Thank you for visiting Nature.com. The browser version you are using has limited CSS support. For the best experience, we recommend that you use an updated browser (or disable Compatibility Mode in Internet Explorer). In the meantime, to ensure continued support, we will render the site without styles and JavaScript.
The market share of adsorption refrigeration systems and heat pumps is still relatively small compared to traditional compressor systems. Despite the huge advantage of using cheap heat (instead of expensive electrical work), the implementation of systems based on adsorption principles is still limited to a few specific applications. The main disadvantage that needs to be eliminated is the decrease in specific power due to low thermal conductivity and low stability of the adsorbent. Current state of the art commercial adsorption refrigeration systems are based on adsorbers based on plate heat exchangers coated to optimize cooling capacity. The results are well known that decreasing the thickness of the coating leads to a decrease in the mass transfer impedance, and increasing the surface area to volume ratio of the conductive structures increases power without compromising efficiency. The metal fibers used in this work can provide a specific surface area in the range of 2500–50,000 m2/m3. Three methods for obtaining very thin but stable coatings of salt hydrates on metal surfaces, including metal fibers, for the production of coatings demonstrate for the first time a high power density heat exchanger. The surface treatment based on aluminum anodizing is chosen to create a stronger bond between the coating and the substrate. The microstructure of the resulting surface was analyzed using scanning electron microscopy. Reduced total reflection Fourier transform infrared spectroscopy and energy dispersive X-ray spectroscopy were used to check for the presence of the desired species in the assay. Their ability to form hydrates was confirmed by combined thermogravimetric analysis (TGA)/differential thermogravimetric analysis (DTG). Poor quality over 0.07 g (water)/g (composite) was found in the MgSO4 coating, showing signs of dehydration at about 60 °C and reproducible after rehydration. Positive results were also obtained with SrCl2 and ZnSO4 with a mass difference of about 0.02 g/g below 100 °C. Hydroxyethylcellulose was chosen as an additive to increase the stability and adhesion of the coating. The adsorptive properties of the products were evaluated by simultaneous TGA-DTG and their adhesion was characterized by a method based on the tests described in ISO2409. The consistency and adhesion of the CaCl2 coating is significantly improved while maintaining its adsorption capacity with a weight difference of about 0.1 g/g at temperatures below 100 °C. In addition, MgSO4 retains the ability to form hydrates, showing a mass difference of more than 0.04 g/g at temperatures below 100 °C. Finally, coated metal fibers are examined. The results show that the effective thermal conductivity of the fiber structure coated with Al2(SO4)3 can be 4.7 times higher compared to the volume of pure Al2(SO4)3. The coating of the studied coatings was examined visually, and the internal structure was evaluated using a microscopic image of the cross sections. A coating of Al2(SO4)3 with a thickness of about 50 µm was obtained, but the overall process must be optimized to achieve a more uniform distribution.
Adsorption systems have gained a lot of attention over the past few decades as they provide an environmentally friendly alternative to traditional compression heat pumps or refrigeration systems. With rising comfort standards and global average temperatures, adsorption systems may reduce dependence on fossil fuels in the near future. In addition, any improvements in adsorption refrigeration or heat pumps can be transferred to thermal energy storage, which represents an additional increase in the potential for efficient use of primary energy. The main advantage of adsorption heat pumps and refrigeration systems is that they can operate with low heat mass. This makes them suitable for low temperature sources such as solar energy or waste heat. In terms of energy storage applications, adsorption has the advantage of higher energy density and less energy dissipation compared to sensible or latent heat storage.
Adsorption heat pumps and refrigeration systems follow the same thermodynamic cycle as their vapor compression counterparts. The main difference is the replacement of compressor components with adsorbers. The element is able to adsorb low pressure refrigerant vapor at moderate temperatures, evaporating more refrigerant even when the liquid is cold. It is necessary to ensure constant cooling of the adsorber in order to exclude the enthalpy of adsorption (exotherm). The adsorber is regenerated at high temperature, causing the refrigerant vapor to desorb. Heating must continue to provide the enthalpy of desorption (endothermic). Because adsorption processes are characterized by temperature changes, high power density requires high thermal conductivity. However, low thermal conductivity is by far the main disadvantage in most applications.
The main problem of conductivity is to increase its average value while maintaining the transport path that provides the flow of adsorption/desorption vapors. Two approaches are commonly used to achieve this: composite heat exchangers and coated heat exchangers. The most popular and successful composite materials are those that use carbon-based additives, namely expanded graphite, activated carbon, or carbon fibers. Oliveira et al. 2 impregnated expanded graphite powder with calcium chloride to produce an adsorber with a specific cooling capacity (SCP) of up to 306 W/kg and a coefficient of performance (COP) of up to 0.46. Zajaczkowski et al. 3 proposed a combination of expanded graphite, carbon fiber and calcium chloride with a total conductivity of 15 W/mK. Jian et al4 tested composites with sulfuric acid treated expanded natural graphite (ENG-TSA) as substrate in a two-stage adsorption cooling cycle. The model predicted COP from 0.215 to 0.285 and SCP from 161.4 to 260.74 W/kg.
By far the most viable solution is the coated heat exchanger. The coating mechanisms of these heat exchangers can be divided into two categories: direct synthesis and adhesives. The most successful method is direct synthesis, which involves the formation of adsorbing materials directly on the surface of heat exchangers from the appropriate reagents. Sotech5 has patented a method for synthesizing coated zeolite for use in a series of coolers manufactured by Fahrenheit GmbH. Schnabel et al6 tested the performance of two zeolites coated on stainless steel. However, this method only works with specific adsorbents, which makes coating with adhesives an interesting alternative. Binders are passive substances chosen to support sorbent adhesion and/or mass transfer, but play no role in adsorption or conductivity enhancement. Freni et al. 7 coated aluminum heat exchangers with AQSOA-Z02 zeolite stabilized with a clay-based binder. Calabrese et al.8 studied the preparation of zeolite coatings with polymeric binders. Ammann et al.9 proposed a method for preparing porous zeolite coatings from magnetic mixtures of polyvinyl alcohol. Alumina (alumina) is also used as binder 10 in the adsorber. To our knowledge, cellulose and hydroxyethyl cellulose are only used in combination with physical adsorbents11,12. Sometimes the glue is not used for the paint, but is used to build the structure 13 on its own. The combination of alginate polymer matrices with multiple salt hydrates forms flexible composite bead structures that prevent leakage during drying and provide adequate mass transfer. Clays such as bentonite and attapulgite have been used as binders for the preparation of composites15,16,17. Ethylcellulose has been used to microencapsulate calcium chloride18 or sodium sulfide19.
Composites with a porous metal structure can be divided into additive heat exchangers and coated heat exchangers. The advantage of these structures is the high specific surface area. This results in a larger contact surface between adsorbent and metal without the addition of an inert mass, which reduces the overall efficiency of the refrigeration cycle. Lang et al. 20 have improved the overall conductivity of a zeolite adsorber with an aluminum honeycomb structure. Gillerminot et al. 21 improved the thermal conductivity of NaX zeolite layers with copper and nickel foam. Although composites are used as phase change materials (PCMs), the findings of Li et al. 22 and Zhao et al. 23 are also of interest for chemisorption. They compared the performance of expanded graphite and metal foam and concluded that the latter was preferable only if corrosion was not an issue. Palomba et al. have recently compared other metallic porous structures24. Van der Pal et al. have studied metal salts embedded in foams 25 . All previous examples correspond to dense layers of particulate adsorbents. Metal porous structures are practically not used to coat adsorbers, which is a more optimal solution. An example of binding to zeolites can be found in Wittstadt et al. 26 but no attempt has been made to bind salt hydrates despite their higher energy density 27 .
Thus, three methods for preparing adsorbent coatings will be explored in this article: (1) binder coating, (2) direct reaction, and (3) surface treatment. Hydroxyethylcellulose was the binder of choice in this work due to previously reported stability and good coating adhesion in combination with physical adsorbents. This method was initially investigated for flat coatings and later applied to metal fiber structures. Previously, a preliminary analysis of the possibility of chemical reactions with the formation of adsorbent coatings was reported. Previous experience is now being transferred to the coating of metal fiber structures. The surface treatment chosen for this work is a method based on aluminum anodizing. Aluminum anodizing has been successfully combined with metal salts for aesthetic purposes29. In these cases, very stable and corrosion-resistant coatings can be obtained. However, they cannot carry out any adsorption or desorption process. This paper presents a variant of this approach that allows mass to be moved using the adhesive properties of the original process. To the best of our knowledge, none of the methods described here have been previously studied. They represent a very interesting new technology because they allow the formation of hydrated adsorbent coatings, which have a number of advantages over the frequently studied physical adsorbents.
The stamped aluminum plates used as substrates for these experiments were provided by ALINVEST Břidličná, Czech Republic. They contain 98.11% aluminium, 1.3622% iron, 0.3618% manganese and traces of copper, magnesium, silicon, titanium, zinc, chromium and nickel.
The materials chosen for the manufacture of composites are selected in accordance with their thermodynamic properties, namely, depending on the amount of water that they can adsorb/desorb at temperatures below 120°C.
Magnesium sulfate (MgSO4) is one of the most interesting and studied hydrated salts30,31,32,33,34,35,36,37,38,39,40,41. The thermodynamic properties have been systematically measured and found to be suitable for applications in the fields of adsorption refrigeration, heat pumps and energy storage. Dry magnesium sulfate CAS-Nr.7487-88-9 99% (Grüssing GmbH, Filsum, Niedersachsen, Germany) was used.
Calcium chloride (CaCl2) (H319) is another well-studied salt because its hydrate has interesting thermodynamic properties41,42,43,44. Calcium chloride hexahydrate CAS-No. 7774-34-7 97% used (Grüssing, GmbH, Filsum, Niedersachsen, Germany).
Zinc sulfate (ZnSO4) (H3O2, H318, H410) and its hydrates have thermodynamic properties suitable for low temperature adsorption processes45,46. Zinc sulfate heptahydrate CAS-Nr.7733-02-0 99.5% (Grüssing GmbH, Filsum, Niedersachsen, Germany) was used.
Strontium chloride (SrCl2) (H318) also has interesting thermodynamic properties4,45,47 although it is often combined with ammonia in adsorption heat pump or energy storage research. Strontium chloride hexahydrate CAS-Nr.10.476-85-4 99.0–102.0% (Sigma Aldrich, St. Louis, Missouri, USA) was used for the synthesis.
Copper sulfate (CuSO4) (H302, H315, H319, H410) is not among the hydrates frequently found in the professional literature, although its thermodynamic properties are of interest for low temperature applications48,49. Copper sulfate CAS-Nr.7758-99-8 99% (Sigma Aldrich, St. Louis, MO, USA) was used for the synthesis.
Magnesium chloride (MgCl2) is one of the hydrated salts that has recently received more attention in the field of thermal energy storage50,51. Magnesium chloride hexahydrate CAS-Nr.7791-18-6 pure pharmaceutical grade (Applichem GmbH., Darmstadt, Germany) was used for the experiments.
As mentioned above, hydroxyethyl cellulose was chosen because of the positive results in similar applications. The material used in our synthesis is hydroxyethyl cellulose CAS-Nr 9004-62-0 (Sigma Aldrich, St. Louis, MO, USA).
Metal fibers are made from short wires bonded together by compression and sintering, a process known as crucible melt extraction (CME)52. This means that their thermal conductivity depends not only on the bulk conductivity of the metals used in the manufacture and the porosity of the final structure, but also on the quality of the bonds between the threads. The fibers are not isotropic and tend to be distributed in a certain direction during production, which makes the thermal conductivity in the transverse direction much lower.
The water absorption properties were investigated using simultaneous thermogravimetric analysis (TGA)/differential thermogravimetric analysis (DTG) in a vacuum package (Netzsch TG 209 F1 Libra). The measurements were carried out in a flowing nitrogen atmosphere at a flow rate of 10 ml/min and a temperature range from 25 to 150°C in aluminum oxide crucibles. The heating rate was 1 °C/min, the sample weight varied from 10 to 20 mg, the resolution was 0.1 μg. In this work, it should be noted that the mass difference per unit surface has a large uncertainty. The samples used in TGA-DTG are very small and irregularly cut, which makes their area determination inaccurate. These values can only be extrapolated to a larger area if large deviations are taken into account.
Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectra were acquired on a Bruker Vertex 80 v FTIR spectrometer (Bruker Optik GmbH, Leipzig, Germany) using an ATR platinum accessory (Bruker Optik GmbH, Germany). The spectra of pure dry diamond crystals were measured directly in vacuum before using the samples as a background for experimental measurements. The samples were measured in vacuum using a spectral resolution of 2 cm-1 and an average number of scans of 32. Wavenumber range from 8000 to 500 cm-1. Spectral analysis was performed using the OPUS program.
SEM analysis was performed using a DSM 982 Gemini from Zeiss at accelerating voltages of 2 and 5 kV. Energy dispersive X-ray spectroscopy (EDX) was performed using a Thermo Fischer System 7 with a Peltier cooled silicon drift detector (SSD).
The preparation of metal plates was carried out according to the procedure similar to that described in 53. First, immerse the plate in 50% sulfuric acid. 15 minutes. Then they were introduced into 1 M sodium hydroxide solution for about 10 seconds. Then the samples were washed with a large amount of distilled water, and then soaked in distilled water for 30 minutes. After preliminary surface treatment, the samples were immersed in a 3% saturated solution. HEC and target salt. Finally, take them out and dry them at 60°C.
The anodizing method enhances and strengthens the natural oxide layer on the passive metal. The aluminum panels were anodized with sulfuric acid in a hardened state and then sealed in hot water. Anodizing followed an initial etching with 1 mol/l NaOH (600 s) followed by neutralization in 1 mol/l HNO3 (60 s). The electrolyte solution is a mixture of 2.3 M H2SO4, 0.01 M Al2(SO4)3, and 1 M MgSO4 + 7H2O. Anodizing was carried out at (40 ± 1)°C, 30 mA/cm2 for 1200 seconds. The sealing process was carried out in various brine solutions as described in the materials (MgSO4, CaCl2, ZnSO4, SrCl2, CuSO4, MgCl2). The sample is boiled in it for 1800 seconds.
Three different methods for producing composites have been investigated: adhesive coating, direct reaction, and surface treatment. The advantages and disadvantages of each training method are systematically analyzed and discussed. Direct observation, nanoimaging, and chemical/elemental analysis were used to evaluate the results.
Anodizing was chosen as a conversion surface treatment method to increase the adhesion of salt hydrates. This surface treatment creates a porous structure of alumina (alumina) directly on the aluminum surface. Traditionally, this method consists of two stages: the first stage creates a porous structure of aluminum oxide, and the second stage creates a coating of aluminum hydroxide that closes the pores. The following are two methods of blocking salt without blocking access to the gas phase. The first consists of a honeycomb system using small aluminum oxide (Al2O3) tubes obtained in the first step to hold the adsorbent crystals and increase its adhesion to metal surfaces. The resulting honeycombs have a diameter of about 50 nm and a length of 200 nm (Fig. 1a). As mentioned earlier, these cavities are usually closed in a second step with a thin layer of Al2O(OH)2 boehmite supported by the alumina tube boiling process. In the second method, this sealing process is modified in such a way that the salt crystals are captured in a uniformly covering layer of boehmite (Al2O(OH)), which is not used for sealing in this case. The second stage is carried out in a saturated solution of the corresponding salt. The described patterns have sizes in the range of 50–100 nm and look like splashed drops (Fig. 1b). The surface obtained as a result of the sealing process has a pronounced spatial structure with an increased contact area. This surface pattern, along with their many bonding configurations, is ideal for carrying and holding salt crystals. Both structures described appear to be truly porous and have small cavities that appear to be well suited for retaining salt hydrates and adsorbing vapors to the salt during operation of the adsorber. However, elemental analysis of these surfaces using EDX can detect trace amounts of magnesium and sulfur on the surface of boehmite, which are not detected in the case of an alumina surface.
The ATR-FTIR of the sample confirmed that the element was magnesium sulfate (see Figure 2b). The spectrum shows characteristic sulfate ion peaks at 610–680 and 1080–1130 cm–1 and characteristic lattice water peaks at 1600–1700 cm–1 and 3200–3800 cm–1 (see Fig. 2a, c). ). The presence of magnesium ions almost does not change the spectrum54.
(a) EDX of a boehmite coated MgSO4 aluminum plate, (b) ATR-FTIR spectra of boehmite and MgSO4 coatings, (c) ATR-FTIR spectra of pure MgSO4.
Maintaining adsorption efficiency was confirmed by TGA. On fig. 3b shows a desorption peak of approx. 60°C. This peak does not correspond to the temperature of the two peaks observed in TGA of pure salt (Fig. 3a). The repeatability of the adsorption–desorption cycle was evaluated, and the same curve was observed after placing the samples in a humid atmosphere (Fig. 3c). The differences observed in the second stage of desorption may be the result of dehydration in a flowing atmosphere, as this often leads to incomplete dehydration. These values correspond to approximately 17.9 g/m2 in the first dewatering and 10.3 g/m2 in the second dewatering.
Comparison of TGA analysis of boehmite and MgSO4: TGA analysis of pure MgSO4 (a), mixture (b) and after rehydration (c).
The same method was carried out with calcium chloride as adsorbent. The results are presented in Figure 4. Visual inspection of the surface revealed minor changes in the metallic glow. The fur is barely visible. SEM confirmed the presence of small crystals evenly distributed over the surface. However, TGA showed no dehydration below 150°C. This may be due to the fact that the proportion of salt is too small compared to the total mass of the substrate for detection by TGA.
The results of surface treatment of the copper sulfate coating by the anodizing method are shown in fig. 5. In this case, the expected incorporation of CuSO4 into the Al oxide structure did not occur. Instead, loose needles are observed as they are commonly used for copper hydroxide Cu(OH)2 used with typical turquoise dyes.
The anodized surface treatment was also tested in combination with strontium chloride. The results showed uneven coverage (see Figure 6a). To determine if the salt covered the entire surface, an EDX analysis was performed. The curve for a point in the gray area (point 1 in Fig. 6b) shows little strontium and a lot of aluminum. This indicates a low content of strontium in the measured zone, which, in turn, indicates a low coverage of strontium chloride. Conversely, white areas have a high content of strontium and a low content of aluminum (points 2–6 in Fig. 6b). EDX analysis of the white area shows darker dots (points 2 and 4 in Fig. 6b), low in chlorine and high in sulfur. This may indicate the formation of strontium sulfate. Brighter dots reflect high chlorine content and low sulfur content (points 3, 5, and 6 in Fig. 6b). This can be explained by the fact that the main part of the white coating consists of the expected strontium chloride. The TGA of the sample confirmed the interpretation of the analysis with a peak at the characteristic temperature of pure strontium chloride (Fig. 6c). Their small value can be justified by a small fraction of salt in comparison with the mass of the metal support. The desorption mass determined in the experiments corresponds to the amount of 7.3 g/m2 given off per unit area of the adsorber at a temperature of 150°C.
Eloxal-treated zinc sulfate coatings were also tested. Macroscopically, the coating is a very thin and uniform layer (Fig. 7a). However, SEM revealed a surface area covered with small crystals separated by empty areas (Fig. 7b). The TGA of the coating and substrate was compared to that of pure salt (Figure 7c). Pure salt has one asymmetric peak at 59.1°C. The coated aluminum showed two small peaks at 55.5°C and 61.3°C, indicating the presence of zinc sulfate hydrate. The mass difference revealed in the experiment corresponds to 10.9 g/m2 at a dehydration temperature of 150°C.
As in the previous application53, hydroxyethyl cellulose was used as a binder to improve the adhesion and stability of the sorbent coating. Material compatibility and effect on adsorption performance was assessed by TGA. The analysis is carried out in relation to the total mass, i.e. the sample includes a metal plate used as a coating substrate. Adhesion is tested by a test based on the cross notch test defined in the ISO2409 specification (cannot meet the notch separation specification depending on the specification thickness and width).
Coating the panels with calcium chloride (CaCl2) (see Fig. 8a) resulted in uneven distribution, which was not observed in the pure aluminum coating used for the transverse notch test. Compared to the results for pure CaCl2, TGA (Fig. 8b) shows two characteristic peaks shifted towards lower temperatures of 40 and 20°C, respectively. The cross-section test does not allow for an objective comparison because the pure CaCl2 sample (sample on the right in Fig. 8c) is a powdery precipitate, which removes the topmost particles. The HEC results showed a very thin and uniform coating with satisfactory adhesion. The mass difference shown in fig. 8b corresponds to 51.3 g/m2 per unit area of the adsorber at a temperature of 150°C.
Positive results in terms of adhesion and uniformity were also obtained with magnesium sulfate (MgSO4) (see Fig. 9). Analysis of the desorption process of the coating showed the presence of one peak of approx. 60°C. This temperature corresponds to the main desorption step seen in the dehydration of pure salts, representing another step at 44°C. It corresponds to the transition from hexahydrate to pentahydrate and is not observed in the case of coatings with binders. Cross section tests show improved distribution and adhesion compared to coatings made using pure salt. The mass difference observed in TGA-DTC corresponds to 18.4 g/m2 per unit area of the adsorber at a temperature of 150°C.
Due to surface irregularities, strontium chloride (SrCl2) has an uneven coating on the fins (Fig. 10a). However, the results of the transverse notch test showed uniform distribution with significantly improved adhesion (Fig. 10c). TGA analysis showed a very small difference in weight, which must be due to the lower salt content compared to the metal substrate. However, the steps on the curve show the presence of a dehydration process, although the peak is associated with the temperature obtained when characterizing pure salt. The peaks at 110°C and 70.2°C observed in Figs. 10b were also found when analyzing pure salt. However, the main dehydration step observed in pure salt at 50°C was not reflected in the curves using the binder. In contrast, the binder mixture showed two peaks at 20.2°C and 94.1°C, which were not measured for the pure salt (Fig. 10b). At a temperature of 150 °C, the observed mass difference corresponds to 7.2 g/m2 per unit area of the adsorber.
The combination of HEC and zinc sulfate (ZnSO4) did not give acceptable results (Figure 11). TGA analysis of the coated metal did not reveal any dehydration processes. Although the distribution and adhesion of the coating have improved, its properties are still far from optimal.
The simplest way to coat metal fibers with a thin and uniform layer is wet impregnation (Fig. 12a), which includes preparation of the target salt and impregnation of metal fibers with an aqueous solution.
When preparing for wet impregnation, two main problems are encountered. On the one hand, the surface tension of the saline solution prevents the correct incorporation of the liquid into the porous structure. Crystallization on the outer surface (Fig. 12d) and air bubbles trapped inside the structure (Fig. 12c) can only be reduced by lowering the surface tension and pre-wetting the sample with distilled water. Forced dissolution in the sample by evacuating the air within or by creating a solution flow in the structure are other effective ways to ensure complete filling of the structure.
The second problem encountered during preparation was the removal of the film from part of the salt (see Fig. 12b). This phenomenon is characterized by the formation of a dry coating on the dissolution surface, which stops the convectively stimulated drying and starts the diffusion stimulated process. The second mechanism is much slower than the first. As a result, a high temperature is required for a reasonable drying time, which increases the risk of bubbles forming inside the sample. This problem is solved by introducing an alternative method of crystallization based not on concentration change (evaporation), but on temperature change (as in the example with MgSO4 in Fig. 13).
Schematic representation of the crystallization process during cooling and separation of solid and liquid phases using MgSO4.
Saturated salt solutions can be prepared at or above room temperature (HT) using this method. In the first case, crystallization was forced by lowering the temperature below room temperature. In the second case, crystallization occurred when the sample was cooled to room temperature (RT). The result is a mixture of crystals (B) and dissolved (A), the liquid part of which is removed by compressed air. This approach not only avoids the formation of a film on these hydrates, but also reduces the time required for the preparation of other composites. However, the removal of liquid by compressed air leads to additional crystallization of the salt, resulting in a thicker coating.
Another method that can be used to coat metal surfaces involves the direct production of target salts through chemical reactions. Coated heat exchangers made by the reaction of acids on the metal surfaces of fins and tubes have a number of advantages, as reported in our previous study. The application of this method to fibers led to very poor results due to the formation of gases during the reaction. The pressure of the hydrogen gas bubbles builds up inside the probe and shifts as the product is ejected (Fig. 14a).
The coating has been modified through a chemical reaction to better control the thickness and distribution of the coating. This method involves passing an acid mist stream through the sample (Figure 14b). This is expected to result in a uniform coating by reaction with the substrate metal. The results were satisfactory, but the process was too slow to be considered an effective method (Fig. 14c). Shorter reaction times can be achieved by localized heating.
To overcome the disadvantages of the above methods, a coating method based on the use of adhesives has been studied. HEC was selected based on the results presented in the previous section. All samples were prepared at 3% wt. The binder is mixed with salt. The fibers were pretreated according to the same procedure as for the ribs, i.e. soaked in 50% vol. within 15 minutes. sulfuric acid, then soaked in sodium hydroxide for 20 seconds, washed in distilled water and finally soaked in distilled water for 30 minutes. In this case, an additional step was added before impregnation. Immerse the sample briefly in a dilute target salt solution and dry at approximately 60°C. The process is designed to modify the surface of the metal, creating nucleation sites that improve the distribution of the coating in the final stage. The fibrous structure has one side where the filaments are thinner and tightly packed, and the opposite side where the filaments are thicker and less distributed. This is the result of 52 manufacturing processes.
The results for calcium chloride (CaCl2) are summarized and illustrated with pictures in Table 1. Good coverage after inoculation. Even those strands with no visible crystals on the surface had reduced metallic reflections, indicating a change in finish. However, after the samples were impregnated with an aqueous mixture of CaCl2 and HEC and dried at a temperature of about 60°C, the coatings were concentrated at the intersections of the structures. This is an effect caused by the surface tension of the solution. After soaking, the liquid remains inside the sample due to its surface tension. Basically it occurs at the intersection of structures. The best side of the specimen has several holes filled with salt. The weight increased by 0.06 g/cm3 after coating.
Coating with magnesium sulfate (MgSO4) produced more salt per unit volume (Table 2). In this case, the measured increment is 0.09 g/cm3. The seeding process resulted in extensive sample coverage. After the coating process, the salt blocks large areas of the thin side of the sample. In addition, some areas of the matte are blocked, but some porosity is retained. In this case, salt formation is easily observed at the intersection of the structures, confirming that the coating process is mainly due to the surface tension of the liquid, and not the interaction between the salt and the metal substrate.
The results for the combination of strontium chloride (SrCl2) and HEC showed similar properties to the previous examples (Table 3). In this case, the thinner side of the sample is almost completely covered. Only individual pores are visible, formed during drying as a result of the release of steam from the sample. The pattern observed on the matte side is very similar to the previous case, the area is blocked with salt and the fibers are not completely covered.
In order to evaluate the positive effect of the fibrous structure on the thermal performance of the heat exchanger, the effective thermal conductivity of the coated fibrous structure was determined and compared with the pure coating material. Thermal conductivity was measured according to ASTM D 5470-2017 using the flat panel device shown in Figure 15a using a reference material with known thermal conductivity. Compared to other transient measurement methods, this principle is advantageous for porous materials used in the current study, since the measurements are performed in a steady state and with a sufficient sample size (base area 30 × 30 mm2, height approximately 15 mm). Samples of the pure coating material (reference) and the coated fiber structure were prepared for measurements in the direction of the fiber and perpendicular to the direction of the fiber to evaluate the effect of anisotropic thermal conductivity. The specimens were ground on the surface (P320 grit) to minimize the effect of surface roughness due to specimen preparation, which does not reflect the structure within the specimen.