Research and Findings in Engineering Sciences 2025 - III

RESEARCH AND FINDINGS IN ENGINEERING SCIENCES 2025 - III Editor Sarp Korkut SÜMER Lyon 2025 RESEARCH AND FINDINGS IN ENGINEERING SCIENCES 2025 - III Editor Sarp Korkut SÜMER Lyon 2025 Research and Findings in Engineering Sciences 2025 - III Editor • Prof. Dr. Sarp Korkut SÜMER • Orcid: 0000-0001-7679-6154 Cover Design • Motion Graphics Book Layout • Motion Graphics First Published • October 2025, Lyon e-ISBN: 978-2-38236-938-8 DOI: 10.5281/zenodo.17408988 copyright © 2025 by Livre de Lyon All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior written permission from the Publisher. The author or authors of the relevant section are responsible for any copyright infringement that may occur due to the images and graphics used in the book. The editor or publisher does not assume responsibility in this regard. Publisher • Livre de Lyon Address • 37 rue marietton, 69009, Lyon France website • http://www.livredelyon.com e-mail • [email protected] i PREFACE Engineering is the act of creating beneficial products for humans from natural resources using scientific and mathematical principles, experience, and creativity. With this approach, engineering brings together various disciplines to offer innovative solutions that simplify our lives and increase the quantity and quality of production. Today, considering the changing climate and evolving needs, interdisciplinary interaction has become more important, and studies that bring together different engineering disciplines have become more common. This nine-chapter book, “Research and Findings in Engineering Sciences 2025 - III,” brings together experimental, theoretical, and compilation studies conducted by scientists and engineers from various disciplines. i would like to thank our esteemed chapter authors for their scientific contributions to the book and engineering science, and the academics who contributed to the evaluation of these studies. i hope it will be useful to all interested readers, especially those in the engineering field. Editor Prof. Dr. Sarp Korkut SÜMER iii CONTENTS PREFACE i CHAPTER I. iNVESTiGATiON ON FiNNED-HEAT SiNK THERMAL PERFORMANCE ViA ANALYTiCAL AND NUMERiCAL METHODS 1 Ali GÜLSÜN & Şahin GÜNGÖR CHAPTER II. PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON ALGORiTHM (HEOA) 17 Emine BAŞ & Muhammet Ramazan ÇOPUR CHAPTER III. FREEZE–THAW DURABiLiTY AND ULTRASONiC CHARACTERiZATiON OF FiNE-GRAiNED WHiTE MARBLES FROM WESTERN TURKEY 49 Mustafa Yavuz ÇELİK CHAPTER IV. MODELLiNG NANOREFRiGERANT FLOWS iN EJECTORS: DETAiLED COMPUTATiONAL PROCEDURES FOR CONSTANT PRESSURE AND CONSTANT AREA EJECTORS 75 Candeniz SEÇKİN CHAPTER V. A BRiEF OVERViEW OF iNTELLiGiENT FOOD PACKAGiNG TECHNOLOGY 101 Demet AYDINOĞLU CHAPTER VI. BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL APPROACHES, REGULATiONS, AND SUSTAiNABiLiTY PERSPECTiVES 117 Ayfer ERGİN & M.Fatih ERGİN iv   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii CHAPTER VII. EVALUATiON OF CHANGES iN CAPiLLARY WATER ABSORPTiON DUE TO SALT AND SODA WATER CONCENTRATiONS iN POROUS BUiLDiNG STONES 141 Mustafa Yavuz ÇELİK CHAPTER VIII. THERMAL MANAGEMENT OF PV PANEL SYSTEMS ViA EXTENDED SURFACES AND COOLiNG JACKETS 163 Şahin GÜNGÖR CHAPTER IX. BiODiESEL PRODUCTiON iN PACKED-BED REACTORS USiNG Rhizomucor miehei LiPASE iMMOBiLiZED ON FiBROUS SUPPORTS 177 Nedim ALBAYRAK & Mehmet Erhan KANISLI iNVESTiGATiON ON FiNNED-HEAT SiNK THERMAL PERFORMANCE . . .   7 The Corsair ML120 Pro fan [Corsair user manual] produces an airflow rate of 75 CFM (0.0354 m³/s). The resulting air velocity is: 2 2 0.12 · 0.0113 2 fan Am π  = =   (3) fan Q vA = (4) According to the available data, the Reynolds (Re) number, Nusselt (Nu) number, and convective heat transfer coefficients are given as follows [Shah, R. K., & Sekulic, D. P. (2003); Thulukkanam, K. (2024)]: vH Re ρ µ = (5) Nu = 0.664 · Re½ · Pr1/3 (6) · air fin Nu k hH = (7) The above equations give us a convective heat transfer coefficient of h = 40.27 W/m2K. On the other hand, fin calculations have been carried out to design an effective thermal control system. In these calculations, the perimeter of each fin and the cross-sectional area of each fin were taken into account. ( ) 2 fin fin PHt = + (8) c fin hP mAk = (9) ( ) ( ) tanh fin f fin mH mH η = (10) Surface area of a single fin: , 2 fin single fin cpu A HL= (11) Total fin area, open base area, and total heat transfer surface area are given as follows: , fins fins fin single A NA= (12) base cpu fin fin cpu A A NtL= − (13) 8   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii total fins base AAA= + (14) Overall efficiency and processor temperature are calculated via: 1 · ( 1 ) fins of total A A ηη  =−−   (15) ( ) · · cpu cpu amb total o Q TT hA η  = +    (16) The determined parameters based on the above equations are shown in Table 2. Table 2: Analytical parameters used in the numerical investigations. Parameter Value Unit Number of Fins 13 - Base Width 0.040 m Fan Area 0.0113 m² Air Velocity (v) 3.13 m/s Reynolds Number 5763 - Nusselt Number 3.34 - Heat Transfer Coefficient (h) 40.27 W/m²·K Fin Efficiency 0.98 - Total Fin Area 0.08775 m² Open Base Area 0.00521 m² Total Area 0.0929 m² Overall Efficiency 0.979 - Processor Temperature 78.57 °C in the numerical point of view, heat transfer from the processor system to the finned heat sink system has been investigated by using ANSYS Fluent. in this phase, the parameters and results of the analytical solution were used in design modelling, mesh generation, and definition of boundary conditions. The intel Core i7-14700 processor dimensions are 45 mm × 37.5 mm. iNVESTiGATiON ON FiNNED-HEAT SiNK THERMAL PERFORMANCE . . .   9 Figure 1: Fluent structural mesh view. The model created for the analysis matched these dimensions. The fins of the heat sink were designed with a height of 30 mm, a thickness of 1.5 mm, and a spacing of 2 mm, with a total of 13 fins placed within the available area. The fins have a rectangular cross-section and are straight. The design demonstrates cooling performance through natural and forced convection assisted by the fan. in numerical analysis, mesh structure is a critical factor for accuracy. The sweep method was used to ensure accurate results. Mesh quality was optimized. The system provided a contact surface between the processor and the heat sink, and a high-density distribution of elements was achieved in this area (Fig. 1). Furthermore, the “Quad Dominance” method was chosen to prioritize rectangular elements on the surface. Boundary condition definitions in the ANSYS Fluent are given as follows: - Heat sink outer surfaces: The outer surfaces of the entire volume of the heat sink are named. The boundary conditions on the surface are determined as 298.15 K and the convective heat transfer coefficient is defined as the analytical result of h = 40.27 W/m² K. 10   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii - Processor walls: in this section, the contact surface of the processor is named. The boundary conditions on the surface are determined as 298.15 K and h = 40.27 W/m² K. - Processor interior: This section represents the inner volume of the processor. in this region, there is volumetric heat generation. in the “Cell Zone Conditions” section of ANSYS Fluent Setup, based on the processor’s power consumption of 65 W and the calculated processor volume, the heat flux was defined. The calculated volumetric heat generation is as follows: cpu vol cpu Q qV ′′ = (17) corresponding to 6 8.754 10 × W/m3. The material of both the processor and the heat sink was selected as aluminium. 3. Results and Discussion in this section, the results of analytical and numerical methods are compared. The closeness of the results of the methods is stated and the differences that occur after the application are examined for their reasons. In addition to the calculated temperatures, the target temperature was determined as 40°C or lower for the processor. Accordingly, an additional process was carried out to determine the convective heat transfer coefficient. The system was simulated in ANSYS using the CHT (Conjugate Heat Transfer) method. in this way, both the solid and fluid domains were solved in detail. For the fluid domain, air was selected as the working fluid, and the flow was defined as laminar. As a result of the simulation: • Maximum Processor Temperature: 75.8 °C • Minimum Surface Temperature: 35.7 °C When these results are compared with the analytical results obtained using MATLAB (where the calculated processor temperature was 78.57 °C), it is observed that the consistency between the results is at a high level. The maximum processor temperature difference was calculated to about 2.77°C, which shows that the approach is consistent. The reasons for the temperature difference are the ANSYS mesh sizes, the assumptions of the boundary conditions, and the iNVESTiGATiON ON FiNNED-HEAT SiNK THERMAL PERFORMANCE . . .   11 solution method algorithm used in the program combined with the modelling of the contact surfaces. Figure 2: CPU temperature variation with respect to air velocity. Figure 2 indicates that processor temperature decreases as air velocity increases, confirming that higher flow rate and heat transfer coefficient improve cooling. At 3.13 m/s, a sharp temperature drop occurs about 3 m/s, demonstrating that the selected fan provides sufficient airflow for the design. When the existing system was examined from a design perspective, it provided the opportunity for a rapid evaluation in terms of blade geometry and fan selection before CFD simulations. in addition to these calculations, another scenario was planned in this research. The objective of this plan is to reduce the system temperature further, with a new target set at 40 °C. For this purpose, an analytical solution was carried out with MATLAB using the reference equations. As a result, the convective heat transfer coefficient was calculated as h = 144.3 W/m² K. This value was kept constant for other parameters and the existing boundary conditions were applied with ANSYS Fluent as in the previous analysis. This allowed for investigation of the effect of the heat transfer coefficient on system behaviour. As a result of the simulation, Figure 3b shows that the maximum processor temperature is about 42 °C, and the minimum surface temperature is approximately 35.7 °C. Again, as in the previous step, the temperature difference between the two strategies is small enough to indicate that the results were 12   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii consistent. The calculated values and results of these procedures are presented in Table 3. Table 3: Analytical and numerical temperatures at various “h” values. Case Description h (W/ m²·K) Analytical Temperature Numerical (ANSYS Fluent) Temperature (Min–Max) Based on initial Fan Specifications 40.27 78.57 °C 69.1 °C – 75.8 °C Based on Target Temperature of 40 °C 144.33 40 °C 35.7 °C – 42.0 °C Figure 3: Temperature distribution for the cases of a) h = 40.27 W/m²·K and b) h = 144.33W/m²·K 4. CONCLUSION In this work, a finned heat sink has been designed for the purpose of strict thermal management of the intel Core i7-14700 processor. The thermal behaviour and performance of the system are examined using both analytical and numerical methods. As a result of the analytical calculations performed in MATLAB, the convective heat transfer coefficient was calculated as h = 40.27 W/m²·K based on the specifications of the Corsair ML120 PRO fan. First, the processor temperature was analytically calculated as 78.57 °C. Following these calculations, the heat transfer coefficient, the processor dimensions in the given dataset, the number of fins to be added on the plate that will be directly mounted onto the processor, the spacing between the fins, and the dimensions of iNVESTiGATiON ON FiNNED-HEAT SiNK THERMAL PERFORMANCE . . .   13 the fins have been determined. Next, the real case processor is transferred into the numerical world via ANSYS Fluent software. Note that identical boundary and initial conditions have been applied to the computational domain. As a result of the simulations, the maximum processor temperature is obtained as 75.8°C. According to the calculated and numerically obtained thermal data, the results are consistent for the 65 W processor heat sink design. in addition to these operations, an alternative scenario was planned during the research. This strategy is based on the existing system design, yet a target temperature of 40°C was set as the optimal operating temperature level, and the heat transfer coefficient required to reach this temperature was calculated. The heat transfer coefficient was determined and applied to the ANSYS Fluent interface as a boundary condition. The required convective heat transfer coefficient was analytically calculated as 144.33W/m² K. In this strategy, the maximum processor temperature is obtained maximum of 42°C, and the mean temperature is less than the threshold temperature level. Consequently, the current work uncovers that the thermal behaviour and performance of cooling systems can be predicted and accurately modelled by using a combination of analytical and numerical methods. Furthermore, it can be concluded that parameters such as fin geometry, fin arrangement, fan flowrate and pressure drop performance, and convective heat transfer coefficient have a direct influence on the cooling performance of the processors. References Alam, M. W., Bhattacharyya, S., Souayeh, B., Dey, K., Hammami, F., Rahimi-Gorji, M., & Biswas, R. (2020). CPU heat sink cooling by triangular shape micro-pin-fin: Numerical study. International Communications in Heat and Mass Transfer, 112, 104455. Cengel, Y. A., & Ghajar, A. J. (2020). (TEE-815): Heat and Mass Transfer: Fundamentals and Applications (Chap No. 15, 16 & 17 Includes). McGraw-Hill Education. Chu, W. X., Tsai, M. K., Jan, S. Y., Huang, H. H., & Wang, C. C. (2020). CFD analysis and experimental verification on a new type of air-cooled heat sink for reducing maximum junction temperature. International Journal of Heat and Mass Transfer, 148, 119094. Corsair ML120 PRO 120mm PWM Fan Teknik Özellikleri: https://www. corsair.com/us/en/p/case-fans/co-9050040-ww/ml120-pro-120mm-pwm- 14   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii premium-magnetic-levitation-fan-co-9050040-ww?srsltid=AfmBOooTyBXGp iU6ABn4x00TRNUUStSJMttg443Gstoje1zq8Qbq-WDowling, A., Jiang, L., Cheng, M. C., & Liu, Y. (2025). Regulating CPU temperature with thermal-aware scheduling using a reduced order learning thermal model. Future Generation Computer Systems, 166, 107687. Freegah, B., Hussain, A. A., Falih, A. H., & Towsyfyan, H. (2020). CFD analysis of heat transfer enhancement in plate-fin heat sinks with fillet profile: investigation of new designs. Thermal Science and Engineering Progress, 17, 100458. Fu, C., Zhang, W., Zhou, X., Shen, Q., & Wu, T. (2025). Stochastic optimization of photovoltaic-integrated data centers with hybrid cooling and waste heat recovery for district energy supply. Renewable Energy, 124248. Gurav, R. B., Purohit, P., Tamkhade, P. K., & Nalavade, S. P. (2023). Computational and analytical study on CPU heat sink cooling by single and double stack air-foil micro pin fins. Materials Today: Proceedings, 92, 6-10. Jeng, T. M. (2015). Combined convection and radiation heat transfer of the radially finned heat sink with a built-in motor fan and multiple vertical passages. International Journal of Heat and Mass Transfer, 80, 411-423. Gurav, R. B., Purohit, P., Tamkhade, P. K., & Nalavade, S. P. (2023). Computational and analytical study on CPU heat sink cooling by single and double stack air-foil micro pin fins. Materials Today: Proceedings, 92, 6-10. Haywood, A. M., Sherbeck, J., Phelan, P., Varsamopoulos, G., & Gupta, S. K. (2015). The relationship among CPU utilization, temperature, and thermal power for waste heat utilization. Energy Conversion and Management, 95, 297-303. Intel® Core™ i7-14700 Technical Specifications: https://www.intel.com/ content/www/us/en/products/sku/236781/intel-core-i7-processor-14700-33mcache-up-to-5-40-ghz/specifications.html Jeng, T. M. (2015). Combined convection and radiation heat transfer of the radially finned heat sink with a built-in motor fan and multiple vertical passages. International Journal of Heat and Mass Transfer, 80, 411-423. Kirchhofer, M., Krieger, M., & Hofer, D. (2025). Experimental and numerical investigations of forced convection impingement heat transfer on plate-fin heat sinks using three different electronic cooling fans. Applied Thermal Engineering, 258, 124665. Kosdere, O., Sert, Z., & Altun, O. (2024). investigation of thermal performance at forced convection in plate-fin heat sink. Energy, 307, 132621. iNVESTiGATiON ON FiNNED-HEAT SiNK THERMAL PERFORMANCE . . .   15 Li, C., & Li, J. (2023). Thermal characteristics of a flat plate pulsating heat pipe module for onsite cooling of high power server CPUs. Thermal Science and Engineering Progress, 37, 101542. Ma, X., Basem, A., Singh, P. K., Dara, R. N., Almadhor, A., Hajri, A. K., Babiker, S. G. (2025). Efficient cooling capability in microchannel heat sink reinforced with Y-shaped fins: Based on artificial neural network, genetic algorithm, Pareto front, and numerical simulation. Case Studies in Thermal Engineering, 68, 105936. Naduvilakath-Mohammed, F. M., Jenkins, R., Byrne, G., & Robinson, A. J. (2023). Closed loop liquid cooling of high-powered CPUs: A case study on cooling performance and energy optimization. Case Studies in Thermal Engineering, 50, 103472. Pathak, S., Luitel, A., Pasha, A. A., Reddy, D. S. K., Algarni, S., Alqahtani, T., & irshad, K. (2025). ML–augmented CFD–framework for heat-sink and fanspeed optimization in gaming workstation. International Journal of Mechanical Sciences, 110752. Rajan, A., Kanakkillam, A., Parasuram, A., KS, A., AR, A., & PR, S. (2025). Optimisation of the heat sink’s fin perforation shape using numerical methods for natural and forced convection in electronic cooling applications. Progress in Engineering Science, 2(2), 100084. Shah, R. K., & Sekulic, D. P. (2003). Fundamentals of heat exchanger design. John Wiley & Sons. Siahchehrehghadikolaei, S., Gholinia, M., Ghadikolaei, S. S., & Lin, C. X. (2022). A CFD modeling of CPU cooling by eco-friendly nanofluid and fin heat sink passive cooling techniques. Advanced Powder Technology, 33(11), 103813. Thulukkanam, K. (2024). Heat Exchangers: Classification, Selection, and Thermal Design. CRC Press. Vasilev, M. P., Abiev, R. S., & Kumar, R. (2021). Effect of circular pinfins geometry and their arrangement on heat transfer performance for laminar flow in microchannel heat sink. International Journal of Thermal Sciences, 170, 107177. Wu, J., Gao, L., Wang, L., Zhou, X., & Gao, J. (2025). A CPU-parallel thermal topology optimization framework for arbitrary structures using flood fill algorithm and implicit identification model. Computer Methods in Applied Mechanics and Engineering, 444, 118165. Yang, M., Mathew, G., Nemati, H., & Moghimi, M. A. (2024). A novel approach for active cooling of a battery at cell level: Air-cooled mini-channel 16   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii heat sink, enhanced with intermittent metal foam. Journal of Energy Storage, 81, 110374. Zhao, H., Ye, H., Liang, H., Liu, P., Asai, Y., Miyamoto, H., ... & Ho, J. Y. (2025). Development of conduction optimized heat sinks with enhanced fluid flow path for passive cooling applications. International Journal of Heat and Mass Transfer, 244, 126963. PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   23 HEOA Pseudo code and Flowchart: Algorithm 1: The Pseudo code of HEOA (Lian and Hui, 2024) Starting Stage: Logistic chaotic mapping is initialized according to Equation (1). The fitness value of the population to be obtained is calculated. For i = 1 to Maxiter For m = 1 to N Human Exploration Phase: If i <= (1 / 4) * Maxiter Then Xm is updated according to Equation (2). End Stage of Human Development: Else Leaders: For m = 1 to LeaderNumber (LS) Xm is updated according to Equation (6). End Explorers: For m = (LeaderNumber (LS) + 1) to (LeaderNumber (LS) + Number of Explorers (KS)) Xm is updated according to Equation (8). End Followers: For m = (LeaderNumber (LS) + Number of Explorers (KS) + 1) to (LeaderNumber (LS) + Number of Explorers (KS) + Number of Followers (TS)) Xm is updated according to Equation (9). End Losers: For m = (LeaderNumber (LS) + Number of Explorers (KS) + Number of Followers (TS) + 1) to N Xm is updated according to Equation (10). End Border Control and Updates are performed. End End Return Best 24   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Figure 1: The flow chart of HEOA (Lian and Hui, 2024) PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   25 3. Parameter Analyses Performed on 13 Standard Test Functions of HEOA: All studies were performed on a device with hardware and software features shown in Table 1. Parameter analysis was performed on 13 standard test functions of HEOA algorithm. Table 2 shows mathematical equations, search spaces and minimum fitness values of 13 single-mode and multi-mode standard test functions. it is important to use single-mode and multi-mode test functions in the study. Sometimes, algorithms that show superior success in single-mode test functions cannot have sufficient exploitation and discovery capabilities in multimode ones. Therefore, single-mode and multi-mode test functions frequently used in the literature were preferred while performing parameter analysis of HEOA. Table 1: Computer settings Name Feature Hardware CPU Core i7 Frequency 2.60GHz RAM 16 GB Software Operating System Windows 11 (64-bit) Language and iDE MATLAB R2022A 26   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Table 2: Mathematical Equations of Single and Multimodal Test Functions (F1-F7) Func. No Mathematical Equation Search Area Fmin F1(x) 2 1 n i i x = ∑[-100, 100] 0 F2(x) 11 nn ii ii xx = = + ∑∏ [-10, 10] 0 F3(x) 2 11 ni j ij x = −     ∑∑ [-100, 100] 0 F4(x) m a x { , 1 i x in ≤≤ [-100, 100] 0 F5(x) ( ) 12 22 1 1 100 ( 1) n ii i i xx x − + =  − +−   ∑ [-30, 30] 0 F6(x) ( ) 2 1 5 n i i x = + ∑[-100, 100] 0 F7(x) ( ) 4 1 [0,1) n i i i x random = ×+ ∑[-1.28, 1.28] 0 F8(x) ( ) 1 ( sin ) n ii i xx = − ∑[-500, 500] -418.9829 × dimension F9(x) ( ) 2 1 10cos 2 10 n ii Ý xx π =  −+  ∑ [-5.12, 5.12] 0 F10(x) 2 1 1 1 20exp( 0.2 1 exp( cos(2 )) 20 ) n i i n i i x n xe n π = = −− − ++ ∑ ∑ [-32, 32] 0 F11(x) 2 11 1cos( 1 400 nn i i ii x xi = = −+ ∑∏ [-600, 600] 0 PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   27 F12(x) ( ) ( ) ( ) ( ) ( ) 12 1 1 2 2 1 1 10sin 1 1 10 1 ,10,100,4 , n i i in n i i yy nsin y y x π π π µ − = + =  +−     + +−   + ∑ ∑ ( ) 1 1 , , , , 4 ( ) 0 ( ) i ii m ii i m ii x y x akm kx a x a ax a kxa x µ + = +  −>  = −< <  −−  [-50, 50] 0 F13(x) 22 1 1 2 22 1 0.1sin (3 ) ( 1) [1 sin (3 1)] ( 1) [1 sin (2 )]) ( , 5,1 00, 4) n i i i nn n i i xx x xx x π π π µ = = +− ++ +− + + ∑ ∑ [-50, 50] 0 3.1. Population Number Analysis for HEOA in the population number analysis of the HEOA algorithm, results were obtained on 13 classical test functions with dimension size (dim) = 30, maximum number of iterations (Maxiter) = 300, and 20 runs. The results are shown in Table 3 and Table 4. The superior results are marked in bold. According to the results shown in Table 3, it was understood that the success rate was higher when the population number was determined as 100. Comparable results could not be obtained in the F1, F2, F3, F4, and F10 functions. The best results were obtained with the values of N=70 for F5, N=90 for F6, N={80, 90, 100} for F7, N=100 for F8, N=60 for F9, N={90, 100} for F11, N=100 for F12, N={80, 90, 100} for F13. in the following studies, studies were carried out with the population value set to 30 instead of 100 due to the slow operation of the algorithm. 28   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Table 3: HEOA’s Population (N) = {10, 20, 30, 40, 50} analysis results (F1-F13) Func. No N10 20 30 40 50 F1(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F2(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F3(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F4(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F5(x) Best 2.98 1.79 0.96 1.24 0.73 Median 10.17 5.02 5.03 2.14 3.05 Mean 9.92 5.59 5.17 4.67 3.23 Worst 19.67 12.17 8.27 28.45 6.56 Standard Deviation 5.20 2.94 2.34 6.16 1.79 F6(x) Best 0.83 0.07 0.02 0.00 0.05 Median 6.42 3.27 0.89 1.29 0.16 Mean 6.14 3.17 1.21 1.54 0.38 Worst 7.50 7.03 3.45 4.98 1.18 Standard Deviation 1.65 1.86 0.96 1.52 0.36 PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   29 F7(x) Best 0.01 0.00 0.00 0.00 0.00 Median 0.06 0.04 0.02 0.03 0.02 Mean 0.08 0.05 0.04 0.04 0.02 Worst 0.26 0.19 0.13 0.17 0.05 Standard Deviation 0.07 0.05 0.04 0.04 0.02 F8(x) Best -6547.04 -7552.92 -7295.72 -7550.94 -7524.60 Median -4869.26 -5720.30 -6060.27 -6361.33 -6458.71 Mean -4912.00 -5725.13 -5915.17 -6168.87 -6404.45 Worst -3679.40 -4408.47 -3713.46 -4493.95 -5389.16 Standard Deviation 743.84 834.61 947.89 802.62 605.02 F9(x) Best 0.00 29.97 29.97 29.96 30.00 Median 31.22 32.02 30.94 30.95 30.52 Mean 36.52 33.59 32.16 32.28 31.89 Worst 63.17 47.97 43.65 45.02 45.28 Standard Deviation 17.53 4.80 3.25 3.69 4.34 F10(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F11(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.07 Mean 0.03 0.04 0.05 0.05 0.09 Worst 0.50 0.34 0.26 0.23 0.23 Standard Deviation 0.11 0.10 0.08 0.07 0.07 F12(x) Best 0.00 0.11 0.00 0.00 0.00 Median 1.55 1.05 0.43 0.41 0.31 Mean 1.81 1.54 0.88 0.73 0.48 Worst 4.58 4.21 3.75 3.12 1.27 Standard Deviation 1.41 1.24 1.18 0.82 0.44 F13(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.01 0.00 0.01 0.00 0.00 Mean 0.01 0.01 0.00 0.00 0.00 Worst 0.02 0.03 0.01 0.01 0.01 Standard Deviation 0.01 0.01 0.00 0.00 0.00 30   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Table 4: HEOA’s Population (N) = {60, 70, 80, 90, 100} analysis results (F1-F13) Func. No N60 70 80 90 100 F1(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F2(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F3(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F4(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F5(x) Best 0.54 0.50 0.69 0.62 0.59 Median 3.69 2.29 2.47 1.77 1.82 Mean 5.89 2.55 4.09 3.23 3.41 Worst 54.54 6.33 33.62 28.39 28.38 Standard Deviation 11.66 1.60 7.10 6.00 6.02 F6(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.09 0.10 0.04 0.03 0.01 Mean 0.20 0.30 0.07 0.06 0.05 Worst 0.70 3.14 0.36 0.28 0.44 Standard Deviation 0.25 0.70 0.09 0.09 0.10 PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   31 F7(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.01 0.01 0.01 0.01 0.01 Mean 0.02 0.01 0.01 0.02 0.01 Worst 0.08 0.04 0.04 0.05 0.04 Standard Deviation 0.02 0.01 0.01 0.01 0.01 F8(x) Best -8214.83 -8085.66 -8150.38 -8284.61 -8310.11 Median -6949.87 -7136.34 -6561.69 -6909.65 -7193.94 Mean -6735.08 -6953.81 -6639.50 -6917.59 -7231.39 Worst -4763.83 -5632.96 -5263.87 -5673.03 -5838.25 Standard Deviation 961.55 718.74 796.29 597.23 862.03 F9(x) Best 0.00 0.00 30.14 30.06 29.96 Median 30.42 30.48 30.45 30.52 30.20 Mean 28.99 29.02 30.54 30.64 30.84 Worst 31.46 36.26 31.54 33.07 36.02 Standard Deviation 6.83 7.02 0.39 0.71 1.59 F10(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F11(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.01 0.01 Mean 0.03 0.05 0.04 0.03 0.03 Worst 0.21 0.28 0.20 0.17 0.17 Standard Deviation 0.06 0.09 0.06 0.04 0.04 F12(x) Best 0.00 0.00 0.01 0.00 0.00 Median 0.23 0.12 0.23 0.04 0.11 Mean 0.40 0.27 0.27 0.19 0.13 Worst 1.70 1.04 0.60 1.25 0.48 Standard Deviation 0.51 0.31 0.20 0.33 0.12 F13(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.01 0.01 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 32   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii 3.2. Maximum Iteration Number Analysis for HEOA A detailed analysis process was performed for the maximum iteration value of the HEOA algorithm. In the maximum iteration analysis of the HEOA algorithm, results were obtained on 13 classical test functions with dimension size (dim) = 30, population number (N) = 30, and 20 repetitions (runs). The results are shown in Table 5. The superior results are marked in bold. in the evaluation made according to the results shown in Table 5, it was understood that the success rate was higher when the maximum iteration number (Maxiter) was determined as 5000. As expected, the success rate increased in general for the relevant functions as the maximum iteration number was increased. Table 5: HEOA Max. Iteration (Maxiter) = {100, 300, 500, 1000, 5000} Analysis Results (F1-F13) Func. No Max. Iteration 100 300 500 1000 5000 F1(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F2(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F3(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F4(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   39 F7(x) Best 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 F8(x) Best -7896.23 -7073.91 -7894.77 -7140.03 Median -6419.36 -6055.04 -6178.31 -6207.75 Mean -6358.35 -5943.40 -6234.65 -6063.07 Worst -4200.72 -4761.48 -4178.54 -3985.52 Standard Deviation 941.10 580.21 1026.11 885.88 F9(x) Best 30.04 0.00 30.07 29.92 Median 30.61 30.87 30.93 30.54 Mean 31.31 31.63 35.25 32.84 Worst 40.07 48.85 62.81 53.54 Standard Deviation 2.36 9.01 9.05 5.69 F10(x) Best 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 F11(x) Best 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 Mean 0.11 0.07 0.04 0.07 Worst 0.57 0.43 0.28 0.48 Standard Deviation 0.17 0.12 0.08 0.13 F12(x) Best 0.01 0.00 0.00 0.02 Median 0.39 0.44 0.33 0.44 Mean 0.81 0.94 0.56 0.90 Worst 3.46 4.42 2.10 4.20 Standard Deviation 0.00 1.27 0.70 1.17 F13(x) Best 0.00 0.00 0.00 0.00 Median 0.00 0.01 0.00 0.00 Mean 0.00 0.01 0.00 0.01 Worst 0.00 0.02 0.01 0.02 Standard Deviation 0.00 0.00 0.00 0.01 40   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii 3.5. Analysis of HEOA with Best Parameter Properties in Different Dimensions The analysis of HEOA algorithm in different dimensions with the best parameters was performed. The results were obtained on 13 classical test functions with HEOA algorithm skip coefficient ratio = 10000, population number (N) = 30, maximum iteration number of 300, and 20 repetitions (runs). The results are shown in Table 8. According to the results shown in Table 8, the success decreases as the size increases in the evaluation. PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   41 Table 8: HEOA’s Analysis Results in Different Dimensions (F1-F13) Func. No Dimension (dim) 10 30 50 500 1000 F1(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F2(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F3(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F4(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F5(x) Best 0.07 0.88 1.65 118.28 533.43 Median 1.07 2.53 9.45 521.44 224.93 Mean 4.11 3.67 10.66 500.18 199.48 Worst 8.63 14.06 21.00 896.55 2083.15 Standard Deviation 4.16 3.19 5.22 211.02 399.25 F6(x) Best 0.00 0.02 0.24 50.65 64.32 Median 0.00 1.16 5.29 110.34 224.93 Mean 0.01 2.14 5.08 105.28 199.48 Worst 0.15 6.25 11.61 124.66 248.64 Standard Deviation 0.07 2.00 3.37 21.73 58.94 F7(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 42   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii F8(x) Best -3449.8 -10267.2 -16824.8 -142077 -330800 Median -2448.4 -6376.01 -9805.6 -85630.8 -136530 Mean -2478.8 -6589.8 -9338.2 -80510.3 -155310 Worst -1539.9 -3149.6 -5714.8 -34901 -57842 Standard Deviation 478.91 2246.17 2524.26 27393.2 71678 F9(x) Best 0.00 0.00 50.48 538.97 1331.72 Median 10.07 30.66 51.42 792.42 1489.87 Mean 7.76 28.07 55.81 798.66 1501.34 Worst 12.06 33.73 89.24 1123.26 1720.87 Standard Deviation 4.62 9.67 10.91 114.60 104.59 F10(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.00 0.00 0.00 0.00 0.00 Worst 0.00 0.00 0.00 0.00 0.00 Standard Deviation 0.00 0.00 0.00 0.00 0.00 F11(x) Best 0.00 0.00 0.00 0.00 0.00 Median 0.00 0.00 0.00 0.00 0.00 Mean 0.04 0.00 0.00 0.00 0.00 Worst 0.74 0.00 0.00 0.00 0.00 Standard Deviation 0.17 0.00 0.00 0.00 0.00 F12(x) Best 0.00 0.00 0.00 0.27 0.36 Median 0.00 0.37 0.47 1.14 4.14 Mean 0.10 0.60 0.63 1.39 4.65 Worst 0.77 1.63 1.68 4.32 19.01 Standard Deviation 0.21 0.58 0.50 1.04 4.66 F13(x) Best 0.00 0.00 0.00 0.11 0.31 Median 0.00 0.00 0.00 0.26 0.72 Mean 0.00 0.00 0.01 0.30 0.86 Worst 0.00 0.01 0.02 0.62 1.64 Standard Deviation 0.00 0.00 0.0 0.14 0.37 3.6. Analysis of HEOA with Best Parameter Properties on CEC-2017 Test Functions The analysis of HEOA algorithm with the best parameters in different dimensions was performed. The results were obtained on 29 CEC-2017 test functions with HEOA algorithm skip coefficient ratio = 10000, population number (N) = 30, maximum iteration number of 300, and 20 repetition runs. CEC-2017 test function definitions are shown in Table 9 (Awad et al., 2016). The results are shown in Table 10. PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   43 Table 9: CEC-2017 Test Functions (Awad et al., 2016) No Functions F* Single Mode Functions F1 Shifted and Rotated Bent Cigar Function 100 F3 Shifted and Rotated Zakhrov Function 300 Simple Multi-Mode Functions F4 Shifted and Rotated Rosenbrock’sFunction 400 F5 Shifted and Rotated Rastrigin’s Function 500 F6 Shifted and Rotated Expanded Scaffer’s F6 Function 600 F7 Shifted and Rotated Lunacek Bi_Rastringin’s Function 700 F8 Shifted and Rotated Non-Continuous Rastringin’s Function 800 F9 Shifted and Rotated Levy Function 900 F10 Shifted and Rotated Schwefel’s Function 1000 Hybrid Functions F11 Hybrid Function 1 (N = 3) Zakhrov, Rosenbrock’s, Rastrigin’s 1100 F12 Hybrid Function 2 (N = 3) High-Conditioned Elliptic, ModifiedSchwefel’s, Ben Cigar 1200 F13 Hybrid Function 3 (N = 3) Ben Cigar, Rosenbrock’s, Lunacek Bi_Rastringin’s 1300 F14 Hybrid Function 4 (N = 4) High-Conditioned Elliptic, Ackley, Schaffer’s F7, Rastringin’s 1400 F15 Hybrid Function 5 (N = 4) Ben Cigar, HGBat, Rastringin’s, Rosenbrock’s 1500 F16 Hybrid Function 6 (N = 4) Expanded Schaffer’s F6, HGBat, Rosenbrock’s, Modified Schwefel’s 1600 F17 Hybrid Function 6 (N = 5) Katsuura, Ackley, Expanded Griewank’s plus, Rosenbrock’s, Schwefel’s, Rastringin’s 1700 F18 Hybrid Function 6 (N = 5) High-Conditioned Elliptic, Ackley, Rastringin’s, HGBat, Discus 1800 F19 Hybrid Function 6 (N = 5) Bent Cigar, Rastringin’s, Griewank’s plus Rosenbrock’s, Weierstrass, Expanded Schaffer’s F6 1900 F20 Hybrid Function 6 (N = 5) HappyCat, Katsuura, Ackley, Rastringin’s, Modified Schwefel’s, Schaffer F7 2000 44   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Composition Functions F21 Composition Function 1 (N = 3) Rosenbrock’s, High Conditioned Elliptic, Rastringin’s 2100 F22 Composition Function 1 (N = 3) Rastringin’s, Griewank’s, Modified Schwefel’s 2200 F23 Composition Function 1 (N = 4) Rosenbrock’s, Ackley, Modified Schwefel’s, Rastringin’s 2300 F24 Composition Function 2 (N = 4) Ackley, HighConditioned Elliptic, Griewank’s,Rastringin’s 2400 F25 Composition Function 3 (N = 5) Rastringin’s, HappyCat, Ackley Discus, Rosenbrock’s 2500 F26 Composition Function 4 (N = 5) Expanded Schaffer’s F6, Modified Schwefel’s, Griewank’s, Rosenbrock’s,Rastringin’s 2600 F27 Composition Function 5 (N = 6) HGBat, Rastringin’s, Modified Schewel’s, Bent Cigar, High-Conditioned Elliptic,Expanded Schaffer’s F6 2700 F28 Composition Function 6 (N = 6) Ackley, Griewank, Discus, Rosenbrock, HappyCat, Expanded Schaffer’s F6 2800 F29 Composition Function 7 (N = 3) F15,F16, F17 2900 F30 Composition Function 8 (N = 3) F15, F18, F19 3000 Search Area: [ ] 100,100 dim − PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   45 Table 10: HEOA’s Analysis Results in Different Dimensions (CEC-2017 Test functions) Func. No Dimension Size: 10 30 50 F1(x) Mean 5624743599.44 43523641083.50 101301101574.79 Standard Deviation 2062129443.59 7346510263.05 9220547279.07 F3(x) Mean 16871.95 88967.18 219236.40 Standard Deviation 2533.57 4065.42 12396.08 F4(x) Mean 676.77 8442.30 27964.36 Standard Deviation 166.48 2317.33 6319.25 F5(x) Mean 596.30 940.16 1236.62 Standard Deviation 23.84 32.63 26.91 F6(x) Mean 652.50 694.62 703.35 Standard Deviation 10.68 6.68 2.99 F7(x) Mean 827.38 1494.45 2116.11 Standard Deviation 22.99 38.21 34.33 F8(x) Mean 860.58 1164.12 1590.81 Standard Deviation 11.06 21.04 26.77 F9(x) Mean 1869.63 13269.28 42458.61 Standard Deviation 349.38 1285.82 3684.92 F10(x) Mean 2837.96 9424.18 15826.08 Standard Deviation 247.44 458.83 584.95 F11(x) Mean 4204.82 10677.91 24678.30 Standard Deviation 2240.36 2407.95 2918.21 F12(x) Mean 27966114.93 6705888089.07 57174109399.94 Standard Deviation 21943842.05 2331800181.13 12828024036.91 46   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii F13(x) Mean 279825.64 3551180181.54 30932770933.49 Standard Deviation 422028.17 2002975782.13 7504294981.36 F14(x) Mean 5019.44 5290728.28 48922041.90 Standard Deviation 5392.63 2937645.23 31203182.53 F15(x) Mean 12301.24 494201586.30 3357710722.85 Standard Deviation 5060.65 285857811.73 1974257301.45 F16(x) Mean 2208.51 5366.40 8780.47 Standard Deviation 129.42 785.89 1139.12 F17(x) Mean 1896.47 3396.56 7384.81 Standard Deviation 103.08 756.29 3171.41 F18(x) Mean 3425022.89 50329143.67 162384846.48 Standard Deviation 5525018.27 29843693.13 66716891.73 F19(x) Mean 1594386.45 233936864.23 1526558769.35 Standard Deviation 4617620.63 160277352.49 887187049.61 F20(x) Mean 2288.15 3294.60 4423.19 Standard Deviation 89.33 158.05 224.49 F21(x) Mean 2389.81 2726.28 3160.47 Standard Deviation 17.56 41.61 69.46 F22(x) Mean 2585.30 9140.46 17172.15 Standard Deviation 188.67 1276.46 671.05 F23(x) Mean 2754.72 3639.30 4630.35 Standard Deviation 34.25 282.20 375.93 F24(x) Mean 2800.78 3560.10 4382.39 Standard Deviation 115.34 136.74 239.72 PARAMETER ANALYSiS OF HUMAN EVOLUTiON OPTiMiZATiON . . .   47 F25(x) Mean 3181.77 4202.74 11861.93 Standard Deviation 107.36 297.92 1210.05 F26(x) Mean 3847.38 10305.87 17168.46 Standard Deviation 397.35 1202.29 892.50 F27(x) Mean 3195.84 3200.01 3200.01 Standard Deviation 11.05 0.00 0.00 F28(x) Mean 3299.98 3300.01 3300.01 Standard Deviation 0.09 0.00 0.00 F29(x) Mean 3436.00 6969.07 24416.29 Standard Deviation 107.26 799.84 10721.94 F30(x) Mean 5338061.42 651767752.86 4064876761.06 Standard Deviation 4360735.04 468450473.73 1888969381.22 4. Conclusions Optimization is a technology that allows a process to achieve the best result by using the available resources in the best way. Metaheuristic algorithms do not aim to reach the optimal result, but to reach a result close to the optimal result in a shorter time compared to classical algorithms. in this study, a detailed parameter analysis was performed on the classical test functions of the Human Evolution Optimization Algorithm (HEOA), a new metaheuristic algorithm. The inspiration for HEOA is the extraordinary adaptability of human evolution and the ability to find the best solutions in complex environments. Human evolution has served as a driving force for human survival and progress, and has demonstrated the effectiveness of natural selection and adaptation. HEOA was first introduced by Lian and Hui in 2024. It was noticed that no detailed parameter analysis was performed for HEOA in the original study. The focus of this study is to select the best parameter settings for HEOA. Population number analysis, maximum iteration analysis, skip coefficient ratio, and population distribution ratio parameters were analyzed on 13 standard test functions of HEOA, and the results are presented in the tables. Thus, this study will be helpful in parameter selection for researchers who will use HEOA. 48   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii References Dokeroglu, T., Sevinc, E., Kucukyilmaz, T., & Cosar, A. (2019). A survey on new generation metaheuristic algorithms. Computers & industrial Engineering, 137, 106040. Holland, J. H. (1975). Adaptation in natural and artificial systems. University of Michigan Press. Karaboga, D. (2005). An idea based on honey bee swarm for numerical optimization (Technical Report-TR06). Engineering Faculty, Computer Engineering Department, Erciyes University. Kennedy, J., & Eberhart, R. (1995). Particle swarm optimization. Proceedings of iEEE international Conference on Neural Networks, 4, 19421948. Kirkpatrick, S., Gelatt, C. D., & Vecchi, M. P. (1983). Optimization by simulated annealing. Science, 220(4598), 671-680. Lian, J., & Hui, G. (2024). Human evolutionary optimization algorithm. Expert Systems with Applications, 241, 122638. Nesterov, Y. (1983). A method for unconstrained convex minimization problem with the rate of convergence O(1/k^2). Doklady Akademii Nauk SSSR, 269, 543-547. Rajabioun, R. (2011). Cuckoo Optimization Algorithm. Applied Soft Computing, 11(8), 5508-5518. Storn, R., & Price, K. (1997). Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces. Journal of Global Optimization, 11(4), 341-359. Yang, X. S. (2012). Hunger games search: An efficient metaheuristic algorithm for global optimization. international Journal of Bio-inspired Computation, 4(2), 78-84. Yildiz, B. S., Kumar, S., Panagant, N., Mehta, P., Sait, S. M., Yildiz, A. R., Pholdee, N., Bureerat, S., & Mirjalili, S. (2023). A novel hybrid arithmetic optimization algorithm for solving constrained optimization problems. Knowledge-Based Systems, 271, 110554. Awad, N., Ali, M., Liang, J., Qu, B., Suganthan, P. (2016). “Problem definitions and evaluation criteria for the CEC 2017 special session and competition on single objective real-parameter numerical optimization”. Zhengzhou University, Zhengzhou. FREEZE–THAW DURABiLiTY AND ULTRASONiC CHARACTERiZATiON . . .   55 Fig. 3. Schematic representation and measurement of ultrasound transmission velocity according to the structural properties of the rock. 3. Experimental Results and Discussions 3.1. Chemical Analysis Chemical analysis conducted on Afyon sugar, Muğla sugar, Muğla white, and Uşak white marbles provides valuable insights into their composition. The results presented in Table 2 reveal that CaO was the predominant component in all the tested marbles, with a significant proportion ranging from 51.7% to 55.6%. This finding is consistent with a typical marble composition, in which calcium carbonate (CaCO3) is the primary mineral. The analysis also showed that the concentrations of SiO2, Al2O3, and Fe2O3 in all marbles were relatively low and fell within the standard ranges. These components are commonly found in trace amounts as impurities in natural stones and do not significantly affect their overall composition. One notable disparity among the marbles was the higher MgO content observed in the Uşak white marble than in the others. The MgO concentration in the Uşak white marble was measured to be 2.78%, which is notably higher than the range of 0.18% to 0.29% found in the remaining marbles. This high MgO content can be attributed to the process of dolomitization, where magnesiumrich fluids infiltrate limestone during its formation, replacing some calcium carbonate with magnesium carbonate (dolomite). This geological phenomenon accounts for the variation in the composition of the Uşak white marble and distinguishes it from the other tested marbles. 56   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Table 2. Chemical composition of the tested marbles Composition (%)Afyon sugar Muğla sugar Muğla white Uşak white SiO20.60 0.08 0.06 0.47 Al2O30.30 0.03 0.01 0.05 Fe2O30.05 0.28 0.03 0.21 MgO 0.18 0.29 0.22 2.78 CaO 55.20 54.99 55.6 51.70 Na2O 0.05 0.01 0.02 0.18 K2O 0.05 0.04 0.01 0.08 LOi 43.50 42.09 43.86 43.85 3.2. Mineralogical and Petrographic Analysis 3.2.1. Polarizing Optical Microscope Analyses The mineralogical composition of the marbles was assessed through petrographic studies conducted on thin sections, and the results are presented in Figure 4, which includes images captured using a polarizing microscope. Microscopic examination of the Afyon sugar, Muğla sugar, Muğla white, and Uşak white marbles revealed that calcite was the main mineral present in all marbles. Polysynthetic twinning was also observed in the examined marbles. The marbles exhibited a granoblastic texture, characterized by equidimensional calcite crystals in each sample. Notably, the calcite grain boundaries in Afyon sugar and Uşak white marbles appeared indented because of the pressure experienced during their formation. This feature positively impacts the compressive strength of marble, enhancing its resistance to atmospheric effects when used in outdoor environments. In contrast, the boundaries of the calcite grains in the Muğla and Muğla white marbles were found to be more regular. These marbles exhibited larger and smoother grain borders, which in turn decreased the resistance of the marble against external influences and facilitated its decomposition over time. FREEZE–THAW DURABiLiTY AND ULTRASONiC CHARACTERiZATiON . . .   57 Fig. 4. Thin section views of marbles: Afyon sugar (A-B), Muğla sugar (C-D), Muğla white (E-F), Usak white (G-H), (A, C, E, G crossed nicols and B, D, F, H single nicols) 3.2.2. Grain Size Analysis in the Microscope The grain size of the marbles is a critical factor that significantly influences their properties and applications. Grain size is crucial in measuring marbles, as it directly impacts various aspects such as strength, weathering resistance, suitability for specific uses, brightness, and polishing properties. The economic value of marble tends to increase as the grain size decreases, aligning with improvements in the physical and mechanical properties. However, coarse grain sizes and flat grain boundaries result in marbles with a lower strength and reduced resistance to atmospheric effects. In contrast, marbles with acceptable 58   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii grain sizes and intricate grain boundaries exhibit enhanced strength and increased durability against weathering. The classification of marbles based on grain size encompasses four categories: very fine (<0.1 mm), fine (0.1–2 mm), medium (2–5 mm), and coarse (>5 mm) grains (Çelik, 2004). The textural features of marble provide valuable information regarding the size, shape, and distribution of calcite grains, which constitute the primary components of these rocks. Furthermore, they offer insights into geological events, such as diagenesis and metamorphism, which have influenced the formation of marbles. The characteristics of the grain boundaries influence the marble texture. Marbles with non-smooth and indented grain boundaries exhibited softer natures. in such marbles, the presence of intricate grain boundaries resembling lace-like indentations extending into adjacent grains adds complexity to the structure. Conversely, marbles with coarse grain-size distributions and flat grain boundaries tended to have lower strengths. However, marbles with fine grain size distributions and indented grain boundaries demonstrate improved strength (Figure 5) (Çelik et al., 2011). Fig. 5. Microscopic examination of marbles depicting the relationship between grain size and grain boundaries: (A) fine-grained with lace-like indented grain boundaries, (B) coarse grains with smooth grain boundaries, (C) large grains with lace-like indented grain boundaries, and (D) fine-grained with smooth grain boundaries (Çelik et al., 2011). Particle size analysis of the examined marbles was conducted using thin sections, which enabled the counting of calcite grains in each marble sample. A minimum of 30 calcite grains were counted to ensure representative results. The obtained data regarding grain size were analyzed, and the average grain sizes for Afyon sugar, Muğla sugar, Muğla white, and Uşak white marbles were determined as 0.40 mm, 0.85 mm, 1.72 mm, and 0.88 mm, respectively (Table 3). Afyon Sugar exhibited the smallest grain size among these marbles, while Muğla white marble demonstrated the largest grain size. However, it is FREEZE–THAW DURABiLiTY AND ULTRASONiC CHARACTERiZATiON . . .   59 worth noting that all examined marbles fall within the category of fine-grained marbles, as their grain sizes range between 0.1 mm and 2 mm. The grain size of the marbles is important for evaluating their properties and potential applications. In this study, the marbles analyzed exhibited relatively small grain sizes, indicating their fine-grained nature. Fine-grained marbles possess desirable physical and mechanical properties such as improved strength and enhanced resistance to weathering. Classifying these marbles as finegrained further suggests their suitability for various uses, including construction and decoration. Table 3. The average grain size of marbles SampleGrain size (mm) max. min. mean Afyon sugar 0.78 0.23 0.40 fine-grained Muğla sugar 1.44 0.30 0.85 fine-grained Muğla white 2.88 0.38 1.72 fine-grained Uşak white 1.77 0.27 0.88 fine-grained 3.3. Mechanical and Physical Properties To evaluate the physical properties of Afyon sugar, Muğla sugar, Muğla white, and Uşak white marbles, various tests were conducted, including density, water absorption by weight, water absorption by volume, open pores, ultrasound wave velocity, and frost loss. Six samples were prepared from each marble group, and the average results are presented in Table 3. The determination of physico-mechanical properties plays a vital role in the context of extensive mining and civil engineering, including the construction of dams, tunnels, and mines. These properties provide essential information regarding the behavior and characteristics of materials, enabling informed decision-making and ensuring the integrity and success of large-scale projects. Moreover, natural stones used in construction must possess specific properties suitable for their intended usage areas. Properties such as porosity, water absorption, and density are closely related to the mechanical strength of rocks, and low-density and highly porous rocks tend to be less stable. The water absorption values, whether by weight or volume, remained within a relatively narrow range across all marbles. Open pores in the marbles influenced the ultrasound wave velocities, with higher pore numbers correlating with lower wave velocities. The recorded density values indicate minor 60   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii variations among the different marble types. The densities of the tested marbles were 2706 kg/m³ for Afyon sugar, 2700 kg/m³ for Muğla sugar, 2702 kg/m³ for Muğla white, and 2722 kg/m³ for Uşak white marble. The water absorption values, weight, and volume ranged between 0.08% and 0.25% for all marbles. The observed pore rates fell within the range of 0.21% to 0.25%. The ultrasound wave velocities varied depending on the number of pores in the marbles. The highest number of pores was found in Muğla sugar marble, which exhibited an ultrasound wave speed of 3.35 km/s. Conversely, Uşak white marble, which had the least number of pores, demonstrated the highest ultrasound wave velocity of 4.68 km/s. Knowledge of these physical properties is crucial for making informed decisions regarding the selection and utilization of marble in various industries including construction, mining, and civil engineering. By considering factors such as the density, water absorption, and porosity, stakeholders can ensure the appropriate use of marbles based on their mechanical strength and stability requirements. Table 3. Average physico-mechanical properties of tested marbles (6 samples used in each experiment) Afyon sugar Muğla sugar Muğla white Uşak white Density (kg/m3) 2706 2700 2702 2722 Water absorption by weight (%) 0.08 0.09 0.09 0.08 Water absorption by volume (%) 0.23 0.25 0.24 0.21 Porosity (%) 0.23 0.25 0.24 0.21 Ultrasound pulse velocity (km/s) 4.48 3.35 4.30 4.68 Frost resistance 0.027 0.034 0.031 0.026 3.3.1. Ultrasound Pulse Velocity Experiment Ultrasound techniques are utilized to determine various technological properties of rocks, offering the advantages of easy application and nondestructive testing. Properties such as rock type, texture, grain size and shape, porosity, density, water content, and anisotropy significantly influence ultrasonic wave velocity. in addition, factors such as weathering, alteration zones, and joint properties (e.g., water presence, filling material, roughness, strike, and slope) are crucial parameters that affect the ultrasound velocity. To investigate the changes in the ultrasound properties of Afyon sugar, Muğla sugar, Muğla white, and Uşak white marbles, ultrasonic wave velocity FREEZE–THAW DURABiLiTY AND ULTRASONiC CHARACTERiZATiON . . .   61 measurements were conducted on dry, saturated, and frozen samples. The ultrasound wave velocity data obtained are shown in Fig. 6. Across all samples, the measured ultrasound velocities gradually increased under the dry, saturated, and freezing conditions. The primary factor contributing to these increases was water infiltration into the marble pores, followed by subsequent freezing and ice formation. Precisely, the ultrasound velocity of dry Afyon sugar marble was measured at 4.48 km/s, which increased to 5.56 km/s in the saturated state and 5.74 km/s in the frozen state. This represents an increase of 23.89% and 28.12% in ultrasound velocity for Afyon sugar marble, respectively. Similarly, the increase rates for Muğla sugar marbles were 48.49% and 71.48%, for Muğla white marbles were 27.65% and 31.84%, and for Uşak white marbles were 4.15% and 8.29%, respectively. The smallest increase was observed in the Uşak white marbles, while the highest increase was found in the Muğla sugar marbles. These findings align with the pore ratios observed in the samples, indicating that the sound velocity increases proportionally with the filling of porous structures and cracks in the marble. in conclusion, this study demonstrates that the ultrasound wave velocities of Afyon sugar, Muğla sugar, Muğla white, and Uşak white marbles are influenced by moisture conditions, with higher velocities observed in the saturated and frozen states than in the dry state. These results underscore the importance of water saturation and ice crystallization in modifying the acoustic properties, with implications for assessing freeze-thaw durability using non-destructive methods. The obtained data contribute to our understanding of marble behavior under different environmental conditions, and can aid in the selection and application of marbles in various engineering and construction projects. Ultrasound velocity measurements conducted on Afyon sugar, Muğla sugar, Muğla white, and Uşak white marbles revealed increased ultrasound properties under saturated and frozen conditions compared to the dry state. increases in ultrasound velocity were observed in all marble types with varying magnitudes. The highest increase in ultrasound velocity was observed in the Muğla sugar marbles, indicating significant filling of the porous structure and cracks. This finding aligns with the higher pore ratios of the marbles. Conversely, the Uşak white marbles exhibited the smallest increase in ultrasound velocity, suggesting a lower pore ratio and a relatively denser structure. The results highlight the influence of the marble’s porous structure and cracks on its ultrasound properties. The filling of these voids led to an increase in the ultrasound velocity, indicating a potential improvement in the mechanical properties of the marbles. Understanding these changes in the 62   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii ultrasound properties can aid in the characterization and assessment of marble materials and inform their use in various applications. Fig. 6. Change of ultrasound properties of marbles in the dry, water-saturated, and frozen state. 3.3.2. Freeze-Thaw Experiments The objective of the freeze-thaw tests was to assess the ability of white marbles from the Aegean region of Turkey to withstand cold climate conditions. These tests play a vital role in comprehending natural stones’ durability and longterm performance in outdoor applications, given their heightened vulnerability to the adverse impacts of freeze-thaw cycles. The marble samples were subjected to 25 freeze-thaw cycles throughout the testing process, replicating seasonal fluctuations and the repetitive freezing and thawing mechanism. The freeze-thaw cycles exposed the marbles to alternating temperature variations, causing the water within their pores to expand and contract. The results obtained from the freeze-thaw tests yield valuable insights into the response of the white marbles under cold climate conditions. Visual observations and measurements were conducted to evaluate the extent of damage and modifications in physical properties, including surface roughness, weight loss, and dimensional stability. These findings provide crucial information regarding the behavior and performance of the marbles when exposed to freezethaw cycles, aiding in assessing their suitability for outdoor applications. FREEZE–THAW DURABiLiTY AND ULTRASONiC CHARACTERiZATiON . . .   63 3.3.2.1. Changes in Ultrasound Pulse Velocity Freeze-thaw tests of the samples were carried out according to TS EN 12371 (2003). The changes occurring during the freeze-thaw cycles of the examined marbles were investigated using ultrasound. For this purpose, the ultrasound pulse velocity rates of the frozen and thawed samples were measured. The changes in ultrasound pulse velocity rates at the end of the freeze-thaw cycles in marbles are shown in Figure 7. The frozen ultrasound data of all samples were higher than the saturated data. The most significant difference between the ultrasound pulse velocities measured in the saturated and frozen states was measured in the Muğla sugar marble, and the least difference was measured in the Afyon sugar marble. This is consistent with both the porosity and frost loss data. In the Muğla sugar marble, which had the highest porosity, the greatest difference between the ultrasound pulse velocities measured in the saturated and frozen states occurred. Fig. 7. Changes in ultrasound pulse velocity rates in marbles at the end of freeze-thaw cycles (Straight lines are frozen, dashed lines are water-saturated data). (AS: Afyon sugar, MS: Muğla sugar, MW: Muğla white, and UW: Uşak white) Physical changes in the samples during freeze-thaw experiments can cause changes in water absorption and ultrasound data. The ultrasound pulse velocity rates of water-saturated marble samples in freeze-thaw cycles are shown in Fig. 8 on a sample basis. 64   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Fig. 8. Water changes saturated ultrasound pulse velocity according to the freeze-thaw cycles of the tested marble samples. It was observed that all samples experienced a decrease in ultrasound velocity as a result of the freeze-thaw cycles. This reduction can be attributed to the formation of microcracks and an increase in partial pores caused by freezethaw cycles. These factors were also correlated with the decrease in weight and increased water absorption observed in the samples. Comparing the different marble types, it is evident that the extent of the decrease in the ultrasound velocity varied. The highest reduction was observed in the Muğla sugar marble, with a decrease of 6.89%, whereas the Uşak white marble exhibited a minor change, with a decrease of 3.61%. Afyon sugar and Muğla white marbles showed intermediate reductions of 3.65% and 6.20%, respectively. Furthermore, it is noteworthy that the decrease in the ultrasound velocity measured in the frost followed a similar pattern. The reduction percentages at the water-saturated ultrasound speeds were 3.49%, 2.59%, 3.57%, and 1.63% for Afyon, Muğla, Muğla, and Uşak white samples, respectively. These findings highlight the susceptibility of marble samples to damage caused by freeze-thaw cycles. The formation of microcracks and increased pore sizes contributed to the decrease in the ultrasound pulse velocity, indicating a deterioration in the mechanical properties of the marbles. The variations observed among the FREEZE–THAW DURABiLiTY AND ULTRASONiC CHARACTERiZATiON . . .   71 Çelik MY, Yeşilkaya L, Ersoy M, Turgut T (2011) Karbonat kökenli doğal taşlarda tane boyu ile Knoop sertlik değeri arasındaki ilişkinin incelenmesi, Madencilik Dergisi, 50 (2), 29-40. Çelik MY, Akbulut H, Ergül A (2014) Water absorption process effect on strength of Ayazini tuff, such as the uniaxial compressive strength (UCS), flexural strength and freeze and thaw effect, Environ Earth Sci, 71, 4247–4259. Çelik MY (2017) Water absorption and P-wave velocity changes during freeze–thaw weathering process of cross cut travertine rocks. Environ Earth Sci (2017) 76:409 https://doi.org/10.1007/s12665-017-6632-7 Çelik MY, Sert M (2020) Accelerated ageing laboratory tests for the evaluation of the durability of hydrophobic treated and untreated andesite with respect to salt crystallization, freezing–thawing, and thermal shock. Bull Eng Geol Environ 79:3751–3770 (79:7-September 2020). https://doi.org/10.1007/ s10064-020-01776-0 Çelik MY, Korucu MR (2023) The effect of freeze-thaw cycles performed with salty solutions (NaCl and Na2SO4) on carbonate-building stones Bull Eng Geol Environ (2023) 82:64 https://doi.org/10.1007/s10064-023-03072-z Fener M, İnce İ (2015) Effects of the freeze–thaw (F–T) cycle on the andesitic rocks (Sille-Konya/Turkey) used in construction building. J Afr Earth Sci 109, 96–106 Ghobadi MH, Taleb Beydokhti AR, Nikudel MR, Asiabanha A, Karakus M (2016) The effect of freeze–thaw process on the physical and mechanical properties of tuff, Environmental Earth Sciences, 75, 846. Ghobadi MH, Babazadeh R (2015) Experimental studies on the effects of cyclic freezing–thawing, salt crystallization, and thermal shock on the physical and mechanical characteristics of selected sandstones. Rock Mech Rock Eng 48, 1001–1016. https://doi.org/10.1007/s00603-014-0609-6 Gökçe MV, İnce İ, Fener M et al. (2016) The effects of freeze–thaw (F–T) cycles on the Gödene travertine used in historical structures in Konya (Turkey). Cold Reg Sci Technol 127, 65–75 https://doi.org/10.1016/j. coldregions.2016.04.005 Huang S, Yu S, Ye Y, Ye Z, Cheng A (2022) Pore structure change and physico-mechanical properties deterioration of sandstone suffering freeze-thaw actions. Construction and Building Materials 330, 127200 Huigui Li, Xu G, Luo J et al. (2021) Effect of freeze-thaw cycles on microstructure and mechanical properties of limestone. iOP Conf Ser: Earth Environ Sci 861 022058 https://doi.org/10.1088/1755-1315/861/2/022058 72   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Jamshidi A, Nikudel MR, Khamehchiyan M (2016) Evaluation of the durability of Gerdoee travertine after freeze–thaw cycles in fresh water and sodium sulfate solution by decay function models. Eng Geol 202, 36–43, https:// doi.org/10.1016/j.enggeo.2016.01.004. Jamshidi A (2021) Predicting the strength of granitic stones after freeze– thaw cycles: considering the petrographic characteristics and a new approach using petro-mechanical parameter. Rock Mech Rock Eng 54, 2829–2841. https:// doi.org/10.1007/s00603-021-02458-3 Karaca Z, Deliormanli AH, Elci H, Pamukcu C (2010) Effect of freeze– thaw process on the abrasion loss value of stones, international Journal of Rock Mechanics and Mining Sciences, 47, 1207–1211. Liu Q, Huang S, Kang Y, Liu X (2015) A prediction model for uniaxial compressive strength of deteriorated rocks due to freeze–thaw, Cold Regions Science and Technology, 120, 96-107. Luo X, Zhou S, Huang B et al. (2021) Effect of freeze–thaw temperature and number of cycles on the physical and mechanical properties of marble. Geotech Geol Eng 39, 567–582. https://doi.org/10.1007/s10706-020-01513-0 Maľa M, Greif V, Ondrášik M (2022) Pore structure evolution in andesite rocks induced by freeze–thaw cycles examined by non-destructive methods. Sci Rep 12, 8390. https://doi.org/10.1038/s41598-022-12437-5 Martins L, Vasconcelos G, Lourenço PB, Palha C (2016) Influence of the freeze-thaw cycles on the physical and mechanical properties of granites. J Mater Civ Eng 28(5) 04015201. https://doi.org/10.1061/(asce)mt.19435533.0001488 Martínez-Martínez J, Benavente D, García-del-Cura MA (2011) Spatial attenuation: the most sensitive ultrasonic parameter for detecting petrographic features and decay processes in carbonate rocks, Eng Geol, 119 (3–4), 84–95. Martínez-Martínez J, Benavente D, Gómez-Heras M, Marco-Castaño L, García-del-Cura M.A (2013) Non-linear decay of building stones during freeze– thaw weathering processes, Construction and Building Materials, 38, 443–454. Molina E, Cultrone G, Sebastian E, Alonso FJ (2013) Evaluation of stone durability using a combination of ultrasound, mechanical and accelerated aging tests, Journal of Geophysics and Engineering, J. Geophys. Eng. 10,3, 035003. Momeni A, Abdilor Y, Khanlari GR et al. (2016) The effect of freeze–thaw cycles on physical and mechanical properties of granitoid hard rocks. Bull Eng Geol Environ 75, 1649–1656. https://doi.org/10.1007/s10064-015-0787-9 FREEZE–THAW DURABiLiTY AND ULTRASONiC CHARACTERiZATiON . . .   73 Nicholson DT, Nicholson FH (2000) Physical deterioration of sedimentary rocks subjected to experimental freeze–thaw weathering, Earth Surface Processes and Landforms, 25, 1295–1307. Ozcelik Y, Careddu N, Yilmazkaya E (2012) The effects of freeze–thaw cycles on the gloss values of polished stone surfaces, Cold Regions Science and Technology, 82, 49–55. Özbek A (2014) Investigation of the effects of wetting–drying and freezing–thawing cycles on some physical and mechanical properties of selected ignimbrites, Bull Eng Geol Environ, 73, 595–609. Pápay Z, Rozgonyi-Boissinot N, Török Á (2021) Freeze–thaw and salt crystallization durability of silica acid ester consolidated porous limestone from Hungary. Minerals 11(8):824. https://doi.org/10.3390/min11080824 Pápay Z, Török Á (2018) Effect of thermal and freeze-thaw stress on the mechanical properties of porous limestone, Period Polytech, Civ Eng 62(2): 423–428. https://doi.org/10.3311/PPci.11100 Ren JX, Yun MC, Cao XT et al. (2022) Study on the mechanical properties of saturated red sandstone under freeze–thaw conditions. Environ Earth Sci 81, 376. https://doi.org/10.1007/s12665-022-10503-9 Ruedrich J, Kircher D, Siegesmund S (2011) Physical weathering of building stones induced by freeze-thaw action: a laboratory long-term study, Environmental Earth Sciences, 63, 1573–1586. Salvini S, Coletti C, Maritan L. et al. (2023) Petrographic characterization and durability of carbonate stones used in UNESCO World Heritage Sites in northeastern italy. Environ Earth Sci 82, 49. https://doi.org/10.1007/s12665022-10732-y Siegesmund S, Menningen J, Shushakova V (2021) Marble decay: towards a measure of marble degradation based on ultrasonic wave velocities and thermal expansion data. Environ Earth Sci 80, 395. https://doi.org/10.1007/s12665-02109654-y Scherer GW, Valenza JJ (2005) Mechanisms of frost damage. Materials Science of Concrete, Vol. Vii, 209–246. eds. J. Skalny and F. Young (American Ceramic Society) Şengün N, Altındağ R, Mutlutürk M, Karagüzel R, Kıstır R (2008) Kireçtaşlarında donma-çözülme (F-T) çevrimlerinin fiziksel ve mekanik özelliklere etkisi, SDÜ Fen Bilimleri Enstitüsü Dergisi, 12, 128-134. Tan X, Chen W, Tian H, Cao J (2011) Laboratory investigations on the mechanical properties degradation of granite under freeze–thaw cycles, Cold Regions Science and Technology, 68, 130–138. 74   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii TS EN 14579 (2006) Türk Standartları, Doğal taşlardeney metotlarıses ilerleme hızı tayini”, TSE, Ankara. TS EN 1936 (2010) Türk Standartları, Doğal taşlar deney metotları, gerçek yoğunluk, görünür yoğunluk, toplam ve açık gözeneklilik, TSE, Ankara. TS EN 13755 (2006) Türk Standartları, Doğal taşlar-deney metotlarıatmosfer basıncında su emme tayini”, TSE, Ankara. TS EN 12371 (2003) Türk Standartları, Doğal taşlar-deney metotları-dona dayanım tayini”, TSE, Ankara. Török Á, Szemerey-Kiss B (2019) Freeze-thaw durability of repair mortars and porous limestone: compatibility issues. Prog Earth Planet Sci 6, 42. https:// doi.org/10.1186/s40645-019-0282-1 Uğur İ, Toklu HÖ (2020) Effect of multi-cycle freeze-thaw tests on the physico-mechanical and thermal properties of some highly porous natural stones. Bull Eng Geol Environ 79, 255–267. https://doi.org/10.1007/s10064019-01540-z Valenza JJ, Scherer GW (2005) Mechanisms of salt scaling. Mat Struct 38, 479–488. https://doi.org/10.1007/BF02482144 Wang Y, Zhang B, Gao SH, Li CH (2021) Investigation on the effect of freeze-thaw on fracture mode classification in marble subjected to multi-level cyclic loads. Theor Appl Fract Mech 111, 102847 https://doi.org/10.1016/j. tafmec.2020.102847. Wang Y, Zhu C, He M, et al. (2022) Macro-meso dynamic fracture behaviors of Xinjiang marble exposed to freeze thaw and frequent impact disturbance loads: a lab-scale testing. Geomech Geophys Geo-Energy Geo-Resour 8, 154. https://doi.org/10.1007/s40948-022-00472-5 Yavuz H (2011) Effect of freeze–thaw and thermal shock weathering on the physical and mechanical properties of an andesite stone. Bull Eng Geol Environ 70, 187–192. https://doi.org/10.1007/s10064-010-0302-2 Yavuz AB (2012) Durability assessment of the Alaçatı tuff (Izmir) in western Turkey, Environ Earth Sci, 67, 1909-1925. Yavuz H, Altindag R, Sarac S, Ugur i, Sengun N (2006) Estimating the index properties of deteriorated carbonate rocks due to freeze–thaw and thermal shock weathering”, international Journal of Rock Mechanics and Mining Sciences, 43, 767–775. 75 CHAPTER IV MODELLING NANOREFRIGERANT FLOWS IN EJECTORS: DETAILED COMPUTATIONAL PROCEDURES FOR CONSTANT PRESSURE AND CONSTANT AREA EJECTORS Candeniz SEÇKİN1 1(Assoc. Prof.) Marmara University, Faculty of Engineering, Department of Mechanical Engineering E-mail: [email protected] ORCID: 0000-0002-7507-1773 1. Introduction The strong economic development and the growth of nations are propelled by the smart utilization of energy. As global energy demand accelerates, engineers and researchers across disciplines are elevating the conservation and optimization of energy to a central research frontier. in every engineering domain, achieving greater efficiency in energy use — whether through advanced efficiency technologies or by minimizing overall demand — has become a fundamental objective (Gao et al.,2009; Xu et al.,2010; Lam, 2000). Across the diverse areas of energy consumption, HVAC systems represent a considerable share, with cooling processes emerging as the primary contributors to overall energy demand. Rising living standards and the growing demand for human comfort have driven a substantial increase in energy consumption, especially through cooling processes. Globally, the increasing adoption of air-conditioning systems has become a major driver of building energy consumption. According to international Energy Agency (2018), cooling appliances—including air conditioners and electric fans—constitute approximately 20% of electricity use in buildings worldwide, corresponding to roughly 10% of total global electricity consumption. This share is expected to rise further, particularly in regions with 76   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii high population density and growing income levels, with projections indicating that cooling demand could triple by 2050 (international Energy Agency, 2018). Consequently, efforts to curb the overall energy consumption of cooling systems can lead to substantial energy savings on a global scale, addressing both environmental and economic challenges associated with high electricity demand. Enhancing the performance and operational efficiency of cooling units represents a key strategy to achieve such reductions, through the adoption of low energy technologies, optimized system design, and intelligent control strategies. A global analysis of cooling systems reveals that the vapor compression refrigeration cycle stands out as the predominant technology in widespread use. Detailed energy assessments indicate that the compressor, as the core component of this system, is responsible for approximately 80% of its total energy consumption (Westphalen et al.,1996). At this point, it becomes evident that targeted efforts to reduce the energy consumption of compressors in refrigeration cycles hold substantial potential for lowering the overall energy demand of cooling systems. Implementing these efforts enables to achieve significant reductions in global energy use and also to contribute to the sustainable management of energy resources (Harby et al., 2016;Seckin,2018). In this context, the utilization and advancement of ejector technologies—capable of fulfilling the function of a compressor without consumption on electrical power—can be considered as a promising alternative to conventional compression systems, offering potential benefits in terms of energy efficiency and sustainability. Ejector based cooling cycles are classified into two main groups: cycles in which the ejector replaces the compressor (ejector enhanced refrigeration cycle) and cycles in which the ejector is used instead of the throttling valve (ejector expansion refrigeration cycle). The operation of the ejector in an ejector enhanced refrigeration cycle can be summarized as follows: the highpressure flow from the boiler and low-pressure flow from the evaporator enter into the ejector and the resulting mixed flow is discharged from the ejector at an intermediate pressure, which is higher than the evaporator pressure. Thus, the ejector performs the function of compressing the evaporator flow without consuming electrical power. Hence, the ejector can be regarded as a thermally driven compressor with no power requirement (Chen,2012; Rani and Sachdeva, 2015). Additionally, since the ejector operates with heat, low grade energy sources can be utilized as the required energy input to the cooling cycle, in contrast to the electrically powered vapor compression refrigeration cycles. This enables the utilization of low temperature waste heat, facilitates heat MODELLiNG NANOREFRiGERANT FLOWS iN EJECTORS . . .   77 recovery and contributes to the reduction of refrigeration costs. Some additional advantages associated with the use of ejectors in refrigeration processes can be identified as: straightforward configuration, low installation and operational costs and the ability to operate with a wide range of refrigerants (Boumaraf and Lallemand,2009; Seckin,2017). Based on the advantages discussed above, ejector including cycles and studies aimed at improving their efficiency have attracted significant attention from researchers and engineers. A nanorefrigerant is a refrigerant formed by dispersing nanoparticles into a single phase refrigerant. The nanoparticles constitute a very little fraction of the nanorefrigerant, typically accounting for less than 4% of the nanorefrigerant volume (Molana and Wang,2020). The presence of nanoparticles in refrigerants has been shown to markedly enhance their thermophysical properties and heat transfer capabilities, underscoring their potential for superior performance in refrigeration systems (Sanukrishna et al.,2018). The enhanced pool boiling and convective heat transfer performance represents their key advantage, facilitating the design of more compact and economically efficient refrigeration systems. Initially, Choi and Eastman (1995) reported the dispersion of oxide or metal nanoparticles into a refrigerant to produce a homogeneous two phase mixture, which was reported to exhibit enhanced thermophysical properties. Extensive research has been carried out worldwide in this field, revealing consistent findings across most studies: a significant enhancement in heat transfer coefficients with increasing nanoparticle concentration in nanorefrigerants (observed in both pool boiling and convective boiling processes), which results in a reduction in energy consumption within refrigeration systems, reflected by higher COP values (Molana and Wang,-2020;Nair et al.,2018;Pawale et al.,2017; Sanukrishna et al.,2018; Celen et al.,2014). it is observed that as the volume or mass fraction of nanoparticles in a nanorefrigerant increases, the convective heat transfer coefficient also rises, thereby enhancing the overall performance of the cooling system (Sanukrishna et al.,2018; Molana and Wang,2020). On the other hand, the viscosity of nanorefrigerants increases with rising nanoparticle concentration, leading to higher pressure drops and elevated pump/compressor power requirements, which represent a drawback of using nanorefrigerants. However, the negative effect of the additional power requirement is prevailed by the improved heat transfer characteristics of nanorefrigerants, leading to an overall enhancement in COP of refrigeration cycles (Molana and Wang, 2020), as demonstrated in numerous experimental studies (Nair et al., 2018). 78   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii The use of nanorefrigerants in ejector based refrigeration cycles has been addressed in only a very limited number of studies, indicating that this remains as a relatively novel research area. Tashtoush et al. (2017) simulated the nanorefrigerant flow for CuO/R134a and Al2O3/R134a (0–2 wt.%) in an ejector enhanced refrigeration cycle and reported COP increase of the cycle. Khetib et al. (2022) modeled an ejector refrigeration cycle coupled with a lithiumion battery, demonstrating that the system COP rises with increasing CuO nanoparticle volume fraction in the R141b based refrigerant. Li et al. (2024) introduced an innovative ejector based dual evaporator refrigeration system that is integrated to photovoltaic (PV) modules. Nanorefrigerants contain 0.5 wt.% Al₂O₃ nanoparticles in R1234ze, R1234yf, R600a, and R134a base refrigerants. The design of the cycle is a combination of ejector refrigeration cycle and vapor compression refrigeration cycle. Aktemur and Ozturk (2022) conducted a simulation study of a booster added ejector expansion refrigeration cycle using both the pure refrigerant R1270 and CuO/R1270 nanorefrigerant containing 2 wt.% CuO. Aktemur and Ozturk (2023) analyzed a booster added ejector expansion refrigeration cycle using R152a/Cu nanorefrigerant. The cycle’s compressor power supplied by solar panels in İzmir, Turkey. As it is seen here, the number of nanorefrigerant using ejector studies are very limited in the existing literature. In the following sections, an overview of different ejector types (constant area mixing ejector and constant pressure mixing ejector) and their design characteristics are provided in detail. Subsequently, a detailed computational procedure for nanorefrigerant application in these two ejector types is presented. The flowcharts required to model the ejectors with nanorefrigerant flows are provided. 2. Design of Ejectors and Physical Mechanism Ejectors are subject to various classifications depending on their design features. The most common classifications are according to: (i) the nozzle position, (ii) the nozzle design, and (iii) the number of fluid phases (Besagni et al.,2016). ıt is observed that the most commonly used classification and naming of ejectors are based on the nozzle position. According to the nozzle position, the most well known and practically applied types of ejectors are the “constant area mixing ejectors (CAME)” and the “constant pressure mixing ejectors (CPME).” Schematic overview of the CAME and CPME ejectors are seen in the MODELLiNG NANOREFRiGERANT FLOWS iN EJECTORS . . .   79 Figs. 1 and 2, respectively. As presented in the figures, in CPME ejectors, the nozzle exit is located in the “constant pressure part”, whereas in CAME ejectors, the nozzle exit is positioned at the entry of the “constant area mixing part”. A comparison between two types of ejectors reveals that CPME ejectors are widely utilized due to their ability to operate with higher ejector exit pressures, it means, higher compression ratio between the inlet and outlet of the ejector (Tashtoush et al.,2015). CPME ejectors generally exhibit superior performance compared to CAME ejectors, even though CAME ejectors are capable of operating with higher mass flow rates of working fluid flows (which contributes to enhancing the performance of an ejector refrigeration cycle) (Pianthong et al.,2007). Keenan et al. (1950) reported that the CPME performs better than the CAME under identical conditions, while Yapici and Ersoy (2005) observed the contrary when operating under different conditions. As shown in Figs. 1 and 2, the high pressure flow (the first flow of the ejector) and the low pressure flow (the second flow of the ejector) enter the ejector and exit at an intermediate pressure of these flows. In this process, the low pressure flow (second flow) undergoes compression, allowing the ejector to function as a compressor substitute without consuming electrical power. Here, the high pressure flow (first flow) enters a convergent/divergent nozzle, referred to as “first nozzle”, while the low pressure flow (second flow) enters the “second nozzle”. The second nozzle is a simple convergent nozzle. The term “first” refers to the driving or motive flow (it creates the suction effect necessary to draw the second flow into the ejector), whereas the term “second” refers to the driven or passive flow (it is drawn into the ejector by the effect of the first flow) (Chen et al.,2015). Apart from the first and second nozzles, both types of ejectors include a constant area mixing part and a diffuser, as shown in Figs. 1 and 2. The main difference between CAME and CPME is: the presence of “constant pressure part” located between the first and second nozzle exits and the constant area mixing part (Fig. 2), which fundamentally alters the physical mechanism of mixing between CPME and CAME ejectors (Grazzini et al.,2018). 80   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii 5 As shown in Figs. 1 and 2, the high pressure flow (the first flow of the ejector) and the low pressure flow (the second flow of the ejector) enter the ejector and exit at an intermediate pressure of these flows. In this process, the low pressure flow (second flow) undergoes compression, allowing the ejector to function as a compressor substitute without consuming electrical power. Here, the high pressure flow (first flow) enters a convergent/divergent nozzle, referred to as ―first nozzle‖, while the low pressure flow (second flow) enters the ―second nozzle‖. The second nozzle is a simple convergent nozzle. The term ―first‖ refers to the driving or motive flow (it creates the suction effect necessary to draw the second flow into the ejector), whereas the term ―second‖ refers to the driven or passive flow (it is drawn into the ejector by the effect of the first flow) (Chen et al.,2015). Apart from the first and second nozzles, both types of ejectors include a constant area mixing part and a diffuser, as shown in Figs. 1 and 2. The main difference between CAME and CPME is: the presence of ―constant pressure part‖ located between the first and second nozzle exits and the constant area mixing part (Fig. 2), which fundamentally alters the physical mechanism of mixing between CPME and CAME ejectors (Grazzini et al.,2018). Fig. 1. Schematic of constant area mixing ejector (CAME). s Second nozzle tr y-y x-x diff Constant area mixing part Diffuser 1 2 First nozzle f 2 Second nozzle First nozzle 1 x-x y-y tr Constant area mixing part Diffuser Constant pressure part p diff s f Fig. 1. Schematic of constant area mixing ejector (CAME). 5 As shown in Figs. 1 and 2, the high pressure flow (the first flow of the ejector) and the low pressure flow (the second flow of the ejector) enter the ejector and exit at an intermediate pressure of these flows. In this process, the low pressure flow (second flow) undergoes compression, allowing the ejector to function as a compressor substitute without consuming electrical power. Here, the high pressure flow (first flow) enters a convergent/divergent nozzle, referred to as ―first nozzle‖, while the low pressure flow (second flow) enters the ―second nozzle‖. The second nozzle is a simple convergent nozzle. The term ―first‖ refers to the driving or motive flow (it creates the suction effect necessary to draw the second flow into the ejector), whereas the term ―second‖ refers to the driven or passive flow (it is drawn into the ejector by the effect of the first flow) (Chen et al.,2015). Apart from the first and second nozzles, both types of ejectors include a constant area mixing part and a diffuser, as shown in Figs. 1 and 2. The main difference between CAME and CPME is: the presence of ―constant pressure part‖ located between the first and second nozzle exits and the constant area mixing part (Fig. 2), which fundamentally alters the physical mechanism of mixing between CPME and CAME ejectors (Grazzini et al.,2018). Fig. 1. Schematic of constant area mixing ejector (CAME). s Second nozzle tr y-y x-x diff Constant area mixing part Diffuser 1 2 First nozzle f 2 Second nozzle First nozzle 1 x-x y-y tr Constant area mixing part Diffuser Constant pressure part p diff s f Fig. 2. Schematic of constant pressure mixing ejector (CPME). A summary of the different parts of the ejector and the processes occurring within the parts can be presented as follows (Seckin,2017; Chen et al.,2014). First nozzle: The first flow (at high temperature and pressure) enters the first nozzle and expands through the nozzle. At the throat of the nozzle (cross section “tr” in Figs. 1 and 2), choked conditions are established, and the transition from subsonic to supersonic flow regimes is realized. Hence, supersonic velocity is attained at the first nozzle outlet, while its pressure is significantly lower than that at the nozzle inlet. Second nozzle: The low pressure flow at the first nozzle outlet acts as the driving force of the second flow, namely, the first nozzle outlet pressure is lower than the pressure of the second flow, and the second flow is drawn into the MODELLiNG NANOREFRiGERANT FLOWS iN EJECTORS . . .   87 2 , 2 ( 1) 1 BR xx xx yy kM k P Pk −− =+ (39) 2 , ,, 2 ,, , ( 1) ( -1) 2 BR yy BR xx BR xx BR xx BR yy BR xx v kM vkM ρ ρ + = = + (40) ,, (, ) BR yy yy BR yy h fP v= (41) Enthalpy of the nanorefrigerant at y-y cross section (hNR,yy) is computed by Eq. (3) by using the below temperature relation for the cross section y-y. ,, , (, ) BR yy NR yy yy yy BR yy T T T fPh= = = (42) The conservation of energy principle is applied to determine the enthalpy at the diffuser outlet (hNR,diff), and the reduced form of the corresponding equation is presented below. 2 , ,, 2 NR yy NR diff NR yy V hh= + (43) In above equation, VNR,yy is computed by applying the conservation of energy equation between x-x and y-y in Fig. 2. Resuced form of the corresponding equation is given below. 1/2 2 , ,, , 2 ( ) 2 NR xx NR yy NR xx NR yy V Vh h  = +−    (44) On the other hand, by applying the below given procedure between cross sections y-y and diff, the base refrigerant’s enthalpy after the isentropic expansion in the diffuser (hBR,diff,is) is computed. , (,) BR yy yy yy s fT P= (45) ,, , BR diff is BR yy ss= (46) ,, ,, ( , ) BR diff is diff BR diff is h fP s= (47) ,, ,, ,, ( , ) BR diff is NR diff is diff BR diff is T T fP s= = (48) The isentropic enthalpy of the nanorefrigerant at cross section diff (hNR,diff,is) is calculated using Eq. (3). The actual enthalpy of the nanorefrigerant at the diffuser outlet (hNR,diff) is obtained through the nozzle’s isentropic efficiency (ηdiff). 88   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii ,, , ,, NR diff is NR yy NR diff NR yy diff hh hh η − = + (49) Flowchart diagram which illustrates the previously described modeling procedure of the nanorefrigerant using CAME ejector is presented in Fig. 3. 3.2. Modelling of CPME In the modeling of the constant pressure mixing ejector (CPME), it should be noted that, the first and second flows maintain their individual characteristics through the constant pressure section without mixing (Ouzzane and Aidoun,2003; Khalil et al.,2011). As indicated by the term “constant pressure ejector”, the pressures at the first and second nozzle outlets are equal and remain constant throughout the constant pressure section, which spans from the nozzle exits to the inlet of the constant area mixing part (Fig. 2). This means, the pressures of both flows are the same at the nozzles’ exit and at the entrance of the constant area mixing section. To ensure steady operation of the ejector, the second flow undergoes choking at the exit of second nozzle. Consequently, the equal pressures of the first and second flows within the constant pressure part correspond to the critical pressure of the second flow at the exit of the second nozzle (Ouzzane and Aidoun,2003). Since ejectors are very small size devices, the constant pressure part is quite short, and thus, the reduction in cross sectional area within this part is negligible in Fig. 2. Hence, p f s xx yy AAAA A=+= = (50) As previously stated, since the choking process at the second nozzle exit determines the outlet pressures of both nozzles, the modeling starts from the second nozzle. For the analysis of the nanorefrigerant flow through the second nozzle, the nozzle isentropic efficiency (ηs) is included in the computational procedure. Accordingly, the thermodynamic properties of the nanorefrigerant at the second nozzle outlet (s cross section in Fig. 2) are determined by ηs, as expressed below. The subscripts used in the following equations denote the flows at the cross sections labeled in Fig. 2. ( ) , 2 2 2 , BR s fTP= (51) , , ,2 BR s is BR ss= (52) MODELLiNG NANOREFRiGERANT FLOWS iN EJECTORS . . .   89 ( ) ,, ,, , BR s is s BR s is h f Ps= (53) ( ) ,, ,, ,, , BR s is NR s is s BR s is T T f Ps= = (54) Thus, the enthalpy of the nanorefrigerant at the s cross section after the isentropic process (hNR,s,is) is evaluated by means of Eq. (3). Isentropic efficiency (ηs) equation is applied to the nanorefrigerant as presented below to determine the enthalpy at s cross section (hNR,s). , ,2 ,2 , , () NR s NR s NR NR s is hh hh η =−− (55) On the other hand, the velocity of the flow at the s cross section is determined by applying the conservation of energy principle, for which the reduced form of the equation is given below. The critical flow conditions are reached at the second nozzle exit, hence, the velocity of the nanorefrigerant is equal to the local speed of sound at cross section s (Ouzzane and Aidoun,2003; Khalil et al.,2011; Selvaraju and Mani, 2004). ( ) 1/2 , ,2 , [2 ] NR s NR NR s V hh= − (56) ( ) ,, , , BRs NRs s BRs C C f Ph= = (57) To determine hBR,s in Eq.(57), earlier determined hBR,s,is (Eq.(53)) is utilized and isentropic efficiency (ηs) equation is applied to the base refrigerant as presented below. , ,2 ,2 , , () BR s BR s BR BR s is hh hh η =−− (58) To determine cross section area at the second nozzle exit (As), below given mass flow rate equation is applied (based on Eq. (1)). , , , 1 NR s NR s s NR tot w VA m w ρ  = +   (59) For the first nozzle, between the nozzle entrance and the throat (cross section tr in Fig.2), the isentropic efficiency of the first nozzle is applied as follows. ( ) , 1 1 1 , BR s fTP= (60) , , ,1 BR tr is BR ss= (61) ( ) ,, ,, , BR tr is tr BR tr is h fPs= (62) ( ) ,, ,, ,, , BR tr is NR tr is tr BR tr is T T fPs= = (63) 90   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Thus, the enthalpy of the nanorefrigerant at the tr cross section after the isentropic process (hNR,tr,is) is evaluated by means of Eq. (3). Isentropic efficiency (ηf) equation is applied to the nanorefrigerant as presented below to determine the enthalpy at the throat (hNR,tr). , ,1 ,1 , , () NR tr NR f NR NR tr is h h hh η =−− (64) On the other hand, the velocity of the flow at the throat is determined by applying the conservation of energy principle, for which the reduced form of the equation is given below. Since choking takes place at the throat, by definition, the velocity of the nanorefrigerant is equal to the local speed of sound as expressed in Eq. (66). 1/2 , ,1 , [2( )] NR tr NR NR tr V hh= − (65) ,, , () , BR tr NR tr tr BR tr C C fPh= = (66) To determine hBR,tr in Eq.(66), earlier determined hBR,tr,is (Eq.(62)) is utilized and isentropic efficiency (ηf) equation is applied to the base refrigerant as presented below. , ,1 ,1 , , () BR tr BR f BR BR tr is h h hh η =−− (67) To determine cross section area of the throat (Atr), below given mass flow rate equation is applied (based on Eq. (1)). , , , 1 1 NR tr NR tr tr NR tot VA m w ρ  = +   (68) Regarding the part of the first nozzle between the throat (tr cross section) and the nozzle outlet (f cross section in Fig. 2), the following procedure is applied. Since the pressure at the first nozzle exit (Pf) is identical to that at the second nozzle exit (Ps), already determined Pf is incorporated into the computation (Pf = Ps). ( ) , , , BR tr BR tr tr s fh P= (69) ,, , BR f is BR tr ss= (70) ( ) ,, ,, , BR f is f BR f is h fPs= (71) ( ) ,, ,, ,, , BR f is NR f is f BR f is T T fPs= = (72) Thus, the enthalpy of the nanorefrigerant at the f cross section after the isentropic process (hNR,f,is) is evaluated by means of Eq. (3). Isentropic efficiency MODELLiNG NANOREFRiGERANT FLOWS iN EJECTORS . . .   91 (ηf) equation is applied to the nanorefrigerant as presented below to determine the enthalpy at first nozzle exit (hNR,f). , , , ,, () NR f NR tr f NR tr NR f is hh hh η =−− (73) By applying the conservation of energy principle, velocity of the flow at cross section f (VNR,f) is determined. ( ) 1/2 , ,, [2 ] NR f NR tr NR f V hh = − (74) The cross sectional area at the first nozzle exit (Af) is determined using the mass flow rate equation provided below. ,1 , , NR NR f NR f f m VA ρ =  (75) Within the constant area mixing section, the first and second flows preserve their individual characteristics from the nozzle exits up to the cross section x-x (Fig. 2), where the mixing process is fully completed. To evaluate the thermodynamic properties of the mixed stream, the conservation principles of mass, momentum, and energy are employed, as outlined below. in below formulas, Pxx and Axx denote the pressure and cross sectional area of the flow at the x-x cross section, respectively. The friction factor λ, which accounts for the losses due to friction, is conventionally assumed in the range of 0.85–0.9 (Huang et al.,1999). As previously stated, in modelling of the CPMEs, the first and second flows leave the constant pressure part at the same pressure, which is equal to the exit pressure of the second nozzle. Accordingly, when applying the conservation of momentum equation (Eq. (31)) for CPME ejector, the pressure term Pf should be replaced with Ps . The modified form of the equation is given below. Moreover, the relation introduced in Eq. (50) for the area at the x-x cross-section (Axx ) must also be taken into account in the application of the following equation. Below given conservation of momentum equation is applied to determine velocity at x-x (VNR,xx). ( ) ,, , , ,, xx xx NR tot NR xx s f NR f NR f s s NR s NR s PA m V PA m V PA m V λ + = + ++   (76) Under the assumption of uniform flow velocities, the following relation is employed to describe the velocity distribution at the x-x cross section. Subsequently, VBR,xx is inserted into the mass flow rate formulation of the base 92   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii refrigerant fraction of the nanorefrigerant to determine its specific volume at the same cross section (vBR,xx) as detailed below. ,,, NR xx BR xx NP xx VVV= = (77) ( ) , , , , 1 BR xx xx NR tot BR xx BR xx VA mm v ω −= =  (78) Consequently, the specific enthalpy of the base refrigerant at the x-x cross section (hBR,xx) is determined using the following expression. ( ) ,, , BR xx xx BR xx h fPv= (79) An alternative strategy for determining the enthalpy of the base refrigerant (hBR,xx) consists of applying the energy conservation principle to the base refrigerant fraction of the nanorefrigerant, spanning from the nozzle exits to the x-x cross section. The reduced form of the resulting equation is provided below. . 2 22 , ,, , , ,, , , 22 2 NR f NR s NR xx BR f BR f BR s BR s BR xx BR tot V VV m h mh m h     ++ += +              (80) The temperature of the base refrigerant at the x-x cross section is determined by applying the Eq. (81). The uniform flow conditions require the equality TNR,xx = TBR,xx and the enthalpy of the nanorefrigerant at the same cross section (hNR,xx ) is calculated using Eq. (3). ,, (, ) BR xx xx BR xx T fP v= (81) MODELLiNG NANOREFRiGERANT FLOWS iN EJECTORS . . .   93 Det. Atr Eq.(14) Det. NR,1, NR,2 Eq.(1) P 1 , T 1 , P 2 , T 2 , P bp ,  NR,tot , η f , η s , η diff , A f /A tr , ω Set w Set Pf Det. VNR,f , hNR,f, NR,1 Eqs. (15-21) If  NR,1 =  NR,1 NO If V NR,s = C NR,s Det. VNR,s, CNR,s Eqs. (22-29) NO 000 Set Ps Det. h NR,diff Eqs. (43-49) If h NR,diff = h NR,diff If P diff = P bp Det. hNR,xx, Pyy, vNR,yy, hNR,yy Eqs. (37-42) Set Pdiff NO Det. As Eq.(30) Set Pxx Det. hBR,xx Eqs.(31-36) If hBR,xx = hBR,xx NO Set Ptr Det. VNR,tr, CNR,tr Eqs. (6-13) NO If V NR,tr = C NR,tr NO Fig. 3. Flowchart of the CAME ejector computational program. 94   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii The computational procedure employed to determine the flow characteristics in the remaining section of the CPME ejector (i.e., at the y-y cross section and the ejector exit) is analogous to that reported in Section 3.1 (Eqs. (38-49)) and is therefore not repeated here. For CPME ejectors, the application of the equations is outlined in the computational model flowchart in Fig. 4. 4. Conclusion in this study, the modeling procedures for nanorefrigerant using constant area mixing ejector and constant pressure mixing ejector, relevant equations and flowcharts are presented. The aim is to provide a clear framework to estimate the flow conditions resulting from processes in different regions of the ejector and to enable accurate determination of flow properties at key cross sections. For both types of ejectors, the principles of mass, momentum, and energy conservation are applied successively. It is seen that, in the constant pressure mixing ejector, the second nozzle exit pressure is highly determinant; once this pressure is determined, the number of iterations is reduced. in contrast, the constant area mixing ejector requires a higher number of iterations. The working fluid is a nanorefrigerant, and in recent years, the importance of nanofluids in thermodynamic and heat transfer applications has been steadily increasing. The addition of nanoparticles to the base fluid to form nanorefrigerant, enhances heat transfer characteristics, thereby contributing to improved system performance. Within this context, this study integrates nanorefrigerant modeling with ejector modeling, and hence, provides a detailed modeling procedure for both constant area mixing ejector and constant pressure mixing ejector. The developed flowcharts illustrate the calculation steps for determining flow properties at each key cross section. This approach minimizes uncertainties in the computed flow properties and ensures a systematic procedure. In conclusion, the presented framework offers a comprehensive tool for engineers and researchers involved in ejector design and analysis. The combination of explicit formulas and structured flowcharts simplifies the modeling process and allows for straightforward implementation in computational programs. The presented modeling procedure and flowcharts enhance both the theoretical understanding and practical applicability of nanorefrigerants in ejectors. MODELLiNG NANOREFRiGERANT FLOWS iN EJECTORS . . .   95 P 1 , T 1 , P 2 , T 2 , P bp ,  NR,tot , η f , η s , η diff , ω Set w Det. NR,1, NR,2 Eq.(1) Det. VNR,tr, CNR,tr Eqs. (51-58) NO If V NR,s = C NR,s Set Ps Det. As Eq.(59) If V NR,tr = C NR,tr Set Ptr Det. V NR,tr , C NR,tr Eqs. (60-67) NO Det. Atr Eq.(68) Pf = Ps Det. hNR,f, VNR,f , Af Eq.(6975) Set Pxx Det. hBR,xx Eqs.(76-81) If h BR,xx = h BR,xx NO Det. h NR,xx , P yy , v NR,yy , h NR,yy Eqs. (38-42) Set Pdiff NO Det. hNR,diff Eqs. (43-49) If h NR,diff = h NR,diff If P diff = P bp NO Fig. 4. Flowchart of the CPME ejector computational program. 96   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii References Aktemur, C., & Ozturk, i. (2022). Thermodynamic performance enhancement of booster assisted ejector expansion refrigeration systems with R1270/CuO nano-refrigerant. Energy Conversion and Management, 253, 115191. Aktemur, C., & Ozturk, i. (2023). Thermodynamic optimisation of a booster-ejector vapour compression refrigeration system using solar energy and R152a/Cu nano-refrigerant. Applied Thermal Engineering, 229, 120553. Besagni, G., Mereu, R., & inzoli, F. (2016). Ejector refrigeration: A comprehensive review. Renewable and Sustainable Energy Reviews, 53, 373– 407. Boumaraf, L., & Lallemand, A. (2009). Modeling of an ejector refrigerating system operating in dimensioning and off-dimensioning conditions with the working fluids R142b and R600a. Applied Thermal Engineering, 29, 265–274. Braccio, S., Guillou, N., Le Pierrès, N., Tauveron, N., & Trieu Phan, H. (2022). Mass-flow-rate maximization thermodynamic model and simulation of supersonic real-gas ejectors used in refrigeration systems. Thermal Science and Engineering Progress, 37, 1-13. Carrillo, J. A. E., de la Flor, F. J. S., & Lissén, J. M. S. (2017). Thermodynamic comparison of ejector cooling cycles: Ejector characterisation by means of entrainment ratio and compression efficiency. International Journal of Refrigeration, 74, 369–382. Celen, A., Çebi, A., Aktas, M., Mahian, O., Dalkilic, A. S., & Wongwises, S. (2014). A review of nanorefrigerants: Flow characteristics and applications. International Journal of Refrigeration, 44, 125–140. Chen, J. (2012). Investigation of vapor ejectors in heat driven ejector refrigeration systems (Doctoral thesis, Royal institute of Technology, KTH, Stockholm, Sweden). https://www.diva-portal.org/smash/get/diva2:764519/ FULLTEXT02 Chen, J., Havtun, H., & Palm, B. (2014). Parametric analysis of ejector working characteristics in the refrigeration system. Applied Thermal Engineering, 69(1–2), 130–142. Chen, J., Jarall, S., Havtun, H., & Palm, B. (2015). A review on versatile ejector applications in refrigeration systems. Renewable and Sustainable Energy Reviews, 49, 67–90. Choi, S.U.S., & Eastman, J. A. (1995). Enhancing thermal conductivity of fluids with nanoparticles. Proceedings of the ASME International Mechanical A BRiEF OVERViEW OF iNTELLiGiENT FOOD PACKAGiNG TECHNOLOGY   103 Today, with advances in electronics and materials science, as well as the integration of these two sciences, the researches have being carried out more effective, inexpensive, non-toxic, and sustainable smart packaging. In these studies, packaging materials and technologies that deliver reliable results in a wide variety of combinations are being investigated. 2. Classification Intelligient Food Packaging System (IFPS) iFPS can be divided into two main categorizes based on the determination of the internal or external of food packaging environmental conditions (Fig.1). Figure.1. Classification of Intelligient Food Packaging. 2.1. IFPS Based on External Conditions 2.1.1. Data Carriers Data carriers are designed to track every movement of food products from producer to consumer. Although, the primary purpose of their development was not food safety, they indirectly contribute this and quality by providing detailed monitoring of every environment where food packaging is located (Ghaani 104   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii et al., 2016). The most known and used data carriers are barkods and radiofrequency identification (RFID) tags. Barcodes are the cheapest and most common data carriers and are used for product or material identification. Barcodes can also be described as a pattern of bars and spaces symbolizing the 12 digits. The first two of these 12 digits represent the country where the barcode was created, whereas the next four and last six digits show the manufacturer and the product code, respectively. Barcodes are digitally read by an optical barcode reader, enabling the necessary records to be kept (Fig.2.a) ( Ghaani et al., 2016; Sohail et al., 2018). The abovementioned is a type of 1D barcode, one-dimensional and represented only by vertical lines. However, one-dimensional barcode systems have very small data capacities, making them inadequate in some cases. This creates challenges, particularly in industries requiring more complex data storage. To solve this problem, 2D two-dimensional barcodes, which store data both horizontally and vertically, thus enabling keeping more information. The most common type of 2D barcode is QR code. QR code was designed specifically for supply chain management and product tracking (Fig.2b). Figure 2. (a) 1D Barcod Systems; (b) 2D Barcod (QR code) Systems. (Azeredo and Correa, 2021) Radio frequency identification (RFID) systems are quite advanced and useful compared to barcode applications. These systems consist of a microchip, placed on the packaging, which contains an RFiD antenna, and a device that can read it from long distances without physical contact via radio waves (Fig.3) (Lee et al., 2014). RFiD technology allow for the monitoring and recording of all product history, due to movement of chip tags with the product. in addition, RFiD tags, A BRiEF OVERViEW OF iNTELLiGiENT FOOD PACKAGiNG TECHNOLOGY   105 that can be in forms such as disks, glass capsules, and labels, can be combined with several sensors, that are capable of identification in contactless, nonvisual and a wireless. in this way they have the ability to detect changes in food products, such as temperature, pH, humidity and gas etc., and offer the record and sent them to the central control system (Vanderroost et al., 2014) Figure 3. Schematic Represantation of Radio frequency identification (RFID) Systems (Azeredo and Correa, 2021). 2.1.2. Time and temperature indicators The expiration date information on food products is created generally based on the assumption that food products will be stored and distributed under ideal temperature conditions. However this supposition does not reflect the real everytime. Possible temperature fluctuations during storage and transportation can lead to various results from decreasing food quality to serious food spoilage and therefore, the monitorizing the temperature outside of the package is crucial. (Fig.4) (Karel et al., 2003). With this purpose it is used a system called timetemperature indicator. Time-temperature indicators are simple and inexpensive systems that display the temperature history of food as a color change on a specially designed label on the package in these systems, chemicals are incorporated into the label that undergo color changes due to mechanical, chemical, electrochemical, enzymatic, or microbiological reactions under the influence of temperature (Riva et al., 2001). These reactions are irreversible, so the resulting color does not revert to its original state; instead, the color changes over time due to temperature increases. (Fig.5) This provides a time-dependent indication of the temperature to which the food has been exposed and is vital for safe food production and distribution, for particularly essential for various chilled and frozen, particularly meat and dairy products. Some commercial time-tesperature indicators are seen in Fig.6. 106   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Figure.4. The effect of temperature fluctuations on shelf-life. Figure.5. illustration of the preparation of time-temperature indicators. A BRiEF OVERViEW OF iNTELLiGiENT FOOD PACKAGiNG TECHNOLOGY   107 (a) (b) (c) Figure.6. Some commercial time temperature indicators. (a) VitsabTM L5-8 Smart TTi Seafood Label; (b) TTi Monitor MarkTM.; (c) Zebra Temptime TM. 2.2. IFPS Based on Internal Conditions 2.2.1. Freshness indicators Formation of metabolites such as ethanol, carbon dioxide, organic acids, volatile nitrogen and sulfur compounds, which exist as a result of microbial spoilage in foods, are important indicators of decreasing the freshness of food. (Fig.7) (Rokka, et al., 2004). For example, volatile nitrogen compounds are the primary indicator of spoilage in fish. in the light of this information, freshness indicators have been developed by using the chemicals that react with the substances resulting from food 108   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii spoilage within the packaging, causing a color change. Freshness indicators are product-specific. Namely, label that contains different chemicals has the ability to react based on color change with these occuring substances released from various foods (Panjagari, et al.,2021). A vast majority of freshness indicators rely on natural (anthocyanins, betalains, curcumin etc.) and synthetic (bromocresol green and methyl red etc.) pH-sensitive dyes. (Wu et al., 2021). While synthetic dyes are more successful due to their lower cost and more intense color change, natural dyes are preferred, particularly in recent years, due to consumer demand for natural products rather than synthetic ones. (Alizadeh-Sani et al., 2020). The most known commercial freshness indicators, RipeSense® uses palladium-based sensors for pear marketing and FreshTag® which employs detection of total volatile basic nitrogen in poultry (Fig.7). (a) (b) Figure.7. The nost known commercial freshness indicators (a) RipeSense TM; (b) FreshTag TM. A BRiEF OVERViEW OF iNTELLiGiENT FOOD PACKAGiNG TECHNOLOGY   109 2.2.2. Gas and Leak Indicators The presence of oxygen causes spoilage in many foods through various mechanisms. First of all, oxygen promotes the growth of aerobic microorganisms, leading to microbiological degradation of foods. Secondly, it causes ascorbic acid loss in foods containing vitamin C and enzmatic reactions, resulting in decay in fruits and vegetables. Therefore, many foods, especially those requiring extended shelf life, must be packaged in vacuum or modified atmosphere packaging (MAP), preventing from contact with oxygen. MAP is based on the principle of packaging food with a special gas mixture, instead of air. However, even the slightest leakage in the packaging can alter the formulation of the gas mixture, threatening food quality and safety. in both MAP and vacuum packaging, it is essential to maintain the initial characteristics of the packaging internal environment under control until it reaches the consumer. Gas and leak indicators have been developed for this purpose (Heo and Lim, 2024). These indicators, which can be in the form of tablets, labels, prints, or polymer film coatings provide information to the user or consumer about the condition of the food, by changing color with the effects of chemical or enzymatic reactions (Table.1) Table.1. Some indicator Methods Carried out Depending on Metabolyte Type Metabolite İndicator method CO2Colour change in bromotymol compound CO2, SO2, NH4Colour change of xylene blue, bromocresol, phenol phatein on packaging material CO2, NH4, amines, H2S Colour changes in the dyes sensitive to CO2, NH4 and amine and those occured depends on H2S Acetic acid, lactic acid, asetaldeyde, amonia Colour changes in pH dyes and labels E.coli 0157 enteroteoksin Colour change in poyasethyle-based polymers. 2.2.3. Gas Sensors Gas sensors are devices that detect and quantify bacterial metabolites in gaseous form such as carbon dioxide, sulfur dioxide, ethylene, ammonia, ethanol and hydrogen sulphide that occur as a result of food spoilage. (Shaalan et al., 2022). The most common are amperometric oxygen and potentiometric carbon dioxide sensors. The sensors typically consist of two fundamental parts: a receptor that detects and identifies various physical or chemical parameters 110   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii and a transducer that converts these readings into a measurable form of energy. The most important advantage of these sensors is that they do not undergo any chemical reactions during measurement and not consume the analyte due to their chemical inert structure (Wolfbeis et al., 2006). Hovewer, there are some challenges in the practical use. For instance, it is desired the sensors to be compact, flexible, cost-effective, durable, and highly sensitive. They must also comply with strict regulations and be compliant with food safety standards (Vanderroost et al., 2014). 2.2.4. Chemical Sensors In food packaging, toxic matters originates from either food additives, pesticides or monomer migration from the packaging materials. For example, bisphenol-A(BPA, which is one of the biggest responsible for endocrine disorders) can migrates from polycarbonate bottles and epoxy resin.(Karthika et al., 2021) To overcome this issue, it is benefited from chemical sensors. Chemical sensors are devices constitued from a receptor and a transducer and can quickly interact with specific analytes and convert the chemical data they obtain into measurable signals. in this sensors, the receptor is responsible for sampling and is the place where chemical reactions occur, the transducer, on the other hand, is the part like an electrode where conversion of chemical information to electrical signal carry out. (Mustafa and Andreescu, 2018) Based on the transduction mechansim, chemical sensors can be classified as optical, electrical, gravimetric and electrochemical. (Kim et al., 2024). While chemical sensors can be utilized successfully to measure NH3, TMA and H2S especially emitted from chicken and meat spoilage (Senapati and Sahu, 2020), electrochemical sensors are employed to quantify toxic substances. 2.2.5. Biosensors Biosensors are a specialized group of chemical sensors and used only for biological substances, while chemical sensors are suitable every chemical compounds. Biosensors essentially consist of two parts formed from the bioreceptor and the transducer. The bioreceptor, also called the ligand, is the biological binding site where the target molecule is captured. The transducer, on the other hand, is the part that converts the biochemical/physicochemical interactions, resulting from the binding, to the receptor into electrical signals. A BRiEF OVERViEW OF iNTELLiGiENT FOOD PACKAGiNG TECHNOLOGY   111 Biosensors are frequently used for the identification and quantification of many biological substances, including alcohols, amino acids, sugars, lipids, pathogens, and other analytes (Takhistov,,2010; Khan et al., 2025). One of the other most important applications of biosensors is the detection of specific microbial species in foods, particularly pathogenic microorganisms (Ali et al., 2020). For example, biosensors whose surfaces are activated by microbial metabolites such as aflatoxin (Fam et al., 2023) or antigens on the surface of microorganisms such as Escherichia coli (Pebdeni et al., 2022). are used for this aim succesfully. Biosensors that can recognize and quantify metabolites produced during the decomposition process of foods can be divided into six groups based on their operating principles: electrochemical biosensors (Hansen et al., 2006), optical biosensors (Haes et al., 2002), immobilized-based biosensors (Sassolas et al., 2012), piezoelectric biosensors (Zhou et al., 2002) , microbial biosensors (Su et al., 2011), and nanomaterial-based biosensors (Perez-Lopez and Merkoci, 2011). 3. Regulations for Intelligient Food Packaging Since food packaging interacts directly or indirectly with food, it poses a potential risk to food safety and consumer health.(Heckman, 2005) in this context, the framework regulation EC1935/2004 has been published in Europe regarding all substances and materials intended for food contact. Active and intelligent packaging, on the other hand, is required to comply with EC 450/2009, “Regulation on Active and intelligent Substances and Materials intended for Food Contact.” The regulation lists the properties of permitted substances and the requirements that packaging and labeling procedures must meet. Both regulations are applied as mandatory in Europe (CommissionRegulation (EC) No 450/2009). 4. Conclusions intelligient food packaging technology, which targets food safety and quality, is one of the new applications in the food industry. This application, which has emerged and is growing in parallel with the advancement of materials and electronics science, offers both producers and consumers numerous advantages. Some of these advantages include the ability for producers to track products throughout the process and transportation, ease of storage, providing 112   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii consumers with reliable products and thus customer satisfaction, and ensuring consumers are assured of safe food. Despite its significant importance in the food industry, intelligient packaging technology is not widely implemented due to not being subject to any legal obligation. The vast majority of companies view it as an additional system and therefore an additional expense, while consumers have no awareness.In reality, these systems, often considered an additional cost, hold significant potential for both businesses and consumers, protecting food safety and therefore human health, preventing food waste, and ultimately improving quality. As a result of research in this field, it is expected that more effective, lower cost and more applicable and safe packaging materials and systems will be developed and thus, awareness and consciousness about intelligient food packaging technology will increase and that it will be used more widely in the near future, most probably even becoming a legal obligation in the future. References Ali, A. A., Altemimi, A. B., Alhelfi, N., Ibrahim, S. A. (2020). Application of Biosensors for Detection of Pathogenic Food Bacteria: A Review. Biosensors (Basel). 10(6):58. doi:10.3390/bios10060058. Alizadeh-Sani, M., Mohammadian, E., Rhim, J.-W., Jafari, S.M., (2020). pH-sensitive (halochromic) smart packaging films based on natural food colorants for the monitoring of food quality and safety. Trends in Food Science & Technology. 105, 93–144. doi:10.1016/j.tifs.2020.08.014. Avery Dennison, https://label.averydennison.com/eu/en/home.html. Azeredo, H. M. C. and Correa, D. S.(2021). Smart choices: Mechanisms of intelligent food packaging, Current Research in Food Science, 4, 932-936, doi:10.1016/j.crfs.2021.11.016. Biji, K. B., Ravishankar, C. N., Mohan, C. O. and Gopal, T. K. S. (2015). Smart packaging systems for food applications: A review. Journal of Food Science and Technology, 52(10), 6125–6135. doi:10.1007/s13197-015-1766-7. Bottani, E., Montanarii ,R., Volpi, A.(2010). The impact of RFiD and EPC network on the bullwhip effect in the Italian FMCG supply chain, International Journal of Production Economics, 124(2),426-432, doi:10.1016/j. ijpe.2009.12.005. CheckPoint® , CCL industries, https://checkpointsystems.com/ Chen, S., Brahma, S., Mackay, J., Cao, C., Aliakbarian, B. (2020). The role of smart packaging system in food supply chain. Journal of Food Science. 85(3),517-525. doi: 10.1111/1750-3841.15046. BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL . . .   119 ballast water is both a technical necessity for ship safety and a dual threat to ecosystems, human health, and the maritime economy (Jing et al., 2012). This dual nature has brought the issue to the forefront of international maritime policy discussions, accelerating efforts toward global solutions. The International Convention for the Control and Management of Ships’ Ballast Water and Sediments, adopted by the international Maritime Organization (iMO) in 2004, represents the most tangible global step in this direction (iMO, 2004). Furthermore, the IMO’s target to significantly reduce greenhouse gas emissions from shipping by 2050 has positioned ballast water management not only as an ecological safeguard but also as a core component of the sustainable shipping vision. The transfer of species through ballast water not only disturbs ecological balance but also threatens the sustainability of ecosystem services. invasive species suppress native populations, reducing the functionality of critical services such as fisheries, tourism, water quality, and carbon cycling. The economic costs associated with biodiversity loss can be substantial, as expenses for ecosystem restoration and fish stock recovery often exceed the investments required for ballast water management itself. Therefore, ballast water management extends beyond a technical requirement—it is a strategic environmental policy instrument essential for protecting natural capital and ensuring the economic continuity of ecosystem services. This chapter examines the role of ballast water in maritime transportation, its environmental risks, and its economic implications through the lenses of international regulations, technological approaches, and sustainability perspectives. The overarching aim is to frame ballast water management not merely as an environmental burden but as a strategic element for long-term maritime safety, climate objectives, blue economy sustainability, and sustainable shipping. 2. INTERNATIONAL REGULATIONS ON BALLAST WATER Although ballast water is an indispensable practice for ensuring vessel safety, its environmental risks have made the development of internationally coordinated solutions essential. Measures taken individually by different nations are inherently insufficient, given the global nature of maritime transportation— ships can call at ports across multiple continents within a short period of time. For this reason, ballast water management has become one of the most significant areas of international maritime regulation. After years of intensive 120   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii efforts, it was formally consolidated into a global convention under the leadership of the international Maritime Organization (iMO). The iMO adopted the international Convention for the Control and Management of Ships’ Ballast Water and Sediments in 2004, aiming to prevent the spread of harmful aquatic organisms through ships’ ballast water. 2.1. INTERNATIONAL REGULATIONS ON BALLAST WATER The international Convention for the Control and Management of Ships’ Ballast Water and Sediments (BWM Convention), prepared by the international Maritime Organization (iMO), was adopted in 2004 and entered into force in 2017 following extensive preparatory work (IMO, 2004). The primary objective of the convention is to prevent the transfer of harmful and invasive aquatic species through ships’ ballast water, thereby mitigating their negative impacts on marine biodiversity. During this process, the iMO also developed a sciencebased risk assessment framework to evaluate the likelihood of organism transfer between ecosystems (Hewitt et al., 2009). Within the scope of the convention, two main standards are defined: D-1 and D-2. The regional implementation of these standards is illustrated in Figure 2. • D-1 Standard requires ships to exchange their ballast water in open seas, away from coastal ecosystems. This method aims to minimize the risk of transferring organisms from ports to nearshore habitats. However, it presents significant operational challenges under stormy conditions, in areas with dense maritime traffic, or in waters lacking sufficient depth (David et al., 2024). • D-2 Standard, on the other hand, introduces a more stringent and effective approach. According to this standard, ballast water discharged by ships must meet specific biological concentration limits (Vibrio cholerae ≤ 1 cfu/100 mL) established by the IMO. For example, organisms ≥50 µm in size must not exceed 10 individuals per cubic meter, while those between 10–50 µm must be fewer than 10 individuals per milliliter. in addition, strict threshold values have been defined for pathogenic microorganisms such as Vibrio cholerae, Escherichia coli, and enterococci (iMO, 2004; Wright, 2007). BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL . . .   121 Figure 2. Regional implementation map of D-1 and D-2 standards (iMO, 2017). This framework mandates not only the exchange of ballast water during navigation but also its biological treatment through approved ballast water management systems. As a result, ballast water management has evolved into a core environmental regulation directly linked to maritime biosecurity and sustainable shipping objectives. The iMO’s target to reduce greenhouse gas emissions from international shipping by 50% by 2050 further strengthens the convention’s relevance—not only from an environmental but also from a climate strategy perspective. The fundamental differences between the two standards are summarized in Table 1. Thus, while the D-1 standard provides operational flexibility, its effectiveness remains limited, making it a temporary and transitional measure. In contrast, the D-2 standard offers significantly higher biological security and has become the primary and mandatory global benchmark for ballast water management in international shipping (David et al., 2024). 122   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Table 1. Comparison of D-1 and D-2 Standards. Feature D-1 Standard (Ballast Water Exchange) D-2 Standard (Ballast Water Treatment) Definition Exchange of ballast water in open seas (Čampara et al., 2019) Biological treatment of ballast water through iMO-approved systems (Jang et al., 2020) Application Area ≥200 nautical miles offshore and ≥200 m deep; if not possible, ≥50 nautical miles / ≥200 m depth (IMO, 2017) Onboard iMO-approved ballast water treatment systems (Jang et al., 2020) Objective To limit organism transfer by exchanging ballast in areas with low survival probability (David et al., 2024) To reduce living organisms and pathogens below established threshold limits (Jang et al., 2020) Biological Effectiveness Limited; ineffective for small organisms and microbes (David et al., 2024) High; provides strict limits for ≥50 µm and 10–50 µm organisms and pathogens Advantage No additional system required; cost-effective (David et al., 2024) Ensures international biosecurity; effectively prevents invasive species transfer (Jang et al., 2020) Disadvantage Safety risks during rough weather; low microbial removal efficiency (David et al., 2024) High investment and operational costs; requires additional onboard space and energy (Jang et al., 2020) Current Status Transitional; being gradually phased out Mandatory standard implemented by iMO and USCG (Güney, 2022; Jang et al., 2020) 2.2. REGIONAL REGULATIONS Although the iMO Ballast Water Management Convention provides a globally binding framework, several countries and regions have developed additional regulations tailored to their specific environmental priorities. Among these, the practices of the United States Coast Guard (USCG) and the European Union (EU) are the primary regional standards that ships engaged in international maritime operations must comply with. BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL . . .   123 The United States Coast Guard (USCG) introduced its own regulations in 2012, establishing an independent certification process for ballast water management systems separate from the iMO’s D-2 criteria (Guard, 2012). Ships entering U.S. ports must therefore be equipped not only with IMO-approved systems but also with systems tested and certified by the USCG. This dual compliance requirement imposes additional certification costs and technical compatibility challenges for ship operators (Loiacono et al., 2024). The European Union (EU), while largely adopting iMO standards, has developed stricter directives for ecologically sensitive marine regions such as the Baltic Sea, North Sea, and the Mediterranean. Port authorities regularly inspect vessels for compliance with the D-2 standard and impose additional reporting obligations. in this way, EU regulations complement the iMO’s global framework by integrating region-specific environmental safeguards (David et al., 2024). Similarly, the Australian Maritime Safety Authority (AMSA) has incorporated iMO standards into its national legislation and enforces rigorous inspections at ports across the Southern Hemisphere (AMSA, 2021). This demonstrates the increasingly multi-layered regulatory structure characterizing the global maritime sector. A comparative overview of the iMO, USCG, and EU ballast water management frameworks is summarized in Table 2, illustrating the similarities and regional distinctions among their enforcement structures. in conclusion, while the iMO Convention provides a global regulatory foundation, the U.S. and EU frameworks either tighten or complement this foundation according to regional environmental and policy needs. Consequently, ship operators face not only technical but also political and administrative compliance challenges (Güney, 2022; Čampara et al., 2019; David et al., 2024). 124   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Table 2. Comparative Summary of iMO, USCG, and EU Regulations. Feature IMO (BWM Convention) USCG (United States) EU (European Union) Enforcement Adopted in 2004, entered into force in 2017 (iMO, 2017) issued in 2012 (Guard, 2012) implemented from 2015 within EU environmental policies Main Standards D-1 (exchange) and D-2 (treatment) Similar to D-2 but requires USCGcertified systems Based on iMO standards with additional regional requirements System Certification iMO-approved system sufficient iMO + USCG certification required (Loiacono et al., 2024) IMO certification with supplementary oversight by EU environmental authorities Additional Requirements Aims for global compliance and harmonization Stricter controls for ships entering U.S. ports Stricter discharge monitoring in sensitive areas (Baltic, North Sea, Mediterranean) Challenges implementation costs and technical compliance issues Dual certification expenses and high capital investment Administrative burdens and regional regulatory variations 2.3. IMPLEMENTATION CHALLENGES AND COMPLIANCE ISSUES The effectiveness of international and regional regulations on ballast water management largely depends on the technical capability and economic compliance capacity of ship operators. Ballast water treatment systems impose a substantial burden—particularly on small and medium-sized operators—due to their high capital costs, energy consumption, and maintenance requirements (Lloyd’s Register, 2018). Moreover, the simultaneous enforcement of multiple regional regulations increases operational complexity. For example, a vessel on a transatlantic voyage may need to comply with both the iMO’s D-2 standard and the United States Coast Guard (USCG) requirements at the same time (Guard, 2012). The efficiency of these systems depends not only on the applied technology but also on the ship type, operational route, and the physical and chemical characteristics of seawater, as well as climatic conditions. Some systems exhibit performance degradation under environmental factors such as high particulate BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL . . .   125 concentrations, elevated salinity, or warm tropical waters, thereby limiting the practical applicability of uniform standards in real-world operations (Jing et al., 2012). As a result, consistent implementation of iMO standards is often impractical, since local ecosystem variability leads to significant differences in system performance. Under these circumstances, achieving effective international compliance requires more than technological advancement—it also necessitates policy harmonization, financial incentive mechanisms, and sustainable maritime financing tools. In recent years, green financing programs developed by the IMO and the World Bank have aimed to support the compliance efforts of ship operators, particularly in developing countries (World Bank, 2023). Such initiatives are reshaping the balance between technological capability, environmental standards, and economic sustainability, which will play a decisive role in the future of global ballast water management. 2.4. INSPECTION, COMPLIANCE, AND DIGITAL MONITORING SYSTEMS The effectiveness of regulatory standards in ballast water management is ensured not only through the installation of treatment systems but also through continuous inspection and monitoring mechanisms. Within this framework, Port State Control (PSC) represents the primary enforcement mechanism for verifying ships’ compliance with international ballast water discharge standards. During inspections, ballast water samples are collected and analyzed for microbiological, chemical, and physical parameters—including viable organism counts, particle concentration, pH, residual chlorine, and temperature. Measurements are typically verified using fluorometric, microscopic, or ATP (adenosine triphosphate) bioluminescence analysis techniques. The U.S. Coast Guard (USCG) and European port authorities conduct these tests either through portable onboard devices or accredited laboratories, and the results are critical for determining ship certification, re-inspection schedules, and voyage approvals. This systematic monitoring approach strengthens not only regulatory compliance but also transparency in environmental safety and biological risk management. However, several technical and administrative challenges persist within inspection procedures. The lack of standardized sampling conditions, heterogeneous water distribution inside ballast tanks, inconsistencies in equipment calibration, and differences in measurement sensitivity hinder global 126   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii harmonization. Furthermore, the sharing of inspection results in international databases remains limited, restricting the comparability of global monitoring data and the accuracy of risk assessments. Therefore, the effectiveness of inspection mechanisms depends not only on technical capacity but also on institutional coordination, data standardization, and the enhancement of digital reporting infrastructures. In recent years, the maritime industry has witnessed significant progress toward the digitalization of ballast water monitoring and inspection processes. Onboard smart sensor networks and real-time data acquisition systems now allow continuous monitoring of ballast water quality parameters—such as turbidity, microbial density, and residual oxidants. These systems transmit collected data to cloud-based platforms, facilitating instant information exchange between ship operators and port authorities. Additionally, artificial intelligence (AI) and machine learning models are being used to analyze collected data, predict treatment system performance, and proactively detect potential non-compliance cases, enabling more efficient inspection planning. This digital transformation introduces a new paradigm not only in terms of technical accuracy but also regarding transparency, accountability, and regulatory reliability. Recent initiatives such as the iMO’s Ballast Water information Systems (BWiS) and the USCG’s electronic record book (e-log) applications indicate that digital compliance verification is likely to become the future standard. Thus, ballast water management is evolving from traditional field inspections toward a data-driven, AI-assisted environmental governance model (BWMS with integrated Ai diagnostics by Alfa Laval, 2023). 3. BALLAST WATER TREATMENT METHODS: GENERAL APPROACHES In accordance with international regulations, ships are now required to ensure that their ballast water is biologically safe before discharge. This mandate has driven the development of diverse technological and engineering approaches in the maritime industry to minimize environmental risks. Today, ballast water treatment methods are generally classified into four main categories: physical, chemical, physicochemical, and biological. Each method possesses distinct advantages, limitations, and environmental impact profiles. The classification and fundamental principles of these methods are illustrated in Figure 3. BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL . . .   127 Figure 3. Classification and basic principles of ballast water treatment methods. 3.1. PHYSICAL METHODS Physical methods aim to remove particles and organisms in ballast water through mechanical or physical processes. Filtration systems are the most widely used technology in this category. Using multilayer or self-cleaning automatic filters, plankton and other large organisms are effectively trapped (Tsolaki and Diamadopoulos, 2010). However, filtration alone is often insufficient to completely remove microbial pathogens. Sedimentation and hydrocyclone separation systems utilize particle density differences to separate microand macro-organisms from the water. These methods are effective in removing large plankton and suspended solids, but exhibit limited efficiency for smaller microorganisms (David et al., 2024). in recent years, ultrasonic waves and ultraviolet (UV) irradiation have also emerged as promising physical treatment technologies. Ultrasonic treatment weakens or ruptures microbial cell membranes, whereas ultraviolet radiation interferes with DNA replication processes, effectively preventing further microbial propagation (Waite et al., 2003). UV treatment has the advantage of producing minimal chemical by-products; however, its efficiency decreases in highly turbid waters (Hull et al., 2017; Nwigwe and Minami, 2023). 128   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Although physical methods are environmentally friendly due to the absence of chemical residues, they typically require a secondary disinfection step to ensure complete microbial inactivation. 3.2. CHEMICAL METHODS Chemical methods inactivate microorganisms in ballast water through oxidative or biocidal reactions. Chlorination and sodium hypochlorite treatment have long been used for disinfection and are also widely applied in ballast water management (Tamburri et al., 2002). However, the formation of trihalomethanes and other toxic by-products raises significant environmental concerns. Ozonation provides broad-spectrum disinfection due to its strong oxidative potential. Ozone effectively inactivates bacteria, protozoa, and viruses, yet can produce toxic by-products such as bromate during application (Perrins et al., 2006; Herwig et al., 2006). Hydrogen peroxide and peracetic acid are alternative oxidizing agents offering strong oxidative capacity. Nevertheless, their storage stability and operational handling onboard ships pose technical challenges (Summerson et al., 2019). Overall, chemical methods exhibit high disinfection efficiency but present issues regarding residual toxicity, by-product formation, and environmental compliance (Smit et al., 2008). 3.2.1. BY-PRODUCT FORMATION AND REACTION MECHANISM IN CHEMICAL METHODS in ozonation-based disinfection, a critical issue is the conversion of naturally present bromide ions in seawater into bromate species. This reaction sequence is summarized in Equations (1)–(3). This occurs when ozone first oxidizes bromide to hypobromous intermediates, which can subsequently form bromate under alkaline and high-temperature conditions. Although this sequence enhances oxidation strength, uncontrolled ozone dosage or elevated pH may increase bromate accumulation, posing ecological and health concerns (Perrins et al., 2006; Herwig et al., 2006). Br⁻ + O₃ → OBr⁻ + O₂ (1) OBr⁻ + O₃ → BrO₂⁻ + O₂ (2) BrO₂⁻ + O₃ → BrO₃⁻ + O₂ (3) BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL . . .   135 ballast tanks (Kotinis, 2005). Although not yet widely implemented, this concept represents a long-term, sustainable solution with the potential to fundamentally eliminate ballast water-related environmental risks. The application of artificial intelligence (AI) and digital monitoring systems in ballast water management enables real-time operational tracking and treatment system optimization. Machine learning algorithms continuously analyze water quality parameters—such as turbidity, pH, and microbial density—to dynamically adjust system performance (Etemad et al., 2022). These data-driven solutions are increasingly recognized as valuable tools for improving regulatory compliance, particularly within iMO and USCG inspection frameworks. The key ongoing research areas and innovative technologies under development are outlined in Table 4, which highlights the advantages, limitations, and future prospects of each approach. Table 4. Current Research Areas and Future Perspectives in Ballast Water Management Area Description Advantages Limitations References Photocatalytic oxidation Radical generation under UV light using TiO₂ catalyst Residue-free, environmentally friendly Limited commercial application (Tsolaki and Diamadopoulos, 2010) Nanotechnology Nanosilver and nanocomposite membranes Strong antibacterial activity Potential ecotoxicity risk (Melnyk et al., 2024) Ballast-free ship designs Hull optimization eliminating ballast tanks Potential to solve the issue at its source Still at prototype stage (Kotinis, 2005) Artificial intelligence and digital monitoring Ai-based sensor and optimization systems Real-time monitoring and easier compliance High investment cost (Etemad et al., 2022) in conclusion, research on ballast water treatment focuses not only on improving existing technologies but also on developing radically innovative 136   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii solutions. Photocatalytic oxidation and nanotechnology provide environmentally friendly alternatives, while ballast-free ship designs aim to eliminate the problem at its source. Meanwhile, Ai-based solutions are transforming ballast water management from a purely technical operation into a digitally driven field of inspection and optimization. These emerging approaches are expected to play a critical role in meeting future iMO and regional regulatory standards, guiding the transition toward more sustainable and intelligent maritime operations (David et al., 2024). 7. CONCLUSIONS AND RECOMMENDATIONS The concept of ballast water management has shifted beyond its initial technical scope, becoming an essential pillar of maritime sustainability and environmental governance. While essential for maintaining ship stability and safety, ballast water also introduces serious environmental risks, including the transfer of invasive species, disruption of marine ecosystems, and the formation of toxic by-products. Consequently, ballast water management represents a critical balance between environmental protection and economic sustainability in global maritime transport (David et al., 2024). Current treatment technologies exhibit several sustainability-related limitations. Chemical methods are environmentally controversial due to the risk of toxic discharges, while physical methods, though environmentally benign, often lack sufficient biological efficiency. Hybrid systems—offering higher biological performance—remain constrained by high energy demands and investment costs. These challenges clearly underline the need for the development of next-generation technologies that are both eco-friendly and economically viable (Tsolaki and Diamadopoulos, 2010). Recent research highlights the growing trend toward innovation-driven sustainability. Photocatalytic oxidation and nanotechnology-based systems offer high energy efficiency and residue-free treatment potentials (Melnyk et al., 2024). Ballast-free ship designs represent a more radical approach, aiming to eliminate the need for ballast water altogether and thus resolving the problem at its source (Kotinis, 2005). Furthermore, the advancement of Ballast Water information Systems (BWiS) through machine learning–based frameworks could enhance the accuracy of risk assessments, strengthen regulatory decisionmaking, and promote long-term sustainability in maritime transport (Etemad et al., 2022). BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL . . .   137 In light of these findings, ballast water management plays a pivotal role in ensuring the sustainability of global shipping. The maritime industry should move away from short-term, cost-oriented solutions toward long-term environmental strategies. Academic research should prioritize the development of technologies that integrate low ecological impact, high energy efficiency, and regulatory compliance. For policymakers, aligning iMO standards with regional applications and reinforcing technical and financial support mechanisms for developing nations will be essential in achieving global sustainability goals (Čampara et al., 2019; Gollasch et al., 2007). in conclusion, ballast water management will remain an integral pillar of the sustainable maritime transport vision. The protection of marine ecosystems, the achievement of carbon-neutral targets, and the secure continuation of global trade depend on the synergistic collaboration of technological innovation, scientific research, and international cooperation. In this sense, ballast water management is not merely an environmental obligation, but a strategic domain at the heart of future sustainable maritime policies (David et al., 2024). REFERENCES Čampara, L., Frančić, V., Maglić, L. & Hasanspahić, N. 2019. Overview and comparison of the iMO and the US Maritime Administration ballast water management regulations. Journal of marine science and engineering, 7, 283. David, m., Gollasch, s. & Hewitt, c. 2024. Global maritime transport and ballast water management, Springer. Dobbs, F. C. & Rogerson, A. 2005. Ridding ships’ ballast water of microorganisms. Environmental Science & Technology, 39, 259A-264A. Duan, D., Xu, F., Wang, T., Guo, Y. & Fu, H. 2023. The effect of filtration and electrolysis on ballast water treatment. Ocean Engineering, 268, 113301. Endresen, Ø., Sørgård, E., Sundet, J. K., Dalsøren, S. B., isaksen, i. S., Berglen, T. F. & Gravir, G. 2003. Emission from international sea transportation and environmental impact. Journal of Geophysical Research: Atmospheres, 108. Ergin, A. & Sandal, B. 2023. Mobbing among seafarers: Scale development and application of an interval type-2 fuzzy logic system. Ocean Engineering, 286, 115595. Etemad, M., Soares, A., Mudroch, P., Bailey, S. A. & Matwin, S. 2022. Developing an advanced information system to support ballast water management. Management of Biological Invasions, 13, 68. 138   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii Gollasch, S., David, M., Voigt, M., Dragsund, E., Hewitt, C. & Fukuyo, Y. 2007. Critical review of the iMO international convention on the management of ships’ ballast water and sediments. Harmful algae, 6, 585-600. Guard, U. C. 2012. Standards for living organisms in ships’ ballast water discharged in US waters. Federal Register, 77, 17254-17320. Güney, C. B. 2022. Ballast water problem: Current status and expected challenges. Marine Science and Technology Bulletin, 11, 397-415. Herwig, R. P., Cordell, J. R., Perrins, J. C., Dinnel, P. A., Gensemer, R. W., Stubblefield, W. A., Ruiz, G. M., Kopp, J. A., House, M. L. & Cooper, W. J. 2006. Ozone treatment of ballast water on the oil tanker S/T Tonsina: chemistry, biology and toxicity. Marine Ecology Progress Series, 324, 37-55. Hewitt, C. L., Gollasch, S. & Minchin, D. 2009. The vessel as a vector–biofouling, ballast water and sediments. Biological invasions in marine ecosystems: Ecological, management, and geographic perspectives. Springer. Hull, N. M., isola, M. R., Petri, B., Chan, P.-S. & Linden, K. G. 2017. Algal DNA repair kinetics support culture-based enumeration for validation of ultraviolet disinfection ballast water treatment systems. Environmental Science & Technology Letters, 4, 192-196. imo 2004. international Convention for the Control and Management of Ships’ Ballast Water and Sediments. International Maritime Organization. Jang, P.-G., Hyun, B. & Shin, K. 2020. Ballast water treatment performance evaluation under real changing conditions. Journal of Marine Science and Engineering, 8, 817. Jing, L., Chen, B., Zhang, B. & Peng, H. 2012. A review of ballast water management practices and challenges in harsh and arctic environments. Environmental Reviews, 20, 83-108. Kideys, A. E. 2002. Fall and rise of the Black Sea ecosystem. Science, 297, 1482-1484. Kotinis, M. D. 2005. Development and investigation of the ballast-free ship concept, University of Michigan. Lacasa, E., Tsolaki, E., Sbokou, Z., Rodrigo, M. A., Mantzavinos, D. & Diamadopoulos, E. 2013. Electrochemical disinfection of simulated ballast water on conductive diamond electrodes. Chemical Engineering Journal, 223, 516-523. Lakshmi, E., Priya, M. & Achari, V. S. 2021. An overview on the treatment of ballast water in ships. Ocean & Coastal Management, 199, 105296. BALLAST WATER TREATMENT METHODS ON SHiPS: TECHNOLOGiCAL . . .   139 Liltved, H., Tobiessen, A., Delacroix, S., Heiaas, H. & Tryland, I. 2011. Filtration and UV treatment for ships’ ballast water management-water quality challenges and UV-dose requirements. Norwegian Institute for Water Research. IUVA News, 13, 17-21. Loiacono, S., Ceballos-Osuna, L., Scianni, C. & Brown, C. W. 2024. Technical considerations for sampling ballast water to determine compliance with discharge performance standards. Frontiers in Marine Science, 11, 1323989. Melnyk, O. M., Onishchenko, O. A., Shumylo, O. M., Lohinov, V., Pavlova, N. L., Bulgakov, M. P., Shapovalova, i. O. & DOLiA, O. E. 2024. Chemical Methods And Nanotechnology integration iN Ship Ballast Water Management For Maritime Transport Sustainability. Journal of Chemistry and Technologies, 32, 1137-1148. Nwigwe, T. & Minami, K. 2023. Review study of ballast water treatment system: Review of ballast water treatment system technologies. Journal of Maritime Research, 20, 90-97. Perrins, J. C., Cordell, J. R., Ferm, N. C., Grocock, J. L. & HERWiG, R. P. 2006. Mesocosm experiments for evaluating the biological efficacy of ozone treatment of marine ballast water. Marine Pollution Bulletin, 52, 1756-1767. Ricciardi, A. & Macisaac, H. J. 2000. Recent mass invasion of the north American Great Lakes by Ponto–Caspian species. Trends in ecology & evolution, 15, 62-65. Romero‐Martínez, L., Moreno‐Andrés, J., Acevedo‐Merino, A. & NEBOT, E. 2014. Improvement of ballast water disinfection using a photocatalytic (UV‐ C+ TiO2) flow‐through reactor for saltwater treatment. Journal of Chemical Technology & Biotechnology, 89, 1203-1210. Ruiz, G. M., Rawlings, T. K., Dobbs, F. C., Drake, L. A., Mullady, T., Huq, A. & Colwell, R. R. 2000. Global spread of microorganisms by ships. Nature, 408, 49-50. Sandal, Baris & Ergin, A. 2023. Navigating Uncertainty: From Type-1 to interval Type-2 Fuzzy Logic in Decision Making. Academic Studies in Engineering, 79. Shiganova, T. 1998. invasion of the Black Sea by the ctenophore Mnemiopsis leidyi and recent changes in pelagic community structure. Fisheries Oceanography, 7, 305-310. Smit, M. G., Ebbens, E., Jak, R. G. & Huijbregts, M. A. 2008. Time and concentration dependency in the potentially affected fraction of species: the case 140   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii of hydrogen peroxide treatment of ballast water. Environmental Toxicology and Chemistry, 27, 746-753. Stopford, M. 2008. Maritime economics 3e, Routledge. Summerson, R., Bloomfield, N. & Arthur, T. 2019. Treated ballast water and its impact on port water quality. ABARES Canberra, 1-84. Tamburri, M. N., Wasson, K. & Matsuda, M. 2002. Ballast water deoxygenation can prevent aquatic introductions while reducing ship corrosion. Biological Conservation, 103, 331-341. Tsolaki, E. & Diamadopoulos, E. 2010. Technologies for ballast water treatment: a review. Journal of Chemical Technology & Biotechnology, 85, 19-32. Unctad 2023. Review of maritime transport, 2023. Waite, T., Kazumi, J., Lane, P., Farmer, L., Smith, S., Smith, S., Hitchcock, G. & Capo, T. 2003. Removal of natural populations of marine plankton by a large-scale ballast water treatment system. Marine ecology progress series, 258, 51-63. Wright, D. A. 2007. Logistics of shipboard and dockside testing of ballast water treatment systems in the United States. Biological invaders in inland waters: Profiles, distribution, and threats. Springer. 141 CHAPTER VII EVALUATION OF CHANGES IN CAPILLARY WATER ABSORPTION DUE TO SALT AND SODA WATER CONCENTRATIONS IN POROUS BUILDING STONES Mustafa Yavuz ÇELİK (Prof. Dr.), Afyon Kocatepe University, Afyon Vocational School, Department of Marble Technology, Afyonkarahisar, Türkiye E-mail: [email protected] ORCID: 0000-0002-9695-7370 1. Introduction Natural stone is a fundamental building material owing to its durability, aesthetic appeal, and structural integrity. Many historically significant monuments and structures have been constructed worldwide from natural stone. Although natural stones are stronger and more durable than many other construction materials, they are not immune to degradation over time as environmental and atmospheric factors gradually take their toll. Among these factors, water is one of the most significant contributors to deterioration. Water facilitates the weathering of building stones both directly through processes such as wetting-drying and freeze-thaw cycles, and indirectly by transporting soluble salts into the material, leading to salt crystallization and subsequent damage. Capillary water absorption is a key parameter in assessing the penetration of water into natural building stones (Peruzzi et al., 2003). When surface or groundwater comes into contact with porous stones, it tends to rise within the material owing to capillary suction, which is influenced by the porosity of the stone. The extent of this capillary action depends on factors such as pore size, geometry, and connectivity. The capillary water absorption mechanism is governed by the capillary forces, which are directly related to the porous structure of the stone (Vázquez et al., 2010). High capillary water absorption, 142   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii which is often associated with increased porosity, negatively affects the physical and mechanical properties of natural stones. in colder regions, the water absorbed into the stone can freeze, creating ice crystals that expand within the pores. This expansion produces internal stresses that substantially reduce the stone strength. Likewise, soluble salts carried into the stone by capillary action can crystallize, intensifying the deterioration of the material (Tomašić et al., 2011). The capillary water-absorption capacity of natural building stones has been widely investigated. Washburn (1921) laid the theoretical groundwork for capillary flow kinetics in porous media by conceptualizing porous materials as a network of parallel capillary tubes with uniform diameters. Subsequent research has examined the effects of pore dimensions, petrographic features, and structural attributes on capillary absorption, emphasizing their crucial role in the long-term durability of stone materials (Ordoñez et al., 1997; iñigo et al., 2000; Mosquera et al., 2000; Nicholson, 2001; Chabas and Jeannette, 2001; Peruzzi et al., 2003; Karoglou et al., 2005; Moreno et al., 2006; ioannou and Holf, 2009; Vázquez et al., 2010; Tomašić et al., 2011; Juhász et al., 2014; Çelik and Kaçmaz, 2016; Karagiannis et al., 2016; Çelik and Yılmaz, 2018; Çelik and Sert, 2021; İnce, 2021; Çelik and Köken, 2023; Siegesmund et al., 2023; Çelik and Güven, 2024). Vázquez et al. (2010) identified a direct correlation between fracture properties, capillary absorption, and ultrasonic wave velocities in granites. Tomašić et al. (2011) analyzed the capillary absorption capacity of two limestone varieties under dynamic water conditions. Peruzzi et al. (2003) applied two distinct approaches utilizing both absolute and relative capillary index values in their study. Hoffmann and Niesel (1992) explored how the pore structure contributes to stone deterioration. Mosquera et al. (2000) assessed the capillary absorption potential of three granite types, reporting values of 0.24, 0.89, and 1.24 kg/m²s⁰.⁵. Nicholson (2001) investigated the role of pore structure in the weathering processes affecting five different limestone samples. Karoglou et al. (2005) examined the capillary absorption kinetics in various construction materials, including four types of natural stone, two categories of bricks, and six different plasters. Ioannou et al. (2009) developed models to predict the capillary absorption behavior of porous limestone using different fluids, demonstrating that larger pore structures reduce the absorption capacity compared to smaller pores. Şengün et al. (2014) studied the capillary absorption coefficients of 118 types of natural stones and correlated their results with textural properties and other physical characteristics, such as bulk density, apparent and total porosity, and seismic velocity. Juhász et al. (2014) explored both capillary water EVALUATiON OF CHANGES iN CAPiLLARY WATER ABSORPTiON DUE . . .   143 absorption and microbial transport mechanisms in porous limestone. Moreno et al. (2006) analyzed how salt-induced deterioration in granites is linked to capillary absorption. Çelik and Kaçmaz (2016) investigated the static and dynamic capillary absorption behavior of andesite and tuff. Al-Naddaf (2011) studied the capillary absorption capacity of sandstones, particularly its role in salt crystallization within the stone matrix. Karagiannis et al. (2016) examined the relationship between temperature and capillary absorption coefficients in building materials, establishing a direct linear correlation between temperature increases and capillary absorption across all tested materials. Volcanic rocks, such as andesite and tuff, have been widely used as building stones in various regions of Anatolia since the Roman, Seljuk, and Ottoman periods. Numerous structures from these historical eras, constructed using andesite and tuff, still exist in the Afyonkarisar region (Figure 1). Moreover, these stones continue to be used as local construction materials in this region. It is well documented that capillary water absorption has led to significant deterioration of both interior and exterior surfaces in many historical buildings, causing considerable damage to works of high artistic value. To mitigate and prevent such degradation, it is essential to assess the capillary water absorption capacities of building materials, both for restoration efforts in historical structures and for selecting appropriate materials for new construction. Fig. 1. Weathering due to capillary water absorption in historical and contemporary buildings made of Ayazini tuff and İscehisar andesite in Afyonkarahisar (a İmaret Mosque-built AD 1472, b: Mevlevi Mosquebuilt AD 1905, c: Afyon Kocatepe Universitypresent, d: Cürcani Mosque-built AD 1500s). 144   RESEARCH AND FiNDiNGS iN ENGiNEERiNG SCiENCES 2025 - iii This study characterizes the highly porous andesite and tuff used as building stones and examines their physico-mechanical, chemical, and mineralogicalpetrographic properties, as well as their pore structures. Additionally, the capillary water absorption potential was assessed using pure water and solutions with different concentrations of sodium carbonate (Na₂CO₃) and sodium chloride (NaCl). The findings provide valuable insights into the relationship between capillary water absorption and material deterioration in historical structures while also offering essential data for selecting suitable construction materials in environments with varying water conditions. 2. Material and Methods 2.1. Materials The andesite and tuff samples used in this study were collected from active quarries located in the vicinity of Afyonkarahisar. These quarries continue to supply andesite and tuff, which are widely used as building stones. Specifically, the andesite samples were sourced from a quarry on Ağin Mountain, situated to the north of the İscehisar district, whereas the tuff samples were obtained from quarries near Ayazini village, located along the Afyonkarahisar-Eskişehir highway. A location map of the sampling sites is shown in Fig. 2. For the experiments, a sufficient number of cubic samples, each measuring 70 × 70 × 70 mm, were prepared from each type of building stone. Sodium carbonate (Na2CO3) was used to prepare the soda water solution, and rock salt-halite (NaCl) was used to prepare the saltwater solution. Fig. 2. Location map of the quarries where the building stone samples used in the experiments were taken (a), view of the quarries, Ayazini tuff (b), and İscehisar andesite (c).