Next Article in Journal
Incorporating Life Cycle Assessment into Tunneling Technologies: Environmental Performance of TBM and ADECO–RS Methods
Previous Article in Journal
Sustainable Maritime Decarbonization: A Review of Hydrogen and Ammonia as Future Clean Marine Energies
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 17
Issue 24
10.3390/su172411367
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Case Study of Dense Hazardous Gas Dispersion in Large Indoor Spaces: Ventilation Layout Analysis with Modeling

by
Jacek Hendiger
*,
Marta Chludzińska
and
Piotr Ziętek
Air Conditioning and Heating Department, Faculty of Environmental Engineering, Warsaw University of Technology, 00-661 Warsaw, Poland
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(24), 11367; https://doi.org/10.3390/su172411367
Submission received: 9 November 2025 / Revised: 9 December 2025 / Accepted: 10 December 2025 / Published: 18 December 2025
(This article belongs to the Special Issue Healthy Indoor Environments: Air Quality in Sustainable Public and Healthcare Buildings)
Browse Figures
Versions Notes

Abstract

The safety of large indoor workspaces hinges on ventilation layout and airflow organization, particularly for dense contaminants that pool near the floor. This qualitative, full-scale case study evaluates chlorine (Cl2) capture using supporting CFD and visualization experiments in a 20 × 13 × 9 m hall. Four exhaust arrangements—low, mid, high, and all levels combined—were tested under two modes: a single grille at 12,000 m3/h and three co-located grilles at 4000 m3/h each (total 12,000 m3/h), with and without an auxiliary supply (2000 m3/h). Removal performance was sensitive to exhaust elevation: low-level extraction consistently confined the plume near the floor, while distributing the same total flow across three levels achieved comparable or improved capture; mid/high extraction was less effective. A practical extraction radius of ≈5 m was identified, and the auxiliary supply improved outcomes only when steering the plume toward the low grille. CFD results showed that, regardless of the lower grille’s duty, the inlet concentration at the low grille was about twice that at the middle grille and more than four times that at the upper grille; in the three-grille configuration, the upper grille received negligible contaminant. These full-scale findings provide geometry-first guidance for dense-gas control in high-ceiling, large-volume spaces.

1. Introduction

Among the wide range of indoor pollutants, a critical group comprises gas-phase contaminants that are heavier than air and have low diffusivity, which in enclosed spaces tend to accumulate and persist near the floor. Typical examples include chlorine (Cl2), hydrogen sulfide (H2S), LPG vapors (propane–butane), and heavier volatile organic compounds (e.g., toluene, styrene, naphthalene), commonly encountered across water and wastewater treatment, garages, paint and varnish production, and composites manufacturing, and whose uncontrolled releases often lead to poisonings and fires [1,2,3]. Because their spatial spread is limited and architectural barriers can hinder air exchange, pronounced intra-room concentration gradients may arise, with contaminants remaining predominantly in the floor-adjacent zone [4,5,6]. From the perspective of public and occupational health, this elevates exposure risks in the breathing zone, while from a sustainability perspective, it supports ventilation strategies that minimize both health risks and unnecessary energy use [7,8].
Prior research consistently shows that removal effectiveness depends on (i) exhaust elevation and (ii) source-to-exhaust distance [4,5,9,10,11,12]. For dense contaminants that do not readily reach the upper part of the room, low-level exhaust—often supported by a directed supply jet—outperforms high-level exhaust [4,5,9,10,11,12]. This behavior arises from the highly limited spatial reach (capture range) of the exhaust for floor-layer plumes [13,14,15]: as the source-to-exhaust distance increases, effectiveness declines and clearance time grows [4,5,9,16]. Supply air is not always beneficial: a properly directed floor-level jet can steer the plume toward a low exhaust and improve capture [11,17], whereas excessive or misdirected supply disrupts the near-floor layer, enlarges the contaminated volume, and hinders removal—effects confirmed in industrial halls, garages, technical tunnels, and fire and gas case studies [9,18,19,20,21,22,23,24].
The role of air changes per hour (ACH) is conditional. Increasing ACH improves removal only up to a threshold and primarily if low-level capture is provided [4,5,25]. For heavy gases, benefits converge around ≈10–12 h−1 (with diminishing returns beyond this range) [25], whereas for CO2, beneficial effects can extend to ≈16 h−1 [26]. Moreover, in mixing systems with warm-air supply, breathing-zone concentrations of heavy contaminants may increase even at ACH = 1–3 h−1, and simply raising ACH does not resolve the problem without a low-level exhaust [20]. Accordingly, well-organized airflow (low exhaust and directed supply) can meet health targets at moderate flow rates, supporting sustainable operation by avoiding energy-intensive, high-throughput mixing [27].
Beyond nominal flow rates, response time and control matched to facility geometry are critical [2,19,28,29]. Nevertheless, there remains a shortage of systematic, full-scale qualitative measurements that jointly analyze exhaust elevation, source-to-exhaust distance, and the conditional effectiveness of a supporting directed supply jet [3]. Existing experimental and computational studies often rely on smaller chambers or escalate ACH (e.g., ≈10–12 h−1), rather than disentangling geometric capture effects at moderate ACH, and the concurrent operation of multi-level extraction has rarely been evaluated under large-volume conditions [4,5,9,10,11,12,16,25,26]. Addressing this gap is essential for occupant well-being and sustainable building operation, where design choices should reduce exposure without incurring unnecessary energy penalties [7,8,27].
Accordingly, this study seeks to deliver full-scale, qualitative evidence in a high-ceiling, large-volume hall at a constant, moderate ventilation rate (ACH ≈ 5.1 h−1), with an emphasis on geometry rather than throughput. The objectives are to (i) quantify how exhaust elevation (low/middle/high) affects the containment of dense plumes; (ii) determine whether concurrent, co-located multi-level extraction (3 × 4000 m3/h at the same total flow) improves containment relative to a single exhaust; (iii) qualitatively identify a practical capture radius in a large space (~5 m) to inform the spacing of pickups; and (iv) establish the conditions under which a directed auxiliary supply (2000 m3/h) assists removal versus promotes mixing. The aim is to provide actionable guidance that reduces occupant exposure while avoiding unnecessary ventilation energy use.

2. Materials and Methods

The study employed two methods: numerical simulations and full-scale physical experiments at a test facility. This enabled comparison of results and a more precise analysis of the observed phenomena.
The literature shows that increasing ACH improves the removal of heavy gases only up to a certain level (up to about 10 h−1), after which mixing and disturbance of stratification occur [4,5]. For this reason, a flow rate corresponding to ACH 5 h−1 was adopted in the tests, consistent with the findings of He et al. [30].

2.1. Numerical Analysis

The purpose of the numerical analysis was to assess the dispersion of a harmful substance, heavier than air, within the occupied zone. The flow in a large-area, large-volume room was analyzed, taking into account different configurations of the ventilation system. Chlorine (Cl2) was used in the CFD simulations as a representative substance for heavy contaminants, serving as a model compound to illustrate the transport and dispersion of dense pollutants.
The modeled space was a room measuring 12.86 m × 19.63 m × 9.00 m (average height) with external windows and eleven internal doors. Infiltration through door perimeter leakage was included. The room had a mechanical exhaust ventilation system with three exhaust grilles, each 710 × 500 mm. The grilles were located at different elevations above the floor: the low grille at 0.37 m (to its lower edge), the middle at 3.35 m, and the high at 8.86 m. The total exhaust flow rate was 12,000 m3/h. In individual variants, the exhaust was assumed to operate through a single grille (12,000 m3/h), sequentially changing its elevation. Two simulation variants included concurrent exhaust through all three grilles; in that case, each grille operated at 4000 m3/h. Further analyses also considered an auxiliary displacement supply diffuser. A flat rectangular floor-standing displacement diffuser (1500 mm × 1500 × mm 250 × mm; H × W × D) with a flow rate of 2000 m3/h was placed 10 m from the plane of the low exhaust grille, perpendicular to the axis passing through the centers of the diffuser, the source, and the grille. The purpose of the installed air inlet was to improve the performance of the exhaust ventilation system.
The contaminant source (500 mm × 500 mm × 500 mm) was located on the floor at a distance of 5 m from the low exhaust grille. The choice of a 5 m distance was preceded by preliminary (pilot) studies, which showed that this was the maximum distance at which effective contaminant removal still occurred. Therefore, the final simulations were performed with the source located 5 m away. Similar preliminary tests were also conducted for the 10 m distance of the auxiliary supply airflow.
In the CFD simulations, the size of the pollutant source was defined to represent a realistic release scenario, such as a leak from a tank or a technological device. The source was intentionally kept small enough so that it did not affect the overall flow structure or the direction of contaminant transport, serving only as an initiating point for the dispersion process.
A simulation with the exhaust system off (Case 0) was also conducted.
The scope of numerical simulations is summarized in Table 1. The modeling and testing protocol was designed to ensure reproducibility, transparency, and rigorous cross-validation between simulation and experiment.

2.2. Numerical Model of the Room

A three-dimensional model of the test room was developed for the computations (Figure 1).
The geometric fidelity of the CFD model was 95%. This value refers exclusively to the accuracy with which the physical room geometry was represented in the computational domain, not to the accuracy of the CFD results themselves. The percentage was estimated based on the proportion of real geometric elements that were explicitly included in the model. All components that could affect airflow and contaminant dispersion—such as obstacles, ventilation grilles, and door gaps—were modeled in detail. The computational domain was discretized with an unstructured mesh of approximately 2,500,000 elements, with local refinement in regions where geometric resolution was essential. The mesh density and quality were selected to ensure that the key geometric features from the CAD design were faithfully reproduced, with simplifications applied only where required by standard numerical practices.
The CFD simulations were carried out primarily to illustrate the overall flow behavior of the tracer, rather than to provide detailed quantitative predictions. Due to the flow characteristics, the realizable k-ε turbulence model was employed to describe the velocity and pressure fields, with mesh refinement applied in regions containing small elements. This setup is commonly used for qualitative studies of flow behavior. The turbulent flow equations were complemented by an Eulerian transport model for the contaminant, and all simulations were performed using ANSYS Fluent 21.R1. The flow patterns obtained from the simulations were consistent with those observed in the smoke visualization experiments.

2.3. Boundary and Initial Conditions

The following boundary and initial conditions were applied:
  • Indoor and outdoor air temperature: 20 °C;
  • Temperature of infiltrating and supplied air: 20 °C;
  • Envelope temperature: 20 °C;
  • Indoor atmospheric pressure: 1013.25 hPa;
  • Pressure drop across the door leakage depended on local flow velocity;
  • Chlorine (density 2.95 kg/m3) released at 0.000125 kg/s for 180 s
    (the release time corresponded to the operating conditions of the smoke generator used in the visualization experiments);
  • Ventilation flow rates were constant over time;
  • Makeup air entered via leakage paths.

2.4. Full-Scale Physical Experiments

Measurement study of the influence of source location and source temperature on the extent and rate of contaminant spread in a large-volume facility.
A lot of substances that could be used are heavier than air and tend to accumulate in the lower parts of a room. As in the simulations, dispersion of contaminants heavier than air within the occupied zone was analyzed. The purpose was to determine dispersion conditions under exhaust ventilation operation and to assess the effectiveness of removal. Individual trials were carried out with different configurations of exhaust pickup locations and with/without auxiliary supply. One aim was also to confirm the simulation results; hence, measurements were performed under conditions comparable to the simulations.
Full-scale tests were conducted in a test room of 20 × 13 × 9 m (Figure 2 and Figure 3). The primary technique was flow visualization using a tracer with simultaneous video recording. Heavy smoke from a fog generator was used as the tracer to emulate the behavior of heavy, floor-spreading substances. The experimental visualization was conducted using heavy smoke, allowing safe observation of heavy contaminant flow patterns. In this setup, the smoke generator was operated so that the release duration matched the emission time of chlorine in the CFD model, ensuring comparability between the two approaches. Additionally, the cooled heavy smoke enabled tracing of the general movement and flow directions of contaminants. Although its physical properties differ from those of chlorine, the observed flow patterns provide a consistent and qualitative representation of heavy contaminant transport.
In the visualization experiments, the smoke generator was physically smaller than the pollutant source represented in the CFD model; however, it was mounted on a 500 × 500 mm plenum box to provide a uniform, low-momentum outflow in all directions. This setup allowed the effective size of the tracer release to be comparable to the source used in the simulations, ensuring consistency between the experimental conditions and the numerical boundary definition.
Recording was performed by a camera above the source and an additional camera in the lower zone (Figure 4). Observations during tests, together with video analysis, enabled assessment of dispersion conditions in the test room.
During the visualization experiments, the exhaust ventilation system operated at a total capacity of 12,000 m3/h, identical to that used in the simulation studies. The air volume flow rate was measured using an orifice plate system combined with a differential pressure transducer. The orifice plate was a standard commercially manufactured component with an outer diameter of 630 mm and a measurement accuracy not worse than 5%. The differential pressure transducer connected to the orifice plate had a measurement range of 0–500 Pa, a resolution of 1 Pa, and a measurement accuracy of 0.5%. The volume flow rate was calculated using the relation V = k (ΔPm)1/2, where k is the calibration constant of the orifice plate and ΔPm [Pa] is the measured pressure difference. The maximum measurement error of the volume flow rate, determined as the combined uncertainty of the system, was 5.3% and did not exceed 10%, which is sufficient for determining ventilation system performance in accordance with the EN 12599 standard [31].

2.5. Components of the Test Facility

The main components followed from the adopted methodology. The setup included, among others, the following:
  • Devices for regulating and measuring the exhaust airflow, control dampers, and differential pressure transducers;
  • A simulated emission source and heavy-smoke generator;
  • A camera for flow visualization;
  • A movable camera support structure (Figure 4);
  • Multipoint lighting of the test area;
  • Instruments to measure the auxiliary supply flow rate.

2.6. Physical Testing Methodology

Test series differed in source placement, exhaust grille elevation, and the presence of auxiliary supply.
Source distance from the exhaust shaft with grilles—5 m (Figure 5).
Auxiliary supply (Figure 6) distance from the exhaust shaft—10 m.
Exhaust grille elevations (floor to lower grille edge) (Figure 5):
  • Bottom: 0.37 m;
  • Middle: 3.35 m;
  • Upper: 8.86 m.
Sequence of each test series:
  • Opening the relevant damper on the exhaust shaft;
  • Verifying the exhaust airflow rate;
  • Positioning the smoke generator at the source point;
  • Setting up the camera and lighting;
  • Starting video recording devices;
  • Starting the heavy-smoke source;
  • Recording and observing the flow;
  • Noting observations and ending the series.

3. Results

The phenomena associated with the release of a heavy contaminant and its dispersion under exhaust ventilation operation are transient. Therefore, results are shown at selected time instants, although simulations covered broader ranges.
For numerical simulations, characteristic isosurfaces of 100 mg/m3 were selected at times when typical transport behavior was evident for a given case.
The results presented below correspond to an exhaust flow of 12,000 m3/h. Where present, the auxiliary supply operated at 2000 m3/h.
Comments for each series follow from both the numerical analysis and the full-scale observations using heavy smoke.
The presented flow diagrams do not correspond to a specific moment in time; rather, they result from the visualization and observation of the entire smoke movement process as a tracer. It should be noted that a threshold concentration of 100 mg/m3 was assumed in the simulation results, whereas the smoke visualization captured the entire flow. Therefore, the presented results should be considered as complementary to each other. Importantly, the flow directions and characteristics obtained from both the visualization and the simulation were in agreement.
Case 0 shows the results with the ventilation system off (Figure 7).
This case serves as a good baseline for the remaining tests.
With the exhaust system off, the dense substance spreads from the source across the room. The heavier-than-air contaminant remains mainly near the floor, occupying a growing area up to a height of approximately 1–2 m.
Case 1 below shows the results with a low grille exhaust of 12,000 m3/h and no auxiliary supply (Figure 8).
Contaminants are released from the source and directed straight to the low exhaust grille. Dispersion is limited to a small area—about one quarter of the room, mainly between the source and the exhaust, up to 1.0 m height.
Case 2 (Figure 9) below shows the results with a middle grille exhaust of 12,000 m3/h and no auxiliary supply.
The contaminant moves toward the middle grille, spreading over a substantial area, up to 50% of the room floor. The affected zone is confined to the side with the exhaust shaft. The substance rises and is also present above the exhaust level (above 5 m).
Case 3 (Figure 10) below shows the results with a high grille exhaust of 12,000 m3/h and no auxiliary supply.
The substance spreads in all directions within the floor layer, covering about one quarter of the room area. The removal effectiveness is relatively low. A large portion of the room becomes contaminated up to the grille elevation.
Case 1.1 (Figure 11) below shows the results with a low grille exhaust of 12,000 m3/h with an auxiliary supply of 2000 m3/h.
After leaving the source, the contaminant moves directly to the low exhaust grille. The contaminant stream is focused and spreads very little in the room. The hazardous substance is removed relatively quickly by the ventilation system.
Case 2.1 (Figure 12) below shows the results with a middle grille exhaust of 12,000 m3/h with an auxiliary supply of 2000 m3/h.
The substance moves toward the middle grille, occupying a large part of the room, but on the exhaust-shaft side. The contaminant rises quickly and is present above the exhaust point (up to 4.5 m). The affected area is smaller than in Case 2; the plume is held closer to the end wall.
Case 3.1 (Figure 13) shows the results of a high grille exhaust of 12,000 m3/h with an auxiliary supply of 2000 m3/h.
The hazardous substance spreads over about one quarter of the room area. A large part of the room becomes contaminated up to the grille elevation. The auxiliary supply slightly narrows the contaminated area and keeps it closer to the exhaust shaft. Differences between Cases 3 and 3.1 are small. The removal effectiveness remains low. These results underline the limited benefit of auxiliary supply when paired with high-level extraction.
Cases 4 and 4.1 (Figure 14) show the results of the concurrent exhaust through three grilles: low, middle, and high (each 4000 m3/h). Case 4.1 includes an auxiliary supply of 2000 m3/h.
These two cases are presented together due to their distinct character (three simultaneous exhausts).
The hazardous substance is released from the source and moves toward the low grille and, to a small extent, toward the middle grille. The spread of the contaminant in the room is confined mainly to the area between the source and the exhaust shaft, occupying the space up to a height of approximately 3.0–3.5 m.
Because only selected time steps of the simulation are presented, the plots (Figure 15) also include the maximum concentrations of the substance that occurred over the entire observation period. The graphs correspond to three characteristic heights—0.5 m, 3 m, and 5 m—which represent the concentrations in the regions influenced by the lower, middle, and upper ventilation grilles, respectively.
Based on the CFD simulations, the maximum chlorine concentrations occurring at the exhaust grilles during operation were determined. Regardless of whether an auxiliary supply was applied, the lower grille received contaminant concentrations of up to 300–350 mg/m3, including when all three exhaust grilles operated simultaneously. When only the middle grille was active, the maximum concentration reaching the grille was about 150 mg/m3 in both Case 2 and Case 2.1, whereas for the upper grille, it did not exceed 70 mg/m3. Moreover, in Cases 4 and 4.1, chlorine concentrations at the middle grille were around 30 mg/m3, and the contaminant practically did not reach the upper grille, indicating effective capture of the floor-adjacent layer by the lower grille even at its reduced individual flow rate.

4. Discussion

Full-scale experiments showed that low-level exhaust provides the most effective removal of heavy contaminants. In that case, the contaminated area is confined mainly to the zone between the source and the exhaust shaft, up to about 3–3.5 m. Adding an auxiliary supply (2000 m3/h) increased removal effectiveness only when the supply jet directed the contaminant toward the low grille, reducing the transport distance. With middle or high exhaust, contaminants tended to spread over a larger area, and the auxiliary supply provided only partial improvement. These results are consistent with studies [4,5,9], conducted in halls and chambers, which confirm transport within the floor-adjacent layer at low flow velocities. However, such solutions are very sensitive to the location of exhaust grilles and the way the supporting supply is organized; according to [11], these factors are more important than the nominal intensity of ventilation. When exhaust is located in the upper parts of the room, jet nozzles are required to direct contaminants toward the elevated grilles [10]. Consistent with NFPA 55 [32], exhaust for gases heavier than air should be located near the floor—within ≤0.30 m above the slab—and the minimum mechanical ventilation rate is about 18 m3/(h·m2); in our tests, the total exhaust of 12,000 m3/h is about ~46 m3/(h·m2), and the low grille installed at 0.37 m above the floor follows explains the observed advantage of low-level exhaust.
It was also observed that locating the source farther than about 5 m significantly reduces removal effectiveness and is highly sensitive to the alignment with the supply–exhaust flow direction. These observations confirm prior findings [4,5].
The configuration with three parallel grilles (4000 m3/h each) reduced the extent of contamination despite the lower individual flow through the low grille than in the single-exhaust variant. This effect may result from the cooperative action of the exhausts: the low grille removes the main floor-adjacent fraction, while the middle and high grilles remove partially dispersed fractions aloft. It was also beneficial to co-locate all grilles in a single area of the room, which established a more coherent, unidirectional flow and reduced recirculation, consistent with the findings of Han et al. [25]. The analyzed ventilation geometry can also be related to the EN 60079-10-1 standard [33] relating to flammable gases, where the applied solution is a type of local ventilation recommended by the standard and leads to a limitation of the dilution volume and explains the reduction in recirculation and smaller contamination volume.
The benefits of multi-point and multi-level exhaust arrangements are consistent with experimental and numerical studies [5,10,11,22,34].
Supply air did not always improve removal efficiency. Improvement occurred only when the supply jet steered contaminants to the low exhaust grille, limiting the dispersion of the floor layer. In configurations with middle/high exhaust, supply tended to promote dispersion. Similar conclusions were drawn in personal exposure studies [20] and in work [11] on ventilation control, where supporting supply combined with low exhaust was preferred. Also, EN 60079-10-1 [33] indicates that mixing or supply-only systems can enlarge the dilution volume and reduce capture effectiveness; therefore, the directed supply should only support a low-level exhaust, not replace local extract ventilation.
This full-scale, geometry-first case study is and was conducted in a single high-volume hall at a constant, moderate ventilation rate (ACH ≈ 5.1 h−1); thus, it characterizes flow patterns. Prior research on heavy-gas dispersion often uses smaller chambers and/or higher ACH (~10–12 h−1) and single-level extraction, which constrains cross-study generalization [4,5,25]. Future work should deploy multi-height gas-sensor arrays to quantify capture efficiency and clearance time versus source distance/height, perform parametric ACH sweeps (including higher regimes) to probe mixing transitions, test co-located multi-level exhausts with adaptive flow control under thermal stratification and transient releases, and couple CFD with evacuation/consequence models to link layouts to risk-relevant metrics [28].

5. Conclusions

Based on the full-scale experiments and CFD in a 20 × 13 × 9 m hall, together with a comparison to the literature, the following conclusions were formulated, relevant to health protection, occupant well-being, and sustainable design and operation of ventilation systems.
  • For dense contaminants, exhaust elevation governs performance; effective removal requires near-floor capture. Mid-/high-only extraction increases the probability of plume spread and mixing in large-volume rooms.
  • A directed, floor-level supply jet is beneficial only when paired with low-level exhaust, where it shortens the transport path and stabilizes capture. In combination with mid/high extraction, assisted supply tends to erode the floor layer and should be disabled or carefully limited.
  • Co-locating multi-elevation grilles on a common shaft establishes a coherent pathway and limits recirculation, offering better robustness than a single, high-, or mid-level point exhaust.
  • The effective capture zone is spatially limited. Design and retrofit should space exhaust pickups so that capture zones overlap near likely sources; a conservative capture-radius assumption can be adopted at the concept stage and verified for each facility.
  • Moderate ACH can achieve protective performance when capture geometry is correct; raising ACH without low-level capture yields diminishing or negative returns due to mixing. Prioritize geometry and flow directionality before increasing throughput, aligning safety with energy performance.
This work contributes full-scale evidence. Three exhaust elevations and a distributed variant were compared at constant ACH, which allowed the effect of geometry to be separated from that of flow rate and linked the results to prior studies in halls/chambers and to control strategies aimed at effective removal [4,5,9,10,11,22]. A limitation is the qualitative character of part of the study (visualizations, CFD maps without calibration to absolute concentrations), which hinders direct comparison with exposure thresholds. Analyses pertain to a single geometry and ACH ≅ 5 h. Some conclusions rely on flow analogy between smoke and heavy contaminants; while the flow character is analogous, time and length scales may differ among specific substances [4,5,9,11,20,22]. Future work should include 3D concentration mapping at varied ACH and validation of multi-point control (grille prioritization and flow modulation) under source-location uncertainty.

Author Contributions

Conceptualization, J.H.; methodology, J.H.; software, P.Z.; investigation, J.H., P.Z. and M.C.; data curation, M.C.; writing—original draft preparation, J.H. and M.C.; writing—review and editing, J.H. and M.C.; visualization, J.H. and P.Z.; funding acquisition, M.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available from the corresponding author upon request due to the large file size and the need for specialized software to open and interpret them.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Hou, J.; Gai, W.M.; Cheng, W.Y.; Deng, Y.F. Hazardous Chemical Leakage Accidents and Emergency Evacuation Response from 2009 to 2018 in China: A Review. Saf. Sci. 2021, 135, 105101. [Google Scholar] [CrossRef]
  2. Si, H.; Ji, H.; Zeng, X. Quantitative Risk Assessment Model of Hazardous Chemicals Leakage and Application. Saf. Sci. 2012, 50, 1452–1461. [Google Scholar] [CrossRef]
  3. Hanna, S.; Mazzola, T.; Chang, J.; Spicer, T.; Gant, S.; Batt, R. Gaps in Toxic Industrial Chemical Model Systems: Improvements and Changes over Past 10 Years. Process Saf. Prog. 2022, 41, 151–166. [Google Scholar] [CrossRef]
  4. Dong, L.; Zuo, H.; Hu, L.; Yang, B.; Li, L.; Wu, L. Simulation of Heavy Gas Dispersion in a Large Indoor Space Using CFD Model. J. Loss Prev. Process Ind. 2017, 46, 1–12. [Google Scholar] [CrossRef]
  5. Zhang, Q.; Zhang, X.; Ye, W.; Liu, L.; Nielsen, P.V. Experimental Study of Dense Gas Contaminant Transport Characteristics in a Large Space Chamber. Build. Environ. 2018, 138, 98–105. [Google Scholar] [CrossRef]
  6. Safakar, M.; Syafiie, S.; Yunus, R. CFD Analysis Chlorine Gas Dispersion in Indoor Storage: Temperatures with Wind Velocities Effect Studies. IEEE Int. Conf. Ind. Eng. Eng. Manag. 2014, 2015, 1219–1222. [Google Scholar] [CrossRef]
  7. Li, S. Review of Engineering Controls for Indoor Air Quality: A Systems Design Perspective. Sustainability 2023, 15, 14232. [Google Scholar] [CrossRef]
  8. Cao, Z.; Zhai, C.; Wang, Y.; Zhao, T.; Wang, H. Flow Characteristics and Pollutant Removal Effectiveness of Multi-Vortex Ventilation in High Pollution Emission Industrial Plant with Large Aspect Ratio. Sustain. Cities Soc. 2020, 54, 101990. [Google Scholar] [CrossRef]
  9. Mundt, E. Non-Buoyant Pollutant Sources and Particles in Displacement Ventilation. Build. Environ. 2001, 36, 829–836. [Google Scholar] [CrossRef]
  10. Li, A.; Ren, T.; Yang, C.; Xiong, J.; Tao, P. Numerical Simulation, PIV Measurements and Analysis of Air Movement Influenced by Nozzle Jets and Heat Sources in Underground Generator Hall. Build. Environ. 2018, 131, 16–31. [Google Scholar] [CrossRef]
  11. Wang, T.; Li, A.; Ma, Y.; Zhang, Y.; Yin, H. Enhancing Heavy Gas Capture in Confined Spaces through Ventilation Control Technology. Build. Simul. 2024, 17, 1161–1182. [Google Scholar] [CrossRef]
  12. Ricciardi, L.; Prévost, C.; Bouilloux, L.; Sestier-Carlin, R. Experimental and Numerical Study of Heavy Gas Dispersion in a Ventilated Room. J. Hazard. Mater. 2008, 152, 493–505. [Google Scholar] [CrossRef]
  13. Khan, J.A.; Feigley, C.E.; Lee, E.; Ahmed, M.R.; Tamanna, S. Effects of Inlet and Exhaust Locations and Emitted Gas Density on Indoor Air Contaminant Concentrations. Build. Environ. 2006, 41, 851–863. [Google Scholar] [CrossRef]
  14. Ejaz, M.F.; Kilpeläinen, S.; Mustakallio, P.; Zhao, W.; Kosonen, R. An Experimental Study on the Efficacy of Local Exhaust Systems for the Mitigation of Exhaled Contaminants in a Meeting Room. Buildings 2024, 14, 1272. [Google Scholar] [CrossRef]
  15. Madsen, U.; Breum, N.O.; Nielsen, P.V. Local Exhaust Ventilation—A Numerical and Experimental Study of Capture Efficiency. Build. Environ. 1994, 29, 319–323. [Google Scholar] [CrossRef]
  16. Lee, H.; Awbi, H.B. Effect of Internal Partitioning on Indoor Air Quality of Rooms with Mixing Ventilation—Basic Study. Build. Environ. 2004, 39, 127–141. [Google Scholar] [CrossRef]
  17. Yan, Y.; Li, X.; Tu, J.; Feng, P.; Zhang, J. Characterisation and Analysis of Indoor Tornado for Contaminant Removal and Emergency Ventilation. Build. Environ. 2019, 164, 106345. [Google Scholar] [CrossRef] [PubMed]
  18. To, C.W.; Chow, W.K.; Cheng, F.M. Simulation of Possible Fire and Explosion Hazards of Clean Fuel Vehicles in Garages. Sustainability 2021, 13, 2537. [Google Scholar] [CrossRef]
  19. Wang, R.; Zhang, Z.; Gong, D. Simulation Study on Gas Leakage Law and Early Warning in a Utility Tunnel. Sustainability 2023, 15, 5375. [Google Scholar] [CrossRef]
  20. Jurelionis, A.; Gagyte, L.; Seduikyte, L.; Prasauskas, T.; Ciuzas, D.; Martuzevicius, D. Combined Air Heating and Ventilation Increases Risk of Personal Exposure to Airborne Pollutants Released at the Floor Level. Energy Build. 2016, 116, 263–273. [Google Scholar] [CrossRef]
  21. Volkov, R.S.; Kuznetsov, G.V.; Sviridenko, A.S.; Strizhak, P.A. Impact of Supply and Exhaust Ventilation System on Gas Concentrations in a Compartment during Thermal Decomposition and Combustion of Materials. Fire Saf. J. 2024, 147, 104201. [Google Scholar] [CrossRef]
  22. Ao, Z.; Wang, Z.; Zhou, W.; Qiao, Y.; Wahab, A.; Yang, Z.; Nie, S.; Liu, Z.; Zhu, L. CFD Simulation Based Ventilation and Dust Reduction Strategy for Large Scale Enclosed Spaces in Open Pit Coal Mines—A Case of Coal Shed. Sustainability 2023, 15, 1651. [Google Scholar] [CrossRef]
  23. Siddiqui, M.; Jayanti, S.; Swaminathan, T. CFD Analysis of Dense Gas Dispersion in Indoor Environment for Risk Assessment and Risk Mitigation. J. Hazard. Mater. 2012, 209–210, 177–185. [Google Scholar] [CrossRef]
  24. Persily, A.; Chapman, R.E.; Emmerich, S.J.; Dols, W.S.; Rushing, A. Building Retrofits for Increased Protection Against Airborne Chemical and Biological Releases Building Retrofits for Increased Protection Against Airborne Chemical and Biological Releases. Natl. Inst. Stand. Technol. 2007. [Google Scholar]
  25. Han, O.; Zhang, Y.; Li, A.; Li, J.; Li, Y.; Liu, H. Experimental and Numerical Study on Heavy Gas Contaminant Dispersion and Ventilation Design for Industrial Buildings. Sustain. Cities Soc. 2020, 55, 102016. [Google Scholar] [CrossRef]
  26. Deng, H.Y.; Feng, Z.; Cao, S.J. Influence of Air Change Rates on Indoor CO2 Stratification in Terms of Richardson Number and Vorticity. Build. Environ. 2018, 129, 74–84. [Google Scholar] [CrossRef]
  27. Zabihi, M.; Li, R.; Brinkerhoff, J.R. Influence of Indoor Airflow on Airborne Disease Transmission in a Classroom. Build. Simul. 2024, 17, 355–370. [Google Scholar] [CrossRef]
  28. Wang, J.; Yu, X.; Zong, R. A Dynamic Approach for Evaluating the Consequences of Toxic Gas Dispersion in the Chemical Plants Using CFD and Evacuation Modelling. J. Loss Prev. Process Ind. 2020, 65, 104156. [Google Scholar] [CrossRef]
  29. Qiu, J.; Mao, H.; Jiang, Y.; Zhang, B.; Cai, H. Enhanced Whale Optimization Algorithms with Source Proximity Indicators: Locating Gaseous Pollutants with Time-Varying Release Rates in Weak Airflow Indoors. Sustain. Cities Soc. 2025, 119, 106112. [Google Scholar] [CrossRef]
  30. He, G.; Yang, X.; Srebric, J. Removal of Contaminants Released from Room Surfaces by Displacement and Mixing Ventilation: Modeling and Validation. Indoor Air 2005, 15, 367–380. [Google Scholar] [CrossRef]
  31. EN 12599:2012; Ventilation for Buildings—Test Procedures and Measurement Methods to Hand over Air Conditioning and Ventilation Systems and Air Conditioning Systems for Non-Residential Buildings. European Committee for Standardization: Brussels, Belgium, 2012.
  32. NFPA NFPA 55; Compressed Gases and Cryogenic Fluids Code (2023). National Fire Protection Association: Quincy, MA, USA, 2023.
  33. IEC 60079-10-1:2020; Explosive Atmospheres—Part 10-1: Classification of Areas—Explosive Gas Atmospheres. International Electrotechnical Commission: Geneva, Switzerland, 2021; Volume 114.
  34. Cai, H.; Long, W.; Li, X.; Kong, L.; Xiong, S. Decision Analysis of Emergency Ventilation and Evacuation Strategies against Suddenly Released Contaminant Indoors by Considering the Uncertainty of Source Locations. J. Hazard. Mater. 2010, 178, 101–114. [Google Scholar] [CrossRef] [PubMed]
Sustainability 17 11367 g001
Figure 1. Model of the test room: (a) location of the contaminant source and displacement diffuser -top view, (b) 3D view of the test room model.
Figure 1. Model of the test room: (a) location of the contaminant source and displacement diffuser -top view, (b) 3D view of the test room model.
Sustainability 17 11367 g001
Sustainability 17 11367 g002
Figure 2. Schematic of the test facility, plan view.
Figure 2. Schematic of the test facility, plan view.
Sustainability 17 11367 g002
Sustainability 17 11367 g003
Figure 3. Schematic of the measurement setup, section.
Figure 3. Schematic of the measurement setup, section.
Sustainability 17 11367 g003
Sustainability 17 11367 g004
Figure 4. View of the test facility with the camera support structure.
Figure 4. View of the test facility with the camera support structure.
Sustainability 17 11367 g004
Sustainability 17 11367 g005
Figure 5. Test facility: (a) view of the source and auxiliary supply; (b) shaft with exhaust grilles at elevations: low 0.37 m, middle 3.35 m, high 8.86 m (floor to grille lower edge).
Figure 5. Test facility: (a) view of the source and auxiliary supply; (b) shaft with exhaust grilles at elevations: low 0.37 m, middle 3.35 m, high 8.86 m (floor to grille lower edge).
Sustainability 17 11367 g005
Sustainability 17 11367 g006
Figure 6. Auxiliary supply element supporting contaminant removal.
Figure 6. Auxiliary supply element supporting contaminant removal.
Sustainability 17 11367 g006
Sustainability 17 11367 g007
Figure 7. Case 0—ventilation off: (a) numerical simulation, t = 360 s; (b) experimental result (a selected frame from the full visualization video); (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 0.5 m.
Figure 7. Case 0—ventilation off: (a) numerical simulation, t = 360 s; (b) experimental result (a selected frame from the full visualization video); (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 0.5 m.
Sustainability 17 11367 g007
Sustainability 17 11367 g008
Figure 8. Case 1: (a) numerical simulation, t = 120 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 0.5 m.
Figure 8. Case 1: (a) numerical simulation, t = 120 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 0.5 m.
Sustainability 17 11367 g008
Sustainability 17 11367 g009aSustainability 17 11367 g009b
Figure 9. Case 2: (a) numerical simulation, t = 360 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 3.0 m.
Figure 9. Case 2: (a) numerical simulation, t = 360 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 3.0 m.
Sustainability 17 11367 g009aSustainability 17 11367 g009b
Sustainability 17 11367 g010
Figure 10. Case 3: (a) numerical simulation, t = 240 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 5.0 m.
Figure 10. Case 3: (a) numerical simulation, t = 240 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 5.0 m.
Sustainability 17 11367 g010
Sustainability 17 11367 g011
Figure 11. Case 1.1: (a) numerical simulation, t = 360 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 0.5 m.
Figure 11. Case 1.1: (a) numerical simulation, t = 360 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 0.5 m.
Sustainability 17 11367 g011
Sustainability 17 11367 g012
Figure 12. Case 2.1: (a) numerical simulation, t = 360 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 3.0 m.
Figure 12. Case 2.1: (a) numerical simulation, t = 360 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 3.0 m.
Sustainability 17 11367 g012
Sustainability 17 11367 g013
Figure 13. Case 3.1: (a) numerical simulation, t = 360 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 5.0 m.
Figure 13. Case 3.1: (a) numerical simulation, t = 360 s; (b) flow schematic; (c) concentration [mg/m3], cross-section at a height of 0.2 m; (d) concentration [mg/m3], cross-section at a height of 5.0 m.
Sustainability 17 11367 g013
Sustainability 17 11367 g014aSustainability 17 11367 g014b
Figure 14. Exhaust through three grilles, numerical simulation, t = 360 s: (a) without auxiliary supply, Case 4; (b) with auxiliary supply, Case 4.1; (c) concentration [mg/m3], cross-section at a height of 0.2 m, Case 4; (d) concentration [mg/m3], cross-section at a height of 0.2 m, Case 4.1.
Figure 14. Exhaust through three grilles, numerical simulation, t = 360 s: (a) without auxiliary supply, Case 4; (b) with auxiliary supply, Case 4.1; (c) concentration [mg/m3], cross-section at a height of 0.2 m, Case 4; (d) concentration [mg/m3], cross-section at a height of 0.2 m, Case 4.1.
Sustainability 17 11367 g014aSustainability 17 11367 g014b
Sustainability 17 11367 g015
Figure 15. Maximum concentrations [mg/m3] recorded in the vicinity of the exhaust grilles for all analyzed variants from Case 0 to Case 4.1.
Figure 15. Maximum concentrations [mg/m3] recorded in the vicinity of the exhaust grilles for all analyzed variants from Case 0 to Case 4.1.
Sustainability 17 11367 g015
Table 1. Scope of numerical simulations of contaminant dispersion and ventilation configuration.
Table 1. Scope of numerical simulations of contaminant dispersion and ventilation configuration.
Simulation CaseExhaust Grille LocationAuxiliary Supply
Case 0No ventilationno
Case 1bottom grille
12,000 m3/h		no
Case 2middle grille
12,000 m3/h		no
Case 3upper grille
12,000 m3/h		no
Case 4Exhaust—3 grilles, bottom, middle, upper
3 × 4000 m3/h		no
Case 1.1bottom grille
12,000 m3/h		yes
Case 2.1middle grille
12,000 m3/h		yes
Case 3.1upper grille
12,000 m3/h		yes
Case 4.1Exhaust—3 grilles, bottom, middle, upper
3 × 4000 m3/h		yes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hendiger, J.; Chludzińska, M.; Ziętek, P. Case Study of Dense Hazardous Gas Dispersion in Large Indoor Spaces: Ventilation Layout Analysis with Modeling. Sustainability 2025, 17, 11367. https://doi.org/10.3390/su172411367

AMA Style

Hendiger J, Chludzińska M, Ziętek P. Case Study of Dense Hazardous Gas Dispersion in Large Indoor Spaces: Ventilation Layout Analysis with Modeling. Sustainability. 2025; 17(24):11367. https://doi.org/10.3390/su172411367

Chicago/Turabian Style

Hendiger, Jacek, Marta Chludzińska, and Piotr Ziętek. 2025. "Case Study of Dense Hazardous Gas Dispersion in Large Indoor Spaces: Ventilation Layout Analysis with Modeling" Sustainability 17, no. 24: 11367. https://doi.org/10.3390/su172411367

APA Style

Hendiger, J., Chludzińska, M., & Ziętek, P. (2025). Case Study of Dense Hazardous Gas Dispersion in Large Indoor Spaces: Ventilation Layout Analysis with Modeling. Sustainability, 17(24), 11367. https://doi.org/10.3390/su172411367

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop