Reactive distillation combines the functionality of a continuous reactor with a conventional distillation column. This fusion of unit operations is not new, it has been in limited use since the 1920s (e.g. methyl acetate production), but a crucial development in the 1980s exposed previously untapped potential and led to a rejuvenation of this technology The development was the means to support fine catalyst particles in a manner which allows both effective mass transfer and reaction on the surface of the catalyst. This advance permitted heterogenous reactions to be considered for reactive distillation and introduced the possibility of using hybrid columns which contain both reactive and non-reactive sections. These features have been found to be particularly advantageous for some processes, including the synthesis of methyl tert-butyl ether (MTBE).
MTBE is valuable as a gasoline additive which simultaneously increases the octane rating of the fuel and adds oxygen which promotes cleaner burning. When used in place of lead-based octane enhancers, dual environmental benefits are realised: a reduction in atmospheric lead concentrations and reduced emissions of carbon monoxide and other smog forming chemicals. Since the 1970s, the worldwide consumption of MTBE has increased significantly and many new facilities have been constructed to support the growing market (Kirschner, 1996; Riddle, 1996). The success of hybrid reactive distillation in this service has been clearly evident and the technology is now accepted as the best which is currently available.
Ethyl teri-butyl ether (ETBE) is an alternative to MTBE which offers a further environmental advantage: the primary raw material, ethanol, can be produced from renewable resources. ETBE also has superior fuel properties, including a higher octane rating and a low volatility, which enhance its potential. Although MTBE is now widely used, ETBE has remained little more than a novelty and commercial production has been extremely limited. However, a reduction in the cost of ethanol compared with methanol (via subsidies or otherwise) could make ETBE the market leader provided the technological base for commercialisation was available.
The combination of reaction and separation within a single unit operation not only reduces the overall capital cost of a process but provides process benefits in some cases. These benefits arise from the constant recycling of reactants to the reaction zone which increases the conversion of the limiting reactant in an equilibrium limited reaction. A secondary benefit is the increased energy efficiency which results from directly utilising the heat of reaction for fractionation. Reactive distillation is usually justified on steady-state process results but the increased process complexity reduces operability and controllability compared with a conventional two-stage process. Consequently, the full steady-state benefits may not be realisable in dynamic operation, particularly if regular disturbances are likely.
To date, effective, specialised control schemes have not been developed for reactive distillation and a lack of understanding of the process fundamentals has hindered the optimisation of reactive columns. The main source of difficulty in developing an effective control system is the uncertainty of control objectives. It is desirable to simultaneously produce both a high reactant conversion and a high product purity (i.e. it should exactly duplicate both functions of the two-stage process) but the operating conditions which maximise reactant conversion do not necessarily coincide with the conditions which maximise the product purity. Thus, the dual objectives are not entirely compatible.
More recently, reports of multiple steady states in the reactive distillation of MTBE (e.g. Nijhuis et al., 1993; Jacobs and Krishna, 1993) have revealed a further complexity with unique implications for process control. There is a substantial potential for the operation of columns where multiplicity exists to be adversely affected by unwanted transitions between parallel steady states. Output multiplicity has been confirmed at an experimental level in azeotropic distillation (Güttinger et al., 1997), and there appear to be parallels with reactive distillation, but the experimental evidence of multiplicity in reactive distillation has not yet been produced. Several authors have speculated on the physical cause(s) of the simulation-based multiplicities (e.g. Jacobs and Krishna, 1993; Hauan et al., 1995; 1997) and a wide range of explanations has been proposed, but the underlying and fundamental aspects have not been elaborated and generally only mechanistic descriptions are available. It is clear that this phenomenon is not yet fully understood although there are at least two promising lines of research which appear to offer broadly applicable explanations with strong foundation.
Experimental work on reactive distillation (and particularly hybrid reactive distillation) has been limited to work connected with patent applications and several bench-top studies that focus on the catalyst support mechanism. Although there has undoubtedly been work in the private sector to support process designs, data is not widely available to test modelling techniques and control strategies. There are significant opportunities to make a substantial contribution to the current body of knowledge in each of the areas described above.
1.2 Motivation and Significance of this Work
At a fundamental level, this study was motivated by a perceived need to address Australia's worsening urban pollution and smog levels which are caused in a large part by emissions from automotive engines. While our air quality still exceeds that of most developed countries, we risk losing this advantage through continued neglect and apathy with respect to controlling the composition of gasoline and, therefore, the composition of vehicle emissions. Our standards are already 10-20 years behind those of countries such as the USA, Japan and the Scandinavian nations in several key areas (Furzer, 1994).
The failure to keep the legislation regarding the composition of Australian gasoline in line with recent developments and international recommendations can be attributed (at least partly) to the insular nature of the local oil refining industry and an over-reliance on overseas technology which has stifled local research and development activities. Consequently, the Australian public and environmental movements have failed to fully appreciate the decrease in air quality and fail to recognise the technological alternatives that are available to redress this problem. Local research has a valuable role in promoting these issues and can help to initiate change by identifying technically sound and economically viable approaches to reducing toxic emissions and improving air quality.
This study was also motivated by several technical issues. Hybrid reactive distillation is a relatively new development and there are still significant gaps in the current body of knowledge, particularly concerning specialised control strategies and the notable phenomenon of multiplicity. Although effective research has been completed on the chemistry of reactive distillation and the control of conventional distillation, the intersection of these activities has attracted little prior attention. With expertise available in the associated areas, the current work was an opportunity to direct effort towards this aspect of reactive distillation using a combination of simulation and experimentation.
A relatively wide focus was assumed for this research to permit significant contributions to be made in several areas. In particular, an improved understanding of the behaviour of hybrid reactive distillation columns and the associated design issues will help to reduce the technical risk in developing this technology (especially locally), the study of control issues will allow for safer and more profitable operation of these columns and the experimental work will provide the basis for further work which should clearly demonstrate that this technology is both feasible and potentially profitable.
This thesis was undertaken with the objective of making technical contributions to the body of knowledge associated with fuel ether production in hybrid reactive distillation columns in the following areas:
• effective modelling techniques for both the steady state and dynamic simulation of reactive distillation columns;
• the understanding of obscure operating characteristics which are specific to reactive distillation;
• a design strategy for hybrid reactive distillation;
• the implications of extremely non-linear behaviour for the control of reactive distillation systems;
• the potential for multiple steady states in hybrid reactive distillation and the causes and implications of this phenomenon;
• the construction of pilot scale facilities for the synthesis of ETBE from ethanol and a locally available hydrocarbon source using non-commercial packing arrangements.
Was this article helpful?