• For Contributors +
• Journal Search +
Journal Search Engine
ISSN : 1598-6721(Print)
ISSN : 2288-0771(Online)
The Korean Society of Manufacturing Process Engineers Vol.21 No.4 pp.1-13
DOI : https://doi.org/10.14775/ksmpe.2022.21.04.001

# Development of Criteria for Predicting Delamination in Cabinet Walls of Household Refrigerators

Jin Seong Park*, Sung Ik Kim*, Gun Yup Lee**, Jong Rae Cho*#
*Dept. of Mechanical Engineering, National Korea Maritime & Ocean University
#Corresponding Author : cjr@kmou.ac.kr Tel: +82-51-410-4298, Fax: +82-51-405-4790
25/10/2021 14/11/2021 28/01/2022

## Abstract

Household refrigerator cabinets must undergo cyclic testing at -20 °C and 65 °C for quality control (QC) after their production is complete. These cabinets were assembled from different materials, including acrylonitrile butadiene styrene (ABS), polyurethane (PU) foam, and steel plates. However, different thermal expansion values could be observed owing to differences in the mechanical properties of the materials. In this study, a technique to predict delamination on a refrigerator wall caused by thermal deformation was developed. The mechanical properties of ABS and PU foams were tested, theload factors causing delamination were analyzed, delamination was observed using a high-speed camera, and comparison and verification in terms of stress and strain were performed using a finite element model (FEM). The results indicated that the delamination phenomenon of a refrigerator wall can be defined in two cases. A method for predicting and evaluating delamination was established and applied in an actual refrigerator. To determine the effect of temperature changes on the refrigerator, strain measurements were performed at the weak point and the stress was calculated. The results showed that the proposed FEM prediction technique can be used as a basis for virtual testing to replace future QC testing, thus saving time and cost.

# 냉장고 캐비닛 벽면에서 발생하는 박리현상 예측을 위한 평가 기준 개발에 관한 연구

박 진성*, 김 성익*, 이 건엽**, 조 종래*#
*한국해양대학교 기계공학과
**LG전자

## 초록

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

## 1. Introduction

Many countries have implemented an energy consumption efficiency rating labeling system, which encourages the development of high-efficiency products to replace products that consume large amounts of energy and have a high supply rate in household and industrial sectors. Because of their ability to operate all year, home refrigerators, which have a high supply rate, account for a large proportion of the total energy consumed at home[1]. As a result, improving the efficiency of refrigerators can have a significant impact on energy savings. To improve the energy consumption efficiency rating of home refrigerators, their adiabatic load must be reduced, and the refrigeration cycle efficiency must be maximized.

To reduce the insulation load, among other things, the amount of heat transfer through the outer and inner walls must be reduced by increasing the insulation of a refrigerator cabinet[2,3]. A refrigerator cabinet is made of a variety of materials, each with its own set of mechanical and thermal properties. As a result of the thermal deformation caused by the temperature difference between the inside and outside of the refrigerator, the refrigerator cabinet is not only distorted at an overall scale, but there is also local bulging on the inner and outer walls that comprise the cabinet wall. This is a severe issue that occurs in refrigerators requiring high thermal insulation and can have a significant impact on quality control owing to the deterioration in refrigerator performance[4,5].

Yang et al.[6] predicted the deformation of a refrigerator panel and the unfilled portion of polyurethane (PU) foam by taking into account the chemical and physical behaviors that occur when manufacturing PU foam, which is used as an insulator for refrigerators. They conducted a study using this to reduce the deformation of the refrigerator cabinet.

By using finite element analysis, Kwak et al.[7] confirmed that the bulging phenomenon in cabinet ducts was caused by the elastic deformation of the duct and the non-uniformity of the foaming pressure, among other various parameters, in a refrigerator cabinet made of acrylonitrile butadiene styrene (ABS), PU foam, and steel plates. As a result, to significantly reduce bulging, a reinforcing plate was installed between the refrigerator’s outer plate and the duct.

Zhai et al.[8] used finite element analysis to investigate a method to reduce refrigerator cabinet bulging in considering the different temperatures of the refrigerating room, freezing room, and outdoor environment by applying a bead shape inside the refrigerator.

Many studies have investigated the thermal deformation of cabinets considering their insulation properties to be directly related to the performance of home refrigerators. However, no study has focused on the cause of refrigerator wall bending owing to the delamination of dissimilar materials, as a result of thermal deformation.

Therefore, this study used a quantitative evaluation method to predict the cause of bulging, delamination, and cracking in refrigerators due to temperature changes. The physical properties of ABS and PU foam were determined using specimens of a fully assembled refrigerator. To determine the cause of bulging, specimen tests were performed to analyze the delamination phenomenon of dissimilar materials. Furthermore, strain measurements were carried out to determine the stresses generated on the inner walls of the refrigerator in order to investigate the causes of cracks occurring on refrigerator walls under the test temperature conditions.

## 2. Mechanical property test of refrigerator cabinet materials

### 2.1 PU foam

A tensile test was carried out to investigate the effect of temperature changes on the tensile strength of PU foam and ABS, two primary materials used in refrigerator cabinets, as well as to determine their mechanical properties. Because it is difficult to tighten the grip during tensile tests, compression tests are typically performed[9]. However, a tensile test was performed in this study to evaluate the effect of PU foam insulation on the tensile load, which causes the refrigerator to bulge due to thermal deformation.

Fig. 1 shows the shape of the testing specimen(ASTM D624-14) and the 100 N testing machine. To measure the strain in the tensile test, a tensile load was applied to the specimen at a rate of 2 mm/min, and strain data were collected by attaching a three-axis strain gage (KYOWA KFGS-5-120-D17-11) at the center of the gauge length. Fully built prototype refrigerator cabinet products were used to obtain specimens. Three specimens were prepared for each temperature condition to ensure accurate test results[10]. Considering the test temperature conditions used for quality assurance, which range from -25 °C to 65 °C for 48 h, a total of five representative temperature conditions were selected and tested.

Fig. 2(a) shows the tensile stress as a function of temperature. Tensile strength values obtained under the same temperature conditions ranged from a minimum of 0.25 MPa to a maximum of 0.47 MPa; unlike general steel, the tensile test results of these specimens appeared to be dispersed. PU foam is made by combining polyol and isocyanate with water, which undergo a chemical reaction, and then subjecting them to a rapid volume expansion and curing process[11–15]. As a result, the densities vary depending on the position and direction of the filling inside the cabinet during curing; this is known to cause the variation in the tensile strength of PU foam with a porous structure[16–19]. Fig. 2(b) shows a nominal stress–strain diagram based on the minimum tensile strength of the three specimens that were subjected to tensile tests at different temperatures. Fig. 2(c) shows the measured elastic modulus, and Table 1 summarizes the results. As shown in Fig. 2(b), the tensile strength is similar at a low temperature (-20 °C) and room temperature (25 °C). By contrast, at 45 °C and 65 °C, the stress increases according to the strain. The elastic modulus is also confirmed to increase with increasing temperature.

### 2.2 ABS

ABS has excellent impact resistance and mechanical properties and is primarily used for interior and exterior car parts as well as home electronic products. Moreover, it has excellent moldability and is used as a material for the inner case of refrigerators with complex structures. ABS RS670, the material used in this study, is a vacuum-formed sheet-type raw material that is used inside a refrigerator cabinet[20].

Fig. 3 shows the INSTRON 4469 universal tester and an electric heating furnace capable of controlling temperatures up to 100 °C. Tests were performed by establishing four representative temperatures ranging from -25 °C to 65 °C. The test specimen was stored in an electric furnace for 24 h under each temperature condition to heat it to a uniform temperature inside.

The load was controlled at a rate of 5 mm/min, and Type I test specimens were prepared according to ASTM D638-14[19]. To ensure the accuracy of the test results, three specimens were used for investigating each temperature, and the strain data were collected using the testing machine’s built-in extensometer.

Fig. 4 shows the results of the nominal stress–strain curves obtained via the ABS tensile test conducted at various temperatures. Linear behavior was confirmed at all temperatures tested up to a nominal strain of 0.02. The stress rapidly decreased and nonlinear behavior was maintained; however, fracture occurred at a strain of 0.04.

Table 2 shows the calculated yield strength and elastic modulus at a strain of 0.02. Except at -20 °C, the tensile strength values at room temperature, 45 °C, and 65 °C were around 38–40 MPa, indicating that temperature change had little effect on tensile strength.

## 3. Experiments and analysis of interfacial delamination on cabinet wall

### 3.1 Test method for delamination strength measurements

A refrigerator cabinet is constructed by sandwiching PU foam between a steel plate and a molded ABS. As shown in Fig. 5, a three-layer sandwiched laminate structure composed of PU foam is formed between the steel sheet and ABS through this process. When the cross-section between the steel sheet and the ABS covering the PU foam is examined, it is clear that the filled PU foam is in contact with both the materials as it hardens, and each layer of material can only be separated by applying an external force.

For quality assurance, a fully built refrigerator cabinet is tested in a -20 °C–65 °C environment. By observing the outer and inner parts of the refrigerator after the test, local bulging in the ABS and steel sheets is noted. This phenomenon is confirmed to be caused by the delamination of the PU foam from the refrigerator wall interface. At this point, micro gaps created reduce the refrigerator’s insulation performance and cause shape defects in the interior and exterior of the product.

Here, basic experiments and FE analysis were carried out to develop the criterial of interfacial separation between dissimilar materials such as PU foam and ABS and PU foam and steel sheets in a refrigerator cabinet under a high-temperature environment. The exper iment was carried out by dividing a specimen into ABS and PU foam sections from the refrigerator cabinet’s inner wall and steel and PU foam sections from the refrigerator cabinet’s outer wall to observe the differences between the inner and outer walls.

The test load was assumed to be a condition in which interfacial separation between dissimilar materials occurs as a result of tensile and shear forces[20].

Fig. 6 shows the specifications of the specimens and the equipment used in the tensile test. The tensile test was carried out using the INSTRON 4469 universal testing machine, and a tensile load was applied at 2 mm/min for load control. The specifications of the specimens and the equipment used for the shear test are shown in Fig. 6(b). A 100 N tensile/compression tester (JSV H1000) capable of fixing the grip portion of the specimen under eccentric conditions was used, and the load control was the same as that for the tensile test.

The specimens used in the test were made from the refrigerator cabinet’s wall.

### 3.2 Experimental results of delamination strength measurements

Fig. 7 and Table 4 summarize the stress results at fracture for each specimen due to tensile and shear forces. In Cases 1 and 3, the shear strength at the two interfaces of the refrigerator cabinet was approximately 0.15 MPa. In for Cases 2 and 4, the tensile strength in steel side of cabinet wall was higher than ABS side; however, this varied depending on the sampling location of the specimen. The tensile strength ranged from 0.19 to 0.31 MPa, similar to the tensile test results in Section 2.1. Delamination due to tension was lower than the stress at which fracture occurred in the PU foam tested in Section 2.

A closer examination of the results by load type revealed that delamination occurred at a relatively low shear force compared to the tensile force. Fig. 8 shows the fractured specimens and the displacement–stress curve of Case 1. For specimen A-1, the PU foam was damaged, rather than the interface, due to poor adhesion. Delamination occurred on the ABS side of specimens A-2 and A-3, with no PU foam residue remaining due to fracture. The displacement–stress curve indicated that stress increased with a constant slope and rapidly reduced after initial cracking; delamination occurred at the same time.

PU foam residue due to fracture was confirmed at the interface on the ABS side in specimens A-4 and A-5, and delamination occurred at the interface. In addition, the displacement–stress curve indicated that the stress increased with a constant slope in both specimens, resulting in the formation and progression of initial cracks. After that, the stress reduced by a certain amount, and fracture occurred in the PU foam, confirming the resulting stress inflection point.

According to the results of this experiment, the shear force had a greater impact on delamination at the refrigerator cabinet interface than the tensile force acting along the vertical direction of the refrigerator wall. By contrast, PU foam fracture was confirmed to be caused by the normal tensile load rather than shear separation at the interface. Furthermore, the interfacial state of the specimen and the load–stress curve confirmed that two types of delamination were caused by the shear load acting on the specimen.

### 3.3 Testing and analysis of delamination mode

A high-speed camera was used in real time to conduct an in-depth investigation of the delamination caused by the shear load at the ABS–PU foam interface, as shown in Fig. 9. Images were captured at 250 frames per second using the high-speed camera (Photron FASTCAM SA4 500K-M1). The shear test was performed at a speed of 2 mm/min using a 100 N tester.

Fig. 10 shows the time–load results of the shear test, while Fig. 11 shows the appearance of the specimen at a time point corresponding to the load inflection point.

The shear load caused initial cracks at the interface, and the contact area between the ABS and PU foam decreased as a result of continuous shear loading, resulting in interface delamination.

According to the time–load results in Fig. 10 and the specimen shape observed in Fig. 11, an initial crack formed at the ABS–PU foam interface at points (a) and (b) after the test began. When the specimen was initially sheared, PU foam deformed, and a tensile force acted on the interface between the PU foam and ABS.

The ABS continued to move after time (c), and the PU foam was subjected to tensile forces at the upper end and compressive forces at the lower end due to shear deformation. The greatest load occurred at this time. After time (d), the load rapidly decreased, and the open angle of the interface due to cracks increased at time points (e) to (g), resulting in distortion and tearing at the top of the PU foam owing to shear deformation.

Finite element analysis was used to predict the stress and strain that would occur under the same conditions as the experiment. ANSYS mechanical APDL 2019 R1 was used to conduct the analysis, which took into account the material nonlinearity applied with a large deformation as well as the properties of the PU foam obtained from the earlier tests. SOLID185 (3D Solid) elements were used to create a grid for analysis.

Fig. 12 shows the analysis result of reaction force caused by the load applied to the shear peeling specimen. The maximum reaction force in the analysis and the experimental load of the point (c) in Fig. 10 are almost same value. Therefore, it can be confirmed that the delamination starts at the point (c) in Fig. 10. The stress and strain were examined at the point of maximum reaction force.

Fig. 13 summarizes the stress and strain changes that occur in PU foam over time under the same conditions as those considered in the shear load test for each component.

Here, x represents the component perpendicular to the PU foam and ABS interface, and xz represents the interface and the horizontal plane.

The maximum shear strain of 0.204 and the maximum shear stress of 0.188 MPa were calculated by tracing the points in the analysis up to at point (c) in Fig. 10. The stress distribution along the path from the start (0 mm) to the end (20 mm) of the PU foam interface at point (c) is shown in Fig. 14. The maximum shear stress was 0.15 MPa in the 5–15mm area corresponding to the center of the PU foam, excluding the stress at the beginning and end of the PU foam.

Based on the results of the delamination test and the analysis in this section, as well as the experimental results in Section 2, Equations (1) and (2) can be established, taking the safety factor into account.

Equation (1) is used for evaluating the delamination occurring at the PU foam interface by the shear force, and Equation (2) is used for evaluating the delamination due to PU foam fracture occurring during interfacial separation caused by the initial cracks.

$Shear peeling: τ m a x > 0.15 MPa and γ m a x > 0.2$
(1)

$Fracture of PU foam: σ 2 s t > 0.25 MPa$
(2)

## 4. Stress measurement and analysis of for verification of delamination

The delamination criteria in Equations (1) and (2) were validated using experiments and finite element analysis. The strain generated on a refrigerator wall after continuous exposure to high temperatures for an extended period of time was measured in the experiment. Furthermore, the validity of the analysis model was confirmed by comparing the experimental results with the results of a finite element analysis for the entire refrigerator model.

### 4.1 Experimental method

A strain gauge was used to measure the stress in order to evaluate the delamination of the refrigerator that occurs in a high-temperature environment. Two refrigerator cabinets(Model A and Model B) were selected.

To measure the stress generated in discontinuous arase such as bends inside the refrigerator wall, a three-axis strain gauge was installed. Strain correction according to temperature changes was considered for each material, as shown in Fig. 15, using a dummy gauge for accurate strain measurement.

To examine the effect of long-term exposure to a high temperature environment, the temperature was initially maintained from the ambient temperature Tref to the high-temperature region T1 for a certain period of time as shown in Fig. 16.

### 4.2 Experimental results

The three-axis strain gauge used in the experiment was a rectangular rosette-type gauge. The equations for converting the measured strain to the von-Mises stress are as follows:

$ε A = ε x ⋅ ε B = 1 2 ( ε x + ε y ) + 1 2 γ x y , ε C = ε y$
(3)

$ε 1 , 2 = ε A + ε C 2 ± 1 2 ( ε A − ε B ) 2 + ( ε B + ε C ) 2$
(4)

$σ 1 = E ( 1 − υ 2 ) ( ε 1 + υ ε 2 ) , σ 2 = E ( 1 − υ 2 ) ( ε 2 + υ ε 1 )$
(5)

$σ e q v = 1 2 { σ 1 2 + σ 2 2 + ( σ 2 − σ 1 ) 2 }$
(6)

Fig. 17 shows the linear expansion coefficients of ABS and PU foam calculated using the strain data from the dummy gauges installed for temperature compensation according to the temperature range. In a low temperature environment of –20~20 °C, ABS had a 1.7 times higher coefficient of linear expansion than PU foam. In a high temperature environment of 20~45 °C, ABS was found to have a 2.7 times higher coefficient of linear expansion than PU foam.

Fig. 18 shows the equivalent stress calculated using the strain measured at each point of temperature change over time. Model A had a maximum equivalent stress of 12.7 MPa, while Model B had a maximum equivalent stress of 14.4 MPa. The stress generated decreased as the exposure time under a high-temperature environment increased in all measurement areas except G2 of Model B. This is thought to be a result of the structural stabilization and stress relaxation phenomena occurring in a refrigerator cabinet made using various materials.

Furthermore, the residual stress after 30 h of cooling to room temperature following long-term exposure to a high-temperature environment was found to be very low. The refrigerator was restored to its original shape following shrinkage and expansion caused by differences in the coefficient of linear expansion during the heating cycle.

The equivalent stress values obtained from the experiment and the finite element analysis for the entire refrigerator model are shown in Fig. 19.

Figs. 20(a) and 20(b) show the stress distribution generated by each model at the G1 sensor position. Model A had a 13% error rate, while Model B had a 17% error rate. The experimental and analysis results were in good agreement, confirming the validity of the analysis models.

The results of applying the delamination evaluation methods(Equations (1) and (2)) based on the analysis results of Models A and B are shown in Fig. 21.

In the PU foam of the refrigerator, the stress distribution is shown for the part where the shear stress is greater than 0.15 MPa. In the case of shear strain, however, a value less than 0.2 is displayed, indicating that delamination does not occur. Furthermore, both models demonstrate values of less than 0.25 MPa for the first principal stress, indicating that no delamination occurs due to PU foam fracture.

To check the delamination in the refrigerator after the experiment was completed, the refrigerator was cut and the interface between ABS and PU foam was observed.

Fig. 22 shows the interfacial state of ABS and PU foam, which were cut from the cabinet of Model A and were separated by applying an external force after the experiment. It was delamination that peeling did not occur at the interface between PU foam and ABS.

## 5. Conclusions

Temperature cycle tests are required for quality control of refrigerator cabinets. An alternative evaluation method using FEM is developed in this study to replace this. The following is a summary of the findings of this study:

• 1. Based on the mechanical properties of PU foam and ABS, the S-S curve required for conducting a finite element analysis was obtained. ABS demonstrated relatively linear behavior up to a nominal strain of 0.02. Fracture occurred after maintaining a constant stress up to a nominal strain of 0.04; meanwhile, the nominal stress decreased slightly.

• 2. When the peel strengths between PU foam and ABS, and PU foam and outer steel plate were measured by shear and tensile loads, the peel strength due to the shear load was found to be lower than that due to the tensile load. Experiments confirmed that the peeling phenomenon at the refrigerator interface was caused by a shear load applied parallel to the wall surface. Using a high-speed camera to observe the peeling phenomenon, the PU foam was observed to be broken with tension at the interface during the initial stage of peeling by shear load.

• 3. Delamination of a refrigerator wall during heating cycle tests is mainly caused by the fact that ABS and PU foam experience different levels of thermal deformation. ABS was found to have a coefficient of linear expansion that was 1.7 times greater at low temperatures and 2.7 times greater at high temperatures than that of PU foam.

• 4. Delamination criteria in terms of strain and stress were established using the experimental results and finite element calculations, and their use was demonstrated by applying them to two refrigerator models.

## Acknowledgement

This research was funded by LG Electronics Co., Ltd., the Ministry of Education's National Research Foundation, and the BK21 project.

## Figure

Test specimen of PU foam and test instrument
Mechanical properties of PU foam obtained through tensile test
Test specimen of ABS and test instrument
Engineering strain–stress curve according to temperature for ABS
Cross-section of the refrigerator cabinet wall
Specimen dimensions and delamination test equipment
Comparison of results for shear and tensile strength
Results of delamination experiment
Observation of shear delamination using a high-speed camera
Trend in shear force with time
Photographs of shear delamination captured using a high-speed camera
Results of reaction force in shear analysis
Results of stress and strain components of PU foam
Stress according to the distance from the interface of PU foam at the time(c)
Schematic of temperature cycle experiments
Conditions for temperature cycle experiments
Comparison of coefficients of linear expansion according to material
Equivalent stress calculated using measured strain according to time for Models A and B
Comparison of stress results of experimental and FEM
Measuring locations and stress distribution at location of G1
Results of applying the delamination prediction method (Models A and B)
Results of applying the delamination prediction method (Models B)

## Table

Summary of tensile test results for PU foam
Summary of tensile test results for ABS
Test conditions for delamination experiments
Summary of results of delamination experiments

## Reference

1. Kim, J. K., Roh C. G., Kim, H. and Jeong, J. H. "An experimental and numerical study on an inherent capacity modulated linear compressor for home refrigerators," International Journal of Refrigeration, Vol. 34, pp. 1415-1423, 2011.
2. Park, J. K., "Optimization of Heat Insulation System for a Household Refrigerator," Korean Journal of Air-Conditioning and Refrigeration Engineering, Vol. 15, No. 2, pp. 95-103, 2003.
3. Choa, C. S., Choi E. H., Choa, J. R. and Lima, O. K., "Topology and parameter optimization of a foaming jig reinforcement structure by the response surface method," Computer-Aided Design, Vol. 43, pp. 1707-1716, 2011.
4. Baek, J. M., Won, J. R., Lee B. H. and Kim, J. H., "A study on contribution to UNFCCC caused by reinforcement of efficiency standard for residential refrigerators," The Korean Institute of Electrical Engineers, pp. 226-228, 2008.
5. Jeong, G. E., Kang, P., Youn, S. K., Yeo, I., Song, T. H., Kim, J. O., Kim, D. W. and Kuk, K., "Study of Structural Stiffness of Refrigerator Cabinet Using the Topology Optimization of a Vacuum Insulated Panel (VIP)," Journal of the Korean Society for Precision Engineering, Vol. 32, No. 8, pp. 727-734, 2015.
6. Yang, W. J., Lee, G. Y. and Park, S. H., "Analysis on Chemical and Physical Behaviors of Polyurethane Foam for Prediction of Deformation of Refrigerator Panels," International Journal of Precision Engineering and Manufacturing, Vol. 20, pp. 2041– 2049, 2019.
7. Cho, J. R., Kwak, G. Y. and Jung, J. H., "A Study on the Deformation near Duct of Refrigerator Cabinet," The Korea Society of Mechanical Engineers, pp. 250-253, 2008.
8. Zhai, J., Cho, J. R., Jeon, W. J. and Kim, J. H., "The Study for Bead Effect in Inner Case on Thermal Deformation of Refrigerator," Journal of the Korean Society for Precision Engineering, Vol. 28, No. 1, 96-101, 2011.
9. Mills, N. J., “Polymer Foams Handbook,” Elsevier, pp. 86-107, 2007.
10. ASTM D624-14, "Standard Test Method for Tear Strength of Conventional Vulcanized Rubber and Thermoplastic Elastomers," 2014.
11. Fishback, T. L., Reichel, C. J. and Lee, T. B., “Polyol composition having good flow and water blown rigid polyurethane foams made thereby having good dimensional stability,” US Patent, No. 5686500, 1997.
12. Landrock, A. H., "Handbook of Plastic Foam; Types, Properties, Manufacture and Applications," United States of America by Noycs PubIications, pp. 11-15, 1995.
13. Saint-Michel, F., Chazeau, L., Cavaillé, J. Y. and Chabert, E. "Mechanical properties of high density polyurethane foams: I. Effect of the density," Composites Science and Technology, Vol. 66, pp. 2700-2708, 2006.
14. Gama, N. V., Ferreira, A. and Barros-Timmons, A., "Polyurethane Foams: Past, Present, and Future," Materials, Vol., 11, No. 10, p. 1841, 2018.