to pass the nozzle four times. The surface

bsh =

⎛ s̈⎞ Bo ⎜sin β − ⎟ g⎠ ⎝

(1)

where Bo is the dimensionless Bond number defined as Bo = ρgL2/γ, where ρ is the density of water, g is the gravitational acceleration constant, L is the drop radius, γ is the surface tension of water in air, β is the surface inclination, and s̈ is the vector length of the drop acceleration. 2.3.4. Scanning Electron Microscopy (SEM). A field emission scanning electron microscope (FE-SEM; Hitachi S4800, Tokyo, Japan) was used to study the morphology of the sprayed silica surfaces. In order to improve the image quality, the samples were coated with a 5 nm thick Pt/Pd layer using an agar high resolution sputter coater (Cressington 208 HR, Cressington Scientific Instruments Ltd., Watford, U.K.). 1061


Article


between 30 and 50 °C are lacking. The difference in the DSC curves indicates that most of the unknown impurities in the AKD were removed by the cleaning operation. 3.2. Mass of Extracted AKD from scCO2. The solubility of AKD is highly dependent on the pressure and temperature of the scCO2; the mass of extracted AKD was greatest at the highest pressure studied, p = 25 MPa, and lowest temperature used, T = 67 ± 0.6 °C, with a value of 4.5 ± 1.1 g/L. The mass of extracted AKD decreased with increasing temperature: 4.1 ± 0.5 g/L at 71 ± 1.1 °C, 3.7 ± 0.5 g/L at 77 ± 1.7 °C, 2.4 ± 0.2 g/L at 81 ± 2.1 °C, and 2.6 ± 0.6 g/L at 87 ± 0.3 °C. The trend is similar at the lower pressure of 20 MPa, with values of 0.9 ± 0.2 g/L at 82 ± 2.0 °C and 1.4 ± 0.1 g/L at 66 ± 0.6 °C. The extracted amount is shown as a function of temperature at the two different pressures in Figure 3, where the error bars

2.3.5. Topographic Measurements. Height maps and the surface roughness of coated silica surfaces were evaluated using a Microprof 200 (FRTFries Research & Technology GmbH, Bergisch Gladbach, Germany) noncontact optical profilometer. The instrument is based on confocal microscopy with chromatic aberration: White light is focused on the surface by a measuring head with a strongly wavelength-dependent focal length. The spectrum of the light scattered from the surface generates a peak in the spectrometer. The wavelength of this peak is used to determine the height on the sample. The inplane resolution is 2 μm and the height resolution is approximately 3 nm with a working range of ±150 μm. Contact methods with higher resolution, such as atomic force microscopy (AFM), could not be used due to the limited working range of these instruments. Areas of 500 × 500 μm were analyzed with a sampling distance of 1 μm in the x and y directions in order to obtain topographic maps. One measurement was made on each sample type. The surface roughness (root-mean-squared (RMS) roughness, Sq) was calculated as Sq =

1 n

n

∑ zi 2 i=1

(2)

where zi is the surface height relative to the mean plane at the point i. An area of 200 × 200 μm was analyzed with a sampling distance of 1 μm in the x and y directions; a smaller area than for the topographic maps was analyzed due to the shorter scanning time. Three measurements were performed on each sample type, and the mean value was calculated.

3. RESULTS AND DISCUSSION 3.1. AKD Purification. The AKD pellets were examined with DSC before they were purified in ethanol, and as shown in Figure 2, the pellets start to melt at around 30 °C. The sharp peak shows that AKD has a melting point at 60.7 °C. The purified AKD wax starts to melt at 55 °C, the melting peak is at 62.9 °C, and the low melting point fractions that start melting

Figure 3. Extracted mass of AKD at cell pressures of 20 MPa (filled circles) and 25 MPa (filled squares) as a function of cell temperature. The error bars represent one standard deviation for the temperature in the x direction and the extracted amount of AKD in the y direction.

in the x and y directions show one standard deviation for the measured temperature and extracted mass of AKD, respectively. The extracted amount correlates with the density of scCO2, since the density increases with increasing pressure and decreasing temperature. This is in agreement with a previous solubility study showing that the solubility of AKD is proportional to the density of scCO2.39 This is expected since scCO2 behaves like a hydrocarbon solvent with very low polarizability,42 probably due to the lack of dipole moment. Dividing the extracted mass of AKD at 71 °C and 25 MPa with the approximate density of CO2, ρCO2 = 737 g/L, the calculated value of 5.6 mg of AKD/g of CO2 is higher than the value of 2.7 mg of AKD/g of CO2 previously obtained by cloud point measurements.39 Since the solubility measured by the cloud point method is considered an equilibrium value, we assume that small drops of undissolved AKD are also sprayed from the vessel, explaining the higher amount of extracted AKD by the semicontinuous method compared to the cloud point method. There are also errors associated with solubility measurements in supercritical fluids: synthetic methods commonly underestimate the solubility, while analytical ones may overestimate the solubility. In order to obtain the highest possible yield, it is consequently advantageous to spray at as high a pressure as possible, while simultaneously maintaining a moderate temperature. It is important to note, however, that the AKD must be

Figure 2. Melting point determination of unpurified AKD (solid line) and purified AKD (dashed line) by differential scanning calorimetry (DSC). 1062


Article

Industrial & Engineering Chemistry Research in the molten state during spraying; preliminary tests on AKD powder at a temperature of 40 °C resulted in rapid clogging of the tubing. 3.3. Steady-State Parameters for the Continuous RESS Setup. The pump flow, pump pressure, pressure of CO2, and temperature of CO2 are shown in Figure 4 as functions of spray time for the low temperature, T = 67 °C, and high pressure, p = 25 MPa, conditions. At t = 0, the pump flow is 0 mL/min; CO2

is then pressurized to 25 MPa (pump pressure) and the pressure and temperature of the CO2 are then ∼0.1 MPa and ∼70 °C, respectively. When the valve leading into the pressure vessel is opened and CO2 passes through the system, the pump flow reaches its maximum value of ∼50 mL/min within 5 s. The flow starts to decrease after approximately 10 s and levels off to a flow of ∼15 mL/min after 1 min of spraying. It remains stable until the pump is emptied after ∼4 min. The pump pressure drops from 25 to ∼18 MPa within the first 5 s, but it subsequently starts to increase and after approximately 10 s it has returned to 20 MPa. After 60 s the pump pressure is back at the starting value of 25 MPa. The pressure of the CO2 increases monotonically with time; after 10 s it is ∼10 MPa and after 60 s it has reached 25 MPa. Both pump pressure and pressure of the CO2 remain stable at 25 MPa until the pump is empty. The temperature of the CO2 shows a more complicated behavior: during the first 10 s the temperature decreases to approximately 65 °C, between 10 and 30 s it rises to ∼71 °C, and it subsequently decreases to a stable value of ∼67 °C after 90 s. The temperature then remains stable until the pump is empty. The initial temperature decrease is probably due to the expansion of high-pressure CO2 entering the cell. The gas inside the vessel is compressed by the entering CO2, and this leads to less CO2 expansion and the drop in temperature is halted. Simultaneously, the gas inside the vessel is compressed, and this leads to an increase in temperature between 10 and 30 s of spraying. After approximately 60 s, the pressure of CO2 is roughly equal to the pump pressure, leading to a stable temperature. The pump flow, pump pressure, pressure of CO2, and temperature of CO2 for the other spray parameters, i.e., high temperature and high pressure, low temperature and low pressure, and high temperature and low pressure, all show a similar behavior and can be seen in the Supporting Information (Figures S1−S3). The CO2 temperatures and pump flows at steady state for the different spray parameters are summarized in Table 1. The stability of the flow, pressure of CO2, and temperature of CO2 after ∼90 s of continuous spraying indicates that the process has reached a steady state. In theory, the spray could therefore continue until all the AKD in the cell is extracted, provided that a steady stream of scCO2 is pumped through the vessel. This can be accomplished using an air-driven booster pump or by connecting an additional syringe pump in parallel to the existing pump. One pump would then be able to refill and pressurize CO2 while the other is pumping CO2 through the vessel. 3.4. Spraying of AKD and Surface Evaluation. Static contact angles, θ*, for water on the silica surfaces sprayed under different process conditions are shown in Table 1. All the surfaces exhibited contact angles greater than 150°, indicating superhydrophobicity. In Table 2, the static contact angles for water on uncoated samples and samples spray-coated at a pressure of 25 MPa and a temperature of ∼67 °C with a similar degree of coverage, between 2 and 3 g/m2, can be seen. The contact angles for water on the uncoated samples range from extremely hydrophilic filter paper where the drop is rapidly absorbed, 15 ± 2.5° for glass, 71 ± 1.9° for Al foil, and 74 ± 1.7° for PET film to the hydrophobic PTFE film with a water contact angle of 107 ± 4.6°. The coated samples all exhibit contact angles greater than 160°, indicating that the coating process has rendered them all superhydrophobic.

Figure 4. (a) Pump flow, (b) pump pressure (filled squares) and pressure of CO2 (filled circles), and (c) temperature of CO2 plotted against time for spraying at high pressure (25 MPa) and low temperature (67 °C). The valves are opened at t = 0 s, the lines are added as guides to the eye, and the error bars indicate one standard deviation. 1063


Article

Industrial & Engineering Chemistry Research Table 1. Steady-State Process Parameters and Surface Properties for Coated Silica Surfacesa press. [MPa] 20 20 25 25 a

temp [°C]

flow [mL/min]

± ± ± ±

10.2 ± 0.35 8.8 ± 0.31 12.2 ± 0.45 10.5 ± 0.15

66 82 67 82

0.6 2.0 0.4 0.7

Sq [μm] 5.9 3.1 4.1 3.9

± ± ± ±

1.07 0.15 1.39 0.64

θ* [deg] 159 151 158 159

± ± ± ±

2.4 6.2 2.5 1.9

bsh 0.027 0.037 0.029 0.038

± ± ± ±

0.002 0.008 0.006 0.009

The errors represent one standard deviation.

containing AKD. The temperature and pressure were then increased to above the critical point of CO2, after which the vessel was closed and allowed to equilibrate at constant temperature and pressure for at least 1 h. The nozzle was subsequently opened and the solution sprayed onto the sample to be coated. This batchwise procedure involves a downtime of at least 1 h between each spray cycle. Our proposed method is semicontinuous with a downtime of a couple of minutes between spray cycles, depending on the refill time for the pump. In the present semicontinuous spray setup approximately 4.5 mg of AKD/mL of CO2 was extracted at the most favorable process parameters (p ∼ 25 MPa and T ∼ 66 °C). Multiplication of this value with the steady-state flow of ∼15 mL/min yields a mass flow of ∼68 mg of AKD/min. We assumed that a minimum coverage of 2 g of AKD/m2 was needed for superhydrophobicity; provided that all AKD was collected on the surface, an area of 0.034 m2/min could theoretically be covered in the present setup. 3.5. Temperature of the Surface Substrates during Spraying. The surface temperature of the substrate during 1 min of spraying at 25 MPa and 71 °C is shown in Figure 6. The maximum temperature measured on the substrate was approximately 32 °C, the minimum temperature was around 28 °C, and the mean temperature of the entire spray area was 30 °C. The surface temperatures for the other spray conditions can be seen in the Supporting Information (Figures S4−S12). These values are also compiled in Table 3, where the lowest measured substrate temperature was 28 °C for a spray pressure of 25 MPa and temperature of 71 °C and the highest substrate temperature was 36 °C for a spray pressure of 20 MPa and temperature of 87 °C. The temperature of the substrate did not increase by more than approximately 15 °C from ambient temperature, showing that the effect of the spray on the substrate is small. In a previous study, the temperature of the center of the jet cone was found to be below −50 °C.44 This discrepancy is presumably due to the difference in nozzle design. In this study the expansion of CO2 occurs to a large extent inside the nozzle, where it is simultaneously heated by the nozzle walls. In order to show that the coating process is able to coat delicate surfaces, a coated sugar cube is shown in Figure 7. A water drop with a volume of approximately 10 μL resting on the surface indicates that the originally highly hydrophilic surface has indeed been rendered superhydrophobic. The drop remained stable on the coating for several minutes without affecting the underlying substrate; it subsequently slid off the surface at a tilt angle below 5° without leaving any residue.

Table 2. Surface Properties for Uncoated and Coated Glass, Paper, Aluminum, PET and PTFE Surfacesa θ* [deg] substrate

untreated

glass paper aluminum PET PTFE

15 ± 2.5 − 71 ± 1.9 74 ± 1.7 107 ± 4.6

RESS treated

surf. coverage [g/m2]

± ± ± ± ±

2.0 2.7 2.5 2.7 2.1

164 163 161 166 160

1.6 2.9 2.4 2.0 2.0

a

The value of the contact angle for the uncoated paper surface was not possible to measure since the drop was rapidly absorbed. The errors represent one standard deviation.

The values for the superhydrophobic sliding resistance, bsh, for the coated silica surfaces can be seen in Table 1. The surfaces sprayed at the lower temperature of ∼66 °C show bsh values of 0.029 ± 0.006 and 0.027 ± 0.002 for pressures of 20 and 25 MPa, respectively. The samples sprayed at a higher temperature of ∼82 °C show bsh values of 0.037 ± 0.008 and 0.038 ± 0.009 for pressures of 20 and 25 MPa, respectively. The bsh values for all surfaces are within experimental error, indicating that the superhydrophobic properties were not significantly affected by the investigated spraying temperatures and spraying pressures. For comparison, the bsh values for the surface sprayed at the lower temperature are similar to those for surfaces obtained by crystallizing AKD from a 10 wt % heptane solution (bsh = 0.028 ± 0.005). All the bsh values are lower than the values obtained by a batchwise spraying procedure (bsh = 0.073 ± 0.008) but somewhat higher than the bsh values measured for fresh lotus leaves (bsh = 0.012 ± 0.005) in a previous investigation.41 FE-SEM micrographs are presented in the insets of Figure 5. The randomly oriented, flaky AKD crystals can clearly be seen in all the samples. The crystals on the surfaces sprayed at a higher pressure appear to be smaller than the crystals on the surfaces sprayed at a lower pressure. The surface sprayed at a low temperature and high pressure shows smaller crystals than the other surfaces, presumably due to the faster rate of nucleation at higher pressure and lower temperature, as has been shown for ibuprofen in scCO2.43 The RMS roughness values, Sq, for the coated silica surfaces are compiled in Table 1, and height maps can be seen in Figure 5. At the lower temperature, the roughness values are Sq = 5.9 ± 1.07 μm and Sq = 4.1 ± 1.39 μm for p = 20 MPa and p = 25 MPa, respectively. At the higher temperature, the roughness values are Sq = 3.1 ± 0.15 μm and Sq = 3.9 ± 0.64 μm, respectively. The roughness values are within experimental error, indicating that the microscale roughness was not significantly affected by the investigated spraying temperatures and spraying pressures. Previously, the most common method for spraying scCO2/ AKD was in brief as follows:38 Liquid CO2 at a pressure of 6 MPa and a temperature of 4 °C was supplied to a closed vessel

4. CONCLUSIONS AND FUTURE WORK We have shown that it is possible to fabricate superhydrophobic coatings by semicontinuous spraying of an AKD/scCO2 mixture using a modified RESS procedure. This would make it possible to apply a coating on larger surface areas owing to 1064


Article


Figure 6. Maximum surface temperature (triangles), average surface temperature (circles), and minimum surface temperature (squares) as functions of time during one spray cycle with a pressure of 25 MPa and a temperature of 71 °C.

Table 3. Surface Temperature as a Function of Cell Temperature and Cell Pressure after 1 min of Spraying 20 MPa

25 MPa

cell T [°C]

max [°C]

mean [°C]

min [°C]

max [°C]

mean [°C]

min [°C]

67 71 77 81 87

34 34 35 35 36

31 31 32 32 32

29 29 30 30 30

32 32 32 33 34

30 30 30 31 31

28 29 29 29 29

Figure 7. A 10 μL water drop resting on an AKD coated sugar cube.

the steady-state operation of the spray compared to the batchwise procedure. It would subsequently be suitable for upscaling. The process can be used to coat a multitude of surfaces including, but probably not limited to, glass, paper, aluminum, PET, and PTFE. Delicate surfaces can be successfully coated since the process does not influence the temperature of the substrate to any great extent and no liquid solvents are used that may remain on the surface. We have shown that, under these conditions, the superhydrophobic properties were not significantly affected by the spraying conditions investigated. The amount of extracted AKD was greater at lower temperature and higher pressure. Future work will focus on scaling up the process by using a continuous CO2 pump or connecting an additional pump in parallel to the existing pump. The possibility of exchanging the relatively small pressure vessel with a larger pressure vessel with a means of continuously supplying molten AKD will also be investigated. We also aim to exchange the spray nozzle for an oval flat jet nozzle in order to further improve the homogeneity

Figure 5. Topographic maps for surfaces sprayed at (a) low pressure and low temperature, (b) low pressure and high temperature, (c) high pressure and low temperature, and (d) high pressure and high temperature. Insets show SEM micrographs of the corresponding surfaces; the length of the scale bar is 5 μm. 1065


Article


(12) Gao, L.; McCarthy, T. J. How Wenzel and Cassie Were Wrong. Langmuir 2007, 23, 3762. (13) Jung, Y. C.; Bhushan, B. Contact Angle, Adhesion and Friction Properties of Micro- and Nanopatterned Polymers for Superhydrophobicity. Nanotechnology 2006, 17, 4970. (14) Lafuma, A.; Quéré, D. Slippery Pre-suffused Surfaces. EPL 2011, 96, 56001. (15) Balu, B.; Breedveld, V.; Hess, D. W. Fabrication of “Roll-off” and “Sticky” Superhydrophobic Cellulose Surfaces via Plasma Processing. Langmuir 2008, 24, 4785. (16) Bhushan, B.; Her, E. K. Fabrication of Superhydrophobic Surfaces with High and Low Adhesion Inspired from Rose Petal. Langmuir 2010, 26, 8207. (17) Yan, Y. Y.; Gao, N.; Barthlott, W. Mimicking Natural Superhydrophobic Surfaces and Grasping the Wetting Process: A review on Recent Progress in Preparing Superhydrophobic Surfaces. Adv. Colloid Interface Sci. 2011, 169, 80. (18) Kwon, Y.; Patankar, N. A.; Choi, J.; Lee, J. Design of Surface Hierarchy for Extreme Hydrophobicity. Langmuir 2009, 25, 6129. (19) Shi, F.; Liu, Z.; Wu, G.-L.; Zhang, M.; Chen, H.; Wang, Z. Q.; Zhang, X.; Willner, I. Surface Imprinting in Layer-by-Layer Nanostructured Films. Adv. Funct. Mater. 2007, 17, 1821. (20) Cho, S. C.; Hong, Y. C.; Uhm, H. S. Hydrophobic Coating of Carbon Nanotubes by CH4 Glow Plasma at Low Pressure, and their Resulting Wettability. J. Mater. Chem. 2007, 17, 232. (21) Zimmermann, J.; Reifler, F. A.; Fortunato, G.; Gerhardt, L.-C.; Seeger, S. A Simple, One-Step Approach to Durable and Robust Superhydrophobic Textiles. Adv. Funct. Mater. 2008, 18, 3662. (22) Xue, C.-H.; Jia, S.-T.; Zhang, J.; Tian, L.-Q. Superhydrophobic Surfaces on Cotton Textiles by Complex Coating of Silica Nanoparticles and Hydrophobization. Thin Solid Films 2009, 517, 4593. (23) Zhang, G.; Wang, D.; Gu, Z.-Z.; Möhwald, H. Fabrication of Superhydrophobic Surfaces from Binary Colloidal Assembly. Langmuir 2005, 21, 9143. (24) Latthe, S. S.; Imai, H.; Ganesan, V.; Rao, A. V. Superhydrophobic Silica Films by Sol−Gel Co-Precursor Method. Appl. Surf. Sci. 2009, 256, 217. (25) Han, D.; Steckl, A. J. Superhydrophobic and Oleophobic Fibers by Coaxial Electrospinning. Langmuir 2009, 25, 9454. (26) Burkarter, E.; Saul, C. K.; Thomazi, F.; Cruz, N. C.; Roman, L. S.; Schreiner, W. H. Superhydrophobic Electrosprayed PTFE. Surf. Coat. Technol. 2007, 202, 194. (27) Guo, Z.; Fang, J.; Wang, L.; Liu, W. Fabrication of Superhydrophobic Copper by Wet Chemical Reaction. Thin Solid Films 2007, 515, 7190. (28) Liu, X.; He, J. One-Step Hydrothermal Creation of Hierarchical Microstructures toward Superhydrophilic and Superhydrophobic Surfaces. Langmuir 2009, 25, 11822. (29) Levkin, P.; Svec, F.; Fréchet, J. M. J. Porous Polymer Coatings: a Versatile Approach to Superhydrophobic Surfaces. Adv. Funct. Mater. 2009, 19, 1993. (30) Teisala, H.; Tuominen, M.; Aromaa, M.; Mäkelä, J. M.; Stepien, M.; Saarinen, J. J.; Toivakka, M.; Kuusipalo, J. Development of Superhydrophobic Coating on Paperboard Surface using the Liquid Flame Spray. Surf. Coat. Technol. 2010, 205, 436. (31) Bondioli, P.; Mariani, C.; Lanzani, A.; Fedeli, E.; Mossa, A.; Muller, A. Lampante Olive Oil Refining with Supercritical Carbon Dioxide. J. Am. Oil Chem. Soc. 1992, 69, 477. (32) Ibáñez, E.; Hurtado Benavides, A. M.; Señoráns, F. J.; Reglero, G. Concentration of Sterols and Tocopherols from Olive Oil with Supercritical Carbon Dioxide. J. Am. Oil Chem. Soc. 2002, 79, 1255. (33) Mendes, R. L.; Nobre, B. P.; Cardoso, M. T.; Pereira, A. P.; Palavra, A. F. Supercritical Carbon Dioxide Extraction of Compounds with Pharmaceutical Importance from Microalgae. Inorg. Chim. Acta 2003, 356, 328. (34) Oschmann, H. J.; Prahl, U.; Severin, D. Separation of Paraffin from Crude Oil by Supercritical Fluid Extraction. Pet. Sci. Technol. 1998, 16, 133.

of the coating. We also intend to evaluate the resistance of these coatings toward various types of abrasion.

■

ASSOCIATED CONTENT

S Supporting Information *

Figures S1−S3 show the pump flow, pump pressure, pressure of CO2, and temperature of CO2 as functions of time for different process conditions. Figures S4−S12 show the maximum surface temperature, mean surface temperature, and minimum surface temperature as functions of time for different process conditions. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Author Contributions

The manuscript was written through the contributions of all authors. All authors have given their approval to the final version of the manuscript. Funding

This work was supported by the Swedish Foundation for Strategic Research (SSF, 2005:0073/13, RMA08-0044) and Cellutech AB. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors would like to thank the Swedish Foundation for Strategic Research (SSF) and Cellutech AB for funding, Dr. Irene Rodriguez-Meizoso at Lund University for valuable comments, Tech. Lic. Andrew Marais at Fibre and Polymer Technology for taking photographs, Margareta Lind at Innventia AB for assistance with the topography measurements, and EKA Chemicals for supplying the AKD.

■

REFERENCES

(1) Gao, L.; McCarthy, T. J.; Zhang, X. Wetting and Superhydrophobicity. Langmuir 2009, 25, 14100. (2) Barthlott, W.; Neinhuis, C. Purity of the Sacred Lotus, or Escape from Contamination in Biological Surfaces. Planta 1997, 202, 1. (3) Marmur, A. The Lotus Effect: Superhydrophobicity and Metastability. Langmuir 2004, 20, 3517. (4) Butt, H.-J.; Roisman, I. V.; Brinkmann, M.; Papadopolos, P.; Vollmer, D.; Semprebon, C. Characterization of Super LiquidRepellent Surfaces. Curr. Opin. Colloid Interface Sci. 2014, 19, 343. (5) Kota, A. K.; Mabry, J. M.; Tuteja, A. Superoleophobic Surfaces: Design Criteria and Recent Studies. Surf. Innovations 2013, 1, 71. (6) Guo, Z.; Liu, W.; Su, B.-L. Superhydrophobic Surfaces: From Natural to Biomimetic to Functional. J. Colloid Interface Sci. 2011, 353, 335. (7) Gao, L.; McCarthy, T. J. Contact Angle Hysteresis Explained. Langmuir 2006, 22, 6234. (8) Zhang, X.; Shi, F.; Niu, J.; Jiang, Y.; Wang, Z. Superhydrophobic Surfaces: From Structural Control to Functional Application. J. Mater. Chem. 2008, 18, 621. (9) Quéré, D. Surface Chemistry: Fakir Droplets. Nat. Mater. 2002, 1, 14. (10) Cassie, A. B. D.; Baxter, S. Wettability of Porous Surfaces. Trans. Faraday Soc. 1944, 40, 546. (11) Milne, A. J. B.; Amirfazli, A. The Cassie Equation: How it is Meant to be Used. Adv. Colloid Interface Sci. 2012, 170, 48. 1066


Article

Industrial & Engineering Chemistry Research (35) Crause, J. C.; Nieuwoudt, I. Fractionation of Paraffin Wax Mixtures. Ind. Eng. Chem. Res. 2000, 39, 4871. (36) Lawandy, N. M.; Smuk, A. Y. Supercritical Fluid Cleaning of Banknotes. Ind. Eng. Chem. Res. 2014, 53, 530. (37) Tom, J. W.; Debenedetti, P. G. Particle Formation with Supercritical Fluidsa Review. J. Aerosol Sci. 1991, 22, 555. (38) Quan, C.; Werner, O.; Wågberg, L.; Turner, C. Generation of Superhydrophobic Paper Surfaces by a Rapidly Expanding Supercritical Carbon Dioxide-Alkyl Ketene Dimer Solution. J. Supercrit. Fluids 2009, 49, 117. (39) Rodriguez-Meizoso, I.; Werner, O.; Quan, C.; Knez, Z.; Turner, C. Phase-Behavior of Alkyl Ketene Dimmer (AKD) in Supercritical Carbon Dioxide. The Implications of using Different Solubility Measurement Methods. J. Supercrit. Fluids 2012, 61, 25. (40) Shibuichi, S.; Onda, T.; Satoh, N.; Tsujii, K. Super WaterRepellent Surfaces Resulting from Fractal Structure. J. Phys. Chem. 1996, 100, 19512. (41) Olin, P.; Lindström, S. B.; Pettersson, T.; Wågberg, L. Water Drop Friction on Superhydrophobic Surfaces. Langmuir 2013, 29, 9079. (42) Hyatt, J. A. Liquid and Supercritical Carbon Dioxide as Organic Solvents. J. Org. Chem. 1984, 49, 5097. (43) Hirunsit, P.; Huang, Z.; Srinophakun, T.; Charoenchaitrakool, M.; Kawi, S. Particle Formation of Ibuprofen-Supercritical CO2 System from Rapid Expansion of Supercritical Solutions (RESS): A Mathematical Model. Powder Technol. 2005, 154, 83. (44) Werner, O.; Turner, C. Investigation of Different Particle Sizes on Superhydrophobic Surfaces Made by Rapid Expansion of Supercritical Solution with In Situ Laser Diffraction (RESS-LD). J. Supercrit. Fluids 2012, 67, 53.

1067


Development of a Semicontinuous Spray Process for the Production of

Recommend Documents