Extractive desulfurization in microchannels with polyethylene glycol 400: An experimental study and mass transfer evaluation

https://doi.org/10.1016/j.cep.2022.109096Get rights and content

Highlights

  • PEG 400 was used as an extraction solvent for EDS in microdevices.

  • An algorithm was created to measure the mass transfer parameters.

  • Relationships between flow patterns, mass transfer, and EDS were investigated.

  • Penetration theory model adequately describes droplet flow pattern.

  • Among the flow patterns studied, the slug pattern presents the best results.

Abstract

Conventional methods for removing sulfur compounds show unsatisfactory performance, especially when treating aromatic compounds. Current studies have directed efforts to investigate extraction desulfurization (EDS) using microdevices to overcome this difficulty. These units are designed with micrometer-sized channels responsible for intensifying heat and mass transfer phenomena. When using EDS, an adequate choice of extraction solvent is essential and among the possibilities, an excellent candidate for extraction solvent is polyethylene glycol (PEG). The present work studied the extractive performance offered by microdevices and the aspects related to mass transfer intensification by adopting PEG 400 as an extraction solvent in the removal of dibenzothiophene. The identification of flow patterns was performed by digital image colorimetry, making it possible to estimate the specific interfacial area (as) and the overall mass transfer coefficient (KLas) more accurately. DBT decontamination behavior was investigated under different volumetric flow rates (71.74–326.58 μL/min), different microchannel lengths (1.0–2.0 m), and a fixed diameter of 0.5 mm. The best result achieved an extractive efficiency of 94.06%, corresponding to a volumetric flow rate of 107.30 μL/min, whose interfacial area and overall mass transfer coefficients related to the slug flow pattern were calculated as 453.69 m−1 and 0.0229 s1, respectively.

Introduction

Environmental problems resulting from human action represent one of the greatest challenges today and for the next decades. In a unique way, most countries try to establish strict regulatory legislation to mitigate the impacts of emissions on the environment. In this scenario, stricter laws that limit the emission of pollutants generated from the burning of fossil fuels should be encouraged, especially with the worsening of climate change resulting from the release of greenhouse gases into the atmosphere, especially sulfur compounds. The latter are also direct sources of acid rain and cause problems to human health, such as respiratory tract diseases, skin irritation, and cardiovascular problems [1].

Sulfur compounds are present in various forms and can be classified into four major categories: mercaptans, sulfides, disulfides, and thiophenes. According to Srivastava et al. [2], the crude oil processing chains are responsible for the emission of approximately 83.45 kT of sulfur oxides (SOx). Crude oil has sulfur compounds in its composition and its current processing chain is insufficient to fully overcome the recalcitrant nature of these compounds. The sulfur content is expressed in mass fractions, with values ranging from 0.1% to levels greater than 5%, depending on the type and source of the crude oil [3].

Catalytic hydrodesulfurization (HDS) is the most mature technology for removing sulfur compounds in oil refineries. The technology consists of converting these contaminants into hydrogen sulfite H2S. The process takes place in a fixed bed reactor operating under high pressure (3.5–7.0 MPa) and temperature (573.2–673.2 K) in the presence of bimetallic catalysts (e.g. Co–Mo or Ni–Mo) [4]. HDS is efficient to transform aliphatic sulfur compounds, however, it presents an unsatisfactory performance in the treatment of aromatic sulfur compounds, such as dibenzothiophene (DBT) and its derivatives [5]. Ali et al. [6] mentioned that HDS can achieve removal rates of 96.9% of DBT from the development of new catalysts for HDS, however, this technology is still incipient and is currently applied only in simulated fuels. While Hamiye et al. [7] reported that the desulfurization process was able to obtain 35% desulfurization in a real fuel from a refinery, obtaining a fuel that is above the sulfur content limit set by current regulations. To overcome this limitation of HDS, alternative desulfurization methods have attracted the attention of researchers.

Extractive desulfurization (EDS) is a liquid–liquid extraction method whose extraction solvent must be immiscible and have a high affinity for the sulfur compound. When compared to HDS, EDS offers advantages such as the absence of complex infrastructure, operation in mild conditions of temperature and pressure, as well as not compromising the chemical identity of the fuel. The pioneer studies in EDS proposed the use of packed columns to improve sulfur removal performance, but there is a recent trend towards the use of new reactor designs. Inspired by the concepts of process intensification, reactor miniaturization has become an important subject in the field of EDS. Microdevices are units with micrometer-sized channels that enhance mass and heat transfer phenomena. The microdevices offer high values of specific interfacial area [8], which allow achieving similar performances with residence times several times shorter than packed columns. The rapid execution of experiments and the use of small amounts of chemicals are also positive aspects of microdevices [9]. On the industrial scale, microdevices already show their potential for enabling scale-up without compromising the performance observed at smaller scales.

Recent studies on EDS in microdevices have been striving to consolidate the technology, and many of them were guided by environmental factors for the choice of solvents. Green solvents are a class of compounds that are considered biodegradable and generally have low toxicity and low corrosivity. Some works have shown surprising results in desulfurization in microdevices using green solvents. However, the choice of deep eutectic solvents or ionic liquids sometimes overlooks technical gaps. Ionic liquids are recognized for involving multiple steps in their preparation and generally have a high cost [10]. On the other hand, the regeneration of eutectic solvents must involve the use of systems that avoid compromising the integrity of the compound by heating [11]. it is worth mentioning that the magnitude of the desulfurization achieved by deep eutectic solvents practically does not differ from the results obtained by more common and cheaper solvents, such as polyethylene glycol. Polyethylene glycol (PEG) is a linear polyether with hydroxyl groups at its ends. Due to its particular chemical structure, PEG is an excellent candidate for extraction solvent, since it has multiple sites for establishing hydrogen bonds with sulfur compounds [12]. This polymer has also been used in the synthesis of deep eutectic solvents. Chen et al. [13] achieved 90% benzothiophene removal and full DBT decontamination (100%) after treating a simulated fuel contaminated with 500 ppm sulfur compounds using PEG 200, while Kianpour and Azizian [14] achieved 98% DBT removal from fuel after three cycles of PEG extraction. Al-Azzawi et al. [10] reported that there were no significant differences between the performance of a eutectic solvent and PEG 200, demonstrating that PEG itself is a possible candidate for extraction solvent in EDS. Despite the importance of these results, there are still gaps that need to be clarified, especially those related to mass transfer phenomena occurring in microfluidics. Determining the specific interfacial area is also often neglected in EDS studies and its exact determination can support discussions about process performance.

Thus, the present work investigated the performance of desulfurization by extraction, as well as aspects of mass transfer intensification offered by the microdevice. Polyethylene glycol with an average molecular mass of 400 Da (PEG 400) was adopted as an extraction solvent for the removal of DBT from a model fuel. Slug and droplet flow patterns were mapped using different volumetric flow rates. The identification of flow patterns was performed by digital image colorimetry to estimate the specific interfacial area and the mass transfer coefficient more accurately. Furthermore, the DBT removal behavior was investigated under different volumetric flow rates and different microchannel lengths. All these aspects will dictate the feasibility of PEG 400 as an extraction solvent in EDS experiments.

Section snippets

Materials

Polyethylene glycol 400 (>99.5%) was used as a solvent in the EDS experiments and was purchased from Merck (Germany). The model fuel was represented by n-hexane with a mixture of isomers (98%), which was purchased from Synth (Brazil). Dibenzothiophene (DBT) (98%) and dichloromethane (>99.5%) were purchased from Sigma-Aldrich-Chemical (USA).

Experimental apparatus

The experimental apparatus (Fig. 1) used for the EDS tests consisted of a syringe pump, 10 mL glass syringes, a T-type connector, and polytetrafluoroethylene

Mapping of flow patterns

In microdevices, four distinct flow patterns are expected: slug flow, slug-droplet transition pattern, droplet flow, and parallel flow, where the latter is obtained only in two-phase aqueous systems [9].

The reason why the performance of microdevices is so distinct is related to the phenomena associated with mass transfer, which results from the contributions of internal circulation and interfacial diffusion. The behavior of the system and how these contributions affect the process vary

Conclusion

To evaluate the extractive efficiency of PEG 400 in diesel desulfurization, a microdevice was designed to investigate the process. The present study successfully evaluated the use of PEG 400 in EDS for a simulated fuel. From the obtained results, it was clear that the slug flow pattern presents the best results, which was obtained for a volumetric flow rate of 107.30 μL/min with an extractive index of 94.06%.

The experimental volumetric flow rate range of the slug flow pattern was from 71.74 to

Nomenclature

asSpecific interfacial area (m2/m3)
CaCapillary number
CConcentration (mol/L)
EEfficiency Factor (%)
KLMass transfer coefficient (m/s)
ReReynolds number
τResidence time (s)
tTime (s)
VVolume (m3)

CRediT authorship contribution statement

Luiz E.P. Santiago: Conceptualization, Methodology, Software, Data curation, Writing – original draft, Visualization, Investigation, Supervision, Writing – review & editing. Maxwell G. Silva: Data curation, Writing – original draft, Visualization, Investigation, Supervision, Writing – review & editing. Eledir V. Sobrinho: Software, Validation. Juan A.C. Ruiz: Software, Validation. Carlos E.A. Padilha: Visualization, Investigation. Domingos F.S. Souza: Software, Validation.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Luiz Eduardo Pereira Santago reports equipment, drugs, or supplies was provided by Federal University of Rio Grande do Norte. Luiz Eduardo Pereira Santago reports a relationship with Federal University of Rio Grande do Norte that includes: employment. There are no additional relationships or activities to declare.

Acknowledgments

The authors thank the research funding agencies in Brazil CAPES and CNPQ for all the support provided. In addition, we thank the Chemistry Institute from UFRN for providing the infrastructure to carry out this work.

References (27)

Cited by (3)

View full text