Original Articles
Unmixed combustion: an alternative to fire

https://doi.org/10.1016/S0010-2180(99)00136-4Get rights and content

Abstract

A new form of combustion has been studied with properties that are greatly different from combustion via fire. Called unmixed combustion, it occurs when fuel and air alternately pass over a catalyst that undergoes oxidation and reduction, storing oxygen from the air and delivering it to the fuel. Examples of catalysts include finely divided Cu/CuO or FeO/Fe2O3 supported on γ-alumina. The full heat of combustion of the fuel is released, the fuel is converted to CO2 and water, and the air is depleted of oxygen, all without any need for the fuel and air to mix. Thus unmixed combustion is an alternative to fire, another way of using fuel and air to generate heat. The properties and characteristics of unmixed combustion are different from those of combustion by fire in a number of ways, some obvious, some subtle, suggesting a number of applications where conventional combustion is not or cannot be used. One example are situations in which it is difficult to provide complete mixing but in which complete combustion is required; e.g., rotary kiln incinerators. These incinerators have a failure mode called “puffing” to which unmixed combustion may be relevant. In the area of pollution control unmixed combustion is capable of burning natural gas and pyridine with zero NOx production; of burning sulfur-containing fuels in a manner that facilitates subsequent removal of the SO2; and of burning coal in a manner that directly provides sequestration ready CO2. Greatly enhanced delivery of heat to surfaces, rapidly supplying heat for cold starting engines, and direct generation of dry inert gases are also possible. Unmixed combustion also allows the delivery of heat uniformly throughout a volume. This method of heat delivery is potentially useful for supplying heat to endothermic reactions carried out in packed beds of catalyst. Experimental evaluation of this technique for steam reforming shows that while conventional steam reforming is a strongly endothermic reaction with an unfavorable equilibrium, the use of unmixed combustion allows a redefinition of the system’s thermodynamics, making the reaction weakly exothermic with a more favorable equilibrium.

Introduction

Despite its historic importance, fire is simply one of the possible methods of using fuel and air to generate useful heat. Other methods include fluid bed combustion and combustion with catalysts that accelerate chemical reaction rates. In this paper we report an examination of a different kind of catalytic combustion, i.e., combustion in which the function of the catalyst is not to increase chemical reaction rates but to facilitate mass transfer. While catalysts of this second kind are less common than rate accelerating catalysts in industry, they are of great importance in biological systems (e.g., hemoglobin). We call this second kind of catalytic combustion unmixed combustion because the use of a mass transfer catalyst can eliminate the need to mix the fuel and air before, during, or after the combustion event.

In unmixed combustion the fuel and air alternately pass over a mass transfer catalyst consisting of a metal or metal oxide that undergoes a reversible change in oxidation state during each portion of the reaction cycle [1]. For example, consider the reversible oxidation of Cu to CuO or of FeO to Fe2O3 that can occur as these materials are alternately exposed to air and a gaseous fuel. These reactions are facilitated if these materials are dispersed on high-surface-area alumina or a similar ceramic support in a packed-bed reactor. The fuel is oxidized to CO2 and H2O, and the air is depleted of oxygen, but during the combustion process the two need not mix. If a valve on the outlet of the reactor directs the two sets of reaction products in different directions, mixing need never occur.

The scientific interest in fluid bed combustion and combustion with rate accelerating catalysts has been based on their present and potential practical importance. Each of these alternatives to fire are phenomena with unique properties quite different from fire. Because of these unique properties, they are useful in a variety of specialized applications; i.e., they allow one to do things that would be difficult or impossible with fire. Unmixed combustion, too, has a unique set of properties. It is the authors’ hope that unmixed combustion will also prove to be a useful alternative to fire in one or another application.

In this paper we report a preliminary survey of the properties of unmixed combustion as they relate to its possible uses for a number of applications. The topics to be reported include preventing puffing in rotary kilns incinerators, delivery of heat to surfaces, providing heat for cold starting, ignition, eliminating NOx production during combustion, preventing SO2 and CO2 emissions from coal combustion, regenerating beds of spent sorbent, and supplying heat to endothermic catalytic reactions. In order to avoid moving back and forth among these diverse topics, this paper is not divided into the usual experimental methods, results, discussion, and conclusions sections. Instead the paper is divided into sections each of which treats a topic, providing introductory information, a description of the experimental methods used, a report of the results obtained, and a discussion of the significance of the results.

Section snippets

Preventing puffing in rotary kiln incinerators

One of fire’s limitations to which unmixed combustion may be relevant is the problem of incomplete combustion due to incomplete mixing; i.e., combustion via flames provides complete combustion only if the fuel is mixed with at least stoichiometric air, and in a timely manner. In some systems this can be difficult; e.g., rotary kiln incinerators which have the problem of “puffing”. As discussed by Wendt and Linak [2] when a rotary kiln incinerator is used to dispose of volatile liquids, the air

Delivery of heat to a surface

Unmixed combustion has also been shown to provide very high heat transfer rates in situations that may be of industrial significance. While flames can provide extremely high volumetric rates of heat release, in many applications it is then necessary to recover heat from the hot gases thus produced. In raising steam, for example, this means transferring heat out of the hot gases to the surfaces of steam tubes, through the steam tubes and into the boiling water. While the latter processes are

Providing heat for cold starting

Cold starting is a problem for a wide variety of systems and devices. For example, diesel engines can be impossible to restart in cold environments; a problem that truck drivers address by stopping only at places at which they can plug in their electric block heaters. Also, 80% of the NOx and unburned hydrocarbons emitted by automobiles equipped with three-way catalyst system occurs during starting while the catalyst is warming to its operating temperature [3]. Since finely divided metals are

Ignition

If an unmixed combustion process is to ignite, the catalyst must be at a sufficiently high temperature so that the fuel will reduce it. A series of experiments was done to define this threshold. In these experiments a sample of an unmixed combustion catalyst was cemented to the end of a Type K thermocouple inside a tube furnace. A timer-controller three-way solenoid valve was used to expose the sample to an alternating flow of fuel and air as the furnace’s temperature was increased. The

Generation of inert gas for airplane fuel tanks

Stoichiometric combustion followed by drying is a method frequently used for converting air into an “inert” gas. This method of inert gas generation is limited, however, to situations in which it is either practical to dry the postcombustion gases or acceptable that they be wet. The design of airplane fuel tanks requires that they be vented to the atmosphere, breathing out air and fuel vapors as the plane ascends, breathing in air as it descends. For civilian aircraft the mixture of air and

NOx emissions

One significant disadvantage of fire is that it releases a variety of pollutants to the atmosphere. One of these, NOx, can be controlled to low levels by combustion modification techniques for fuels that do not contain chemically bound nitrogen. Even in this favorable case, however, it is not possible to completely eliminate the production of NOx, which is generated via the prompt NOx mechanism. In the case of fuels that do contain bound nitrogen, only limited control can be achieved over NOx

Heat generation within a volume: sorbent regeneration

Sorbent beds are frequently used to remove impurities from gases. When the bed approaches saturation it can, in principle, be regenerated by heating. In some applications, however, this is difficult. When heat is input to a sorbent bed from the outside, the outermost sections of the bed insulate the inner sections, making the regeneration a slow process. As a method of generating heat, unmixed combustion has the advantage that the amount of heat an unmixed combustion catalyst produces per cycle

Supplying heat to endothermic catalytic reactions

Many industrial processes involve endothermic reactions that are carried out in packed beds of catalysts. Here again the heat transfer situation can be difficult. As an example let us consider the steam reforming process. Steam reforming is a process, which operates at elevated pressures and at temperatures in the 600°C to 800°C range. In this process a catalyst, usually nickel on a refractory support, is used to carry out the endothermic conversion of steam and natural gas or other light

Unmixed combustion of sulfur-containing fuels

When wet scrubbing is used to control SO2 emissions from the combustion of high sulfur fuels, the cost of the reagents consumed is a minor part of the total cost, e.g., 4.2% in the example shown by McKnight [10]. Thus the overall cost of scrubbing flue gas to remove SO2 is insensitive to the amount of SO2 which is to be removed but nearly proportional to the amount of flue gas which must be scrubbed to remove it. If, with a suitably chosen catalyst, an unmixed combustor could burn a

Prevention of CO2 and SO2 emissions from coal combustion

Coal, while it is the most abundant fossil fuel resource in the United States, has a number of difficult problems. The flue gas produced by coal combustion is erosive, making the use of coal as a gas turbine fuel difficult. In addition to the criteria pollutants, CO, SO2 and NOx, coal combustors emit mercury, submicron particles coated with toxic heavy metals [11], furans, dioxins, and polynuclear aromatic compounds [12]. Furthermore, coal has the disadvantage that its CO2 emissions per BTU

Conclusions

In this paper the authors have demonstrated a number of applications in which unmixed combustion has properties greatly different from fire and for which unmixed combustion may prove a useful alternative to combustion via fire. These include providing complete combustion in applications in which it is difficult to provide complete mixing (e.g., rotary kiln incinerators); combustion with zero NOx production; high heat transfer density; combustion of high sulfur fuels in which the volume of flue

Acknowledgements

Support for this research was provided by the NSF SBIR program under Grants ISI-9261121, ISI-3960042, DMI-9560194, and DMI-9560237, by the EPA SBIR program under contract 68D20093, the DARPA SBIR program under contracts DAAH01-94-C-R071 and DAAH01-95-R162, U.S. Air Force SBIR program under contract F41624-97-C009 and U.S. Department of Energy contract DE-FC02-97EE50488, and by Energy and Environmental Research Corporation.

References (19)

  • H Freund et al.

    Combust. Flame

    (1982)
  • The first reference to this approach as a general method of combustion was apparently R. K. Lyon, Preprints ACS Fuels...
  • J.O.L Wendt et al.

    Combust. Sci. Technol.

    (1988)
  • Kirk-Othmer Encyclopedia of Chemical Technology, Vol. 5, 4th ed., pp....
  • C Chaouki et al.

    Ind. Eng. Chem. Res.

    (1944)
  • J.R Rostum-Nielsen
  • M Jacoby

    Chem. Eng. News

    (1999)
  • Che, S. C., and Bredehoft, R. L., Preprints, Division of Petroleum Chemistry, 210th National Meeting, American Chemical...
  • Williams, R., U.S. Patent 1,938,202, 1993; see also E. Gorin and W. B. Retallick, U.S. Patent...
There are more references available in the full text version of this article.

Cited by (163)

View all citing articles on Scopus
View full text