FCC gasoline desulfurization by pervaporation: Effects of gasoline components

https://doi.org/10.1016/j.memsci.2007.01.029Get rights and content

Abstract

Detailed investigations of the influence of fluid catalytic cracking (FCC) gasoline components on membrane performance have been rarely conducted. This is related to the fact that gasoline is a complex mixture composed of hundreds of components. It is also necessary to investigate the composition change of gasoline components during membrane treatment, since the research is helpful in analyzing the octane number loss. Typical hydrocarbon and sulfur components were studied for their effect on pervaporation (PV) performance and swelling degree through polyethylene glycol (PEG) membranes. The results demonstrated that the increasing aromatics, alkenes and sulfur content in feed led to the increased flux and decreased sulfur enrichment factor due to the exacerbated membrane swelling. However, alkanes and cycloalkanes had little influence on membrane swelling and PV performance. Membrane performance comparison for model compounds and gasoline feed was also conducted to examine above conclusion. Hydrocarbon group, sulfur species distribution and octane number of feed and permeate sample were obtained by gas chromatography (GC) analysis, and the results suggested that aromatics content increased and other hydrocarbon groups content changed little through PV process. Additionally, the overall analysis indicated that higher sulfur enrichment factor and lower flux were usually achieved with lower sulfur level feed.

Introduction

Refiners worldwide are rushing to develop their strategy for economically and reliably meeting clean fuel regulation by the view of Zhao et al. [1]. Application of membrane technology in petrochemical field provides a newly efficient approach for the separation of organic mixture and has gained increasing attention of membrane and petrochemical field. Gasoline desulfurization by membrane process is a newly emerged technology. The process offers a number of potential advantages over conventional sulfur removal process.

In our previous work [2], [3], the PEG membrane material was selected and modified for FCC gasoline desulfurization by PV and the detailed technology background introduction was involved also. The PEG membrane has excellent sulfur removal performance with higher sulfur enrichment factor (7.3 for model compounds feed) than reported polydimethyl siloxane (PDMS) membrane material (3.9 for model compounds feed) [4], [5].

Little has been reported on the influence of gasoline components on membrane material and PV performance for FCC gasoline desulfurization. With regard to previous research, the influence of aromatic and olefinic compounds on the polydimethyl siloxane/ polyacrylonitrile (PDMS/PAN) composite membranes performance was reported [6] and only simple model compounds were discussed. The PV performance of some membrane materials in five component model feed (1-pentene, 2,2,4-trimethylpentane, methylcyclohexane, toluene, and thiophene) and refinery naphtha feed was also reported [7]. For the PERVAP1060 commercial membrane, it was observed that the sulfur enrichment factor (calculated by total sulfur content) for naphtha feed was lower than that for five component model feed, but no further analysis about the influence of gasoline components on membrane material was provided. Until recently the report about the influence of gasoline components on membrane is quite insufficient.

However, FCC gasoline is a complex mixture composed of hundreds of compounds, as is described in this paper, and it is necessary to analyze the detailed interaction between membrane performance and gasoline components. The aim of this paper is to present a detailed analysis of the effect of various hydrocarbon and sulfur components on PEG membrane performance through the PV and swelling experiments. Meanwhile, by gas chromatography (GC) analysis, membrane performance comparison for two kinds of typical FCC gasoline was conducted which provided further understanding to the effect of gasoline components. Octane number loss for gasoline during membrane treatment will also be discussed. All these investigations will provide helpful suggestions for the newly-emerged membrane desulfurization technology.

Section snippets

Materials

Membrane materials were all purchased from the Shanghai Reagent Factory. Gasoline feed with higher sulfur content level was obtained from Shenghua Refinery (China) and lower sulfur content level from Dushanzi Petrochemical Plant (China). Both the gasoline feed were after alkali cleaning process. Four kinds of model compounds feed were used as are shown in Table 1. And the model feed sulfur content was about 1200 μg/g in control of thiophene if no special declaration. All chemicals used were of

Results and discussion

Table 2 gives the hydrocarbon group GC analysis results for the FCC gasoline from Shenghua Refinery (China). FCC gasoline is a complex mixture composed of hundreds of compounds, in which four typical hydrocarbon groups are alkanes, cycloalkanes, aromatics and alkenes. Different stability and interaction with the membrane, which influence the membrane performance severely, exist in above various hydrocarbon groups.

According to the hydrocarbon groups content in Table 2, four kinds of model

Conclusions

PEG was an excellent sulfur removal membrane material. The investigations of the influence of gasoline components on PEG membrane performance and the discussion of PV performance and octane number loss for various gasolines have been rarely conducted before.

The PV and swelling experiments results demonstrated that the increasing aromatics and alkenes content led to the increased flux and decreased sulfur enrichment factor due to the exacerbated membrane swelling. Likewise, alkanes and

Acknowledgements

The authors gratefully acknowledge support of this work by Program for New Century Excellent Talents in University (NCET) of Ministry of Education of China.

References (13)

There are more references available in the full text version of this article.

Cited by (0)

View full text