Selective Allowance of Precipitation from Oversaturated Solution Using Surface Structures.

Precipitation is a well-known phenomenon commonly observed in salt ponds. However, it causes pipe clogging in industrial sites, which can be resolved by controlling the direction of precipitation. Herein, we propose a method to control the precipitation direction by changing the structures and properties of the solid surface. Bare, nanostructured, microstructured, and micro/nanostructured surfaces were immersed in the same saturated aqueous NaCl solution, and the heights at which precipitation occurred in the different specimens were compared. On bare and nanostructured surfaces, NaCl deposits as a flat layer on the surface, while on micro and micro/nanostructured surfaces, it forms a thick deposit in a direction perpendicular to the surface. When the same experiment was conducted on surfaces made by patterning different structural surfaces, the precipitates did not spread on the surface with microscale structures. We believe that this novel approach may prove useful in solving the problems caused by precipitation.

Introduction

The phenomenon of precipitation is widely used for both salt production and lithium production from salt ponds and salt lakes.[1−3] While precipitation is highly beneficial for salt and lithium production, it can also cause chronic pipe clogging.[4,5] Clogging of pipes may reduce the efficiency of heat exchangers or even block common pipe flow, resulting in significant losses.[6,7] Therefore, many researchers have investigated various ways to prevent precipitation.[8−10]

Precipitation is a chemical reaction that causes solid particles to separate from a liquid. The crystal size of the precipitates is influenced by the surface energy and temperature of the liquid.[11] The surface energy between solids and liquids is closely related to surface wettability, the extent of which is given by the contact angle (CA), i.e., the angle formed by a liquid droplet on a solid surface. The CA is determined by the interplay between the surface energies of the gas, liquid, and solid under thermodynamic equilibrium, and it decreases with increasing roughness, in accordance with the Wenzel state.[12,13]

Many researchers have investigated methods to control surface wettability by modifying surface energy and surface roughness.[14−20] In particular, surface modification has been used to improve the efficiency of condensation heat transfer by applying an antifrosting mechanism in heat exchangers.[21,22]

There have been many studies to learn more about controlling the liquefaction and formation of solids in gas; however, there has been little research on the inhibition and/or induction of precipitation utilizing the roughness of a surface. Herein, we propose a method to control the precipitation direction by changing the structure and properties of the solid surface. To the best of our knowledge, this is the first study on how to actively control the area of precipitation in an environment where precipitation is bound to occur. In this study, we also closely investigate the growth of precipitates and propose a new method to solve the problem of precipitation in pipelines.

Results and Discussion

Principle of Nucleation

For precipitation to occur, it is imperative for the liquid to be oversaturated. The crystal nuclei are first generated in the liquid or on the surrounding surface. Precipitation in the liquid is known as homogeneous nucleation, while that on the surrounding surface is known as heterogeneous nucleation. For homogeneous nucleation (Figure S1), the Gibbs free energy change after crystallization can be determined by the following equationwhere VS is the solid volume, GvL and GvS are the Gibbs free energies per unit volume of the liquid and solid states, respectively, ASL is the surface area between the solid and liquid phases, and γSL is the surface tension between the solid and liquid phases.[23] The Gibbs free energy difference can be determined by assuming the radius of the generated nucleus as r and using the following equationwhere ΔGv is the difference in the Gibbs free energy per unit volume between the liquid and the solid. To determine the critical radius r* from the maximum change in Gibbs free energy, the value of r is calculated at the local maximum by differentiating ΔG(r) with respect to r.[24]

The theoretical critical homogeneous nucleation radius (r*) and the homogeneous Gibbs free energy change (ΔGhom*) can be obtained as follows

If the radius of the generated nucleus is less than r*, the Gibbs free energy may decrease as the nucleus redissolves. If it is larger than r*, the Gibbs free energy may decrease as the nucleus grows. The change in Gibbs free energy for heterogeneous nucleation (Figure S2) can be determined using the following equationwhere ΔGhet* is the Gibbs free energy change for heterogeneous nucleation, which is related to the homogeneous nucleation as followswhere S(θ) is <1. The Gibbs free energy change for heterogeneous nucleation is greater than that for homogeneous nucleation. In addition, the former occurs more spontaneously than the latter. Therefore, for our experiment, we induced salt precipitation on the aluminum surface, a heterogeneous surface, and analyzed the precipitation behavior.

Principle of the Capillary Phenomenon

When a capillary tube is placed in a liquid, the level of liquid in the tube becomes higher or lower than that outside the tube, which is called the capillary phenomenon. The height of the liquid in the tube is determined by the attraction between the molecules and that between the molecule and the wall of the tube. The level of liquid in the capillary tube can be determined as followswhere γ is the surface tension of the liquid, θ is the contact angle, ρ is the density of the liquid, r is the radius of the capillary tube, and g is the gravitational acceleration.[25] According to eq , the smaller the radius of the tube, the greater is the height of liquid in the tube.

Characterization of Aluminum Surfaces

The structure of each surface was observed by scanning electron microscopy (SEM), and the results are shown in Figure . The bare surface (BS) has a clean surface (Figure a), while the nanostructured surface (NS) has a structure comprising nanoflakes (Figure b). The microcubic structure is well-formed on both microstructured surfaces (MS) and micro/nanostructured surface (MNS) (Figure c,d, respectively). The approximate structure sizes of the NS and MS are 100 nm and 1 μm, respectively. Only Al and Mg components were detected when the surface components were analyzed by energy-dispersive spectrometry (EDS) because an aluminum oxide film was formed on both BS and MS (Figure e–g). In contrast, an aluminum hydroxide layer was formed on NS and MNS, showing the presence of Mg and O components (Figure f,h). This result can be verified by measuring the water CAs (WCAs). WCA refers to the angle formed by solid specimens and 5 μm deionized (DI) water droplets in the atmosphere at room temperature. The BS and MS of the aluminum oxide film with relatively low surface energy have high WCAs (Figure a,c), while the NS and MNS of aluminum hydroxide surface with high surface energy have very low WCAs (Figure b,d).

Figure 1

Characterization of various surfaces. Scanning electron microscopy (SEM) image and water contact angles (WCAs) of (a) bare surface, (b) nanostructured surface, (c) microstructured surface, and (d) micro/nanostructured surface. Energy-dispersive spectroscopy (EDS) results of (e) bare surface, (f) nanostructured surface, (g) microstructured surface, and (h) micro/nanostructured surface.

Precipitation Behavior of Different Surface Structures

An experiment was also conducted to observe the growth of salt precipitates on the BS, MS, NS, and MNS surfaces. The experiment was conducted for 144 h, and the precipitation process was observed every 24 h. On the BS (Figure a) and NS (Figure b) surfaces, NaCl particles are greatly spread out along the surface. In contrast, on the MS (Figure c) and MNS (Figure d) surfaces, the NaCl particles are thick and do not spread out significantly.

Figure 2

Results of the precipitation experiment over time. (a) Bare surface, (b) nanostructured surface, (c) microstructured surface, and (d) micro/nanostructured surface.

The precipitation height was measured to analyze the BS, NS, MS, and MNS surfaces (Figure a–d, respectively). The height of the precipitation spread was measured on the basis of the center line of the specimen, determined by the precipitation test results for each structure (Figure e). The area of the precipitation spread was measured through the image processing of the photos of precipitation test results for each structure (Figure f). The mean and standard deviation were calculated from the results of five repeated experiments (Figure g,h). The means and standard deviations of the precipitation heights on the BS, NS, MS, and MNS surfaces are 10.21 ± 0.94, 6.539 ± 0.78, 1.148 ± 0.13, and 1.003 ± 0.15 cm, respectively. The means and standard deviations of the precipitation areas on the BS, NS, MS, and MNS surfaces are 52.18 ± 5.15, 38.72 ± 4.78, 5.74 ± 0.45, and 5.21 ± 0.49 cm2, respectively.

Figure 3

Precipitation experiment results on (a) bare surface, (b) nanostructured surface, (c) microstructured surface, and (d) micro/nanostructured surface. (e) Precipitation height measuring method; (f) precipitation area measuring method; (g) precipitation heights on the bare, nanostructured, microstructured, and micro/nanostructured surfaces; and (h) precipitation areas on the bare, nanostructured, microstructured, and micro/nanostructured surfaces.

Precipitation Behavior of Patterned Surface Structures

Bare·nanostructured (B·N), bare·microstructured (B·M), bare·micro/nanostructured (B·MN), and nanostructured·micro/nanostructured (N·MN) patterned surfaces were manufactured using an acid-corrosion-resistant Kapton tape. The growth of salt precipitation on the patterned surfaces of B·N, B·M, B·MN, and N·MN was observed experimentally. The experiment was conducted for 144 h. As in the previous experiments, precipitation proceeded along the surface of the BS and NS; in contrast, it did not spread on the MS and MNS. On the B/N surface, the precipitation proceeds from the BS to the NS and the crystal growth continues (Figure a). On the remaining B·M (Figure b), B·MN (Figure c), and N·MN (Figure d) surfaces, the crystals do not spread because of the surface patterning boundary. The height of the precipitation spread was measured on the basis of the center line of the specimen, which was obtained from the precipitation test results for each structure (Figure e). The area of the precipitation spread was measured through the image processing of the photos of precipitation test results for each structure (Figure f). The mean and standard deviation were calculated from the results of five repeated experiments (Figure g,h). The means and standard deviations of the precipitation height on the B·N, B·M, B·MNS, and N·MNS surfaces are 9.071 ± 1.05, 4.213 ± 0.373, 4.049 ± 0.263, and 4.178 ± 0.324 cm, respectively. The means and standard deviations of the precipitation areas on the B·N, B·M, B·MNS, and N·MNS surfaces are 46.12 ± 5.69, 21.04 ± 1.87, 20.12 ± 1.45, and 20.45 ± 1.62 cm2, respectively.

Figure 4

Precipitation experiment results on (a) bare·nanostructured surface, (b) bare·microstructured surface, (c) bare·micro/nanostructured surface, and (d) nano·micro/nanostructured surface. (e) Precipitation height measuring method; (f) precipitation area measuring method; (g) precipitation heights on the bare·nanostructured, bare·microstructured, bare·micro/nanostructured, and nano·micro/nanostructured surfaces; and (h) precipitation areas on the bare·nanostructured, bare·microstructured, bare·micro/nanostructured, and nano·micro/nanostructured surfaces.

Hence, it can be concluded that the crystal growth behavior on BS and NS surfaces was different from that on MS and MNS surfaces with microcubic structures. In addition, the results revealed that the microstructure suppressed the spread of the crystals on the patterned surfaces.

According to the heterogeneous nucleation theory, precipitation by evaporation begins at the point where the aluminum surface meets the saturated aqueous NaCl solution. The surface tension of the saturated aqueous NaCl solution was 83.8 ± 0.28 mN/m, and the size of the nuclei produced therein has been reported to vary from several to tens of nanometers.[26] The precipitation process of BS and NS is shown in Figure . In the first step, numerous nuclei grow parallel to the surface boundary (Figure a). Then, a capillary force acts along the narrow gap between the grown nuclei, pulling the solution in the positive z-direction (Figure b). Afterward, similar to the first step, precipitation occurs at the interface (Figure c), and similar to the second step, a capillary action occurs in the positive z-direction (Figure d). The precipitation proceeds along the surface and involves a series of processes, and the crystal grows much higher than the original water level. Therefore, in BS and NS, the precipitation in the horizontal direction is much higher than that in the vertical direction of the aluminum surface.

Figure 5

Three-dimensional schematic diagram of bare surface (BS) and nanostructured surface (NS) precipitation. (a) Precipitation on the boundary between the solution and the surface. (b) Capillary action in the space between crystals in the positive z-direction. (c) Precipitation on the changed boundary. (d) Capillary action in the space between new crystals in the positive z-direction.

MS and MNS with microcubic structures also begin to precipitate at the point where the aluminum surface and the saturated aqueous NaCl solution come into contact. The different precipitation results are attributed to the microcubic structure. The solution moves in the enlarged part because of the capillary force acting along the crystallized surface (Figure a). Because the capillary action acts in the positive x-direction, the boundary line of the solution can continue to move (Figure b). When this process is repeated and reaches the end of the microcubic structure (Figure c), the capillary force can no longer raise the solution in the positive z-direction, and gravity and capillary force act in the negative z-direction (Figure d). Therefore, the crystals can no longer rise upward and precipitate; instead, they thicken downward and grow (Figure e). In other words, MS and MNS containing microstructures show much higher precipitation in the direction perpendicular to the surface than in the direction horizontal to the surface.

Figure 6

Three-dimensional schematic diagram of microstructured surface (MS) and micro/nanostructured surface (MNS) precipitation. (a) Precipitation on the boundary between the solution and the surface. (b) Capillary action in the space between crystals in the positive x-direction. (c) Precipitation on the changed boundary. (d) Capillary action in the space between new crystals in the negative z-direction. (e) Precipitation on the changed boundary.

Finally, the results confirm that selective precipitation is possible using surface microstructures. In particular, a large structure on the order of several microns was found to influence the direction of the capillary phenomenon, thereby suppressing the width and height, as well as the spread, of the crystals. Therefore, it can be concluded that a flat surface induces the spreading of the precipitation, while a microstructured surface suppresses it.

Conclusions

This study showed that the precipitation phenomenon of a saturated aqueous NaCl solution can be controlled by modifying the surface structure. The industrial aluminum surface was modified to obtain nanostructured, microstructured, and micro/nanostructured surfaces. The structure, wettability, and elemental composition of each surface, including those of the bare surfaces, were analyzed, and precipitation experiments were conducted on the fabricated surfaces. While the precipitated crystals spread widely along the bare and nanostructured surfaces, the spread was inhibited on the microstructured and micro/nanostructured surfaces, which also contained microcubic structures.

The microscale structure of the surface suppresses crystal growth by preventing the solution from spreading along the surface. Patterning the microstructures on the surface can selectively allow or suppress precipitation along the patterned shape. For example, the microsized structure is patterned only on the letter string NSCS. After that, if it is immersed in the saturated aqueous NaCl solution, salt will be precipitated in all regions except for the NSCS region containing the microstructure, as shown in Figure . The study results provide a better understanding of the precipitation phenomenon at the solid–solution interface. Moreover, induction and inhibition of precipitation by changing the surface structure, which has been observed and described on the basis of our findings, is a novel functional phenomenon. In addition, the results of this study can suggest new possibilities to solve the pipe clogging problem by allowing selective precipitation by modifying the surface structure inside the tube.

Figure 7

(a) Microsized structure patterning surface on the NSCS letter string. (b) Surface precipitation results after immersion in the saturated aqueous NaCl solution.

Experimental Section

Materials

An aluminum sheet (99.5%, Chunwoo metal, Republic of Korea) was used as a material for preparing the experimental specimens. Sodium hydroxide (NaOH; 99.5%), sodium chloride (NaCl; 99.5%), hydrochloric acid (HCl; 35.0–37.0%), and ethyl alcohol (94.5%) were supplied by Samchun Chemical (Republic of Korea).

Methods

The surfaces used in the experiment were processed with aluminum to a width of 5 cm and a length of 15 cm. The bare surfaces (BS) were prepared by washing the untreated aluminum surface for 5 min using an ultrasonic cleaner in an ethanol environment. The nanostructured surfaces (NS) were prepared by immersing the washed aluminum surface in a 1 M NaOH aqueous solution at 25 °C for 1 min to remove surface impurities. To prepare the nanostructure using gelling aluminum, the aluminum surface was placed in boiling deionized (DI) water (over 90 °C) for 10 min and dried. The microstructured surfaces (MS) were prepared by immersing the washed aluminum surface in a 1 M HCl solution at 80 °C for 1 min to scrape the surface, followed by washing with DI water and drying. The micro/nanostructured surface (MNS) was prepared by shaping the microstructured surface with a HCl solution and immersing the washed surface in a 1 M NaOH aqueous solution at 25 °C for 5 s to gelate the surface, placing it in boiling DI water (over 90 °C) for 10 min, followed by drying.

Surface Precipitation

A glass beaker (Corning) was used for all surface precipitation experiments. A constant-temperature-and-humidity chamber (SH SCIENTIFIC, Republic of Korea) was used to maintain the experimental environment. A forced circulation dryer (SH SCIENTIFIC) was used to dry each surface. The prepared aluminum surfaces were dried at 60 °C for 5 min for all surface precipitation experiments. A saturated NaCl solution was prepared by mixing NaCl and water (0.359:1 m/m). An experiment was conducted to observe the crystal growth process for each surface. The BS, NS, MS, and MNS and a saturated aqueous NaCl solution were prepared. Under the same temperature and humidity conditions, each surface was first immersed 15 mm below the surface in 120 mL of a saturated aqueous NaCl solution, and then the crystallization process was observed by periodically photographing the growth of crystals on the surface (Figure a,b).

Figure 8

Schematic of the precipitation experiment using (a) surface structures and (b) patterning surface structures.

Characterization

Field-emission scanning electron microscopy (FE-SEM; SU6600, Hitachi, Japan) was used to examine the surface morphologies, and energy-dispersive spectroscopy–electron backscatter diffraction (EDS–EBSD; Pegasus, EDAX) was used to examine the surface elemental composition. CAs were measured using a CA analyzer (SmartDrop, Femtofab Co., Republic of Korea), on the basis of the averages of five experiments conducted using 5 μL droplets. A digital caliper (CAS, Republic of Korea) was used to measure the precipitation height.