Numerical investigation of coal dust migration in sealed tunneling environments: A CFD-based study

Abstract

Effective dust control is crucial for improving air quality and worker safety in underground mining operations, particularly in sealed excavation faces where traditional ventilation methods are insufficient. Understanding the behavior of coal dust in such environments is essential for developing targeted control strategies. Transitioning from ventilated to sealed tunneling dramatically changes dust dispersion dynamics. High-concentration dust (5,000 mg/m³) primarily diffuses along the roof toward the sidewalls and distal regions of the tunnel, rather than being dominated by airflow velocity and volume. Initial gas composition and dust particle size significantly impact dust concentration, with changes ranging from − 99% to + 1,216% and − 72% to + 169%, respectively. In contrast, methane emissions and temperature variations have limited effects on dust concentration. Based on these insights, we propose specific dust suppression strategies for sealed tunneling environments, offering practical solutions for dust control in similar mining scenarios.

Introduction

Coal dust is one of the by-products of underground coal mining operations, and its generation and dispersion are pervasive throughout the entire coal mining system, including production, transportation, storage, and tunnel excavation stages^1,2,3. Coal dust not only significantly increases the risk of pneumoconiosis among underground workers^4,5 but also poses a severe explosion hazard due to its combustible and volatile nature, potentially leading to casualties and equipment damage when improperly managed^6,7,8. Tunnel excavation faces, which are prone to frequent coal dust accidents⁹, often fail to effectively dilute and remove dust due to the forced ventilation systems employed¹⁰, resulting in persistent safety hazards^11,12. Therefore, elucidating the mechanisms of coal dust generation and migration and implementing targeted dust suppression measures are critical for ensuring safety in confined underground spaces.

Previous studies have addressed these issues by exploring the mechanisms of dust generation and developing dust control technologies. In terms of dust generation mechanisms, Lu et al.¹³ investigated the impact of cutting head angles of a roadheader on coal dust generation rates, finding that the shortest duration from initiation to steady-state dust generation was 45 s. Cao et al.¹ analyzed the influence of coal-rock hardness, moisture content, and distribution on dust generation rates using particle size analysis. Li et al.¹⁴ explored the relationship between rotational speed (35–80 rev/min) and coal wall fragmentation, revealing that the number of tensile fractures, coal dust particle counts, and fragmentation volume all increased with rising rotational speed. Wang et al.¹⁵ developed a discrete element model for gas-bearing coal-rock under confining pressures of 4–10 MPa, considering stress-gas coupling. Zeng et al.¹⁶ summarized the effects of various excavation techniques, including methods, roadheader types, and excavation speeds, on dust generation efficiency.

Regarding dust migration, diffusion, and control, Wang et al.¹⁷ investigated the spatiotemporal patterns of coal dust diffusion and designed a low-surface-tension spray dust suppression device for roadheaders. Hu et al.¹⁸ introduced the concepts of continuous release period (CRP) and sporadic release period (SRP) for coal dust, optimizing the layout of dust extraction devices based on different release periods. Zhang et al.¹⁹ explored the relationship between forced ventilation airflow and coal dust particle transport, revealing that dust concentration was higher in strong flow regions at the excavation face and that dust distribution segregated based on particle size. Guo et al.²⁰ combined numerical simulations with field measurements to characterize the spatiotemporal evolution of coal dust diffusion under single forced ventilation conditions. Zheng et al.²¹ employed the response surface method to study the relationship between outlet air velocity, distance from the outlet to the excavation face, and roadheader height with dust concentration at the operator’s position and in the breathing zone of pedestrians, achieving significant improvements in dust pollution control. Geng et al.²² examined the impact of air curtains on coal dust concentration, finding that air curtains with airflow rates of 273–600 m³/min could limit coal dust diffusion to within 9.6 m. Xu et al.²³ studied the transport and diffusion patterns of coal dust under different excavation face inclinations, proposing technical improvements for airflow velocity adjustments. Liu et al.²⁴ investigated the distribution of coal dust at different cross-sections along the excavation face axis, determining the optimal air-water mist ratio for different droplet sizes in spray dust suppression. Chen et al.²⁵ explored the transport patterns of multiple dust sources at the excavation face during simultaneous excavation and spraying operations, finding that the distance required to dilute dust concentration to 100 mg/m³ increased from 52 m to 55 m when multiple dust sources were present.

Previous studies on dust migration and control at excavation faces have primarily relied on computational fluid dynamics (CFD) simulations combined with field validations. These approaches leverage the suction and repulsion forces generated by forced airflow to partition the excavation space into vortical regions, thereby altering the natural trajectories of coal dust and confining it to targeted zones. Complementary dust suppression techniques–such as air curtains^26,27, sprays^28,29, foam injection^30,31, coal seam pre-injection^32,33, and dry filtration^34,35–have been integrated to achieve localized dust concentration control. While these innovations have yielded positive outcomes, challenges persist, including poor generalizability, high costs, and reliance on rigid airflow parameters. Specifically, current dust control systems require airflow rates to remain within a narrow-predefined range; deviations (sudden increases or decreases) disrupt dust distribution patterns, compromising suppression efficacy. Furthermore, fixed airflow regimes may inadvertently escalate risks of fire and methane explosions, underscoring a critical paradigm shift: dust control is evolving from airflow-dependent to airflow-induced hazards.

The advent of sealed tunneling techniques (no-ventilation systems)^36,37 addresses these limitations by decoupling dust migration from forced airflow. By eliminating oxygen, this method not only reduces ventilation costs but also mitigates traditional dust-related safety risks. Following the removal of the dominant airflow that governs coal dust migration, the role of excavation conditions in the coal dust transport mechanisms within the sealed environment remains unclear: the effects of methane emission rates, initial gas compositions, temperature, and coal dust particle sizes on coal dust migration patterns under the disturbance of roadheader components during operation still need to be analyzed and presented. In this study, we constructed a geometrically scaled model based on real-world excavation data to investigate the spatiotemporal evolution of dust concentrations and flow fields in sealed tunneling environments. Utilizing a CFD-based discrete phase model (DPM)^38,39,40 and dynamic meshing techniques^41,42, we analyzed the impacts of initial gas composition, dust particle size, methane emissions, and operational temperatures on dust distribution. This work serves as a foundational exploration of dust behavior in sealed excavation faces. The findings not only inform targeted dust confinement strategies but also provide theoretical guidance for optimizing engineering practices in similar environments.

Impact of methane emission rates on coal dust migration

Distribution of coal dust and flow field at the sealed excavation face under different methane emission rates

A sealed excavation face model with dimensions of 25 m (length) × 4 m (width) × 3 m (height) was established to simulate real-world operational conditions. Key simulation parameters are summarized in Table 1. The k–ε Realizable turbulence model was selected, and dynamic meshing was applied to the cutting head and scraper regions to account for fluid disturbances caused by their rotational motion during excavation.

Table 1 Simulation parameters.

The dynamic mesh technique relies on coordinate transformation. It also involves grid velocity equations:

$$\:\begin{array}{c}\left\{\begin{array}{c}x=x\left(\xi\:,t\right),\\\:y=y\left(\eta\:,t\right),\\\:z=z\left(\zeta\:,t\right),\end{array}\right.\end{array}$$

(1)

Where $\:\left(x,y,z\right)$ denote physical coordinates, $\:\left(\xi\:,\eta\:,\zeta\:\right)$ represent computational coordinates, and $\:t$ is time.

$$\:\begin{array}{c}{{U}_{g}=\frac{{\partial\:}_{x}}{{\partial\:}_{t}},V}_{g}=\frac{{\partial\:}_{y}}{{\partial\:}_{t}},\:{W}_{g}=\frac{{\partial\:}_{z}}{{\partial\:}_{t}}\end{array}$$

(2)

The dynamic mesh region was refined to a resolution of 0.05 m, while other regions used a 0.1 m grid. The total mesh count reached 800,000 cells, and grid independence was verified to ensure computational accuracy.

Near the outer edge of the roadheader’s cutting head, a 1–1.2 m-diameter annular zone is established. Via surface-jet technology, anthracite particle clusters are injected at an initial velocity of 10 m/s and a concentration of 0.001 kg/m³·s⁻¹, after being stirred by the rotating dynamic-mesh regions of the cutting head and scraper, the particles disperse into the sealed excavation face.

Particle trajectories were tracked using a Lagrangian approach, incorporating coupled interactions between discrete particles and the continuous fluid phase. The particle motion equation, derived from Newton’s second law, accounts for multiple forces:

$$\:\begin{array}{c}{m}_{p}\frac{{du}_{p}}{{d}_{t}}={F}_{d}+{F}_{g}+{F}_{b}+{F}_{l}+{F}_{t}+{F}_{v}\end{array}$$

(3)

Where: $\:{m}_{p}$: particle mass; $\:{u}_{p}$: particle velocity; $\:{F}_{d}$: drag force (dominant term); $\:{F}_{g}$: gravitational force; $\:{F}_{b}$: buoyancy force; $\:{F}_{l}$: lift force; $\:{F}_{t}$: thermophoretic force; $\:{F}_{v}$: virtual mass force;

The drag force ($\:{F}_{d}$) is determined by:

$$\:\begin{array}{c}{F}_{d}=\frac{\frac{1}{2}{C}_{d}{\rho\:}_{f}{\left({u}_{f}-{u}_{p}\right)}^{2}\pi\:{d}_{p}^{2}}{4}\end{array}$$

(4)

where: $\:{C}_{d}$: drag coefficient (dependent on Reynolds number $\:{Re}_{p}$); $\:{\rho\:}_{f}$: fluid density; $\:{d}_{p}$: particle diameter; $\:{u}_{f}$: fluid velocity;

The drag coefficient $\:{C}_{d}$ is expressed as:

$$\:\begin{array}{c}{C}_{d}=\frac{24}{{Re}_{p}}\left(1+0.15{Re}_{p}^{0.687}\right)\end{array}$$

(5)

Where $\:{u}_{f}$ is the fluid dynamic viscosity.

The lift force ($\:{F}_{l}$) arises from fluid velocity gradients:

$$\:\begin{array}{c}{F}_{l}=\frac{1}{2}{C}_{l}{\rho\:}_{f}\left|{u}_{f}-{u}_{p}\right|\left({u}_{f}-{u}_{p}\right)\cdot\nabla\:{u}_{f}\cdot\:\frac{\pi\:{d}_{p}^{2}}{4}\end{array}$$

(6)

Where $\:{C}_{l}$ is the lift coefficient.

Particles were injected into the domain via predefined sources, with initial positions, velocities, and size distributions specified. Trajectories were computed by integrating Eq. (3) using adaptive time-stepping:

$$\:\begin{array}{c}\varDelta\:t=\frac{L}{\left|{u}_{f}-{u}_{p}\right|\cdot\:Step\:Length\:Factor}\end{array}$$

(7)

Where $\:L$ is a characteristic length scale.

Using the Discrete Phase Model (DPM) for gas-solid two-phase flow, the distribution of coal particles within a sealed excavation face was investigated over a period of 100 s, under conditions of 300 K, a methane emission rate of 7.2 m³/min, a gas composition of 50% methane and 50% nitrogen, and a particle size of 10 μm. Given the large number of actual coal dust particles, the DPM model employed the Particle Parcel Method to approximate the particle behavior. The excavation coal face was selected as the inlet for the uniform influx of methane into the sealed space.

Fig. 1

Dust concentration profiles.

The results are shown in Fig. 1. As illustrated, the high-concentration region of anthracite dust is primarily concentrated in the 0–10 m zone at the front of the excavation face. As the coal dust diffuses towards the side walls and the sealed exit, its concentration gradually decreases from 5000 mg/m³.

As illustrated in Fig. 2, panels (a) to (d) show the coal dust concentration distribution corresponding to methane emission rates of 3.6, 5.4, 9, and 10.8 m³/min, respectively. Coal dust is generated at the cutting head of the excavation face and migrates towards the deeper part of the tunnel along the roof. During this process, it diffuses towards the side walls and undergoes gravitational settling⁴³. When it reaches the sealed wind door, the migration of coal dust is obstructed, leading to accumulation at the exit. Changes in ventilation conditions cause the excavation tunnel to be filled with high-concentration coal dust.

Fig. 2

Dust concentration profiles under methane emission rates (3.6, 5.4, 9.0, and 10.8 m³/min).

For methane emission rates below 7.2 m³/min [(a) and (b)], high-concentration coal dust is more concentrated in the front section of the excavation face (0–10 m). In the z = 10–15 m region, the coal dust concentration is relatively lower and more unevenly distributed. In contrast, for emission rates above 7.2 m³/min [(c) and (d)], the high-concentration coal dust band gradually extends from the front section to the middle Sects. (10–15 m) of the excavation face.

The distribution of the flow field at the excavation face under various methane emission rates was investigated, as shown in Fig. 3. It was observed in (a) to (d) that as the methane emission rate increased from 3.6 m³/min to 10.8 m³/min, the fluid velocity in the mixing zone within 0–2 m of the excavation face rose from 1.18 m/s to 1.25 m/s.

Fig. 3

Flow field distribution under methane emission rates (3.6, 5.4, 9.0, and 10.8 m³/min).

The flow state of the mixed fluid within the sealed excavation face was examined via the Reynolds number $\:Re$:

$$\:\begin{array}{c}Re=\frac{{\rho\:}_{0.5}u{D}_{h}}{{\mu\:}_{0.5}}\end{array}$$

(8)

Where, $\:{\rho\:}_{0.5}$ denotes the average density of the mixed fluid within the sealed excavation face, taken as 0.895 kg/m³; $\:{\mu\:}_{0.5}$ represents the average dynamic viscosity of the mixed fluid, adopted as 1.445 × 10⁻⁵ Pa·s; $\:u$ signifies the velocity of the mixed fluid; and $\:{D}_{h}$ stands for the hydraulic diameter (for a rectangular cross-section).

$$\:\begin{array}{c}{D}_{h}=\frac{4A}{P}\end{array}$$

(9)

Where, $\:A$ represents the cross-sectional area of the excavation face, taken as 12 m², and $\:P$ denotes the wetted perimeter of the excavation face, adopted as 14 m. Calculations reveal that within the 0–2 m zone at the front end of the sealed excavation face, the Reynolds number (Re) of the mixed fluid ranges from 2.51 to 2.66 × 10⁶, indicative of a distinctly turbulent flow.

To investigate the driving mode of the mixed fluid flow within the sealed excavation face, the Richardson number ($\:{R}_{i}$) is introduced:

$$\:\begin{array}{c}{R}_{i}=\frac{g\varDelta\:{\rho\:}_{0.5}}{{\rho\:}_{0.5}{u}^{2}}\end{array}$$

(10)

Where, g denotes the acceleration due to gravity, taken as 9.8 m/s², and $\:\varDelta\:{\rho\:}_{0.5}$ represents the density difference between the mixed fluid and the newly influx gas, adopted as 0.245 kg/m³. Calculations indicate that when the methane emission rate is 3.6m³/min, $\:{R}_{i}$ is 1.927, and when it is 10.8m³/min, $\:{R}_{i}$ is 1.717. After the gas from the coal body enters the sealed excavation face, it mixes with the initial gas within the face under the influence of the excavation machine’s operation. Following the cessation of ventilation due to technical characteristics, the mixed fluid does not exhibit a settling flow dominated by inertial forces. At this juncture, the buoyancy-driven diffusion resulting from the gas density difference takes the lead in governing the mixed fluid flow: migrating towards the rear of the enclosed space along the roof of the excavation face, the conveyor belt of the excavation machine, and the gaps on both sides.

To investigate the influence of methane emission rate on the flow of the mixed fluid in the sealed excavation face, the buoyancy flux $\:B$ is introduced:

$$\:\begin{array}{c}B=\frac{g{Q}_{s}\varDelta\:{\rho\:}_{0.5}}{{\rho\:}_{0.5}}\end{array}$$

(11)

where $\:{Q}_{s}$ is the methane emission rate per unit time:

$$\:\begin{array}{c}{Q}_{s}=\frac{Q}{60}\end{array}$$

(12)

where $\:Q$ is the methane emission rate in the sealed excavation face, in m³/min. The calculated results of the buoyancy parameters are shown in Table 2:

Table 2 Buoyancy parameters under methane emission rates (3.6, 5.4, 9.0, and 10.8 m³/min).

As analyzed in conjunction with Fig. 3, the methane emission rate $\:Q$ increases from 3.6 m³/min to 10.8 m³/min, and correspondingly, the buoyancy flux $\:B$ also increases proportionally from 0.161 N/m² to 0.483 N/m². The increase in buoyancy flux results in a greater buoyant force experienced by the mixed gas flow within the sealed excavation face over a unit of time, thereby leading to a higher velocity (1.18 m/s to 1.25 m/s) in the mixing zone from 0 to 2 m. As shown in (a), when the mixed fluid travels from the front section of the tunnel (0–10 m) towards the middle and rear sections, its velocity decreases due to the widening of the flow path⁴⁴. The fluid converges and forms a vortex region in the 8–10 m section of the tunnel. After the vortex region stabilizes, the mixed fluid flows along the edge of the vortex, adhering to the side walls of the excavation tunnel, and bypasses the vortex in a pincer-like motion before converging again downstream. As illustrated in (b)-(d), the increase in methane emission rate causes the vortex region in (a) to shift deeper into the excavation face: the convergence point of the mixed fluid moves from 12.5 m to 14.5 m in the middle section of the excavation face, resulting in a more dispersed distribution of coal dust in the middle Sects. (10–15 m) of the excavation face.

Distribution of coal dust concentration at the sealed excavation face under different methane emission rates

The excavation face space was uniformly divided into 100-unit spaces, each 2.5 m (length) ×0.4 m (width)× 3 m (height), and the number of coal dust clusters in each unit space was counted. As shown in Fig. 4, panels (a) to (d) depict the coal dust distribution under methane emission rates of 3.6, 5.4, 9.0, and 10.8 m³/min, respectively. The different color gradient bars represent the distance between the coal dust sampling range and the left boundary of the excavation face. The x-axis indicates the length of the excavation face, the y-axis indicates the width of the excavation face, and the z-axis represents the number of coal dust clusters.

Fig. 4

Distribution of dust clusters under methane emission rates (3.6, 5.4, 9.0, and 10.8 m³/min).

The increase in methane emission drives coal dust away from the excavation face within the sealed space. The coal dust clusters in the front Sects. (0–7.5 m) gradually decrease, while those in the middle Sects. (10–15 m) correspondingly increase. Influenced by this phenomenon, the sparse coal dust concentration zones located at 12.5 m in the middle section and 15–17.5 m in the rear section also shift backward accordingly.

Taking the coal dust concentration distribution at the excavation face with a methane emission rate of 7.2 m³/min as the reference standard, a contour map was created to investigate the impact of methane emission rates on the distribution of coal dust particle clusters within the excavation face, as shown in Fig. 5. In conjunction with the vortex zone formed in the flow field at 8–10 m as shown in Fig. 3, which traps some coal dust and interrupts the continuity of coal dust transport carried by the methane flow. In panel (a), two coal particle cluster accumulation zones are formed centered at coordinates (x = 1.2 m, z = 10 m) and (x = 2.4 m, z = 17.5 m), with an increase in particle cluster numbers exceeding 7.5%, reaching up to 43%. Meanwhile, two regions of coal dust sparsity are formed centered at (x = 1.2 m, z = 15 m) and (x = 3.2 m, z = 15 m), with a decrease in particle cluster numbers ranging from − 7.5% to −38%.

Fig. 5

Contour map of dust distribution under methane emission rates (3.6, 5.4, 9.0, and 10.8 m³/min).

The increase in methane emission weakens the vortex’s ability to capture coal dust, causing some of the previously trapped coal dust to begin shifting towards the middle section of the excavation face. In panel (b), the extent of the two accumulation zones centered at (x = 1.2 m, z = 10 m) and (x = 2.4 m, z = 17.5 m) is reduced, with an increase in particle cluster numbers ranging from 7.5% to 19%. The coal dust sparsity zones centered at (x = 1.2 m, z = 15 m) and (x = 3.2 m, z = 15 m) show a decrease in particle cluster numbers ranging from − 7.5% to −18%. Lower methane emission rates reduce the flow velocity and coal dust transport capacity of the mixed fluid composed of excavated methane and the initial gas in the tunnel⁴⁵. This results in a discontinuity in coal dust concentration distribution, with the 10–15 m section behind the roadheader acting as a boundary for the transport of coal dust.

As the methane emission continues to increase, its capacity to transport coal dust overcomes the vortex’s restriction. The coal dust surmounts the vortex zone between 8 and 10 m and begins to be transported in large quantities towards the side walls and the middle section of the excavation face at 10–15 m. In panel (c), coal dust accumulation zones are primarily centered around the coordinates of x = 1.2 m, z = 15 m and x = 2.4 m, z = 12.5 m, with an increase in coal dust particle numbers ranging from 7.5% to 20%. New coal dust sparse zones are formed around the coordinates of x = 0.4 m, z = 10 m, x = 1.6 m, z = 10 m, and x = 2.0 m, z = 7.5 m, with a decrease in coal dust particle numbers ranging from − 7.5% to −19%.

After the methane emission rate reaches 10.8 m³/min, the vortex control becomes completely ineffective. In the front and middle sections of the excavation face from 0 to 15 m, coal dust is transported towards the side walls with the fluid, and some coal dust accumulates in the rear section of the excavation face from 15 to 25 m. Coal dust accumulation zones are formed at the front, middle, and rear sections of the excavation face, centered around the coordinates of x = 0.8 m, z = 7.5 m, x = 3.2 m, z = 12.5 m, x = 1.2 m, z = 15 m, and x = 3.2 m, as shown in panel (d), with an increase in coal dust accumulation ranging from 7.5% to 43%. Meanwhile, the coal dust sparse zones in the front-middle Sects. (0–15 m) expand with a decrease in coal dust particle numbers ranging from − 7.5% to −38%. Due to the enhanced transportation capacity of the mixed fluid, a new coal dust sparse zone is formed in the rear section of the excavation face (15–25 m).

As can be seen from Fig. 5, coal dust particles generated by the excavation process are transported towards the middle and rear sections of the excavation face from three directions: above the roadheader, below the cutting head and conveyor belt, and along the side walls. The magnitude of the methane emission rate has a significant impact on the diffusion and transportation of coal dust within the enclosed excavation face. When the methane emission rate decreases from 7.2 m³/min to 3.6 m³/min, the transportation distance of the mixed fluid is insufficient, causing coal dust particles to mainly accumulate in the front Sects. (0–10 m) of the excavation face. Conversely, when the methane emission rate increases from 7.2 m³/min to 10.8 m³/min, the transportation capacity of the mixed fluid is significantly enhanced, allowing coal dust particles to bypass the front section and vortex zone, gradually accumulating and settling in the middle and rear Sects. (10–25 m) of the excavation face.

Impact of initial gas composition on coal dust migration

Distribution of coal dust and flow field at the sealed excavation face under different methane-nitrogen ratios

During the initial phase of excavation, nitrogen injection is employed to inert the enclosed excavation face, thereby displacing the original air-methane mixture. Once the inerting process is complete, the enclosed space at the excavation face is filled with nitrogen, with the underground air being entirely replaced and methane content reduced to negligible levels, thus fulfilling the prerequisite safety conditions for ventilation-free excavation. A unidirectional restricted connection device is installed to maintain stable pressure within the enclosed excavation face. Methane subsequently emanates from the excavation face and surrounding coal walls, gradually increasing its proportion in the methane-nitrogen mixture.

Fig. 6

Dust concentration profiles under methane-nitrogen ratios (1:99, 20:80, 40:60, 60:40, 80:20, and 99:1).

The initial gas composition at the excavation face was set to methane-nitrogen ratios of 1:99, 20:80, 40:60, 60:40, 80:20, and 99:1 to investigate the effects of gas composition on the transport and distribution patterns of coal dust within the sealed excavation space, as shown in Fig. 6. In panel (a), coal dust clusters with concentrations exceeding 5000 mg/m³ are not only distributed in the front-middle section of the excavation face (0–15 m) but are also transported by the mixed fluid to the rear section of the excavation face (20–25 m).

In panel (b), with a methane-nitrogen ratio of 20:80, high-concentration coal dust clusters are concentrated within the 0–15 m section of the excavation face. Compared to the scenario in panel (a), when the methane concentration rises to 20%, the coal particle concentration begins to become sparse in the 15–25 m section of the excavation face, and the number of high-concentration methane clusters is significantly reduced. In panel (c), with a methane concentration of 40%, high-concentration coal particle clusters are concentrated along the transport path within the 0–15 m section, with almost no aggregated high-concentration coal dust clusters in the rear Sects. (15–25 m) of the excavation face.

As shown in panels (d) and (e), when the methane concentration reaches 60% and 80%, respectively, high-concentration coal dust clusters are further confined to the front Sects. (0–10 m) of the excavation face. In panel (f), where methane concentration is dominant at 99%, the transport of coal dust within the enclosed excavation space is highly restricted. Almost all coal dust particles, including high-concentration clusters, are concentrated in the front section of the excavation face and the working area of the roadheader.

As depicted in Fig. 7, the temporal and spatial distribution patterns of methane concentration within the sealed excavation space are examined.

Fig. 7

Methane distribution under methane-nitrogen ratios (1:99, 20:80, 40:60, 60:40, 80:20, and 99:1).

To investigate the ease of gas migration in the initial gas environment of the excavation face during the tunneling process:

The mixed density of gas in the sealed excavation space, $\:{\rho\:}_{mix}$, is given by:

$$\:\begin{array}{c}{\rho\:}_{mix}={c}_{m}{\rho\:}_{m}+{c}_{n}{\rho\:}_{n}\end{array}$$

(13)

where $\:{c}_{m}$ is the proportion of methane gas, $\:{c}_{n}$ is the proportion of nitrogen gas, $\:{\rho\:}_{m}$ is the density of methane, taken as 0.65 kg/m³, and $\:{\rho\:}_{n}$ is the density of nitrogen, taken as 1.14 kg/m³.

The mixed dynamic viscosity of gas in the sealed excavation space, $\:{\mu\:}_{mix}$, is given by:

$$\:\begin{array}{c}{\mu\:}_{mix}={c}_{m}{\mu\:}_{m}+{c}_{n}{\mu\:}_{n}\end{array}$$

(14)

where $\:{\mu\:}_{m}$ is the dynamic viscosity of methane, taken as 1.11 × 10⁻⁵Pa·s, and $\:{\mu\:}_{n}$ is the dynamic viscosity of nitrogen, taken as 1.78 × 10⁻⁵Pa·s.

The kinematic viscosity, $\:\nu\:$, is given by:

$$\:\begin{array}{c}\nu\:=\frac{{\mu\:}_{mix}}{{\rho\:}_{mix}}\end{array}$$

(15)

The Schmidt number, $\:{S}_{c}$, is given by:

$$\:\begin{array}{c}{S}_{c}=\frac{\nu\:}{D}\end{array}$$

(16)

where $\:D$ is the mass diffusivity, taken as 2.24 × 10⁻⁵ m²/s.

The flow parameters in the sealed excavation space were calculated for different methane-nitrogen ratios, as shown in Table 3.

Table 3 Flow parameters under Methane-Nitrogen ratios (1:99, 20:80, 40:60, 60:40, 80:20, and 99:1).

In Figure (a), with a methane-nitrogen ratio of 1:99, the initial gas environment provides a significant density difference, triggering buoyancy-driven convection. After being released from the working face, methane migrates along the roof, forming a stable stratified shear flow. Meanwhile, the Schmidt number $\:{S}_{c}$=0.679 indicates that the mass diffusion rate of methane gas is greater than the momentum diffusion rate. Macroscopically, this results in a diffusion-dominated effect, with noticeable mixing effects only in the 0–5 m zone at the front end of the excavation face.

In Figure (b), with an initial methane ratio of 20%, the density difference begins to decrease, weakening the buoyancy-driven effect. A gentle and stable shear flow is only evident in the 20–25 m zone at the back end of the excavation face. The increased $\:{S}_{c}$ value reduces the methane diffusion effect, expanding the mixing effect influence zone to the 0–10 m range. In Figure (c), there is no longer any distinct stratified flow, and the mixing effect influence zone has extended to the middle front section of the excavation face, covering the 0–15 m area.

In Figures (d) and (e), methane holds a dominant position in the initial gas composition of the sealed excavation face. When fresh gas rushes in, the buoyancy-driven effect no longer dominates gas migration. Combined with the continuously increasing $\:{S}_{c}$ value, the gap between mass diffusion rate and momentum diffusion rate further narrows. Throughout the entire excavation face, the incoming methane flow is mixing with the initial gas. In Figure (f), the mixing effect dominates the gas flow process. However, at this point, the high concentration of methane forms a concentration boundary layer that hinders the penetration of fresh methane, thereby restricting the mixing zone to the 0–10 m range at the front end of the excavation face.

The distribution of the flow field at the excavation face under various gas ratios was investigated, as shown in Fig. 8. After emanating from the excavation face, methane migrates towards the middle and rear sections of the excavation space, both horizontally and vertically along the roof, due to its intrinsic gas properties. Simultaneously, the initial gas within the excavation face is disturbed by the excavation activities, particularly in the front section and around the excavation machine. This disturbance causes the affected gas to merge into the migrating methane flow.

Fig. 8

Flow field distribution under methane-nitrogen ratios (1:99, 20:80, 40:60, 60:40, 80:20, and 99:1).

In Figure (a), the mixed fluid velocity in the disturbed working area of the excavation face is 1.48 m/s, while in the middle section’s disturbed working area, the mixed fluid velocity is 0.7 m/s. During this stage, the migration of methane gas is primarily dominated by diffusion. A distinct parallel shear flow exists in the middle and rear sections of the excavation face (10–25 m), which can continuously convey a substantial amount of coal dust towards the back of the excavation face.

As shown in Figures (b)-(e), when the methane concentration increases from 1% to 80%, the mixed fluid velocities in the front and middle sections of the excavation face decrease to 0.93 m/s and 0.44 m/s, respectively. During this phase, as the proportion of methane in the mixed gas gradually increases, the mode of methane gas migration shifts from diffusion to mixing. The parallel shear flow is progressively replaced by a shear-free flow, resulting in a continuous reduction in the ability to convey coal dust. When the initial methane concentration approaches 100%, the flow of the incoming methane exhibits distinct mixing characteristics. The velocities of the mixed fluid in the front and middle sections drop to 0.35 m/s and 0.17 m/s, respectively. Under these conditions, the velocity is insufficient to support the long-distance conveyance of coal dust clusters. Consequently, the coal dust, influenced by gravity, settles and accumulates in the range of 0–10 m at the front end of the excavation face.

Distribution of coal dust concentration at the sealed excavation face under different methane-nitrogen ratios

The number of coal dust clusters within the sealed excavation space was counted under various methane-nitrogen ratios, as shown in Fig. 9.

Fig. 9

Distribution of dust clusters under methane-nitrogen ratios (1:99, 20:80, 40:60, 60:40, 80:20, and 99:1).

When the methane concentration is low (1%), coal dust clusters gradually accumulate and distribute along the centerline of the excavation face, extending towards the side walls and the rear of the excavation face. As the methane concentration increases, the distribution of coal dust clusters contracts towards the centerline and the front section of the excavation face. At high methane concentrations (99%), coal dust clusters are predominantly confined to the front section of the excavation face (0–10 m).

Taking the coal dust cluster distribution shown in Fig. 1 as the reference standard, the impact of methane-nitrogen compositions (1:99, 20:80, 40:60, 60:40, 80:20, and 99:1) on the distribution of coal dust clusters within the sealed excavation space was investigated, as shown in Fig. 10. In panel (a), with a methane concentration of 1%, coal dust clusters form accumulation zones centered at coordinates (x = 2 m, z = 12.5 m), (x = 2.4 m, z = 17.5 m), and (x = 0.8 m, z = 22.5 m) in the middle and rear sections of the excavation face (10–25 m), with an increase in cluster density exceeding 25%. Meanwhile, in the front Sects. (0–10 m), coal dust clusters form sparse zones centered at (x = 0.6 m, z = 2.5 m), (x = 3.2 m, z = 10 m), and (x = 3.6 m, z = 2.5 m), with a decrease in cluster density exceeding − 25%. The fresh methane flow induced by excavation activities diffuses rapidly in the high-nitrogen environment, transporting a large number of coal dust clusters into the middle and rear sections of the excavation face within the same time frame.

In panel (b), with the methane concentration increased to 20%, both the accumulation zones with a 25% increase in coal dust clusters and the sparse zones with a −25% decrease shift towards the front of the excavation face. At this stage, methane still does not dominate the gas composition, and the methane gas retains ample space for transport and exchange within the high-nitrogen environment. The diffusion rate of the incoming methane slows down, and its capacity to transport coal dust clusters within the same unit of time gradually weakens. As a result, more coal dust clusters accumulate in the 10–20 m section of the excavation face, with a gradual decrease in distribution density in the rear Sects. (20–25 m).

Fig. 10

Contour map of dust distribution under methane-nitrogen ratios (1:99, 20:80, 40:60, 60:40, 80:20, and 99:1).

In panel (c), with a methane proportion of 40%, the coal dust accumulation zone is primarily distributed in the middle section of the excavation face (10–15 m), centered at coordinates (x = 2 m, z = 12 m). The coal dust sparse zones are mainly located near the side walls in the front and middle sections of the excavation face. In panel (d), where the methane concentration reaches 60%, the sealed excavation space transitions from a high-nitrogen to a high-methane environment. Coal dust clusters accumulate in zones centered at (x = 1.6 m, z = 5 m) and (x = 2.4 m, z = 17.5 m). In the methane-dominated environment, the majority of coal dust clusters are concentrated in the front Sects. (0–10 m) of the excavation face. The influx of fresh methane mixes slowly with the initial gas in the excavation face, significantly weakening its ability to transport coal dust clusters towards the middle-rear sections and side walls. Sparse zones begin to form at coordinates (x = 3.6 m, z = 5 m), (x = 0.4 m, z = 10 m), (x = 2 m, z = 10 m), and (x = 4 m, z = 15 m).

In panel (e), as the methane concentration increases to 80%, coal dust accumulation zones are formed at coordinates (x = 1.6 m, z = 5 m), (x = 1.6 m, z = 12.5 m), and (x = 2.4 m, z = 17.5 m), with the sparse zones expanding along the side walls adjacent to the accumulation zones. The weakened transport capacity of the mixed fluid in the high-methane environment causes coal dust clusters, which would normally be transported towards the middle-rear sections and side walls, to aggregate along the centerline of the excavation face at x = 2 m. When the sealed excavation space is almost entirely filled with methane (99%), as shown in panel (f), the influx of fresh methane barely mixes with the initial gas, effectively halting the transport pathways for coal dust clusters. Consequently, the vast majority of coal dust clusters accumulate in the front Sects. (0–10 m) of the excavation face, with a maximum accumulation increase of up to 1,216% and a minimum sparse decrease of −99%.

When the methane concentration at the excavation face is low, the initial gas flow in the high-nitrogen environment has uniformly distributed molecules. The methane outflow, composed of CH₄ molecules distinct from N₂, mixes smoothly with the initial gas flow. Within the same time frame, the outflowing methane diffuses rapidly and steadily to the middle and rear sections of the sealed excavation space. The fluid velocity during mixing is relatively high and predominantly horizontal, with a strong capacity to transport coal dust clusters. As a result, coal dust is evenly distributed throughout the sealed excavation space, including high-concentration (5000 mg/m³) coal dust zones in the 10–25 m section.

When the excavation space contains a certain amount of methane (20%), the loose arrangement of N₂ molecules in the initial gas flow is disrupted by CH₄ molecules occupying some of the intermolecular spaces. This slows the mixing process between the methane and initial gas flows, reducing the transport velocity of the mixed flow. Consequently, the distribution range of high-concentration coal dust in the 15–25 m section gradually narrows.

At a methane concentration of 40% in the initial gas flow, CH₄ molecules further encroach on the transport pathways and mixing space of the methane flow. The velocity of the mixed flow drops significantly, and coal dust clusters, limited by the reduced transport capacity, struggle to reach the 15–25 m section, accumulating instead in the 0–15 m middle and front sections. As methane concentration exceeds that of nitrogen, the influence of mixed flow transport on coal dust distribution weakens, with gravitational settling of coal dust particles becoming more pronounced. At 60% methane concentration, high-concentration coal dust clusters begin to form discontinuous distributions in the 0–15 m section. At 80% methane concentration, gravitational settling dominates, with coal dust concentrations decreasing along the side walls due to localized settling, and high-concentration coal dust clusters concentrating along the centerline of the 0–20 m section.

At 99% methane concentration, the initial gas flow is dominated by densely packed CH₄ molecules, making it difficult for new methane influx to find intermolecular spaces for transport and mixing⁴⁶. The mixed flow velocity drops to its lowest level (0.17 m/s), and stable laminar flow is unattainable. Meanwhile, gravitational settling of coal dust particles dominates their motion. Together, these factors cause coal dust clusters to settle and accumulate near the front Sects. (0–10 m) of the excavation face.

Impact of particle size on coal dust migration

Distribution of coal dust and flow field at the sealed excavation face under different particle size

Coal particles with diameters of 1 μm, 2.5 μm, 5 μm, and 7.5 μm were selected to investigate the impact of coal dust particle size on transport and distribution within the sealed excavation space, as shown in Fig. 11.

Fig. 11

Dust concentration profiles under particle size (1 μm, 2.5 μm, 5 μm and 7.5 μm).

To analyze and evaluate the ease of transport of coal dust particles of different sizes within the sealed excavation face, the Stokes number ($\:Stk$) is introduced:

$$\:\begin{array}{c}Stk=\frac{{\tau\:}_{p}}{{\tau\:}_{f}}\end{array}$$

(17)

where $\:{\tau\:}_{p}$ is the relaxation time of the coal dust particle (s), and $\:{\tau\:}_{f}$ is the characteristic time of the fluid (s):

$$\:\begin{array}{c}{\tau\:}_{p}=\frac{{\rho\:}_{p}{d}_{p}^{2}}{18{\mu\:}_{0.5}}\end{array}$$

(18)

where $\:{\rho\:}_{p}$ is the density of the coal dust particle: 1400 kg/m³, and $\:{d}_{p}$ is the diameter of the coal dust particle.

$$\:\begin{array}{c}{\tau\:}_{f}=\frac{{D}_{h}}{u}\end{array}$$

(19)

Due to the technical characteristics of the sealed excavation face, the influence of the inertial effect is quantitatively amplified to elucidate the differences in the transport of coal dust particles of different sizes:

$$\:\begin{array}{c}{Stk}_{corr}=\eta\:\cdot\:Stk\end{array}$$

(20)

where $\:\eta\:$ is the environmental enhancement factor, taken as 10³. The calculated results of the transport parameters are shown in Table 4:

Table 4 Transport parameters under particle size (1 μm, 2.5 μm, 5 μm, and 7.5 μm).

In Fig. 11(a), after 100 s of transport within the sealed excavation space, high-concentration coal dust zones are scarcely detected in the middle Sects. (10–15 m), except for the newly excavated coal dust from the cutting head of the excavation machine. The corrected Stokes number $\:{Stk}_{corr}$=0.0017 is much less than 1, indicating that 1 μm coal particles fully follow turbulent diffusion with negligible gravitational inertial settling, predominantly accumulating in the rear Sects. (15–25 m).

As shown in Fig. 11(b), when the coal particle size increases to 2.5 μm, the mixed fluid still dominates the transport of coal dust clusters. At this point, the increase in particle mass and aerodynamic resistance leads to the manifestation of gravitational settling: a small portion of coal dust clusters follows the mixed fluid to the rear section of the excavation face (15–25 m), while the majority of coal dust clusters are enriched in the front Sects. (0–10 m).

In Fig. 11(c), with coal particle size at 5 μm, the influence of gravitational settling begins to accelerate: high-concentration coal dust bands concentrate in the front section of the excavation face (0–10 m) with minimal distribution in other areas. In Fig. 11(d), $\:{Stk}_{corr}$=0.1006, and inertial forces have become one of the main factors affecting the diffusion and transport of 7.5 μm coal dust, with an increased settling velocity of coal dust particles and a sharp increase in coal dust concentration at the front of the excavation face.

Fig. 12

Flow field distribution under particle size (1 μm, 2.5 μm, 5 μm and 7.5 μm).

The impact of different coal dust particle sizes on the flow field distribution at the excavation face was investigated, as shown in Fig. 12.

To investigate the impact of coal dust cluster settling on the flow field within the sealed excavation face, the change in system average density $\:\varDelta\:\rho\:$ is introduced:

$$\:\begin{array}{c}\varDelta\:\rho\:=\eta\:\cdot{\rho\:}_{g}\cdot{u}_{s}\cdot\varnothing\:\cdot\end{array}$$

(21)

where $\:t$ is the time, 100 s; $\:{u}_{s}$ is the coal dust settling velocity (m/s):

$$\:\begin{array}{c}{u}_{s}=g\cdot{\tau\:}_{p}\end{array}$$

(22)

$\:\varnothing\:$ is the volume fraction of coal dust particles:

$$\:\begin{array}{c}\varnothing\:=\frac{{n}_{p}^{{\prime\:}}\cdot{V}_{p}}{{V}_{0}}\end{array}$$

(23)

where $\:{n}_{p}^{{\prime\:}}$ is the average number of coal dust particles in the 0–10 m zone at the front of the excavation face(particles/m³); $\:{V}_{p}$ is the volume of a coal dust particle(m³); $\:{V}_{0}$ is the unit volume.

$$\:\begin{array}{c}{n}_{p}^{{\prime\:}}=\frac{c\cdot{n}_{p}}{{V}_{10}}\end{array}$$

(24)

where $\:{n}_{p}$ is the number of coal dust clusters in the 0–10 m zone at the front of the excavation face, $\:c$ is the number of particles contained in a coal dust cluster, 10⁵. $\:{V}_{10}$ is the volume of the 0–10 m zone at the front of the excavation face, 120m³. The calculated results of the density parameters are shown in Table 5.

Table 5 Density parameters under particle size (1 μm, 2.5 μm, 5 μm and 7.5 μm).

As the coal particle size increases from 1 μm to 7.5 μm, the average density change($\:\varDelta\:\rho\:$) at the front of the excavation face is observed to increase from 3 × 10⁻⁷ to 8.9 × 10⁻³ kg/m³. The settling of coal dust clusters leads to a decrease in the average system density distribution in the 0–10 m zone at the front of the sealed excavation face, which is compensated by the velocity of the continuous phase source through the continuity equation: the velocity of the mixed fluid in the front section of the excavation face increases from 1.08 m/s to 1.14 m/s, and in the middle section, it increases from 0.51 m/s to 0.54 m/s, thereby creating a phenomenon where larger coal dust particles result in higher velocities.

Distribution of coal dust concentration at the sealed excavation face under different particle size

The number of coal dust clusters within the sealed excavation space was counted under different particle sizes, as shown in Fig. 13.

Fig. 13

Distribution of dust clusters under particle size (1 μm, 2.5 μm, 5 μm and 7.5 μm).

When the coal dust particle size was 1 μm, as shown in panel (a), coal dust clusters were predominantly distributed in the middle Sects. (10–15 m) and the rear Sects. (15–25 m) of the excavation face, excluding the dust-generating area near the cutting head of the roadheader. The number of coal dust clusters positively correlated with the transport distance.

In panel (b), with a coal particle size of 2.5 μm, the number of coal dust clusters increased in the front Sects. (0–10 m) of the excavation face, while a significant reduction was observed in the middle Sects. (10–15 m). In panels (c) and (d), where the coal dust particle sizes were 5 μm and 7.5 μm, respectively, the discontinuity in coal dust cluster distribution shifted to the rear Sects. (15–17.5 m) of the excavation face. The coal dust clusters were concentrated along the centerline of the excavation space and gradually distributed towards the side walls.

The changes in coal dust concentration distribution under different particle sizes were examined, with the simulation results in Fig. 1 as the reference standard. The differences in coal dust concentration distribution under various particle sizes are shown in Fig. 14.

Fig. 14

Contour map of dust distribution under particle size (1 μm, 2.5 μm, 5 μm and 7.5 μm).

In panel (a), with a particle size of 1 μm, coal dust accumulation zones with an increase of over 25% are formed, centered at coordinates (x = 0.8 m, z = 22.5 m), (x = 2.4 m, z = 17.5 m), and (x = 3.2 m, z = 17.5 m). The maximum accumulation increase can reach up to 169%. Meanwhile, coal dust sparse zones are primarily concentrated in the front section of the excavation face, centered at (x = 2.4 m, z = 5 m), with a decrease of −25%, and the minimum decrease can reach − 72%.

When the particle size increases to 2.5 μm, as shown in panel (b), coal dust tends to accumulate towards the middle front section of the excavation face, forming two accumulation zones centered at (x = 1.2 m, z = 10 m) in the middle front section and at (x = 1.2 m, z = 17.5 m) and (x = 1.2 m, z = 22.5 m) in the rear section. The sparse zones in the front section gradually decrease in size, while a new sparse zone forms in the middle Sects. (10–15 m) due to the discontinuity in coal dust cluster distribution.

In panel (c), with a coal dust particle size of 5 μm, the area and accumulation increase of the coal dust accumulation zones, centered at (x = 0.8 m, z = 15 m), (x = 2.8 m, z = 17.5 m), and (x = 0.8 m, z = 22.5 m), are significantly reduced. The accumulation increase is limited to 50%. The sparse zones in the front section of the excavation face almost disappear, and the area of the sparse zones in the middle section decreases. In panel (d), with a coal particle size of 7.5 μm, the extent and degree of coal dust accumulation and sparsity are essentially consistent with the reference case.

After being generated during excavation, coal dust particles exhibit different motion states within the excavation face depending on their particle size. The focus of this study is on coal dust clusters in the micron-scale range. When the particle size is small (1 μm), these clusters exhibit some characteristics of diffusion and are less affected by gravitational settling. They remain suspended in the air for a longer period, making them more likely to be transported to the middle and rear sections of the excavation face by the mixed fluid flow¹⁸.

As the particle size increases (2.5 μm, 5 μm), coal dust clusters undergo uniform settling within the excavation face while still being influenced by the entrainment of the mixed fluid. A small fraction of these clusters is transported to the middle and rear sections, while the majority remains distributed in the front Sects. (0–10 m) of the excavation face.

When the particle size reaches 7.5 μm and 10 μm, near the boundary of fine dust, the coal dust clusters begin to exhibit partial accelerated settling characteristics. Dominated by gravity, the vast majority of these clusters are concentrated in the front section of the excavation face and rarely reach the deeper middle and rear sections.

Impact of temperature on coal dust migration

Distribution of coal dust and flow field at the sealed excavation face under different temperature

The impact of temperature variations on coal dust transport and distribution within the sealed excavation space was investigated under excavation face temperatures set at 290 K, 295 K, 305 K, and 310 K, as shown in Fig. 15.

Fig. 15

Dust concentration profiles under temperature (290 K, 295 K, 305 K, and 310 K).

In panels (a) to (d), high-concentration coal dust clusters with a concentration of 5000 mg/m³ are primarily concentrated within the front Sects. (0–10 m) of the excavation face. The transport distance of coal particles within the sealed excavation space is minimally affected by temperature. Examining the horizontal diffusion range of coal dust clusters, under an excavation face temperature of 290 K, the widths of high-concentration coal dust clusters transported to the front positions of 2.5 m, 5 m, and 7.5 m by the mixed fluid are 0.57 m, 0.57 m, and 1.33 m, respectively. When the temperature rises to 310 K, the widths of these clusters at the corresponding positions increase to 0.67 m, 0.95 m, and 1.38 m.

Fig. 16

Flow field distribution under temperature (290 K, 295 K, 305 K, and 310 K).

As shown in Fig. 16, the flow field distribution of methane-nitrogen mixed flow within the sealed excavation space under different temperature ranges is presented.

To investigate the impact of temperature variations on the transport of mixed fluids within the sealed excavation face, the molecular diffusion coefficient $\:{D}_{m}$ and the turbulent diffusion coefficient $\:{D}_{t}$ are introduced:

$$\:\begin{array}{c}{D}_{m}=\frac{{10}^{-7}\times\:{T}^{\frac{3}{2}}}{\sqrt{{M}_{0.5}}}\end{array}$$

(25)

where $\:T$ is the temperature (in Kelvin), and $\:{M}_{0.5}$ is the average molar mass of the mixed fluid, taken as 22 g/mol.

$$\:\begin{array}{c}{D}_{t}=\frac{{\nu\:}_{t}}{{S}_{c}}\end{array}$$

(26)

where $\:{\nu\:}_{t}$ is the turbulent viscosity (in m²/s):

$$\:\begin{array}{c}{\nu\:}_{t}={\rho\:}_{mix0.5}{C}_{\mu\:}\frac{{k}^{2}}{\epsilon\:}\end{array}$$

(27)

Where, $\:{\rho\:}_{mix0.5}$ is the density of the mixed fluid under different temperature conditions; $\:{C}_{\mu\:}$ is the turbulent viscosity constant, 0.09; $\:k$ is the turbulent kinetic energy (in m²/s²); $\:\epsilon\:$ is the turbulent kinetic energy dissipation rate (in m²/s³). The calculated results of the temperature parameters are shown in Table 6:

Table 6 Temperature parameters under temperature (290 K, 295 K, 305 K, and 310 K).

As analyzed in conjunction with Fig. 16, an increase in the temperature of the sealed excavation face leads to a corresponding increase in both the molecular diffusion coefficient $\:{D}_{m}$ and the turbulent diffusion coefficient $\:{D}_{t}$. In comparison, the impact triggered by the thermal motion diffusion of gas molecules ($\:{D}_{m}$) is relatively minor. The rise in temperature intensifies the turbulence of the mixed fluid^47,48, causing a lateral loss of kinetic energy of the mixed fluid along its transport path: starting from the front end of the excavation face at a speed of 1.2 m/s, the velocity in the middle section decreases from 0.66 m/s to 0.54 m/s, while the velocity in the rear section drops from 0.42 m/s to 0.3 m/s. This lateral diversion of the mixed fluid also accounts for the widening of the coal dust cluster band at distances of 2.5–7.5 m from the excavation face, as shown in Fig. 15.

Distribution of coal dust concentration at the sealed excavation face under different temperature

Figure 17 shows the spatial distribution of coal dust clusters at the excavation face under various temperature conditions. The coal dust clusters are symmetrically distributed around the centerline of the excavation face, with the number of clusters gradually decreasing along the horizontal position and as they move laterally towards the side walls. Additionally, a sparse zone of coal dust is present in the rear section of the excavation face, specifically within the 15–17.5 m interval.

Fig. 17

Distribution of dust clusters under temperature (290 K, 295 K, 305 K, and 310 K).

The changes in coal dust concentration under different excavation face temperatures are shown in Fig. 18. In panel (a), compared to the reference condition at 300 K, coal dust accumulation zones with an increase of more than 5% are formed at coordinates (x = 1.2 m, z = 10 m) and (x = 3.6 m, z = 7.5 m). Sparse zones with a decrease of −5% in coal dust concentration are distributed in the middle section of the excavation face at (x = 0.6 m, z = 12.5 m) and (x = 2.2 m, z = 13.5 m), and in the rear section at (x = 0.6 m, z = 20 m) and (x = 3.2 m, z = 20 m).

In panel (b), with the excavation face temperature rising to 295 K, the range of coal dust accumulation zones decreases, and the accumulation increase is reduced. The sparse zones in the middle section of the excavation face shift towards the rear end.

Fig. 18

Contour map of dust distribution under temperature (290 K, 295 K, 305 K, and 310 K).

In panel (c), at a temperature of 305 K, the coal dust accumulation zones that were initially located in the front section of the excavation face at lower temperatures have shifted to the middle Sects. (10–15 m) and are drawn closer to the tunnel walls under the influence of the mixed flow. The sparse zones are concentrated in the region around coordinates (x = 3.6 m, z = 15 m).

In panel (d), continued temperature increase intensifies the transport of coal dust clusters towards the side walls, resulting in multiple accumulation and sparse zones within the excavation face. The accumulation increase can reach up to 18%, while the sparse decrease can reach − 21%. The rise in temperature within the sealed excavation space promotes the migration of coal dust clusters towards the middle-rear sections and side walls of the excavation face. Apparently, on the same timescale, the transport rate towards the side walls is greater than that towards the middle-rear sections. This phenomenon becomes more pronounced after the temperature exceeds 300 K.

Conclusions and innovations

Paradigm shift in dust migration: from single-ventilation to multi-factor synergy

Under forced ventilation, airflow velocity and volume dominate dust transport. Once the excavation face is sealed, aerodynamic driving force collapses, and high-concentration coal dust (peaks of 5000 mg/m³) migrates along the roof before depositing on sidewalls and distal zones. We quantified this transition for the first time and demonstrated that gas-emission rate, initial gas composition, particle size, and temperature—not ventilation parameters—dictate dust behavior in sealed headings. Dust-control strategies thus evolve from single-ventilation to precise multi-factor regulation.

Synergistic control via gas composition and particle size

Simulations reveal that CH₄/N₂ ratio and particle diameter jointly determine dust pathways. At 1–20% CH₄, dust is conveyed to the middle and rear sections with uniform distribution; at ≥ 80% CH₄, dust is confined to the 0–10 m zone near the face, forming a high-concentration core. Meanwhile, 1 μm fines remain suspended and reach the rear, whereas ≥ 5 μm particles settle rapidly, reinforcing frontal accumulation. We therefore propose the “high-CH₄–fine-particle” synergy: elevating methane while tailoring particle size suppresses long-range transport and accelerates local deposition, offering a precise, dual-action suppression route for sealed roadways.

Temperature as a new lever for dust control

Raising temperature from 305 K to 310 K barely changes overall migration distance yet markedly enlarges lateral dispersion, driving dust toward sidewalls. Concurrently, turbulence intensifies and the flow transitions from laminar to turbulent in the middle and rear sections. Traditional ventilation schemes ignore temperature; this study, for the first time, incorporates temperature management into sealed-roadway dust control. Fine-tuning local temperature can actively reshape dust distribution and curtail accumulation, providing a novel theoretical tool and practical pathway.

Targeted suppression framework for sealed excavation

Building on the above mechanisms, we present a targeted suppression framework for sealed headings: optimize initial gas composition, implement size-specific particle control, and apply temperature modulation to achieve localization, reduction, and valorization of dust. The framework delivers an engineering-ready solution for high-concentration dust management in sealed faces, advancing safer and more efficient underground operations.