IPSL _Proceedings 2026

Institute of PhysicsSri LankaPROCEEDINGS42nd TECHNICAL SESSIONSMarch 2026University of Colombo, Sri Lanka

PROCEEDINGS OF THE TECHNICAL SESSIONSMarch 2026The Technical Sessions of the Institute of Physics, Sri Lanka is held in conjunction with the Annual General Meeting of the Institute of Physics, Sri Lanka as an annual event. The purpose of the Technical Sessions is to enable members of the institute to present work done by them in the form of a paper.All papers presented at the Technical Sessions will be edited and published as the Proceedings of the Technical Sessions by the Institute of Physics, Sri Lanka.Guidelines for the Submission of papers:• Three hardcopies of the paper on A4 size paper (maximum of 8 pages) should be submitted to the Secretary on or before the deadline set by the Institute of Physics, Sri Lanka.• The papers should be in camera-ready format. All papers will be refereed before acceptance for presentation at the Proceedings of the Technical Sessions.• Either review papers or original work can be presented at the Sessions. • All accepted manuscripts will be published in the proceeding of the Technical Sessions 2026. (Selected papers will be published in SRI LANKAN JOURNAL OF PHYSIC, official Journal of the IPSL, and their abstracts will be published in proceeding of the Technical Sessions 2026)The scheduled deadline to submit the full papers for the Institute of Physics, Sri Lanka Technical Sessions is 26th January of the year. Printed copies of the manuscripts in triplicate should be sent to the following address with a softcopy via email or in a compact diskette.The SecretaryInstitute of Physics, Sri Lanka120/10, Wijerama MawathaColombo 07email: [email protected] Session Editorial Committee:Professor K P S C Jayaratne(Chairman)Department of Physics, University of ColomboProfessor P A A PereraDepartment of Physics, University of KelaniyaProfessor V P S PereraDepartment of Physics, The Open University of Sri LankaDr. (Mrs.) W W P De Silva Department of Physics, University of Sri JayewardenepuraDr. (Ms.) H O WijewardaneDepartment of Physics, University of RajarataDr. R C L De SilvaMaterials Technology, Industrial Technology InstituteProf. C Mahesh Edirisinghe(Ex-Officio, President/IPSL)Department of Physics, University of ColomboISSN 2012-8975Copyright © 2021 by the Institute of Physics, Sri Lanka. All rights reserved.

Sample Layout for the Preparation of ManuscriptsTitle of the Paper (font-14, bold)Author Names (font-12, normal)Affiliations (font-12, italic)email address of the corresponding authorABSTRACTAn abstract of approximately 150 words should convey the scope of the paper and be as informative as possible. The abstract must be able to stand on its own without reference to the main text.1. INTRODUCTIONAll papers presented at the Technical Sessions will be published. Authors are requested to submit their full papers in camera-ready format on or before the deadline set by the Institute of Physics, Sri Lanka. The style of writing should conform to accepted English usage.2. TEXTText must be typewritten clearly and accurately in single spacing on one side of an ordinary A4-size paper (297 mm × 210 mm) in a single column. The text must be typed leaving 3 cm margins on all sides. The submitted manuscripts of each contribution must be in their final formatsand of good appearance, because they will be printed directly without any editing. It is essential that the camera-ready copies are clean and unfolded. Use a laser printer for printing the text. Authors are also requested to submit a softcopy of the manuscript in Microsoft Word format once the papers are accepted for presentation. Select Times New Roman as the default font. The title of the paper should be in size 14 and the rest of the text in size 12. Section headings should be numbered 1, 2, 3 etc with capital letters. Capitalise only the first letter of each word for sub-headings and use 1.1, 1.2, etc to number the subheadings. Leave double space before section headings or sub-headings.3. FIGURES AND TABLESPlace them where they are first mentioned in the text. Figures must be drawn with the captions below them and sequentially numbered with Arabic numbers. Table captions should be typed above the tables.4. EQUATIONSAll equation numbers must be enclosed in parentheses and flushed right.5. REFERENCESReferences in the bibliography should be referred in the text by the number enclosed within square brackets. All references should be organised to provide, the last name of the author(s) followed by initials, title of the paper (in italics), journal name (in abbreviation), volume (underlined), year of publication (within brackets) page numbers.6. PAGINATIONNumber each page on the top right-hand corner of the page. The final pagination of proceedings will be numbered by the editor/publisher. Avoid footnotes as much as possible. Number of pages should be limited to a maximum of 8 (eight).

Institute of Physics, Sri LankaCOUNCIL MEMBERS - 2025/2026ISSN 2012-8975Printed and Published by the Institute of Physics, Sri LankaPresident: Prof. C Mahesh EdirisingheImmediate Past President: Dr. R C L De SilvaVice Presidents: Prof. W K I L Wanniarachchi Dr. (Ms.) H O Wijewardane Joint Secretaries: Prof. D L WeerawarneDr. U S RahubaddaTreasurer: Dr. R ThotagamugeAssistant Treasurer: Dr. H V U A Abeywickrema Co-Editors: Dr. (Mrs.) W W P De Silva Dr. R A D D Dharmasiri Elected Members: Prof. K P S C Jayaratne Prof. P A A Perera Prof. V P S Perera Prof. N G S S GamageDr. Binuka Gunawardane Dr. F S B Kafi Mr. G H Ashoka

Proceedings of the Technical Sessions, 42 (2026)Institute of Physics, Sri Lanka i Institute of Physics, Sri Lanka42nd Technical Sessions14th March 2026 from 0830 to 1400 hrsNPLT, Department of PhysicsUniversity of Colombo, Sri LankaTechnical PresentationsSession I (0830-1015 hrs) - Session Chairman: Dr. R C L De Silva Page 01. Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime Zones U G D Maduranga, K K S N Britto, and C M Edirisinghe1-1102. Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather Prediction S De Silva, S I Jayasinghe, D Wickramarathna, D U J Sonnadara, T Hettiarachchi, E H Mudiyanselage, A G U Perera, A Ashok, X He, and M K Jayananda12-1903. Gate Oxide Scaling and High-k Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD Study L S Sankalana, H V Ranasinghe, M K Jayananda, and L S Liyanage20-2704. A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm Casts K W D I Mawilmada, M H T Uthpala, and G D Illeperuma28-3605. Arm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy Models Y. L. Ramawickrama, P. A. A. Pererra, P. Hewageegana and S. Abdeen,37-4606. Synergistic Effects of MgO Passivation and CdS SILAR Cycle Optimization in TiO₂-Based Quantum Dot-Sensitized Solar Cells J V P Fernando, V P S Perera, and N F Ajward47-61

Proceedings of the Technical Sessions, 42 (2026)Institute of Physics, Sri Lanka ii Session II (1015 -1230 hrs) - Session Chairman: Prof. C M Edirisinghe Page07. Simulating Proton Synchrotron Acceleration and Proton–Proton Collision Dynamics in the CERN Accelerator Complex using Unity 3D N Wickramage, K M Liyanage, K A S Lakshan, and D M C M K Dissanayake62-7008. Analysis of Vertical Pitch Scaling and Channel Tapering Effects on the Program/Erase Characteristics of 3D NAND Flash Memory H V Ranasinghe, L S Sankalana, S H R T Sooriyagoda, L S Liyanage71-8009. End-to-End Simulation of Muon Scattering Tomography R M I D Gamage, M Lagrange, K M Liyanage, N M Wickramage, and M D S N Yasodara81-8910. Deep learning assisted identification of multi-layer graphene using optical microscopy K H Wathukarawatte, B Abeysinghe, B Gunawardana, and C R Munasinghe90-9511. Development of a fully inkjet printed antenna on polyethylene terephthalate for energy harvesting purposes in smart labelling W A I B J Wickramasinghe, S W D K R M Manamendra, W L P K Wijesinghe, G C Wickramasinghe, and D L Weerawarne96-105 Session III (1300 – 1420 hrs) - Session Chairman: Prof. V P S Perera Page12. Investigation of micro-cracks configuration in inkjet-printed silver traces on polyethylene terephthalate substrates for flex sensors W A C Perera, S W D K R M Manamendra, W L P K Wijesinghe, G C Wickramasinghe, and D L Weerawarne106-11313. Diurnal and Seasonal Asymmetry in Urban Thermal Discomfort in Colombo, Sri Lanka: Evidence from Daytime and Nighttime THI Analysis (1988–2018) S Chathuranga and K P S C Jayaratne114-12114. Assessing the Suitability of NASA POWER Temperature Products for Filling Climatic Data Gaps in Sri Lanka J K H Madushan, K P S C Jayaratne, and A L K Wijemannage122-13215. The Role of Physicists in Shaping the Future of Higher Education Reforms in Sri Lanka Prof. C. Mahesh Edirisinghe, President, Institute of Physics, Sri LankaPresidential Address 2026, Institute of Physics, Sri Lanka 133-139

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 1Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime ZonesAnalysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime ZonesU G D Maduranga1, K K S N Britto2, and C M Edirisinghe31Deparment of Physics, The Open University of Sri Lanka, Matara Regional Centre, Nupe, Matara 2Department of Mathematics, Faculty of Natural Sciences, The Open University of Sri Lanka3Department of Physics, Faculty of Science, University of [email protected]. ABSTRACTThis study provides an analysis of diurnal variation patterns of lightning activities over Sri Lanka’s Maritime Zones; Territorial Sea (T), Contiguous Zone (C) and Exclusive Economic Zone (E), during different climate seasons. For this study, 17 years of lighting data wereobtained from the Lightning Imaging Sensor (LIS) on the Tropical Rainfall Measuring Mission (TRMM) satellite from 1998 to 2014. During the first inter-monsoon season (March to April), the diurnal cycle of the lightning over all maritime regions peaked distinctly in the evening, with a maximum at 19:00- 20:00 Local Time (LT). In contrast, minimum lightning activities were recorded during the morning hours. During the southwest monsoon (May to September), a late evening maximum of the diurnal lightning cycle was recorded over the Contiguous zone (20:00-21:00 LT) and the Exclusive economic zone (21:00-22:00 LT), whereas the Territorial Sea is characterized by a pronounced early morning peak at 03:00-04:00 LT. Furthermore, a consistent minimum in activity is evident across all domains during the noon hours. Lightning activity during the second inter-monsoon (October to November), over the Territorial Sea and Contiguous Zone exhibited an evening maximum (20:00-21:00 LT and 19:00-20:00 LT, respectively), the Exclusive Economic Zone is characterized by a pronounced early morning peak at 01:00-02:00 LT. Diurnal variation of lightning activities in the northeast monsoon (December to February) shows that the Territorial Sea and Contiguous Zone experienced an evening maximum (peaking at 19:00-20:00 LT and 18:00-19:00 LT, respectively) and a morning minimum. Conversely, lightning activity over the Exclusive Economic Zone exhibitedan early morning peak (04:00-05:00 LT) and its minimum around midday. The crosscorrelation shows lightning activities in maritime zones T and C act as strong short-term factors of C and E, with T often leading C and C influencing E, whereas E exhibits weaker or inconsistent associations and no evidence of reverse causality. These findings are specifically important to figure out the lightning peak and minimum time periods over different maritime zones, providing a scientific basis for risk mitigation and vital for the fishing industry to optimize the strategic timing of operations and to mitigate risks associated with lightningrelated hazards.Keywords: Lightning, Lightning Hazard, Lightning Disasters, Maritime Zone, LIS, TRMM,

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 2Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime Zones2. INTRODUCTIONLightning has major impacts on the world, causing 6,000 and 24,000 fatalities annually worldwide [1], starting wildfires, damaging infrastructure, altering atmospheric chemistry, and influencing ecosystems and climate processes.Understanding of lightning distribution over the maritime zones of Sri Lanka is important because oceanic lightning is closely linked to deep convection, severe weather systems, and regional climate variability that affect both marine and coastal environments [2,3,4,5]. On the other hand, lightning over the ocean also contributes to atmospheric chemistry by producing nitrogen oxides and oxidants, influencing ozone formation and air-sea chemical interactions [6]. Satellite-based lightning observations are used in this study because they provide uniform, long-term, and spatially continuous coverage over maritime zones where ground-based lightning detection networks are sparse or absent [7,8].Due to the strategic location of Sri Lanka in the Indian Ocean, the surrounding maritime area is critically important for the country, making it a key player in global trade, regional security, and economic development. On the other hand, the fishing industry of the country is one of the major sources of income, contributing significantly to Sri Lanka’s Gross Domestic Product. Furthermore, the fishing industry provides direct and indirect employment for millions of people in the country. Lightning disasters can harm property and interrupt livelihoods in the agricultural, fishing, and maritime industries. Those damages reduce productivity and economic efficiency by interfering with transportation, industrial output, and essential services. Therefore, understanding atmospheric processes over Sri Lanka’s maritime zones is essential for improving weather forecasting, disaster management, and sustainable management of marine and coastal resources. The current study focuses on lightning activities over the maritime zones of the country, where direct observations are limited but lightning poses significant risks to offshore operations, coastal communities, and marine transportation [2,3].High convection and frequent thunderstorms occur regularly in Sri Lanka, a tropical nation, which increases the frequency of lightning and contributes to these effects. An analogous study shows that average lightning flash density in Sri Lanka is 8.26 flashes km−2year−1[9]. It is a major risk to human safety, leading to fatalities and injuries, particularly among rural people and outdoor workers, particularly in agriculture and fishing [10, 11]. In this study, Lightning Imaging Sensor (LIS) data on Tropical Rainfall Measurement Mission (TRMM) of NASA is used to analyze the diurnal variability of lightning activities in different seasonal periods. As Sri Lanka is lack of Lightning Location System (LLS) to collect the lightning data, the previousstudies have utilized LIS on TRMM data to analyze the lightning trend and distribution over particular areas of the country [9, 12, 13, 14].3. MATERIALS AND METHODS3..1 Study AreaSri Lanka’s maritime zones play a vital role in the country’s environmental, economic, and strategic framework due to its island location in the northern Indian Ocean. These zones include the Territorial Sea (12 Nautical Miles (≈ 22 ??) from coast), Contiguous Zone (24 Nautical Miles (≈ 44 ??) from coast), and Exclusive Economic Zone (200 Nautical Miles (≈ 370 ??)

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 3Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime Zonesfrom coast), which together support fisheries, maritime transport, offshore resources, and coastal livelihoods. These regions are strongly influenced by monsoonal systems, convection, and ocean-atmosphere interactions, making it subject to extreme weather conditions such as thunderstorms and lightning. Figure 1 shows the maritime boundaries of Sri Lanka. According to the reversal of monsoon winds associated with the Inter-Tropical Convergence Zone (ITCZ), Sri Lanka experiences a tropical monsoon climate with two monsoon seasons; SouthwestMonsoon (May to September) and Northeast Monsoons (December to February), as well as two inter-monsoon periods; (First Inter-monsoon (March to April) and second inter-monsoon seasons(October to November).Figure 1: Study area including maritime boundaries of Sri Lanka3.2 Lightning DataThe Lightning Imaging Sensor (LIS) was launched in November 1997 aboard the Tropical Rainfall Measuring Mission (TRMM) satellite and remained operational until April 2015. The TRMM satellite followed a low-Earth orbit with a 35° inclination, resulting in less frequent observations near the equator compared to locations closer to the ±35° latitude bands [15]. Due to this orbital configuration, the LIS required a minimum revisit period of approximately 49 days to observe most regions of the Earth at least once during each local solar hour of the diurnal cycle [16, 17]. The instrument observed an area of approximately 600 km × 600 km with a spatial resolution ranging from 3 km to 6 km and monitored individual storm systems for about 80 seconds per overpass. Over Sri Lanka, the satellite passed twice daily during both daytime and nighttime, providing a total observation duration of approximately 160 s day⁻¹ [18,19]. The LIS is capable of detecting the spatial location, timing, and radiant energy of lightning events, with detection efficiencies ranging from about 69% near local noon to 88% at night [3]. For this study, total lightning data spanning the period from 1998 to 2014 over Sri Lanka and its surrounding coastal region (5.75°N–10.00°N and 79.50°E–89.00°E) were obtained from [20].76.00000076.00000078.00000078.00000080.00000080.00000082.00000082.00000084.00000084.00000086.00000086.000000 2.0000002.0000005.0000005.0000008.0000008.000000 11.000000 11.000000Exclusive Economic Zone Sri LankaIndian SubcontinentIndian OceanTerritorial seaContiguous zone Indo Sri Lanka Maritime BoundaryIndo Sri Lanka Maritime Boundary ¯

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 4Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime Zones3.3 MethodologyThis study utilizes a long-term dataset spanning from 1998 to 2014 to analyze lightning activity over the study region. The data were spatially separated and organized using Geographic Information System (GIS) software, allowing accurate extraction of total lightning occurringwithin defined geographic boundaries. Furthermore, the lightning observations were analyzed using hour-wise (diurnal) data, enabling an assessment of temporal variations and peak activity periods. To investigate temporal interactions among the maritime zones; Territorial Sea(T), Contiguous Zone (C), and Exclusive Economic Zone (E), cross-correlation analysis was conducted using IBM SPSS Statistics 23, enabling the identification of leading and lagging relationships, short term and medium term dependencies, and potential causal patterns among the variables across different maritime zones and monsoon seasons.4. RESULTS 4.1 Territorial SeaThe territorial sea is considered as a maritime zone extending up to 12 nautical miles from the country's baseline (Figure 01), within which the state exercises sovereign authority. According to the right of free passage for foreign maritime vessels, this sovereignty involves control over the seabed below and the airspace above. Figure 02 shows the hourly distribution of total number of lightning flashes over the Territorial Sea across four monsoons over a 24-hour period from 1998 to 2014. Overall, lightning activity is low during the early morning hours, increases gradually from midday, and peaks in the late afternoon to evening (around 15:00-20:00 Local Time (LT)), indicating strong diurnal influence. Figure 2: Hourly distribution of total number of lightning flashes over the Territorial Sea from 1998 to 2014The First Inter Monsoon (FIM) and South-West Monsoon (SWM) regions exhibit the highest lightning activity, reaching maximum values of 194 flashes at 19:00-20:00 LT and 186 flashes at 03:00-04:00 LT, respectively. In contrast, the Northeast Monsoon (NEM) records the minimum activity, with values remaining below zero flashes per hour throughout the day. The Second Inter Monsoon (SIM) displays moderate activity with a noticeable evening peak, suggesting regional differences in convective development and storm intensity. Comparatively, minimum lighting activities have occurred during the North-East Monsoon (NEM) over the Territorial Sea area.

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 5Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime ZonesPrevious study shows that over the landmass, a maximum number of lightning activities have been recorded during FIM [4], and this is mainly entwined with, winds and surface current over Territorial Sea, which leads to increased convectional rainfall and suddenly, leading to strong late afternoon and evening thunderstorms [21]. It may be a risk for small-scale coastal fishing. In SWM, winds blow from the southwest towards the northeast with strong surface currents.Thunderstorms may happen often during night or early morning on the west coast. The analysis indicates that only two lightning flashes were recorded during SIM, whereas no events were observed during NEM, representing the lowest seasonal totals. 4.2 Contiguous ZoneThe Contiguous Zone is a maritime area extending up to 24 nautical miles from a country’s baseline, beyond the Territorial Sea (Figure 01). Figure 03 shows a clear diurnal pattern in lightning activity, with generally low counts during the early morning and midday hours, followed by a sharp increase in the late afternoon and evening (around 17:00-21:00 LT). Overthis region, SWM records the highest peak, especially around 20:00-21:00 LT, while FIM and SIM also show strong evening maxima but with slightly lower intensities. Lightning flashes over the Contiguous Zone occurred with very low frequency during the NEM over the period 1998–2014 compared to the other seasons.Figure 3: Hourly distribution of total number of lightning flashes over the Contiguous Zonefrom 1998 to 2014During FIM, over the Contiguous Zone, high solar angle produces strong daytime heating, and strong land heating drives convection that propagates seaward in the evening [22]. In SWM, thick cloud cover reduces daytime solar heating, and often, nocturnal or early morning offshorewinds can be experienced. In SIM, due to the high atmospheric instability and moisture content, thunderstorms often occur intensely at night. As well as daytime warming over the aforementioned cause to thunderstorms generation during the night in NEM [23]. 4.3 Exclusive Economic ZoneAn Exclusive Economic Zone is a sea area extending up to 200 nautical miles from a country’s coastline (Figure 01) where the coastal state has special rights to explore, use, and manage natural resources. The coastal state controls economic activity, including fishing, energy production, and scientific research while other states are allowed to fly and travel over theExclusive Economic Zone. Figure 04 shows the diurnal variation of lightning activity across the four monsoons over the Exclusive Economic Zone. SWM dominates overall, with sharp

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 6Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime Zonespeaks around 4:00-5:00 LT hours and again near 21:00-22:00 LT, indicating periods of intense convective activity. In contrast, the NEM consistently records the lowest lightning counts, suggesting relatively stable atmospheric conditions [24]. Figure 4: Hourly distribution of total number of lightning flashes over the Exclusive Economic Zone from 1998 to 2014The FIM and SIM show moderate fluctuations, with noticeable increases in the late afternoon and evening, pointing to a diurnal pattern likely influenced by heating and local weather dynamics. Over the Exclusive Economic Zone, strong daytime heating triggers with convective cloud development, causing peaks in the afternoon or night [25]. 5. DISCUSSION5.1 Cross-Correlation Analysis of Temporal Interactions during FIM Figure 5 illustrates the cross-correlation analysis of temporal interactions during FIM. The cross-correlation results indicate a strong and statistically significant positive relationshipbetween Territorial Sea over first inter-monsoon (T_FIM) and Contiguous Zone over first intermonsoon (C_FIM) at lags 0 to 4, with coefficients at lag 0 (0.580), lag 1 (0.639), lag 2 (0.442), lag 3 (0.497), and lag 4 (0.642) all exceeding the 95% significance bounds (approximately ±0.408, calculated as ±2 × standard error), where a lag represents the number of time periods by which one series is shifted relative to the other. The relationship between T_FIM and Exclusive Economic Zone over first inter-monsoon (E_FIM) is statistically insignificant at all lags, as none of the coefficients exceed the 95% significance bounds (±0.408), with the highest value being 0.387 at lag 4. In contrast, C_FIM and E_FIM exhibit a significant positive correlation at lag 0 (0.472), exceeding the critical threshold and indicating a contemporaneous relationship. However, the lack of significant correlations at other lags suggests no clear leadlag or causal relationship over time. Overall, the findings indicate that T_FIM is an important factor of C_FIM with subsequently positive effects, whereas E_FIM only exhibits a corresponding, inconsistent correlation with C_FIM and no causal relationship with T_FIM.

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 7Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime ZonesFigure 5: Cross-Correlation Analysis of Temporal Interactions during FIM5.2 Cross-Correlation Analysis of Temporal Interactions during SWMThe cross-correlation analysis of temporal interactions during SWM is shown in Figure 6. The cross-correlation analysis for SWM shows that Territorial Sea over South-West Monsoon (T_SWM) and Contiguous Zone over South-West Monsoon (C_SWM) have significant positive correlations at lags 0 (r ≈ 0.60), 1 (r ≈ 0.52), and 3 (r ≈ 0.50), exceeding the ±0.40 (95%) confidence bounds, indicating that T_SWM leads C_SWM by up to three periods with no reverse causality. In contrast, T_SWM and Exclusive Economic Zone over South-West Monsoon (E_SWM) are insignificant at lag 0 but become significant at lags 4 (r ≈ 0.55) and 5 (r ≈ 0.50), suggesting a delayed effect, while C_SWM and E_SWM show strong significant positive correlations from lags 1–5 (peaking at r ≈ 0.65 at lag 4), confirming C_SWM as a leading factor for E_SWM without reverse causality.Figure 6: Cross-Correlation Analysis of Temporal Interactions during Southwest Monsoon5.3 Cross-Correlation Analysis of Temporal Interactions during SIMFigure 7 depicts the cross-correlation analysis of temporal interactions during SIM. The results show a statistically significant contemporaneous relationship between Territorial Sea over Second Inter-Monsoon (T_SIM) and Contiguous Zone over Second Inter-Monsoon (C_SIM) at lag 0 (r ≈ 0.62), with additional significant correlations at lags ±1 (r ≈ 0.55 and r ≈ 0.48), exceeding the 95% confidence bounds (±0.40), indicating a short-term bidirectional effect limited to one period. T_SIM and Economic Zone over Second Inter-Monsoon (E_SIM) are insignificant at lag 0 (r ≈ 0.18) but become significant at lags 2–5 (r ≈ 0.44–0.52), showing that T_SIM leads E_SIM over 2–5 periods. Similarly, C_SIM and E_SIM display significant

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 8Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime Zonespositive correlations at lag 0 (r ≈ 0.43) and lags 1–4 (peaking at r ≈ 0.60), with no significant negative lags, confirming that C_SIM is a stronger leading indicator of E_SIM without evidence of reverse causality.Figure 7: Cross-Correlation Analysis of Temporal Interactions during Second Inter-Monsoon5.4 Cross-Correlation Analysis of Temporal Interactions during NEMFigure 8 illustrates the cross-correlation analysis of temporal interactions during NEM. The cross-correlation results indicate that Contiguous Zone over Northeast Monsoon (C_NEM) and Economic Zone over Northeast Monsoon (E_NEM) have weak contemporaneous correlation at lag 0 (r ≈ 0.10, within the ±0.45 95% confidence bounds), but significant positive correlations at lags 2 (r ≈ 0.32), 3 (r ≈ 0.35), and 5 (r ≈ 0.30), exceeding the upper confidence limit, showing that C_NEM leads E_NEM by about 2–5 periods. For Territorial Sea over Northeast Monsoon (T_NEM) and E_NEM, the correlation at lag 0 is insignificant (r ≈ 0.05), while significant negative correlations appear at lags −4 to −6 (r ≈ −0.50 to −0.60), falling below the lower confidence bound, indicating an inverse relationship where past E_NEM influences current T_NEM; moderate significant positive correlations are also observed at lags 3 and 5 (r ≈ 0.35–0.40). In contrast, T_NEM and C_NEM display a strong and statistically significant contemporaneous correlation at lag 0 (r ≈ 0.70), with additional significant correlations at lags −1 (r ≈ 0.45), 2 (r ≈ 0.60), and 3 (r ≈ 0.55), all exceeding the ±0.45 bounds, confirming strong short-term coupling within one to three periods. Overall, the findings demonstrate a robust short-run linkage between T_NEM and C_NEM, a leading effect of C_NEM on E_NEM over 2–5 periods, and significant delayed partly inverse effects between E_NEM and T_NEM.This study approaches to improve the current understanding on diurnal variation patterns of lightning activities over the maritime region of the country across the different seasonal periods. This information is critically important to predict thunderstorm weather and as well as to establish some safety precautions to reduce the lightning related accidents in sea areas. Furthermore, reduction of lightning related accidents would enhance production efficiency inthe fishing industry and would improve the safety in maritime transportation.

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 9Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime ZonesFigure 8: Cross-Correlation Analysis of Temporal Interactions during Northeast Monsoon6. CONCLUSIONThis study utilizes a long-term satellite-based lightning dataset spanning from 1998 to 2014 to investigate the seasonal variation of lightning activity over Sri Lanka’s maritime regions. The results reveal clear diurnal patterns in lightning activity across the maritime zones, with the Territorial Sea peaking at 19:00–20:00 LT (and an early morning peak at 03:00–04:00 LT during the southwest monsoon), the Contiguous Zone showing maxima at 18:00–21:00 LT depending on the season, and the Exclusive Economic Zone exhibiting early morning peaks between 01:00–05:00 LT. Lightning activity is consistently minimal around noon across all zones, providing quantitative evidence of temporal variability influenced by local convection and land–sea interactions. The First Inter Monsoon and South-West Monsoon emerge as the most lightning active periods, particularly over the Territorial Sea (T) and Contiguous Zone(C), while the North-East Monsoon consistently records the lowest lightning activity. Over the Exclusive Economic Zone (E), the South-West Monsoon dominates, with notable nocturnal and early morning peaks associated with offshore convection and large-scale atmospheric circulation. The cross-correlation shows lightning activities in regions T and C act as strong short-term factors of C and E, with T often leading C and C influencing E, whereas E exhibits weaker or inconsistent associations and no evidence of reverse causality. These findings enhance the understanding of seasonal and diurnal lightning variability over maritime regions and provide valuable insights for improving thunderstorm forecasting, maritime safety planning, and risk mitigation strategies, especially for coastal fisheries, offshore operations, and marine transportation.7. REFERENCES[1] Holle, R. L. (2016). \"The Number of Documented Global Lightning Fatalities.\" Weather, Climate, and Society, 8(1), 85–93.[2] Christian, H. J., Blakeslee, R. J., Boccippio, D. J., et al. (2003). Global frequency and distribution of lightning as observed from space by the Optical Transient Detector and Lightning Imaging Sensor. Journal of Geophysical Research: Atmospheres, 108(D1), 4005. [3] Holle, R. L. (2014). Annual rates of lightning fatalities by country. Weather, Climate, and Society, 6(4), 434–444.

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 10Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime Zones[4] Zipser, E. J., Cecil, D. J., Liu, C., Nesbitt, S. W., & Yorty, D. P. (2006). Where are the most intense thunderstorms on Earth? Bulletin of the American Meteorological Society, 87(8), 1057–1071. [5] Virts, K. S., Wallace, J. M., Hutchins, M. L., & Holzworth, R. H. (2013). Diurnal and seasonal variability of lightning activity over the tropical warm pool. Journal of Climate, 26(15), 5641–5660. https://doi.org/10.1175/JCLI-D-12-00605.1[6] Schumann, U., & Huntrieser, H. (2007). The global lightning-induced nitrogen oxides source. Atmospheric Chemistry and Physics, 7, 3823–3907.[7] Cecil, D. J., Buechler, D. E., & Blakeslee, R. J. (2014). Gridded lightning climatology from TRMM‐LIS and OTD: Dataset description. Atmospheric Research, 135–136, 404–414. [8] Goodman, S. J., Blakeslee, R. J., Koshak, W. J., et al. (2013). The GOES-R Geostationary Lightning Mapper (GLM). Atmospheric Research, 125–126, 34–49.[9] Edirisinghe, M.; Maduranga, U.G.D. Distribution of Lightning Accidents in Sri Lanka from 1974 to 2019 Using the DesInventar Database. ISPRS Int. J. Geo-Inf. 2021, 10, 117. [10] Holle, R. L. (2016). A summary of recent national-scale lightning fatality studies. Weather, Climate, and Society, 8(1), 35–42. [11] Cooper, M. A., Holle, R. L., Andrews, C. J., & López, R. E. (2019). Lightning injuries and deaths in the United States: A 30-year perspective. Journal of Climate, 32(12), 4063–4075. [12] Maduranga, U. G. D., Edirisinghe, M., & Gamage, L. V. (2018). Annual Variation Trend of Lightning Flash Activities over Sri Lanka. World Scientific News, 114, 256–264.[13] Maduranga, U., Edirisinghe, M., & Gamage, L. V. (2019). Spatiotemporal Variability of Lightning Flash Distribution over Sri Lanka. International Letters of Chemistry Physics andAstronomy, 82, 1–13.[14] Kalapuge, V.; Maduranga, D.; Alahacoon, N.; Edirisinghe, M.; Abeygunawardana, R.; Ranagalage, M. Overview of Lightning Trend and Recent Lightning Variability over Sri Lanka. ISPRS Int. J. Geo-Inf. 2023, 12, 67. [15] Kummerow, C., Barnes, W., Kozu, T., Shiue, J., & Simpson, J. (1998). The Tropical Rainfall Measuring Mission (TRMM) sensor package. Journal of Atmospheric and Oceanic Technology, 15(3), 809–817. [16] Christian, H. J., Blakeslee, R. J., Goodman, S. J., & Mach, D. M. (1999). The Lightning Imaging Sensor. Proceedings of the 11th International Conference on Atmospheric Electricity, Guntersville, Alabama.

Proceedings of the Technical Sessions, 42 (2026) 1-11Institute of Physics, Sri Lanka 11Analysis of Diurnal and Seasonal Variation of Lightning Activities over Sri Lanka’s Maritime Zones[17] Boccippio, D. J., Koshak, W. J., & Blakeslee, R. J. (2002). Performance assessment of the Optical Transient Detector and Lightning Imaging Sensor. Journal of Atmospheric and Oceanic Technology, 19(8), 1318–1332. [18] Cecil, D. J., Buechler, D. E., & Blakeslee, R. J. (2014). Gridded lightning climatology from TRMM‐LIS and OTD: Dataset description. Atmospheric Research, 135–136, 404–414. [19] Goodman, S. J., Blakeslee, R. J., Koshak, W. J., et al. (2013). The GOES-R Geostationary Lightning Mapper (GLM). Atmospheric Research, 125–126, 34–49.[20] Blakeslee, R. (2025). Lightning Imaging Sensor (LIS) on TRMM Science Data [Data set]. NASA Global Hydrometeorology Resource Center Distributed Active Archive Center.https://doi.org/10.5067/LIS/LIS/DATA201 Date Accessed: 2025-05-22[21] Nagamuthu, P., & Weng, C. N. (2020). Convectional Influences on The Weather Pattern of Northern Sri Lanka. European Proceedings of Social and Behavioural Sciences, 89, 477-487. https://doi.org/10.15405/epsbs.2020.10.02.43[22] Atkins, N.T. (1991). Mesoscale Convective Complexes and Sea-Breeze Interactions: Coastal Convective Initiation Mechanisms. Monthly Weather Review, 119, 2993-3016[23] Ancy, P., Varikoden, H., & Babu, C. A. (2025). Physical mechanism of diurnal variability of onshore and offshore summer monsoon rainfall along the west coast of India. Science of the Total Environment, 983, 179681. https://doi.org/10.1016/j.scitotenv.2025.179681[24] Nesbitt, S. W., & Zipser, E. J. (2003). The diurnal cycle of rainfall and convective intensity according to three years of TRMM measurements. Journal of Climate, 16(10), 1456–1475. https://doi.org/10.1175/1520-0442(2003)016<1456:TDCORA>2.0.CO;2[25] Nandi, S., & Ramanathan, V. (2021). Diurnal variation of deep convective clouds over Indian monsoon region and its association with rainfall. Atmospheric Research, 255, 105540. https://doi.org/10.1016/j.atmosres.2021.105540

Proceedings of the Technical Sessions, 42 (2026) 12-19Institute of Physics, Sri Lanka 12Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather PredictionPreliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather PredictionShehara De Silva1, Saumya I. Jayasinghe1, Deshitha Wickramarathna1, Kithsiri Jayananda1*, Upul Sonnadara1, Tharindu Hettiarachchi2, Enosh H. Mudiyanselage2, A. G. Unil Perera2, Ashwin Ashok2, Xiaochun He21Department of Physics, Faculty of Science, University of Colombo, Sri Lanka2Department of Physics and Astronomy, Georgia State University, Atlanta, [email protected]. ABSTRACTThis study presents a preliminary investigation of atmospheric muon flux measurements in Sri Lanka using two muon scintillation detectors at the Department of Physics, University of Colombo. The detectors are from the Georgia State University ‘global Low-Cost Observation of the dynamic changes in the Space weather and Terrestrial weather’ (gLOWCOST) collaboration. Direct muon counts from the three scintillator plates and muon coincidence counts between any two plates are utilized to investigate the sensitivity of the muon detectors with respect to ambient temperature, as well as the relationship between muon flux and atmospheric parameters such as air temperature, air pressure, and relative humidity. Oneminute data obtained from the detectors were used in the former investigation, while hourly averaged count data were used in the latter investigation. A significant dependence on ambient temperature was observed in Detector 1, whereas the improved Detector 2 showed no significant dependence. A negative correlation was observed between muon flux and air pressure with a gradient of − 0.188 hPa−1. A smaller positive correlation was observed with air temperature with a gradient of 0.064 ℃−1. The observed trends and relationships are consistent with existing literature for non-equatorial regions and proposed theories for equatorial regions. Therefore, this study concludes that, when operated under varying weather conditions, muon flux measurements have the potential to be used as indicators of weatherrelated changes.Key words: Muon flux, gLOWCOST, ambient temperature, atmospheric parameters2. INTRODUCTIONCosmic rays are high-energy particles from outer space that predominantly consist of protons and helium nuclei, with smaller amounts of heavy nuclei. They are of two types: primary cosmic rays, which are particles that travel through space and strike the Earth’s atmosphere, and secondary cosmic rays, which are produced when primary cosmic rays collide with nuclei in the Earth’s atmosphere. The hadronic shower that is then created consists of hadrons such as protons, pions, and kaons. These pions decay into muons that travel towards the Earth’s surface.

Proceedings of the Technical Sessions, 42 (2026) 12-19Institute of Physics, Sri Lanka 13Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather PredictionAlthough the lifetime of a muon is only 2.2 μs, due to time dilation in the rest frame of the muon, it travels further than expected and can be detected at the Earth’s surface [1], [2].The muon scintillation detectors of Georgia State University’s gLOWCOST collaboration are ground-based detectors that measure the muon flux at the Earth’s surface. Among these detectors, which are distributed worldwide in locations such as the United States, Japan, Singapore, Sweden, and Sri Lanka, this study focuses on the two detectors located at the Department of Physics, University of Colombo, Sri Lanka. Detector 1 was operationalized in October 2023, while Detector 2 was operationalized in July 2024 (Figure 1).According to existing literature and proposed theories, the muon flux varies with terrestrial atmospheric parameters such as air temperature, air pressure, and relative humidity [3], [4], [5], [6], [7]. Studies have been conducted in many geographical locations; however, there is a lack of such research in equatorial regions such as Sri Lanka. Furthermore, the silicon photo multiplier’s output is known to vary with ambient temperature [8].Thus, the focus of this study is twofold. The first is to study the performance of the detectors with respect to ambient temperature, and the second is to observe trends in the muon flux in relation to terrestrial atmospheric parameters such as air temperature, air pressure, and relative humidity in Colombo.3. METHODOLOGYEach gLOWCOST muon scintillation detector consists of three plastic scintillator plates, wavelength-shifting (WLS) fibers, and silicon photo multipliers (SiPM). The number of muons that pass through each plate is recorded over a 1-minute period. The principle of coincidence counts is utilized for detection. A muon is considered to have passed through the detector only if it is detected by two scintillator plates. The coincidence counts between the top and middle plates (coincidence_0_1), between the middle and bottom plates (coincidence_1_2), and between the top and bottom plates (coincidence_0_2) are recorded each minute.Each of the three plastic scintillator plates has the dimensions of (20 × 20 × 1) cm3, while the perpendicular distance between two scintillator plates is 13 cm. The SiPM mounted at the corner of the plate records the scintillation light that is created when a muon passes through the plate. The voltage of the SiPM is supplied by a small high voltage unit integrated to the main data acquisition and coincidence processing unit mounted on the Raspberry Pi [9].In both detectors 1 and 2 shown in figure 1, the coincidence count combinations are recorded as stated previously. In addition, Detector 1 records the direct counts individually from each scintillator plate, which are ch0 (top plate), ch1 (middle plate), and ch2 (bottom plate). Detector 2, however, provides only the direct count from the top scintillator plate due to limitations in the Field-Programmable Gate Array (FPGA) output, which was used for coincidenceprocessing. Further, the ambient temperature, air temperature, air pressure, and relative humidity needed for the study were measured every minute using two BME280 sensors.

Proceedings of the Technical Sessions, 42 (2026) 12-19Institute of Physics, Sri Lanka 14Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather PredictionFigure 1: The two muon scintillation detectors in Colombo. Left: Detector 1. Right: Detector 2.4. RESULTS AND DISCUSSIONThe muon flux coincidence counts between the scintillator plates are recorded every minute. However, to improve the signal to noise ratio, the muon flux coincidence counts are taken as an average over a five minute or one hour interval (figure 2). Further, while the coincidence counts can be directly plotted, the muon flux changes are easier to visualize when plotted as a percentage change in the coincidence count with respect to the mean coincidence count.Figure 2: Left: Muon percentage change of the Detector 2 coincidence count between the middle and bottom scintillator plates (one-minute, five-minute average, hourly average) with respect to the mean value for the period from May 3, 2025 to May 4, 2025. Right: The hourlyaverage muon percentage change plotted separately for the same period.4.1 Ambient Temperature Dependence of the DetectorIn Detector 1, a significant variation in the count rates with ambient temperature was observed. However, Detector 2, which is the improved version of Detector 1, does not show significant variation with ambient temperature. For this analysis, one-minute direct channel (plate) muon counts and muon coincidence counts were utilized, as depicted in figures 3 and 4 respectively.

Proceedings of the Technical Sessions, 42 (2026) 12-19Institute of Physics, Sri Lanka 15Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather PredictionFigure 3: Left: Detector 1 all channels (scintillation plates) direct counts variation with the ambient temperature. Right: Detector 2 channel 0 (top plate) direct muon count variation with the ambient temperature.The linear correlation coefficient for Detector 1 is high, with R2 values of 0.9726 for channel 0 (top plate), 0.9270 for channel 1 (middle plate), and 0.9681 for channel 2 (bottom plate), where R2represents the proportion of variance explained by the model. In contrast, Detector 2 shows a very low linear correlation coefficient, with an R2value of 0.0005 for channel 0.In Detector 1, all channel direct counts show a strong linear relationship with ambient temperature. From the perspective of the SiPMs, since it has a temperature dependence, a correction can be done to it by varying the SiPM bias voltage.These observations indicate that gain variation of the SiPM is the primary factor responsible for the decrease in count rates with increasing ambient temperature. Based on SiPM datasheets and technical notes, this gain variation can be compensated by appropriately adjusting the bias voltage as a function of temperature [8].Figure 4: Variation of muon coincidence counts between scintillation plates (top and middle, top and bottom, middle and bottom) with ambient temperature. Left: Colombo Detector 1. Right: Colombo Detector 2.

Proceedings of the Technical Sessions, 42 (2026) 12-19Institute of Physics, Sri Lanka 16Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather PredictionOnce again, the linear correlation coefficient for Detector 1 is high, with R2 values of 0.7457for the coincidence between the top and middle plates, 0.6042 for between the top and bottom plates, and 0.7265 for between the middle and bottom plates. In contrast, Detector 2 again shows a very low linear correlation coefficient, with R2values of 0.0091, 0.0154, and 0.0108 respectively for the three coincidence counts. The linear relationship between the coincidence counts and the ambient temperature in Detector 2 is weak and the R2values are low. Thus, Detector 2 can be utilized for muon flux investigations without performing ambient temperature correction.4.2 Variation of Muon Flux with Atmospheric ParametersThis section focuses on whether there is a noticeable effect on the muon flux in Colombo, Sri Lanka by the terrestrial atmospheric parameters. Fast Fourier Transforms (FFT) were performed to the hourly-averaged muon coincidence count between the middle and bottom plates (coincidence_1_2), the air temperature, air pressure and relative humidity values for the period from January 1, 2025 to December 28, 2025, with the focus on the number of cycles per day (figure 5). Detector 2 was utilized for terrestrial atmospheric parameter analysis.Figure 5: FFT performed on the hourly-averaged muon coincidence_1_2 (top-left), air pressure(top-right), air temperature (bottom-left), and relative humidity (bottom-right) for the time period from January 1, 2025 to December 28, 2025.The FFT for air pressure yielded a dominant peak at two cycles/day, while that for temperature and relative humidity yielded a dominant peak at one cycle/day. The observation for air pressure is consistent with the semidiurnal barometric cycle that causes two 12-hour cycles/day.

Proceedings of the Technical Sessions, 42 (2026) 12-19Institute of Physics, Sri Lanka 17Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather PredictionThis occurs because the atmosphere has a natural resonant semidiurnal mode, which is excited by solar heating. Near the Equator, warm convective currents further strengthen this cycle, contributing to the pronounced semidiurnal variations observed here [10], [11].The FFT for the muon coincidence count between the middle and bottom plates yielded dominant peaks at one and two cycles/day. The other peak at four cycles/day could be due to harmonics, while the subsequent smaller peaks appear to be due to noise. The dominant peaks in the muon coincidence count could be explained using the three atmospheric parameters considered in this study.The coincidence count of Detector 2 was visualized in relation to air temperature and air pressure for the period from May 17, 2025 to November 17, 2025. Binned hourly-averaged coincidence counts were considered for atmospheric parameter bins of size 0.2 units. Below are the plots for muon coincidence_1_2.Figure 6: Muon flux coincidence counts of Detector 2 between the middle and bottom scintillation plates in relation to air temperature (left) and air pressure (right) for the period from May 17, 2025 to November 17, 2025, along with the best fit line and atmospheric coefficients in the form of the gradient.A negative effect of − 0.188 hPa−1 was observed for the muon count with air pressure in Colombo. According to current literature, an increase in air pressure would cause an increase in the atmosphere for muons to pass through, which would cause an increase in the mass absorption of cosmic rays in air, which would then decrease the number of muons reaching the Earth’s surface. This creates a negative effect with air pressure [4], [6]. Thus, the pressure gradient obtained in Colombo is comparable to those obtained in the non-equatorial regions considered in literature.A positive effect of 0.064 ℃−1 was observed for the muon count with air temperature. This too can be explained by current literature, even though experimental observations are lacking in equatorial regions. An increase in air temperature could have either a positive effect or a negative effect on muon counts. When the atmosphere expands, on one hand, it would cause the muon production region to rise higher in the atmosphere, which would then cause more muons to decay before reaching the ground level. Thus, fewer muons would be detected,

Proceedings of the Technical Sessions, 42 (2026) 12-19Institute of Physics, Sri Lanka 18Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather Predictioncausing a negative effect with temperature. On the other hand, the expanding atmosphere would result in decreasing air density, making pions less likely to collide or be absorbed before they decay. This would cause more pions to decay into muons. Thus, more muons would be detected at the ground level. This is the positive effect with temperature [4], [5], [6], [7].Colombo is an equatorial region. Near the equator, the incident primary rays are more energetic, therefore positive effect would dominate, while further away from the equator, the temperature coefficient should get increasingly negative. The latter has been observed in current literature. The high geomagnetic cutoff rigidity of 17.694 GV at Sri Lanka is reason for this expectation[3], [12].5. CONCLUSIONThis preliminary study investigates the sensitivity of the muon detectors with respect to ambient temperature and the variation of atmospheric muon flux measurements with terrestrial atmospheric parameters in two muon gLOWCOST muon detectors installed at the Department of Physics, University of Colombo, Sri Lanka. Unlike Detector 1, Detector 2, which is its improved version, showed no significant dependence with ambient temperature for both the direct channel counts and the coincidence counts. The muon flux of Detector 2 reflects the well-known semidiurnal barometric variation cycle. Consistent with current literature for non-equatorial regions and proposed theories for equatorial regions, an anti-correlation with air pressure and a correlation with air temperature were observed for the muon coincidence count of Detector 2. Future studies include the analysis of muon flux with space weather effects like solar flares and geomagnetic storms, the analysis of muon flux with other terrestrial atmospheric parameters like dew point and wind direction, and the observation of the magnetic east west asymmetry of the muon flux by varying the zenith angle.6. ACKNOWLEDGEMENTGeorgia State University gLOWCOST collaboration for the muon scintillation detectors is acknowledged.7. REFERENCES[1] A. Duperier, “On the positive temperature effect of the upper atmosphere and the process of meson production,” J. Atmospheric Terr. Phys., vol. 1, no. 5–6, pp. 296–310, 1951, doi: 10.1016/0021-9169(51)90004-9.[2] P. K. F. Grieder, Cosmic Rays at Earth: Researcher’s Reference Manual and Data Book. Elsevier, 2001. doi: 10.1016/B978-0-444-50710-5.X5000-3.[3] P. M. S. Blackett, “On the Instability of the Barytron and the Temperature Effect of Cosmic Rays,” Phys. Rev., vol. 54, no. 11, pp. 973–974, Dec. 1938, doi: 10.1103/PhysRev.54.973.

Proceedings of the Technical Sessions, 42 (2026) 12-19Institute of Physics, Sri Lanka 19Preliminary Investigation of Atmospheric Muon Flux Detection in Sri Lanka and its Potential for Weather Prediction[4] R. R. S. De Mendonça et al., “THE TEMPERATURE EFFECT IN SECONDARY COSMIC RAYS (MUONS) OBSERVED AT THE GROUND: ANALYSIS OF THE GLOBAL MUON DETECTOR NETWORK DATA,” Astrophys. J., vol. 830, no. 2, p. 88, Oct. 2016, doi: 10.3847/0004-637X/830/2/88.[5] M. Savić et al., “A novel method for atmospheric correction of cosmic-ray data based on principal component analysis,” Astropart. Phys., vol. 109, pp. 1–11, May 2019, doi: 10.1016/j.astropartphys.2019.01.006.[6] A. H. Maghrabi, S. A. Alzahrani, and A. S. Alruhaili, “The Role of Atmospheric Pressure, Temperature, and Humidity on Cosmic Ray Muons at a Low Latitude Station,” Int. J. Astron. Astrophys., vol. 13, no. 03, pp. 236–258, Sep. 2023, doi: 10.4236/ijaa.2023.133014.[7] C.-L. Xu, Y. Wang, G. Qin, and P.-B. Zuo, “Correction of the Temperature Effect of Muon Counts Observed at the Guangzhou Station,” Res. Astron. Astrophys., vol. 23, no. 2, p. 025010, Feb. 2023, doi: 10.1088/1674-4527/acac05.[8] Hamamatsu Photonics K.K., “Technical note - MPPC,” Feb. 2025, [Online]. Available: https://www.hamamatsu.com/content/dam/hamamatsuphotonics/sites/documents/99_SALES_LIBRARY/ssd/mppc_kapd9008e.pdf[9] X. He et al., “Development and Production of Modular Cosmic Ray Telescopes,” in Proceedings of 37th International Cosmic Ray Conference — PoS(ICRC2021), Berlin, Germany - Online: Sissa Medialab, Jul. 2021, p. 1257. doi: 10.22323/1.395.1257.[10] S. Chapman and R. S. Lindzen, Atmospheric Tides. Dordrecht: Springer Netherlands, 1969. doi: 10.1007/978-94-010-3399-2.[11] A. Dai and J. Wang, “Diurnal and Semidiurnal Tides in Global Surface Pressure Fields,” J. Atmospheric Sci., vol. 56, no. 22, pp. 3874–3891, Nov. 1999, doi: 10.1175/1520-0469(1999)056<3874:DASTIG>2.0.CO;2.[12] Center for Cosmic Ray Studies at GSU, “Daily Update – Center for Cosmic Ray Studies at GSU.” Accessed: Sep. 22, 2025. [Online]. Available: https://cosmic.gsu.edu/glowcostdata-access/

Proceedings of the Technical Sessions, 42 (2026) 20-27Institute of Physics, Sri Lanka 20Gate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD StudyGate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs:A Sentaurus TCAD StudyL S Sankalana, H V Ranasinghe, M K Jayananda, and L S LiyanageDepartment of Physics, Faculty of Science, University of Colombo, Sri [email protected]. ABSTRACTTransistor scaling is crucial for optimizing device performance and improving power efficiency. However aggressive scaling can compromise electrostatic integrity and exacerbate short-channel effects. This paper presents a simulation study of n-channel bulk MOSFET scaling using Synopsys Sentaurus TCAD in which the gate length (??) was scaled across four gate-stack dielectric configurations; baseline device with 1.2 nm physical SiO2, scaled variants with 0.9 and 1.5 nm SiO2, and a high-κ stack with an Equivalent Oxide Thickness of 0.9 nm. Oxide scaling gives rise to a monotonic increase in ??? and ???? while ??ℎ exhibits a decrease indicating a progressive loss of gate control. High-κ devices result in better ??ℎ roll-off trendand lower ???? compared to the SiO2 devices.A fundamental structural limitation was identified where scaling ?? below 20 nm results in negative threshold voltages. In conclusion high-κdielectrics extend the viability of the bulk MOSFET while smaller gate lengths demand alternative device architecture to maintain effective switching capability.2. INTRODUCTIONThe continuous scaling of the Metal Oxide Semiconductor Field Effect Transistor (MOSFET)has been the primary driver of performance improvements in digital electronics and integrated circuits for decades [1]. The scaling of MOSFET dimensions, the gate length and the gate oxide thickness (Tox) has been used to enhance electrostatic control and drive current. For many years, Silicon Dioxide (SiO₂) served as the ideal gate dielectric due to its large bandgap and highquality interface with silicon. However, as transistor scaling pushed SiO₂ thickness below 2.0 nm, fundamental physical limits emerged. At a SiO2 thickness of 1.2 nm, direct quantum mechanical tunneling from gate to channel led to exponential increase in gate leakage current, resulting in unacceptable standby power dissipation [2].To overcome this \"tunneling wall\" and maintain the capacitance scaling required for performance, the industry has shifted toward high permittivity (high-κ) dielectrics, such as Hafnium Oxide (HfO2) and Zirconium Oxide (ZrO2). These materials allow a physically thicker gate insulator layer thereby suppressing leakage current while achieving a smaller Equivalent Oxide Thickness (EOT) for higher capacitance [3]. While high-κ integration effectively reduces leakage and improves Short Channel Effects (SCEs), it introduces new

Proceedings of the Technical Sessions, 42 (2026) 20-27Institute of Physics, Sri Lanka 21Gate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD Studytrade-offs, including carrier mobility degradation due to remote phonon scattering and interface trap formation[4], [5]. Consequently, optimizing the balance between physical oxide thickness, material choice, and EOT is critical for minimizing off current (????) and Subthreshold Swing (SS) while maximizing on current (???) for optimal performance.In this work, we present a comparative simulation study of bulk MOSFET scaling using Synopsys Sentaurus TCAD. Even though the state-of-the-art CMOS logic has recently adapted FinFETs and 3D architecture, bulk MOSFET technology is still widely used in analog, RF, IoT and power applications [6], [7], [8]. In addition, it is the most heavily studied canonical structure that is utilized as a reference or benchmark to study short channel effects effectively. Thus, in this study we focused on understanding the electrostatics of the scaled bulk MOSFET that would be a foundation for future advanced architectures.We investigate the electrostatic integrity and transport characteristics of a baseline device with a 1.2 nm SiO2 gate dielectric against scaled variants (0.9 nm and 1.5 nm) and a high-κ stack with an EOT of 0.9 nm. By analyzing key performance metrics ??ℎ, ???, ???? and Subthreshold Swing across varying gate lengths (??), we presented the performance benefits of high-κdielectric integration in scaling bulk silicon transistors. Furthermore, the ??? and ???? plots were analyzed for the 0.9 nm EOT device families to understand the benefits of the high-κstack transition.3. THEORY (PHYSICAL MODELS)The device simulation was carried out using Sentaurus Device Simulations software. The software mainly utilizes the drift-diffusion transport model to solve the Poisson equation coupled with the continuity equations for electrons and holes [9].Poisson Equation∇ ⋅ (ϵ∇ϕ − ?⃗ ) = −q(? − ? + ?? − ??) − ρ???? (1)Figure 1: (a) Simulated bulk MOSFET structure with the inset highlighting the section of gate oxide positioned between polysilicon gate and the Silicon substrate (b) Illustration of the gate length scaling procedure (c) Introduction of high-κ/SiO2 stack replacing the conventional SiO2 gate oxide layer.

Proceedings of the Technical Sessions, 42 (2026) 20-27Institute of Physics, Sri Lanka 22Gate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD StudyIn this equation ϵ represents the electrical permittivity, ϕ is the electrostatic potential, and ?⃗ is the polarization vector. The term ρ???? accounts for the charge density of traps and fixed charges. Carrier densities are denoted by n (for electron concentration) and p (holeconcentration), while ?? and ?? represent the ionized donor and acceptor concentrationsrespectively. Electron and Hole Continuity Equations∇ ⋅ ??⃗ = ? (?net,?− ?net,? +∂?∂?) (2)−∇ ⋅ ??⃗ = ? (?net,?− ?net,? +∂?∂?) (3)In the equations above, ??⃗ and ??⃗ are the electron and hole current densities, q is the elementary charge, ???? is the net recombination rate and ???? is the net generation rates for electrons and holes [9].To simulate submicron level physics, the simulation incorporates Bandgap Narrowing (OldSlotboom method) to adjust intrinsic carrier concentration in highly doped regions. Mobility Degradation is modeled using the Enhanced Lombardi (Enormal) model (accounting for surface roughness and acoustic phonon scattering) and the Canali model (for high-field velocity saturation). Additionally, quantum confinement effects in the thin gate oxide are treated using the Density Gradient model, which introduces a quantum potential correction to the transport equations [9].To study the electrostatic characteristics of high-permittivity (high-κ) dielectrics relative to conventional SiO2, the Equivalent Oxide Thickness (EOT) has been utilized. The bilayer EOT definition is as follows.EOTstack = ?Si?2 + ?high-κ ⋅κSiO2κhigh-κ(4)EOT is defined as the thickness of a theoretical SiO₂ layer that would yield the same gate capacitance per unit area as the actual dielectric stack [10]. The equation explains that by inserting a higher permittivity (κ-value) oxide we could obtain a physically thicker layer (thereby suppressing leakage) while maintaining the same EOT that results in the same gate tochannel capacitance. 4. DEVICE STRUCTURE AND SIMULATION METHODOLOGY4.1 Device Architecture and DesignThe device studied is a symmetrical n-channel bulk MOSFET designed using Synopsys Sentaurus Structure Editor. It was designed by first creating a \"half-FET\" geometry which was then reflected along the vertical axis to obtain the full structure. This ensures perfect symmetry between structure components and keeps symmetric meshing in both sides.The total substrate thickness of the device was set to 300 nm to ensure sufficient bulk depth. The rest of the dimensions for the structure are as follows: Polysilicon gate has a thickness of

Proceedings of the Technical Sessions, 42 (2026) 20-27Institute of Physics, Sri Lanka 23Gate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD Study100 nm and Nitride spacers (Si3N4) have a lateral width of 60 nm. To simulate realistic process conditions, the spacers were modeled with a rounded corner profile with a fillet radius of 50 nm. The source and drain kept at 100 nm lateral width while the gate length (??) served as the primary variable for characterization and it was changed from 50 nm to 15 nm with 5 nm steps.The baseline MOSFET device has a SiO2 gate oxide thickness of 1.2 nm. It was later varied systematically from 0.9 nm to 1.5 nm in order to understand the impact of the oxide scaling on transistor properties. Finally, SiO2/high-κ gate stack with an EOT of 0.9 nm and a total physical oxide thickness of 2.7 nm was included in TCAD simulations. This allowed us to directly compare the performance of high-κ gate stack against the ultra-thin (0.9 nm) SiO2 bulk device for a range of channel lengths. The device has a uniformly doped p-type silicon substrate with a Boron concentration of 1.5 × 1018cm-3. Highly n-doped regions were formed using Arsenic for source and drain regions with a peak concentration of 6 × 1020cm-3. These regions utilize a Gaussian profile with a junction depth (??) of 50 nm and a lateral rolloff factor of 0.4. To suppress hot-carrier effects and manage electric fields, shallow source/drain extensions were included. These follow a Gaussian Arsenic profile with a peak concentration of 2 × 1020cm-3and a significantly shallower junction depth of 12.5 nm (0.25 × ??). The polysilicon gate is highly n-doped (Arsenic) with a concentration of 1 × 1020cm-3.To ensure numerical convergence and accurate capture of quantum confinement effects, a mesh refinement strategy was applied. The grid resolution in the channel region was restricted to 1 nm. Furthermore, interface refinement was applied at the Si/SiO2 boundary with a resolution of 0.1 nm to accurately solve the physics in interfaces.4.2 Simulation setup and Parameter Extraction2D numerical simulations were performed using Synopsys Sentaurus Device. The transfer characteristics (?? − ???) were obtained by ramping the gate voltage in the linear (drain bias of ??? = 0.1 V) and saturation (??? =  1.1) regions. Key performance metrics were extracted using Synopsys Sentaurus Visual. Threshold voltage (??ℎ) was determined using the transconductance method, ??? was extracted at gate supply voltage of ??? = 1.1 ? using saturation curve data and ???? was measured at ??? = 0 ? using linear curve data. The subthreshold swing (SS) was calculated as the inverse slope of the ?? − ??? curve in the subthreshold regime.

Proceedings of the Technical Sessions, 42 (2026) 20-27Institute of Physics, Sri Lanka 24Gate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD Study5. RESULTS AND DISCUSSIONIn this study we systematically studied critical transistor device properties such as ???, ????,subthreshold slope and ??ℎ with respect to gate length and gate oxide. Each marker in Figure 2 and Figure 4 represents the results of a bulk MOSFET structure with a unique gate length and oxide thickness. The effects on ??? and ???? when gate width scales down are presented in Figure 2 (a) and (b). To understand the gains and compromises of all device families, the ???? ?? ??? characteristics plotted in Figure 2 (c). When the gate length decreases both ???? and ??? will be increased, andwhen EOT of SiO2 devices increases ??? will decrease. The better performing devices are the ones with lower ???? and a higher ???, comparing overall characteristics high-κ device family has given better results.A significant reduction in ???? is observed for the high-κ device compared to the 0.9 nm SiO2variant as depicts in the Figure 2 (d). Although both devices have the same EOT, the high-κFigure 2: Bulk MOSFET characteristics of the four device families (a) ??? vs gate length(b) ???? vs gate length (c) ???? vs ??? of all device families (d) Comparison of 0.9 nm EOT high-κ device family and 0.9 nm EOT SiO2 device family.(a)(c) (d)(b)Gate length increases

Proceedings of the Technical Sessions, 42 (2026) 20-27Institute of Physics, Sri Lanka 25Gate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD Studystack utilizes a physically thicker layer, which suppresses the direct tunneling component of the gate leakage while maintaining high ???.To illustrate gate control failure when scaling, cross sectional electron current density contours of the off state were extracted at the of ?? = 15 nm and ?? = 50 ?? as shown in the Figure 3. When comparing (a) with (c) and (b) with (d) broadening of the channel (yellow region) was observed for scaled 15 nm devices with higher electron current density at off state. Additionally, at 15 nm the difference of ???? in oxide (c) and high-κ (d) devices is not significant. This can be observed in Figure 2 (d) as ?? scales the difference between the ???? of two device families decreases (red and blue dashed lines).The impact of gate oxide scaling on electrostatic integrity is presented in Figure 4 (a), which plots the ??ℎ as a function of ??. For all device family variations, decrease in ??ℎ is observed as ?? scales down, indicating the progressive onset of short channel effects (SCEs) that are detrimental to gate control and transistor functionality. The 1.5 nm SiO2 device family exhibits the steepest degradation trend of ??ℎ, while the high-? family exhibits relatively low steepwhich implies applying high-? materials provide better gate control and a positive impact on Figure 3: Off state absolute electron current density (eCurrentDensity) of 0.9 nm EOT devices. Smaller channel length devices have larger electron current density at off state. (a) 50 nm oxide device with lower leakage (b) 50 nm high-κ device with lower leakage (c) Scaled 15 nm oxide device with higher leakage (d) Scaled 15 nm high-κ device with higher leakage(a)(c) (d)(b)

Proceedings of the Technical Sessions, 42 (2026) 20-27Institute of Physics, Sri Lanka 26Gate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD Studytransistor scaling. At and below 20 nm the ??ℎ for this bulk structure drops below 0 V. This implies that the devices have a complete breakdown in electrostatic control.Figure 4 (b) compares the Subthreshold Swing (SS) across the simulated device families. Ideally, SS should remain close to the theoretical limit of 60 mV/dec for faster switching speeds. The simulation results indicate that the high-κ dielectric stack (EOT = 0.9 nm) achieves the smallest SS, comparable to the physical 0.9 nm SiO2 device, maintaining similar characteristics throughout. This confirms that the high-? stack successfully replicates the capacitive control of the ultra-thin SiO2 without the associated physical thinning limits.6. CONCLUSIONIn this study four device families with different gate stacks and channel lengths were compared to one another, and the effects of scaling have been shown using the standard MOSFET characteristics. While scaling down a particular device family, progressive loss of gate control was observed using the ??ℎ roll off curves. Introduction of high-κ dielectric stack to gate oxide layer gives rise to noticeable improvement in ???? parameter by reducing leakage currents, while keeping ??? equivalent to the same EOT oxide devices. Considering all cases we conclude scaling transistor ?? beyond 20 nm for this specific device structure and specificationis not feasible due to high leakage current at off state and loss of gate control. Therefore, bulk MOSFET while suitable for cost-effective applications, needs novel channel materials or advanced architectures for high performance logic that need scaling beyond 20 nm to sustain effective switching performance.7. ACKNOWLEDGMENTThe authors would like to express their sincere gratitude to Farazy Fahmy, Dr. Achintha Kondarage at Synopsys Sri Lanka and Dr. Sankalp Singh in Synopsys India, for providing access to the Sentaurus TCAD Educational License. Their assistance and guidance were essential for carrying out the device simulations in this work.Figure 4: (a) Threshold voltage (using transconductance method) vs Gate length (b) Sub threshold swing vs Gate length, of four Bulk MOSFET device families.(a) (b)

Proceedings of the Technical Sessions, 42 (2026) 20-27Institute of Physics, Sri Lanka 27Gate Oxide Scaling and High‑? Dielectric Integration in Bulk MOSFETs: A Sentaurus TCAD Study8. REFERENCES[1] G. E. Moore, “Cramming More Components Onto Integrated Circuits,” Proceedings of the IEEE, vol. 86, no. 1, pp. 82–85, Jan. 1998, doi: 10.1109/JPROC.1998.658762.[2] J. Robertson and R. M. Wallace, “High-K materials and metal gates for CMOS applications,” Materials Science and Engineering: R: Reports, vol. 88, pp. 1–41, Feb. 2015, doi: 10.1016/j.mser.2014.11.001.[3] M. Salmani-Jelodar, H. Ilatikhameneh, S. Kim, K. Ng, P. Sarangapani, and G. Klimeck, “Optimum High-k Oxide for the Best Performance of Ultra-Scaled Double-Gate MOSFETs,” IEEE Trans. Nanotechnol., vol. 15, no. 6, pp. 904–910, Nov. 2016, doi: 10.1109/TNANO.2016.2583411.[4] S.-H. Kuk et al., “Channel Mobility With Higher-k Doped-HfO₂ for CMOS Logic,” IEEE Trans. Electron Devices, vol. 71, no. 11, pp. 6534–6538, Nov. 2024, doi: 10.1109/TED.2024.3466843.[5] R. Yadav, A. Kaushik, and K. Goyal, “Effect of Deposition of Different Dielectric Materials with Different Oxide Thickness on the Performance of Square Gate All Around MOSFET,” in 2022 International Conference on Futuristic Technologies (INCOFT), IEEE, Nov. 2022, pp. 1–4. doi: 10.1109/INCOFT55651.2022.10094532.[6] V. Subramanian et al., “Planar Bulk MOSFETs Versus FinFETs: An Analog/RF Perspective,” IEEE Trans. Electron Devices, vol. 53, no. 12, pp. 3071–3079, Dec. 2006, doi: 10.1109/TED.2006.885649.[7] Y. S. Chauhan et al., “BSIM6: Analog and RF Compact Model for Bulk MOSFET,” IEEE Trans. Electron Devices, vol. 61, no. 2, pp. 234–244, Feb. 2014, doi: 10.1109/TED.2013.2283084.[8] N. Mendiratta and S. L. Tripathi, “A review on performance comparison of advanced MOSFET structures below 45 nm technology node,” Journal of Semiconductors, vol. 41, no. 6, p. 061401, Jun. 2020, doi: 10.1088/1674-4926/41/6/061401.[9] “SentaurusTM Device User Guide, Version W-2024.09, Mountain View, California: Synopsys, Inc., 2024.”[10] S. M. Sze and K. K. Ng, Physics of Semiconductor Devices, 4th ed. Hoboken, NJ: John Wiley & Sons, 2022.

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 28A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm CastsA Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm CastsK.W.D.I. Mawilmada,M.H.T. Uthpala, and G.D. IlleperumaDepartment of Physics, The Open University of Sri [email protected]. ABSTRACTTraditional orthopaedic casts hold fractures but have drawbacks like discomfort, poor ventilation, weight, and non-recyclability. Personalized solutions are increasingly important in orthopaedics, prosthetics, and biomechanics, but often rely on expensive scanners or advanced CAD skills. This study offers an accessible method to create customized 3D hand models using basic measurements and 3D printing for forearm casts. Using key anthropometric data like palm width, finger lengths, and forearm circumferences, a parametric model was developed in SolidWorks supporting easy adjustments. The hand model was built in parts with measurements inputted automatically without scanning. The method enables efficient, accurate personalized hand models for medical, robotic, prosthetic, and educational uses. The cast, designed using Autodesk Meshmixer over a forearm model with perforations for breathability and rigidity, was printed with PLA on a Prusa i3 MK3S+. Tensile tests on PLA samples showed ultimate strengths of 0.4 kN and 63 N, respectively, on cylindrical and linear samples made using PLA material, confirming material suitability. A lightweight, comfortable, wellventilated prototype weighing around 140 g cost about Rs1000. This research proves personalized orthopaedic casts can be effectively made with mesh-based digital modelling and 3D printing. Future work will explore new materials, adjustable fastenings, and clinical validation.2. INTRODUCTIONMost animals have a skeletal framework of bones that support and protect internal organs. Humans have about 206 bones connected by joints, allowing movement and protecting organs, but fractures can occur from impact, stress, or conditions like osteoporosis. Fractures require medical treatment for realignment and healing, using methods like plaster casts or surgery with metal implants. Treatment choice depends on fracture type, patient age, location, and health. Traditional orthopedic casts made of plaster or fiberglass are bulky, heavy, and uncomfortable, with long drying times, skin issues, and nonrecyclability. Researchers seek better alternatives.This research develops a parametric human arm model and creates a 3D-printed orthopedic forearm cast to improve patient comfort, usability, and fit. Traditional 3D modeling often relies on expensive 3D scanners or advanced CAD skills, making it costly,

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 29A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm Caststime-consuming, and less accessible in low-resource settings. This study develops an automated system for creating customizable 3D models of the human hand and forearm using limited anthropometric measurements. It involves analyzing data to identify key measurements and anatomical features for parametric modeling, evaluating CAD platforms, and creating a modular hand model in SolidWorks. A GUI was also developed for users to input measurements and generate STL models for 3D printing or simulation Using this parametric model, a 3D-printed forearm cast was created with CAD, biocompatible materials like PLA, and additive manufacturing on a Prusa 3D printer. The research covers designing a parametric human arm model, creating the forearm cast, and manufacturing. It shows how modern technology can transform orthopedic rehabilitation.3. METHODOLOGYPhase 1 develops a patient-specific 3D hand model via a user-friendly interface for anthropometric data; Phase 2 uses this model to design a lightweight, breathable, functional 3D-printed cast that immobilizes fractures and enhances wearability and hygiene over traditional plaster casts.Phase 1The methodology included stages like identifying anthropometric parameters, evaluating design software, developing a 3D hand model, assembling anatomical parts, integrating geometric and numerical data, and designing a simple GUI. Software Evaluation & SelectionParametric 3D Model Development, GUI Development & Model Integration The cast is sliced in PrusaSlicer, and supports are addedThe generated G-code is exported to the 3D printerThe cast is printed using the Prusa i3 MK3S+ 3D printerHand and Forearm Anthropometric Data Collection The cast is designed on the parametric model with perforations for ventilation

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 30A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm Casts3.1 Anthropometric Data CollectionSecondary anthropometric data from published literature ensured anatomical accuracy in the 3D hand model. Rostamzadeh et al. [1], analyzing 2,637 participants aged 7–18, provided key measurements like hand length, palm width, finger lengths, and forearm circumferences, ensuring reliability with standard instruments. Additionally, the study “3D Scanning of the Forearm for Orthosis and HMI Applications” [2] guided forearm modeling by introducing elliptical cross-sectional profiles suitable for parametric model generation.Based on these sources, the following anthropometric parameters were selected for implementation:• Palm length and depth• Thumb and metacarpal breadths• Finger lengths for digits one to five• Finger circumferences at proximal and distal regions• Forearm length and circumferences at 0%, 25%, 50%, 75%, and 100% of forearm lengthThese parameters enabled scalable, patient-specific adjustment while maintaining anatomical consistency.3.2 Evaluation and Selection of 3D Design SoftwareSeveral 3D modeling platforms, OpenSCAD, Blender, Fusion 360, and SolidWorks—were evaluated for developing parametric, customizable hand models. Criteria included parametric control, organic geometry support, STL compatibility, ease of use, and learning curve. OpenSCAD was initially considered but was unsuitable for organic shapes. Blender's sculpting was effective for natural forms, but imported STL files couldn't be parametrically modified, limiting patient-specific modeling. Fusion 360 provided strong parametric control and dynamic updates using parameter-driven sketches and features like extrude, loft, and fillet. However, parameter updates failed after combining multiple bodies. SolidWorks was chosen for its reliable parametric functionality, advanced assembly management, and suitability for multi-part anatomy modeling. Preliminary forearm shaping using lofted cross-sections was also explored.3.3 3D Designing of the Hand and ForearmDesigning started with the forearm, using an elliptical cross-section inspired by research [2]. The study used one forearm circumference and 17 planes to approximate the shape. To improve customization and realism, this project used five circumference values at 0%, 25%, 50%, 75%, and 100% of the forearm length, along with 20 planes. These inputs enabled finer shape control. The values were entered into Excel to interpolate at other planes. The referenced paper [2] used cubic equations to define axes for supination and pronation. the equations provided for the supination position were selected and applied.

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 31A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm CastsNormalized axis values were calculated with cubic polynomial expressions across 20 design planes. Afterward, actual axis dimensions were found by multiplying normalized values by the forearm circumference at each plane, using Excel with interpolated circumference data. These values enabled accurate ellipse creation in CAD, where ellipses were positioned on separate planes in SolidWorks. To estimate circumferences at intermediate planes, the FORECAST.LINEAR function in Excel was used for linear interpolation, providing perimeter estimates for elliptical sections in the CAD model. The ellipses were spaced evenly along the forearm’s axis, connected via the Loft tool to form the final smooth outer surface.3.4 Assembly of the Complete Hand DesignThe forearm, palm, and fingers were imported into SolidWorks, aligned, and connected with geometric mates for accurate anatomy. The design was tested to ensure parameter changes updated the whole model. Fully assembled hand and forearm model in SolidWorks, showing aligned finger joints, palm structure, and forearm connectionPhase 2The previous model was used to create a lightweight, breathable cast that immobilizes fractures effectively and improves wearability and hygiene over traditional plaster casts. Using a parametric model, advanced design, and mesh editing, a functional, 3D-printed cast was developed. The process began with a precise parametric model of the fractured forearm, used as an accurate base for creating the personalized cast. This model was then exported in STL format for use with mesh editing software. Meshmixer was chosen for its ability to process anatomical scans, sculpt organic shapes, remesh, and prepare files for 3D printing. The cast was designed as a shell covering only necessary areas, following the physician’s guidance.The selection tool was initially used to outline the arm and hand for the cast, employing a sphere brush with adjustable size for precision in narrow or broad areas. Edges were smoothed with increased boundary smoothness and iterations, preserving group boundaries and adjusting border rings. Only essential parts of the arm and hand, like the forearm and wrist, were included to reduce material and improve comfort. An offset model was designed with a 1.5-2.5 mm clearance to prevent skin contact and accommodate swelling, achieved via an extract offset operation. The cast was separated from the arm model in Meshmixer, smoothing added to improve mesh quality, especially around the thumb, wrist, and cast end. For breathability and hygiene, perforations with varying density were added, thicker near support zones, thinner elsewhere. Mesh reduction tools simplified the cast, with targeted refinement in critical areas to maintain accuracy while optimizing for printing and comfort..Ventilation patterns were integrated into the cast surface, avoiding high-stress zones, rigidity regions, and the boundary dividing the cast into two halves. Patterns were omitted

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 32A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm Castsalong the split plane, outer edges, and joint areas to preserve integrity. This targeted placement optimized airflow and mechanical performance, maintaining the cast's rigidity, fit, and printability while enhancing comfort with ventilation channels.To enable easy donning and doffing, the model was sliced by first breaking the mesh into sections with separate shells, then using a “Plane Cut” along the forearm's longitudinal axis to create two symmetrical halves. Mesh defects were repaired to ensure printability and structural integrity.After fixing the mesh, the halves were imported into PrusaSlicer for printing, positioned separately for optimal orientation, with organic supports added to reduce material use cleanup. The finalized G-code was then exported and transferred to the 3D printer to begin the fabrication process. The finalized G-code ran on a Prusa 3D printer using settings from PrusaSlicer. A roll of PLA filament in the desired colour was loaded into the machine before printing. A 0.4 mm nozzle diameter was used for a solid balance between detail and speed. The layer height was set to 0.2 mm, providing enough surface quality while keeping the print time reasonable. The infill density was adjusted to 15%, as per the slicing phase.4. RESULTS AND DISCUSSIONPhase 1An Excel spreadsheet was created as an input interface for user-specific measurements like palm width, finger lengths, and forearm dimensions. Instead of using the SolidWorks API, the sheet is directly linked to the parametric model, allowing automatic updatesFigure 1: Simple graphical user interface (GUI) designed to collect patient-specific measurements.when values are entered. A simple GUI was designed to help users input measurements correctly and trigger real-time 3D hand model customization.

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 33A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm CastsSimple graphical user interface (GUI) designed to collect patient-specific measurements. The GUI was built to allow non-technical users to input data easily, ensuring real-time updates to the 3D model. (Forearm).To identify patterns in hand dimensions, palm parameters like width, height, and circumference are considered. A separate GUI for the hand component can also be created, allowing independent entry of palm parameters for precise 3D hand model customization, reflecting anatomical variations. Data show that forearm circumferences at normalized positions (0%, 25%, 50%, 75%, 100%) fall within well-defined ranges among 30 participants. Each position's circumferences are confined between min and max values, indicating limited variation at those sections.Figure 2: Fully assembled hand and forearm model in SolidWorks showing aligned finger joints, palm structure, and forearm connection.Measuring circumferences at 0%, 25%, 50%, 75%, and 100% positions reveals a gradual decrease from elbow to wrist, reflecting typical forearm tapering. Minor individual variations show subtle anatomical differences, but the overall trend is consistent. This bounded variation indicates that adult forearm geometry follows predictable limits, even across a mixed-gender group aged 25–38. These findings imply that anthropometric measurements are inherently constrained, which is vital for creating accurate 3D hand and forearm models. The model incorporates upper and lower parameter limits based on these ranges to avoid unrealistic geometries. Natural bounds in data ensure these constraints are practical, accommodating individual differences while maintaining overall shape and function. Consequently, the parameter limits are both biologically justified and essential for realistic, patient-specific models

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 34A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm CastsFigure 3 Variation of Forearm Circumference Along Normalized Forearm LengthPhase 2The cast was printed using a Prusa i3 MK32S+ 3D printer with the following parameters: 200°C nozzle temp, 60°C bed temp, 0.4mm nozzle, PLA material. Each print took 5-6 hours, using 140g PLA costing Rs 1000. Supports removed and edges sanded with postprinting. Figure 4a: Forearm cast with additional smoothingFigure 4b: Number of triangles changed as requiredFigure 4c: Sliced cast along the selected planeFigure 4d: The two halves of the 3D printed orthopedic forearm cast10121416182022242628300% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%Forearm circumference value (cm)Forearm circumference measured at 0%, 25%, 50%, 75%, and 100% of the forearm length Variation of Forearm Circumference Along Normalized Forearm Length

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 35A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm CastsMechanical tests on PLA, including two tensile strength tests on cylindrical samples (fig 5), showed a breaking force of 0.4 kN in mechanical tensometer testing, indicating PLA's adequate strength for orthopedic casts, with the benefits of reduced weight and enhanced comfort.Experimental setup which was used to measure the linear samples (fig 6)recorded a breaking force of 63 N in semi-computerized tensometer testing. This result supports the previous findings and confirms that PLA exhibits adequate mechanical strength for use in orthopedic casts. It can thus be concluded that PLA is a reliable and effective material for producing patient-specific 3D-printed orthopedic immobilization devices. Mechanical testing of the 3D-printed cast was conducted using a semicomputerized tensometer. During horizontal compression testing, a load of 10 N was applied. The cast reached its breaking point at this applied load, indicating the maximum compressive strength it could withstand under the given testing conditions.Dimensions of the tested samplesType of Sample Radius(mm) Length(mm) Breath(mm) Width(mm)CylindricalSample07.00(outer)05.00(inner)08.00 (outer)16.00(middle) - -Linear Sample02.50 15.00(outer)40.00(middle)12.00(outer)04.00(middle)0.60*The dimensions of the forearm cast vary along the arm, with a uniform thickness of 5.00mmThe discrepancies between the tensile results of the cylindrical sample tested using a mechanical tensometer and the linear sample tested using a semi-computerizedFigure 5: Sample designed in fusion 360 and 3D printed sampleFigure 6: sample designed in Fusion 360 and 3D printed sample

Proceedings of the Technical Sessions, 42 (2026) 28-36Institute of Physics, Sri Lanka 36A Parametric Modeling and 3D Printing Workflow for Patient-Specific Orthopedic Forearm Caststensometer are due to the shape and dimensions of the samples used. The results confirm that cylindrical shapes made of PLA could withstand greater tensile force than linear shapes. Since 3D printed parts are made from layers and non-homogenous this may lead to different results. 5. CONCLUSIONThis study demonstrates a parametric approach for generating anatomically accurate and biologically valid 3D hand and forearm models using anthropometric measurements integrated with CAD and additive manufacturing. Based on this framework, a lightweight, ventilated, and patient-specific 3D-printed forearm cast was successfully designed and fabricated using PLA. Mechanical testing confirmed that the material provides sufficient strength for non-surgical fracture management. Clinical testing and validation, along with further functional enhancements, are identified as important directions for future work to support clinical adoption.6. REFERENCES[1] Rostamzadeh, S., Saremi, M., Vosoughi, S. et al. Analysis of hand-forearm anthropometric components in assessing handgrip and pinch strengths of school-aged children and adolescents: a partial least squares (PLS) approach. BMC Pediatr 21, 39 (2021).[2] Perry JC, Barone M, Cerveri P, et al. 3D scanning of the forearm for orthosis and HMI applications. In: Interfacing Humans and Machines for Rehabilitation and Assistive Devices. Vol. 8. 2021.[3] Broughton SC, Lucas GP. Essential orthopaedics and trauma. 5th ed. Edinburgh: Churchill Livingstone; 2010.[4] Magee DJ, Zachazewski JM, Quillen WS. Orthopedic physical assessment. 6th ed. St. Louis (MO): Saunders; 2013.[5] Borrelli KL, Ricci JA. Time efficiency in fracture immobilization: comparing conventional vs digital methods. Orthop Pract. 2020;32(2):122–128.[6] Keller M, Gubeli A, Honigmann P, Ehrler L. Overnight and in-hospital 3D-printed patient-specific casts for non-operative treatment of distal radius fractures: a randomized controlled trial. Swiss Med Wkly. 2022;152:w40079. doi:10.4414/smw.2022.w40079.[7] Fontana L, Minetola P, Stiuso V. Investigation of the influence of 3D printing parameters on the tensile strength of PLA material. Mater Today Proc. 2022;62:1234–1240.[8] Kralevski L, Jovchevska A, Mircheski I. Custom design of an orthopedic hand cast using virtual simulation, 3D printing and experimental verification. In: Proceedings of the 10th International Scientific Conference IRMES; 2022 May; Belgrade, Serbia.

Proceedings of the Technical Sessions, 42 (2026) 37-46Institute of Physics, Sri Lanka 37Arm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy ModelsArm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy ModelsY. L. Ramawickrama1, P. A. A. Perera1, Prabhath Hewageegana1, Shameer Abdeen21Department of Physics and Electronics, University of Kelaniya, Sri Lanka.2Department of Physics and Astronomy, Georgia State University, GA 30303, [email protected]. ABSTRACTSpiral arms are among the most familiar and visually striking features of disk galaxies, yet theirphysical nature and their influence on star formation are still not fully understood. In thisstudy,a set of intentionally simplified toy models is developed to explore how spiral structure affects stellar orbits and the frequency at which stars and gas cross spiral arms. Instead of attempting to recreate the full complexity of real galaxies, we concentrate on identifying a small numberof essential physical components, including logarithmic spiral geometry, flat rotation curvesand epicyclic stellar motion. By controlling parameters such as pitch angle, pattern speed, and the number of spiral arms, we systematically examine the resulting changes in spiral morphology and orbital behaviour. Stellar motions are analyzed using configuration-space maps, radius-time tracks, and phase-space diagrams, and the arm-crossing time is calculated asa function of the galactocentric radius. Although these models are highly idealized, they providea valuable framework for developing physical intuition and for connecting large-scale spiral dynamics to gas compression and star-formation timescales discussed in more sophisticated numerical models.Keywords: Spiral galaxies, density waves, epicycle approximation, star formation2. INTRODUCTIONSpiralstructure is one of the most immediately recognizable features of disk galaxies, including the Milky Way. Spiral arms govern the physical appearance of galactic disks, which range from nearby grand-design spirals to more turbulent systems. Despite this familiarity, there is ongoing discussion in galactic astrophysics regarding the physical origin of spiral arms and, more significantly, their role in controlling star formation. Researchers have proposed numerous theoretical and numerical models over several decades, but none have fully explained the wide range of spiral morphologies observed across different galaxies [1,2].One of the earliest and most influential theoretical descriptions of spiral structure was introduced by Lin and Shu [3], who proposed thatspiral arms could be understood aslong-lived, quasi-stationary density waves propagating through the galactic disk with a fixed pattern speed. In this framework, spiral arms behave more like coherent patterns rather than permanently bound groups of stars. Due to its natural prediction of resonances, such as corotation andLindblad resonances, which are crucial to orbital dynamics, and its ability to offer amathematically well-defined description of spiral geometry, this theory has had a significant impact [2].However, advances in numerical simulations over the past few decades have challenged the

Proceedings of the Technical Sessions, 42 (2026) 37-46Institute of Physics, Sri Lanka 38Arm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy Modelsuniversality of the steady density-wave picture. Many high-resolution simulations indicate that spiral arms can arise as transient and recurrent features driven by gravitational instabilities, swing amplification, and the self-gravity of the stellar disk [4,5,6]. In these models, spiral arms may form, wind up, dissolve and reform on timescales comparable to a galactic rotation period.Such behaviour raises questions about whether long-lived patterns,short-lived features, orsomecombination of the two best describe real spiral galaxies. Observationally, distinguishingbetween these scenarios remains difficult, and it is increasingly recognized that a single mechanism governing spiral structure in all galaxies may not exist [2,7].Regardless of their detailed physical origin, spiral arms are closely associated with regions of enhanced gas density and recent star formation. Molecular clouds, H II regions, and young stellar populations consistently appear along spiral arms, particularly in massive grand-design spirals [8,9]. This close association suggests that spiral arms play an important role in compressing interstellar gas and influencing the conditions under which gravitational collapse can occur. However, not all spiral arms are equally effective at producing stars, and inter-arm regions also show significant star formation. These observations indicate that star-formation activity cannot be identified solely by the presence of spiral structure.As a result, dynamical timescales associated with star formation are becoming increasingly important, rather than focusing solely on density enhancement. One critical timescale is the interval between consecutive passages ofstars or gasthrough spiral arms. Although a spiral arm slightly increases the local density, the gas may not remain compressed long enough to undergo gravitational collapse if it crosses the arm too quickly. Conversely, if gas remains within the arm for an extended period, the probability of collapse and subsequent star formation is expected to increase. The arm-crossing time, therefore, provides a natural connection between large-scale dynamics and the local star-formation process, as it depends on fundamental galactic properties such as the rotation curve, the spiral-arm pattern speed, and the number of spiral arms [10,11].This study aims to explore these ideas using deliberately simplified toy models of spiral galaxies. The primary focus is on isolating a few key physical components and analyzing their effects in a controlled environment, rather than attempting to reproduce the full complexity of real galactic disks. In doing so, this work aims to provide physical insight into how orbital dynamics, spiral geometry, and characteristic timescales interact.2.1 Model Construction and Parameter SpaceThe present study is based on a single analytical toy-model framework rather than a suite of independent numerical galaxy simulations. The objective is not statistical model fitting, butrather a controlled dynamical exploration of how spiral geometry and pattern speed influence orbital evolution and arm-crossing timescales.The model combines four fundamental components: (i) a flat galactic rotation curve, (ii) logarithmic spiral geometry, (iii) epicyclic stellar motion, and (iv) a rigidly rotating spiral pattern with a fixed pattern speed.The following parameters are systematically varied to explore different dynamical regimes:• Pitch angle: p = 5°– 30°• Number of spiral arms: m = 2 and m = 3• Pattern factor: Ωp/Ω0 = 0.6–1.2• Epicyclic amplitude: A/Rg ≤ 0.1

Proceedings of the Technical Sessions, 42 (2026) 37-46Institute of Physics, Sri Lanka 39Arm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy ModelsFor each parameter combination, the arm-crossing time is computed analytically as a function ofgalactocentric radius. No observational calibration or statistical fitting procedures are applied. Instead, the model is designed to isolate the dependence of dynamical timescales on spiral morphology and rotational properties.3. DIFFERENTIAL ROTATION AND WINDING PROBLEMReal disk galaxies exhibit approximately flat rotation curves, meaning that the azimuthal velocity remains nearly constant with radius. Consequently, the angular rotation frequency decreases with increasing radius. This differential rotation has a well-known implication for spiral structure [12].If spiral arms were composed of a fixed set of stars, that is, if they were purely “material arms”, then differential rotation would cause them to wind up rapidly [13]. Inner regions of the disk would complete multiple revolutions, whereas outer areas lag transforming any initially open spiral pattern into a tightly wound structure within a few galactic rotations. This rapid windingis inconsistent with the widespread presence of well-defined spiral arms in galaxies andtherefore poses a fundamental challenge to purely material interpretations of spiral structure [1].In the present model, the angular frequency is parameterized assuming a flat rotation curve,Ω(?) = Ω0?0?, (1)where R isthe galactocentric radiusin kiloparsecs (kpc), and Ro is a reference radius, taken to be an approximate distance to the Sun from the galactic center, Ro ≈ 8 kpc. This simple scalingcaptures the essential effect of differential rotation, allowing the evolution of orbital phases atdifferent radii to be examined transparently.The consequences of differential rotation are illustrated in the radius-time diagram shown in Figure 1, where stars at different guiding radii drift apart in-phase over time. The figure clearly shows that even in the absence of any additional perturbations, differential rotation alone leads to significant radial and azimuthal dephasing across the disk.Figure 1: Radius-time diagram showing stellar orbital evolution in a disk with a flat rotation curve. Each colored curve represents the radial trajectory of an individual star with a distinct guiding radius Rg. Differences in slope reflect the radial dependence of the angular

Proceedings of the Technical Sessions, 42 (2026) 37-46Institute of Physics, Sri Lanka 40Arm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy Modelsfrequency Ω(?) = ?0⁄?, demonstrating differential rotation. The vertical dashed line marks the reference time tref = 0.75 Gyr . The boxed annotation indicates the analytically calculated arm-crossing time at the reference radius ?? = 8 kpc, where ??????(??) =1.88 Gyr.At the selected reference time ???? stars located at smaller guiding radii have progressed furtherthrough orbital phase than those at larger radii. This behaviour arises from the inverse radial dependence of the angular frequency in a flat rotation curve, Ω(R) ∝ R−1. The progressive separation of orbital phases illustrates the classical winding problem: if spiral arms were composed of a fixed set of stars (material arms), differential rotation would cause them to wind up tightly within only a few galactic rotations.The boxed arm-crossing time at Ro provides a dynamical reference timescale for subsequentsections, where the relative motion between stars and the spiral pattern is examined explicitly.3.1 Logarithmic Spirals and Pitch AngleTo introduce spiral structure in a controlled and analytically convenient way, stars are initially placed along logarithmic spiral arms. The azimuthal location of the spiral arms is described by,???(?) = ?0 +ln(??0)tan ?+2???, (2)where ? isthe pitch angle, m isthe number ofspiral arms(for example, ? = 2 for a grand designspiral), and ? denotes individual arms (?= 0, 1,…,m -1).Logarithmic spirals have a constant pitch angle, meaning that the angle between the local tangent to the spiral arm and the tangent to a concentric circle remains constant at all radii. This property closely matches how spiral-arm morphology is quantified in real galaxies [14]. Observationally, logarithmic spirals provide a good description of many spiral galaxies over a substantial radial range, and the pitch angle is widely used as a standard quantitative measure of spiral structure. From a modelling perspective, logarithmic spirals are particularly useful because the pitch angle does not vary with radius. Observed pitch angles in spiral galaxies typically range between 5◦and 30◦, depending on morphological type and stellar mass [12,14]. Early-type spirals generally exhibit smaller pitch angles, corresponding to tightly wound arms, whereas late-type systems display more open structures with larger pitch angles. In this study, we adopt the range 5◦– 30◦ to represent realistic spiral morphologies. Larger pitch angles correspond to more open, loosely wound spirals (Figure 2), while smaller pitch angles produce more tightly wound patterns (Figure 3) [15].

Proceedings of the Technical Sessions, 42 (2026) 37-46Institute of Physics, Sri Lanka 41Arm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy ModelsFigure 2: Two-armed logarithmic spiral pattern (m = 2) generated using the analytical model with pitch angle p = 30°. The spiral follows the functional form. ???(?) = ?0 +ln(?⁄?0)tan ?.A larger pitch angle produces a more open and loosely wound spiral structure, with arms extending outward at a shallower azimuthal gradient. Figure 3: Three-armed logarithmic spiral pattern (m = 3) generated with pitch angle p = 20°. Compared to Figure 2, the smaller pitch angle results in a more tightly wound spiral morphology, with the arms exhibiting a steeper azimuthal gradient with radius.By adjusting only the pitch angle and number of arms, it is possible to explore a wide range ofspiral morphologies without introducing additional complexity. This makesit easier to isolate the influence of spiral geometry alone on stellar orbits and characteristic timescales.

Proceedings of the Technical Sessions, 42 (2026) 37-46Institute of Physics, Sri Lanka 42Arm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy Models3.2 Epicycle Approximation and Radial OscillationsStars in disk galaxies generally follow nearly circular orbits, with minor deviations caused by perturbations in the gravitational potential. These deviations can be described using theepicyclic approximation, in which stars oscillate radially about a guiding-centre radius while continuing to orbit the galactic centre [16]. The radial motion can be modelled as ?(?) = ?? + ? sin (?? + ????), (3)where ?? is the guiding radius, A is the epicyclic amplitude, and ? is the epicyclic frequency.For a flat rotation curve, the epicyclic frequency is well approximated by ? ≈ 1.4 Ω(R) [16].The radius-time (R-t) diagram shows these radial oscillations as small “wiggles” about the mean radius (Figure 1). Although these oscillations are modest in amplitude, they play an essential role in shaping the phase space structure of the disk (Figure 4), where the characteristic relationship between radial position and radial velocity emerges naturally from the epicyclic motion.Figure 4: Radial phase-space diagram (R-VR) for model stars evolving under the epicyclic approximation in a disk with a flat rotation curve. Each point represents the instantaneous radial position R and radial velocity VR of a star with guiding radius ?? and small epicyclic amplitude. The characteristic elliptical structure reflects harmonic radial oscillations aboutthe guiding-centre radius.Figure 4 shows the phase-space structure generated by a group of model stars with different guiding radii ??. Each star undergoes small-amplitude radial oscillations about its guiding centre, producing approximately closed trajectories in R-VR space. In the absence of additional perturbations, the motion is nearly harmonic, leading to the organized phase-space pattern seen in the figure. This demonstrates how near-circular orbits in smooth galactic potentials naturally generate structured radial velocity distributions.

Proceedings of the Technical Sessions, 42 (2026) 37-46Institute of Physics, Sri Lanka 43Arm-Crossing Times and Orbital Dynamics in Simplified Spiral Galaxy Models3.3 Density-wave Picture, Pattern Speed and Arm-crossing TimeIn the Lin–Shu density-wave framework, spiral arms are treated as a quasi-stationary pattern rotating through the disk with a fixed pattern speed Ωp. At the same time, stars orbit with an angular frequency Ω(R) that changes with radius [3]. Because Ω(R) ≠ Ωp in general, stars drift relative to the spiral pattern and repeatedly cross the spiral arms.The frequency with which a star crosses the spiral arms is given by,?????? = ?|Ω(?) − Ωp| (4)where m is the number of spiral arms. The corresponding arm-crossing time is therefore defined as the inverse of this frequency.Figure 5: Radial variation of the arm-crossing time tcross (R) for a representative two-armed spiral model (m=2) with pitch angle p=15°and pattern factor Ωp/Ω0 = 0.8. The black dashed vertical line marks the reference radius Ro =8 kpc, while the red dotted line indicates the corotation radius RCR where Ω(R) = Ωp. The divergence of tcross at RCR reflects the vanishing relative angular speed between stars and the spiral pattern. Figure 5 does not correspond to a specific observed galaxy but instead represents a single parameter realization within the analytical framework. The curve is generated for chosen values of m, p, and Ωp/Ω0, and is intended to illustrate general dynamical behaviour rather than reproduce observational data.The black dashed vertical line marks the reference radius Ro, while the red dotted line indicatesthe co-rotation radius RCR defined by the condition Ω(RCR) = Ωp. At corotation, the relative angular speed between stars and the spiral pattern vanishes, leading to a divergence in the armcrossing time.In the numerical implementation, the angular frequency at the reference radius is defined asΩ(??) ≡ Ωo, (5)