Analysis of an industrial adsorption process based on ammonia chemisorption: Modeling and simulation

Highlights

• Development of a first-principles model to simulate an industrial adsorption process of ammonia on zinc sulphate-doped activated carbon.

• Characterization of the doped carbon and measurements of ammonia adsorption isotherms and breakthrough fronts.

• Estimability analysis of the unknown parameters involved in the isotherm equation of Sips as well as in the adsorption model from the available experimental measurements.

• Identification of the most estimable parameters and assessment of their accuracy by means of confidence intervals.

• Validation of the identified model by means of additional measurements and appropriate indices, and confirmation by chi-squared test.

Abstract

The paper deals with the development of a one-dimensional model to simulate an industrial adsorption process of ammonia on zinc sulphate-doped activated carbon. It is described by mass balance, thermodynamic and adsorption kinetics equations. Since equilibrium is involved in the model, we started with experimental measurements of ammonia adsorption isotherms on doped activated carbon. A method based on the sensitivity analysis of parameters was used to evaluate the estimability of unknown parameters involved in the Sips adsorption isotherm equation. The estimable parameters were then identified using experimental data at three different temperatures, i.e. 285, 293 and 313 K. Experimental breakthrough fronts at different ammonia concentrations and gas flow rates were then used to determine the overall mass transfer coefficient and the axial dispersion coefficient involved in the model equations, implemented and solved within Comsol Multiphysics® software. Finally, we validated the model by means of four additional breakthrough fronts that were different from those used to identify the parameters. The model predictions and the experimental measurements show a very agreement which is quantified by the performance indices of the model and confirmed by a chi-squared test. The validated model can be used as a predictive tool for the design and optimization of the ammonia adsorption process for air purification boxes used to equip cabins with pressurization and air-conditioning of mechanical devices.

Introduction

The emission of gaseous ammonia is one of the main concerns of composting and organic waste methanisation facilities. Ammonia causes chronic respiratory diseases, such as asthma and occupational chronic bronchitis, when the professional exposure is too high. The Ministry of Labor has established an occupational exposure limit to ammonia at 10 ppm in France (Courtois and Cadou, 2016).

Mechanical machinery used in composting and methanisation plants often operate in an atmosphere where the ambient level of ammonia is higher than this maximum limit. The use of mechanical devices equipped with cabin pressurization and air-conditioning is common in many fields such as building and civil engineering, waste treatment and recycling, and agriculture.

The aim of these cabins is to provide, other than working comfort, protection against various risks potentially present in the changing environment of the device, such as particles and/or gases and toxic fumes. To do so, the cabin is placed in overpressure (> 100 Pa) with respect to the outside via a filtered and purified ventilation system (Bémer et al., 2015). The air flow rate introduced into the cockpit is around 50 to 100 m³.h⁻¹ depending on the manufacturer.

Currently, the installation of an air purification box is very rarely available when purchasing the machine, it is rather included as an adaptation. These air purification boxes, when they are included as protection against multiphase pollutants, are composed of a prefilter layer, followed by a high efficiency filtration layer, then an adsorbent layer. Suppliers offer various configurations, from multilayer annular cartridges to a simple overlay of adapted plane media. While some studies (Bémer et al., 2005; Organiscak and Schmitz, 2012) have assessed the efficiency of cabin pressurization and air-conditioning regarding protection against particles, there are little scientific or technical data on gas and vapor filter boxes. The sector with the most amount of information concerning the adsorption of dangerous gas compounds with regards to preventing occupational hazards is the respiratory personal protective equipment sector and its associated standards (Guimon, 2019).

Activated carbon is the main adsorbent material used on an industrial scale. It is mostly chosen for its large specific surface area and its low cost. Activated carbons doped with inorganic compounds are the main medium used today to equip air purification boxes for machines to purify ammonia, and where the adsorption reactions of pollutant gaseous phases take place. Zinc-sulphate is therefore commonly referenced by manufacturers as their doping agent (Bandosz and Petit, 2009). Currently, manufacturers assess only the performance of the adsorption layer which is generally carried out at relatively high concentration levels. The issue with evaluating the performance of purification boxes used to equip machines is complicated insomuch as the service time is very long (several days) on the one hand, and there are no or very little obligations in terms of normative testing for the purification of the gaseous phase, on the other.

The objective of this paper is to assess and improve the performance of gas and vapor purification in a pilot adsorption column containing the same zinc sulphate-doped activated carbon and operating in the same conditions as the reference air purification box. More specifically, an experimental study on adsorption of ammonia on the zinc sulphate-doped activated carbon is carried out along with the modeling of the phenomena involved in the column. The experiments mainly consist of the characterization of the doped carbon and the measurements of adsorption isotherms and breakthrough fronts of ammonia. It should be noted that only ammonia is considered as the main gaseous pollutant. This simplifying assumption of the gas/vapor phase composition is based on the results of measurement campaigns carried out in France at six methanisation units (Dirrenberger et al., 2019).