Introduction
The electronic nose (e-nose) system analyses the volatile compounds in products by mimicking the human olfactory system and is capable of providing both chemical and sensory information. One of the most important applications of the electronic nose is gas sensors widely used in various industries, including construction, chemical and petrochemical, environmental monitoring, medical and pharmaceutical, agricultural, and food industries, as well as in many other processes where gas monitoring and analysis are essential. Due to the increasing cost of experimental testing, computational fluid dynamics (CFD) can be employed to investigate the effects of key factors in the electronic nose. CFD is a branch of numerical methods used to solve the governing equations describing various flow phenomena. In CFD, different methods and algorithms are utilised to obtain solutions; however, in all cases, the problem domain is discretised into a large number of small elements, and the governing equations are solved for each element. A review of previous studies indicates that simulation processes aimed at achieving reliable results in this field have received considerable attention, and the data obtained from these simulations can be highly accurate and efficient. Moreover, the sensor chamber plays an important role in enhancing the performance, stability, and sensitivity of an electronic nose. Therefore, four different configurations of 3D sensing chambers were simulated using ANSYS FLUENT software, examining both air and CO₂ as fluids. Numerical simulations were carried out to investigate the gas flow behaviour inside these four chambers and specify the optimal chamber design with the best stability time.
Materials and Methods
In the present study, four chamber geometries namely cylindrical, pyramidal, conical, and hemispherical, were designed using CATIA software. Subsequently, three-dimensional simulations were performed using Ansys Fluent software. To predict the fluid flow behaviour, the continuity equation and the Navier–Stokes equations were employed. The boundary conditions were identical for all geometries, and given the equal inlet and outlet cross-sectional areas across all configurations, the only difference among the models lies in their overall geometric structure. The electronic nose chambers were tested with a laminar flow and the SIMPLEC method.
Results and Discussion
The 3-Dimensional simulations were conducted for the four geometries under two fluid injections, air and carbon dioxide, while maintaining identical boundary conditions. To investigate the fluid behaviour inside the geometries, flow field contours were presented at 25 s, 50 s, 75 s, and under approximate stability conditions when the contours became nearly time-invariant. In the air injection case, the fluid inside the hemispherical geometry reached steady-state conditions by 25 s, faster than in the other geometries, and maintained a similar flow pattern at later times. Based on the simulation results, the hemispherical geometry exhibited the best overall performance. This favourable behaviour can be attributed to the geometric dimensions of the hemisphere in the sensor surface region. In contrast, the behaviour of carbon dioxide differed significantly from that of air. Under carbon dioxide injection, similar to the air injection case, the fluid in the hemispherical geometry reached steady-state conditions by 25 s and fully covered the outlet region. In this geometry, the flow attained a nearly uniform distribution at 25 s, and the fluid fully contacted the sensor surface. Also, results showed that small vortices produced near the sensor surface can improve the stability time, and they can help to bring the fluid to the surface faster.
Conclusion
In the present study, four geometries were investigated under two injection conditions: air and carbon dioxide. The evaluations were conducted to identify the optimal geometry in terms of fluid flow behaviour, sensor surface coverage, and flow stability. The results demonstrated that the hemispherical geometry exhibited the best performance under both air and carbon dioxide injection conditions. This indicates that providing a larger fluid–sensor contact area along with a balanced geometric configuration can lead to improved fluid behaviour and overall system performance.