WO2015103436A1

WO2015103436A1 - Method and apparatus for extending flammability limits in a combustion reaction

Info

Publication number: WO2015103436A1
Application number: PCT/US2014/073086
Authority: WO
Inventors: Joseph Colannino; James K DANSIE; Jesse C. Dumas
Original assignee: Clearsign Combustion Corporation
Priority date: 2013-12-31
Filing date: 2014-12-31
Publication date: 2015-07-09
Also published as: CN105765304A; US20160363315A1; EP3090210A1; CN105765304B

Abstract

A method for controlling a combustion reaction includes introducing fuel and oxidizer into a combustion volume at a ratio that is outside a range defined by an upper flammability or stability limit and a lower flammability or stability limit of the fuel, and producing a modified range defined by a modified upper flammability or stability limit and a modified lower flammability or stability limit of the fuel, by applying an electric field across a flame supported by the fuel and oxidizer, the ratio falling within the modified range.

Description

METHOD AND APPARATUS FOR EXTENDING FLAMMABILITY LIMITS IN A COMBUSTION

REACTION

CROSS REFERENCE TO RELATED APPLICATION

The present application claims priority benefit from U.S. Provisional Patent Application No. 61/922,430, entitled "METHOD AND APPARATUS FOR

EXTENDING FLAMMABILITY LIMITS IN COMBUSTION REACTION", filed December 31 , 2013; which, to the extent not inconsistent with the disclosure herein, is incorporated by reference.

BACKGROUND

Combustion requires fuel, an oxidant such as oxygen, and an ignition source. For any given fuel, there is a range of fuel/oxygen ratios within which combustion can occur or be sustained. That range is defined by the flammability limits of the particular fuel, i.e., the lowest and highest ratios at which the fuel is flammable. Typically, variables are fixed at standard values unless specifically defined otherwise. For example, the fuel and oxidizer may be specified to be at 25 degrees C and one bar of pressure, absolute (100 kPa). The oxidizer may be specified to be oxygen in air.

Flammability limits are expressed as a percentage of fuel within a volume of air. For example, the lower flammability limit of gasoline (100 octane) is 1 .4%, i.e., a mixture containing 1 .4% gasoline and 98.6% air. This is the lowest, or leanest concentration of gasoline that is combustible. At the other end of the range, the upper flammability limit of gasoline is 7.6%, representing the richest concentration that is combustible.

Stability limits are analogous to flammability limits. Flammability limits are considered properties of the fuel - that is, a flammability limit is device- independent. Stability limits, by comparison, are the actual combustible limits realizable by a given device such as an actual burner or combustor. In industrial burners, stability limits often govern the safe operation of the burner. The lower stability limit is the most fuel-lean composition whose combustion a given burner can support, while the upper stability limit is the most fuel-rich composition whose combustion a given burner can support. In practical combustion equipment, the upper and lower stability limits define the stable combustion operating range for a given burner or combustion device.

SUMMARY

According to an embodiment, a method includes introducing fuel and air into a combustion volume in a first ratio that is outside a range of fuel

concentrations between an upper stability limit and a lower stability limit, igniting the fuel, and producing a modified range of fuel concentrations defined by a modified upper stability limit and a modified lower stability limit of the fuel by applying an electric field across a flame supported by the fuel and air, wherein the first ratio is within the modified range.

According to an embodiment, a combustion system includes a burner configured to support a combustion reaction, first and second electrodes positioned and configured to apply an electric field across the combustion reaction supported by the burner, and a voltage supply, operatively coupled to the first and second electrodes, and configured to supply voltage signals to the first and second electrodes. A controller is configured to detect the combustion reaction at or near either of an upper stability limit or a lower stability limit of the fuel, and to control the voltage supply to supply voltage signals to the first and second electrodes sufficient to produce a modified upper stability limit and/or a modified lower stability limit of the fuel to extend stability of the combustion reaction.

According to an embodiment, a method for controlling a combustion reaction includes receiving, via a data interface, a command to establish a particular fuel stability limit, reading data corresponding to a fuel parameter (such as pressure and/or temperature, for example), and determining, as a function of the data, (e.g., algorithmically calculating or looking up) second data

corresponding to a signal selected to cause a particular voltage to be applied to a physical electrode operatively coupled to the combustion reaction. The second data is converted into the signal used to drive a voltage amplifier. The method further includes outputting from the voltage amplifier and transmitting to the electrode, electric current at the particular voltage. The applied electric field causes the fuel to combust at the particular stability limit responsive to exposure to the energized physical electrode.

According to an embodiment, a low NOx burner includes a physical flame holder configured to receive a particular fuel and oxidant mixture at a particular condition and an electrode configured to apply an electric field to the fuel and oxidant mixture, the electric field being selected to cause the fuel and oxidant to undergo a combustion reaction at the particular mixture. The fuel and oxidant are characterized by a leaner mixture than would react in combustion at the particular condition without being exposed to the electric field.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagram of a combustion system according to an embodiment. FIG. 2 is a flow chart depicting a method of operation of a combustion system, according to an embodiment. FIG. 3 is a diagram showing elements of a test system employed by the inventors to investigate and prove principles described and claimed herein, according to an embodiment.

FIG. 4 is a ternary mixture diagram depicting three fuels in experimental mixture space.

FIG. 5A-B are ternary diagrams illustrating the widening of flammability limits in the ternary hydrogen-methane-propane (H₂-CH -C3H₈) mixture space in the presence of an electric field.

FIG. 6A-C are ternary diagrams showing the contours for flammability limit changes determined by experiment.

FIG. 7 is a graph showing enhancement of rich flammability limits.

FIG. 8A-B includes ternary diagrams showing percent change contours for lean and rich flammability limits per equations. DETAILED DESCRIPTION

In the following detailed description, reference is made to the

accompanying drawings, which form a part hereof. In the drawings, similar symbols typically identify similar components, unless context dictates otherwise. Other embodiments may be used and/or other changes may be made without departing from the spirit or scope of the disclosure.

As used in the specification and claims, the term lower flammability limit (LFL) is used to refer to the leanest concentration of a given fuel in air that is combustible under industry standard measurement conditions, i.e., fuel and air at 25° C and at 1 bar of pressure, absolute. The term lower stability limit (LSL) is used to refer to the leanest concentration of a given fuel in air that is combustible under actual operating conditions for a given burner. Upper flammability limit (UFL) is used to refer to the richest concentration of a given fuel in air that is combustible, likewise under industry standard measurement conditions. Upper stability limit (USL) is used to refer to the richest concentration of a given fuel in air that is combustible under actual operating conditions of a given burner.

The terms modified lower flammability limit (MLFL) and modified upper flammability limit (MUFL), as well as the more general term modified flammability limit (MFL) are used to refer to the respective flammability limits as modified using the structures and/or methods disclosed hereafter. The terms modified lower stability limit (MLSL) and modified upper stability limit (MUSL), as well as the more general term modified stability limit (MSL) are used to refer to the respective stability limits as modified using the structures and/or methods disclosed hereafter.

Other terms that are related to, or synonymous with terms defined above may also be used hereafter, the meanings of which will be clear in view of the context.

It has been long understood in the art that flammability limits of any given fuel are substantially immutable. Tables setting forth the flammability limits of selected or common fuels can be found in many combustion engineering texts and general references. Such tables are relied upon by designers of combustion systems when calculating the parameters of individual system components to ensure that systems perform as intended.

The inventors have discovered that the flammability limits of many fuels can be modified by application of an electric field to a flame. The inventors have also discovered that the stability limits of actual combustion equipment can be modified by application of an electric field to a flame. Essentially, while an electric field is present across a flame, the LFL and UFL of the particular fuel or mix of fuels no longer represents the flammability limits of that fuel. Modified limits are imposed, in which the MLFL is lower, i.e., leaner that the LFL, while the MUFL is higher, i.e., richer than the UFL. Likewise, while an electric field is present across a flame, the LSL and USL of the particular fuel or mix of fuels no longer represents the stability limits of that fuel for a given burner. Modified limits are imposed, in which the MLSL is lower, i.e., leaner that the LSL, while the MUSL is higher, i.e., richer than the USL. The degree to which the flammability or stability limits are modified varies according to the fuel, and appears also to be a function of the strength of the electric field.

As previously noted, flamnnability limits are generally expressed as values corresponding to specific standard temperatures and pressures. Of course, as the temperature and/or pressure changes, so also will the flamnnability limits change. This is very well understood in the art. Many combustion systems are configured to operate at pressures and/or temperatures that are far removed from the nominal values associated with the standard tables. Furthermore, many combustion systems are not configured to operate using a mix of air and fuel, but instead employ other oxidizers, or further dilute the oxygen by introducing recirculated flue gas, etc. Nevertheless, the flamnnability limits of a given fuel can be calculated for any reasonable combination of temperature, pressure, and oxygen concentration. Thus, the applicable standard flamnnability limits are considered to apply, as revised to accommodate the varied conditions.

The inventors have found that experimental results related to upper and lower flamnnability limits and modified upper and lower flamnnability limits at standard conditions are predictive of upper and lower stability limits and modified upper and lower stability limits in given burners. FIG. 1 is a diagram showing a combustion system 100 which is configured to modify the flamnnability or stability limits of fuels employed, according to an embodiment. The combustion system 100 includes a burner 102 and a combustion control system 104. The burner 102 is configured to support a combustion reaction 106 and includes a fuel nozzle 108, an oxidizer conduit 1 10, a fuel source 1 12 and an oxidizer source 1 14. The fuel source 1 12 and oxidizer source 1 14 are coupled to the fuel nozzle 108 and oxidizer conduit 1 10, respectively, via corresponding transmission conduits 1 16. The fuel source 1 12 and oxidizer source 1 14 are each configured to be controllable to regulate the volume or flow rate of fuel and oxidizer, respectively.

Details of the fuel source 1 12 and oxidizer source 1 14 are not provided, inasmuch as such systems are very well known in the art, and can be arranged in any of a large number of configurations, depending upon factors such as, for example, the size, capacity, intended purpose, complexity, and projected duty cycle of the combustion system 100. Additionally, in many cases, the

combustion system may be subject to governmental regulations relating to emissions, which further affect the design of the system. According to an embodiment, the oxidizer source 1 14 can be configured, for example, to control oxygen input and/or air temperature by introducing recirculated flue gas.

According to an embodiment, the burner 102 is configured to premix fuel and air, then to emit the mixture from the nozzle 108. According to another embodiment, the fuel is ejected with some force from the nozzle 108 in a form that entrains air from the oxidizer conduit 1 10. According to an alternate embodiment, the oxidizer is forced into the combustion volume with a blower. According to some embodiments, only one of the fuel or oxidizer is regulated, the other being supplied at a substantially constant rate, etc. Other system designs and variations are well known and wholly within the abilities of one having ordinary skill in the art.

According to an alternative embodiment, the fuel and oxidizer are premixed to a selected mixture, and the combustion reaction is supported by the premixed fuel and oxidizer. In an embodiment, fuel and oxidizer are premixed to a selected mixture that falls outside the normal flammability or stability limits of the fuel. As will be appreciated, a mixture of fuel and oxidizer that are outside the flammability or stability limits is nominally non-combustible. Such a mixture can be considered a safe mixture because a flame will not propagate into a mixer containing the mixture. The mixture combusts only where and when the mixture is exposed to an appropriate electric field.

According to an embodiment, a mixture of fuel and oxidizer is ejected from the burner 102 at a ratio that is below the LFL or LSL of the fuel, i.e., too lean to maintain combustion. An electric field is applied, as described, for example, in more detail below, which produces a modified LFL or LSL that is below the current ratio, rendering the mixture combustible, but only while the electric field is present. Removal of the field again renders the mixture non-flammable. As the mixture flows from the burner, it can only become leaner, and thus cannot become flammable absent the electric field. A failsafe is thus provided, significantly reducing the danger of accidental combustion.

The combustion control system 104 includes first and second electrodes 1 18, 120, a sensor 122, a voltage source 124, and a control unit 128. The first electrode 1 18 is positioned downstream from the burner 102 adjacent to the combustion reaction 106. A surface of the burner nozzle 108 is adapted for operation as the second electrode 120. Connectors 130 couple the voltage source 124 to the first and second electrodes 1 18, 120, and the control unit 128 to the sensor 122, the voltage source 124, the fuel source 1 12, and the oxidizer source 1 14. The voltage source 124 is configured to apply a voltage difference across the combustion reaction 106 via the first and second electrodes 1 18, 120. The sensor 122 represents one or more individual sensors, each configured to measure one or more characteristics of the combustion reaction 106 and to supply corresponding signals to the control unit 128. For example, the sensor 122 can be configured to measure characteristics such as temperature, oxygen concentration, luminosity, combustion byproducts, electrical charge, etc.

Structures configured to electrically connect components or assemblies shown in the drawings are depicted generically as connectors 130, inasmuch as electrical connectors and corresponding structures are very well known in the art, and equivalent connections can be made using any of a very wide range of different types of structures. The connectors 130 can be configured to carry high-voltage signals, data, control logic, etc., and can include single conductors or multiple separately-insulated conductors. Additionally, where a voltage potential, control signal, feedback signal, etc., is transmitted via intervening circuits or structures, such as, for example, for the purpose of amplification, detection, modification, filtration, rectification, etc., such intervening structures are considered to be incorporated as part of the respective connector.

In operation, fuel is supplied and regulated to the fuel nozzle 108 by the fuel source 1 12, while the oxidizer is similarly supplied and regulated by the oxidizer source 1 14. The control unit 128 monitors selected parameters or characteristics of the combustion reaction 106, and controls the fuel source 1 12 and oxidizer source 1 14 to keep the selected characteristics within defined limits. Additionally, various aspects of the combustion reaction 106 can be controlled by application of electrical energy via the first and/or second electrodes 1 18, 120. Finally, when the control unit 128 receives sensor signals indicating that the combustion reaction 106 is operating near its upper or lower flammability or stability limit, the control unit 128 is configured to control other elements of the combustion system 100 to bring operation of the combustion reaction to a point further removed from the flammability or stability limits, or to modify the flammability or stability limits to be farther from the current point of operation.

For example, the control unit may detect that the combustion reaction 106 is unstable, or that it repeatedly blows out, requiring periodic re-ignition; or the control unit may determine that a leaner fuel/air mixture is required in order to obtain a selected emissions value, which results in producing an unstable flame, etc. Upon detection of operation at a fuel/air ratio that approaches the LFL or UFL of the particular fuel, the control unit 128 is configured to control the voltage source 124 to apply a voltage difference across the combustion reaction 106, via the first and second electrodes 1 18, 120, in order to establish an electric field across the flame. The magnitude of the voltage difference to be applied can be determined by reference to a lookup table or by calculation based on the fuel type, degree of desired modification of the flammability or stability limits, and/or on other predetermined factors. Alternatively, the control unit 128 can be configured to simply control the voltage source 1 10 to adjust the magnitude upward until signals from the sensor 122 indicate that the combustion reaction 106 is in stable operation.

FIG. 2 is a flow chart depicting a simple process 200 for controlling a combustion reaction, according to an embodiment. At 202, a fuel/air mixture is supplied to a combustion reaction. This will typically be in the form of a fuel and air mixture, but can include any appropriate oxidizer, and can include other known components that affect the fuel to oxygen ratio, including, for example, recirculated flue gas. At 204, a determination is made whether the fuel/air ratio of the mixture is outside the range defined by the LFL and UFL, or alternatively the LSL or USL. If the determination is made that the ratio is within the range (the NO path), the process returns to step 202, at which the fuel/air mixture continues to be supplied to the combustion reaction.

If, at step 204, it is determined that the fuel/air ratio is outside the range (the YES path) defined by the upper and lower flammability or stability limits of the particular fuel, the process proceeds to step 206, where the applicable flammability or stability limits are modified by application of an electrical field across the combustion reaction. Following step 206, the process returns to step 202 and repeats.

According to an embodiment, step 206 can include selecting an adequate magnitude of the electric field to modify the flammability or stability limits to a degree sufficient to encompass the current fuel/air ratio. According to an alternative embodiment, the magnitude of the electric field is increased

incrementally with each repetition of the cycle, so that multiple cycles of the process may be required before the flammability or stability limits have been sufficiently modified.

According to a further embodiment, the process can include a step in which a previously applied electric field is reduced incrementally each time the NO path is taken from step 204, or, alternatively, after a predetermined number of times that the NO path is taken. In this way, the strength or magnitude of the electric field is maintained near the minimum value necessary for proper operation, and is removed when no longer necessary.

FIG. 3 is a diagram showing elements of a test system 300 used by the inventors in experiments performed to demonstrate the principles described with reference to the previous embodiments. The test system 300 includes a burner 302, a fuel/air control system 304, and a combustion control system 306.

The burner 302 includes a sintered bronze plate 308, a cooling coil 310, and a plenum chamber 312 defined by a plenum chamber wall 313. The bronze plate 308 is porous, configured to permit fuel and air to pass from the plenum chamber 312 below the plate 308 to the upper side of the plate. A quartz cylinder 314 is positioned above and surrounding the bronze plate 308. The cooling coil 310 includes a coolant inlet 316 and a coolant outlet 318. During operation of the test system, water is pumped through the cooling coil 310 to control the temperature of the bronze plate 308 and to prevent transmission of heat from a combustion reaction 320 above the plate 308 to the plenum chamber 312, below. A fuel inlet 322 is provided to permit introduction of a fuel/air mixture into the plenum chamber 312.

The fuel/air control system 304 includes fuel sources A, B, and C, each configured to provide a respective type of fuel. An air source 324 is also provided. A respective valve/flow meter 326 is associated with each of the fuel sources A, B, and C, as well as with the air source 324. The air source 324 and each of the fuel sources A, B, and C are coupled, via their respective valve/flow meter, with a mixer 328, which is in turn coupled, via a master valve/flow meter 326e to the fuel inlet 322.

The combustion control system 306 includes a voltage supply 330 operatively coupled to a stainless steel mesh electrode 332 positioned above the quartz cylinder 314. The voltage supply 330 is also coupled to the bronze plate 308, and is configured to apply a voltage difference between the electrode 332 and the bronze plate 308.

During operation of the tests, the inventors controlled the valve/flow meters 326a-d to regulate the type and mixture of fuel, and the ratio of fuel to air. These elements were combined by the mixer 328, and the total volume of fuel/air mixture 334 introduced into the plenum chamber 312 was controlled by the master valve/flow meter 326e. The mixture 334 was ignited as it passed through the bronze plate 308, and the tests were conducted as described in detail below.

EXAMPLES

A modest electric field significantly widened flammability or stability limits (henceforth limits) for hydrogen-methane-propane (H₂-CH -C3H₈) fuel blends simulating refinery fuel gas. The lean limit was decreased by 2.7 to 5.9% while the rich linnit was increased by 5.6 to 14.1 %, giving an overall widening of the limits of 8.5 to 20.3%, depending on the fuel.

The rich limit correlates well with the square root of the H/C molar ratio in the fuel (i² = 99.0%). Enhancement of the rich limit may come from better transport of the oxidizing species from the air side. One possibility is the attraction of H₃O⁺ ions from the air side to the grounded nozzle (fuel side) to provide additional OH to the flame. This, in turn, would enhance CO oxidation: CO + OH = CO2 + H, and thus the overall reaction rate of the hydrocarbons.

The change in lean limit is about half of that of the rich limit. Possibly, this is because CO is present in much lower concentration in lean flames, therefore, the enhancement of CO oxidation is presumably less beneficial. The change in the lean limit also correlates negatively with the C3H8 concentration (r² = 99.0%) and to a much lesser extent with the addition of H₂ concentration (or positively to CH₄ addition) to the model (1² = 99.98%).

As to propane's effect, one possibility is that CsH₃ ⁺ impeded oxidation at the lean limit (see Section 5). Since C3H₃ ⁺ correlates with C3H8 concentration, this effect would be more pronounced with propane-rich fuels. The reason that an increase in the hydrogen concentration slightly reduces the relative widening of the lean limit is less clear. One possibility is that at the lean limit H₂ scavenges OH generated by H₃O⁺. Another possibility is CH is actually the important moiety (and as C3H8 increases, H₂ must, of mathematical necessity, decrease in a ternary mixture).

Flammability and stability limits were found to be significantly enhanced in an electric field.

1. Apparatus

The apparatus included a 51 mm diameter sintered bronze disk through which a premixed fuel and air mixture flowed. An added quartz tube atop the burner isolated the flame from the surroundings. The quartz tube had an inner diameter of 56 mm. A circular stainless steel screen electrode was positioned 8 mm above the quartz tube. An electric potential between the screen electrode and the flame was maintained at 10 kV, generating an electric field of 1 .2 kV/cm. Cooling water was used to stabilize the flame.

A circular stainless steel screen electrode was placed above the quartz tube resting on an electrically grounded burner. Hydrogen, methane, propane, and air were individually fed through four OMEGA(TM) thermal flow meters (not shown, available from Omega Engineering, Inc., Stamford CT, USA) to generate the desired fuel flow rates. Each meter was electrically floated with an inverter and a battery so as to keep the meters from being shocked by high voltage. The gases were then blended and controlled by valves downstream of the fuel blend but upstream of the burner (not shown). The valve panel was grounded for safety. The fuels were fed from a tank farm; air was fed from an air compressor. Ambient humidity was neglected in the calculations (and constituted about 1 % of the air by volume). 2. Procedure

To investigate the fuel lean and rich conditions, the fuel feed was held constant while increasing or decreasing the airflow until reaching blowout. In the case of the upper limit, the air was decreased so as to make the mixture more fuel rich. In the case of the lower limit, the air was increased so as to make the mixture more fuel lean. Blowout was defined as the condition where the flame blew completely out of the quartz tube. The blowout limit was determined without charge under constant fuel flow before testing with charge. While the burner can withstand higher flows, the airflow meter had a maximum 50 SLPM capacity. Thus, the airflow was set to 80% of maximum (to give some margin for leaner limits under an electric field) and then decreased the fuel to give the maximum lean condition. The airflow was then reduced, the burner was relit, and the flame was charged, increasing the airflow until blowout was achieved. In all cases, leaner limits were achieved under electrically charged conditions. For the rich limit, the airflow was reduced little by little until the flame blew out. This procedure was then repeated under charged conditions. In all cases, the charged condition gave richer flammability limits. 3. Data and Results

Table 1 shows properties of the tested fuels and mixtures. They represent fuel mixtures comprising a maximum of 50% hydrogen. In Table 1 H/C is the molar hydrogen to carbon ratio in the fuel, LHV is the lower heating value in

BTU/scf, AFT is the adiabatic flame temperature, Φ is the stoichiometric fuel/air ratio, τ is the theoretical air/fuel ratio (defined at stoichiometric ratio Φ = 1 ), %

Fuel is the volume fuel concentration in the air fuel mixture, LL is the lean flamnnability or stability limit in terms of the percent of fuel in the fuel-air mixture, and RL is the rich flamnnability or stability limit expressed on the same basis. In Table 1 , no electric field has been applied and the rich and lean flamnnability limits represent native fuel properties as measured by the apparatus. Such limits show good general agreement with literature values for flamnnability limits despite literature values being measured in more exacting apparatuses.

Table 1, Fuel and Mixture properties

Fuel Fuel/Air Mixture Properties

Composition Properties Φ = 1 Flammability

Code H₂ CH₄ C3H8 H/C LHV AFT τ %Fuel LL, % RL, %

0-100-0 0 100 0 4.00 912 2223 9.52 9.50 5.2 14.4

0-0-100 0 0 100 2.67 2385 2265 23.81 4.03 2.4 8.7

0-50-50 0 50 50 3.00 1649 2244 16.67 5.66 3.1 10.9

50-50-0 50 50 0 6.00 593 2346 5.95 14.38 4.8 19.9

50-0-50 50 0 50 3.33 1330 2369 13.10 7.09 2.8 13.1

25-50-25 25 50 25 3.60 1121 2294 11.31 8.12 3.7 13.9

FIG. 4 is a ternary mixture diagram depicting the fuels in experimental mixture space. The circles show the experimental blends that were investigated, mapping the white portion of FIG. 4. Hydrogen blends above 50% were not investigated (shaded gray). Table 2 below contrasts fuel properties with and without the presence of a modes electric field (1 .2 kV/cm). Table 2, Flammability Limits With and Without an Electric Field

From inspection, one sees that the LL and RL differ for the electric field off (0 kV/cm) and on (1 .2 kV/cm); for example, the lean flammability limit for

methane (first row) decreased from 5.1 % methane to 4.8% and the rich

flammability limit increased from 14.4 to 15.8%. In general, the effect of the

electric field was to widen the flammability limits for all fuels and blends with the lean limit becoming leaner and the rich limit becoming richer. In order to

compare lean and rich limits, we defined a change parameter (χ_Γ, χ/), per

Equation 1 .

where χ χ/ are the fractional change of the rich and lean limits, respectively,

,e, ,e is the limit (rich or lean, respectively) expressed as the fuel fraction in the presence of the electric field, and λ_Γ, λ/ is the normal flammability/stability limit

(no electric field).

Under this definition, widening of the lean limit is negative while widening of the rich limit is positive. To calculate a total widening over the entire range we modified Equation (1 ) as follows.

(2) λ,— λ,

where χ_τ is the total fraction of change in the flammability or stability range (rich - lean) with and without electrical charge, _r,_e is the rich limit in the presence of the electric field, λ_Γ is the rich limit in the absence of the electric field, __e is the lean limit in the presence of the electric field, and λ/ is the lean limit in the absence of the electric field.

FIGS. 5A-B include two ternary diagrams illustrating the widening of

flannnnability limits in the ternary H₂-CH -C3H₈ mixture space in the presence of an electric field. A modest electric field significantly widened flannnnability limits.

The ternary diagram (FIG. 5A shows the normal (uncharged) flannnnability limits that were measured at each point in the mixtures space. The ternary diagram

FIG. 5B shows a wider flannnnability region in the presence of an electric field with the lean flannnnability limit becoming leaner and the rich flannnnability limit

becoming richer.

4. Analysis

Contour lines for six fuel blends can be fit exactly with a mixture model of the form

3 2 3

y =∑¾¾ +∑∑¾^■¾ (³) k=l j<k k=l

where y is the response of interest (e.g., lean or rich flannnnability limit, etc.), ^',/ are indexes for the three fuel components (1 < j<k < 2; 1 < k≤ 3), Zj,z_k are the

3

fuel components i.e.,

= H₂, z₂ = CH₄, z₃ = C3H8.∑¾ = 1 , because the fuel

k=l

components must sum to 100%, Ck, Cjk are the associated coefficients for the

pure components and blends (ci , c₂, C3, and Ci₂, C13, c₂3, respectively).

Note that Equation (3) contains no error term, nor is it possible to deduce one as six mixture points will determine six coefficients with perfect certainty.

Thus, Equation (3) fits the data exactly, whether or not they contain errors.

FIGS. 6A-C include three ternary diagrams showing the contours for

flannnnability limit changes deduced in this way. The series shows percent

changes in flannnnability or stability limits in ternary mixture coordinates. The

diamonds indicate the data points with actual percent changes indicated above each point. In FIG. 6A, the lines and negative values indicate the percent

change in lean limits. In FIG. 6B, the lines and positive values indicate the

percent change in rich limits. FIG. 6C gives the contours for total percent change over the entire range. Overall, a 1 .2 kV/cm electric field widened limits between 8.5 and 20.3%, depending on the fuel. Because of the way these contours were derived they are purely empirical and agree exactly with the values of the data set.

In order to perform statistical tests the number of coefficients must be less than the number of fuel blends tested; that is, fewer than six and preferably only two or three. Equations (4) and (5) fit the bill for the rich and lean change fractions, respectively. χ = α₀ + α₁χ + ε

Where χ_Γ , χ_/ are the fractional change in the rich and lean limits, respectively, x is the H/C ratio of the fuel, zi is the fraction of H₂ in the fuel, z₃ is the fraction of C3H8 in the fuel, ao-2, 60-33 are the respective coefficients, and ε is the error term.

The rich limits were found to be a function of H/C ratio alone. The model has the following statistics.

Table 3, ANOVA for Equation (4)

Source DF SS MS F p

Model 1 16,373 16373 381 .3 < 0001

Residual 4 172 42.9 1² 0.990

Total 5 16,5451 361 .4 r_p ² 0.978 Table 3 shows the analysis of variance (ANOVA) for equation (4). The model contains 1 degree of freedom (DF), leaving 4 DF to estimate the error, and thus comprising a total of 5 degrees of freedom. In actuality, the model may be said to contain 2 degrees of freedom: a₀, and av, however, if the model were not significant - termed the null hypothesis - then all data points are replicates and would be best expressed by a mean value. Since the null hypothesis contains 1 degree of freedom (the mean) it is subtracted from the degrees of freedom of the model to give a net 2 degrees of freedom, which is what is reported in the analysis of variance table. Furthermore, it may be said that there are actually 6 DF for the model corresponding to the six data points collected. However, if the model was not significant, one would average all six values with a single mean. Since the mean represents 1 degree of freedom, the net degrees of freedom is actually 5.

Table 3 entries have the following meanings: the sum of squares (SS) column shows the variance proportional to each respective source of variance. If the model were to fit the data exactly, it would be equal to the Total SS with zero Residual SS. The mean square (MS) column is derived by dividing the SS column by the DF column, excluding the bottom number which will be discussed later. The ratio of Model MS with the Residual MS gives an F ratio - in this case of 381 .3. If the model were no better at explaining the variance than chance, F would be ~1 . Since 381 .3 » 1 , the model is statistically differs from chance deviation. The probability, p, that the F ratio is significant is given by the value p < .0001 . Thus, the F ratio of 381 .3 is estimated to occur by chance less than 1 time in 10,000. In other words, the model is statistically significant with >99.99% certainty (1 - 0.0001 = 0.9999). Generally, a model is considered statistically significant if p < 0.05, which is the case here, r² is the ratio of the Model SS / Total SS. If the model fits the data exactly the r² = 1 and the model explains 100% of the data variation. In the present case, r² = 0.990; that is, the model explained 99.0% of the total variation. The bottom number is the PRESS statistic (predicted sum of squares). It is not derived from the ANOVA, but may be used to make an inference about the predictive power of the model (as opposed to the correlative power of the model given by r²). A predictive estimate, r_p ² , was calculated by subtracting from 1 the ratio of PRESS / Total SS. In this case r_p ² = 0.978 and infers that model is likely to have good predictive power. With knowledge that the model is statistically significant, the data was examined further to derive coefficient estimates for each of the model terms (Table 4). Table 4, Statistics for Equation (4)

Term Est Std Err t ratio P

a₀ -90.28 9.67 -9.34 0.0007 ai 48.178 2.47 19.53 < 0001

The Table 4 entries have the following meaning: The Term shows the respective coefficients for Equation (4). The estimate column (Est) gives the least squares value for each coefficient. The standard error (Std Err) gives the uncertainty of the associated coefficient. For example, ao is estimated to be - 90.28 ± 9.67. Thus the estimate was many times larger than the standard error and likely to be statistically significant. The t ratio column is the estimate divided by the standard error. One prefers to see |t| » 1 as is the case here. The p value gives the probability that a particular t ratio may occur by chance. For ao, the p value is 0.0007, meaning there is merely a 0.07% probability that a t ratio of -9.34 may occur by chance. In general we reject the null hypothesis if p < 0.05, as is the case here.

FIG. 7 is a graph showing that the enhancement of the rich limits correlated well with square root of the H/C ratio (Equation 4).

Tables 5 and 6 give the associated statistics for Equation 5

Table 5, ANOVA for Equation (5)

Source DF SS MS F P

Model 3 8.7047 2.9016 3243.4 0.0003

Residual 2 0.0018 0.0009 r² 0.9998

Total 5 8.7065 0.1553 r ² 0.9822 Table 6, Statistics for Equation (5)

Term Est Std Err t ratio P VIF bo -5.449 0.0264 -206.69 < 0001

bi 0.596 0.0597 9.97 0.0099 1 .20

bz 4.013 0.0416 96.4 0.0001 1 .39

633 -3.120 0.1099 -28.4 0.0012 1 .44

All of the model coefficients are statistically significant at p < 0.05, both the correlation coefficient and the predictive coefficient are very near 1 . The variance inflation factor (VIF) characterizes the correlation among factor variables. If there is no correlation (desired) then VIF = 1 . The relationship between VIF and r² is r² = 1 - 1/VIF. Thus, VIFs of 1 .20, 1 .39, and 1 .44 correspond respectively to fs of 0.167, 0.281 , and 0.306 - all quite meager, meaning that the factor disposition is well dispersed in mixture space with little collinearity. Equation (5) generates the contours shown in FIGS. 8A-B.

FIGS. 8A-B include two ternary diagrams showing the percent change contours for lean and rich flannnnability limits per equations (5) and (4). FIG. 8A shows the percent decrease in the lean limits as a function of mixture fraction based on Equation (5). FIG. 8B shows the increase in rich limits according to Equation (4). In general, the contours compare well to the exact contours (FIG. 6).

5. Conclusions

Flannnnability limits can be defined as "the state at which steady

propagation of the planar premixed flame... fails to be possible." It has been long known that increased temperature widens flannnnability limits. Increased pressure also widens flannnnability limits because it increases the fuel and oxygen concentrations. Increases in oxygen concentration widen flannnnability limits, particularly on the rich side because the additional oxygen can react under rich conditions whereas on the lean side oxygen is not the limiting reagent. If an electric field enhances the fuel concentration at the lean side or the oxidant concentration on the rich side, then it would likewise widen the overall

flammability limit.

Two facts are now apparent. The rich limit widens in direct proportion to the square root of the H/C molar ratio. The lean limit widens in opposition to the C3H8 concentration and to a much lesser extent with the hydrogen concentration.

While various aspects and embodiments have been disclosed herein, other aspects and embodiments are contemplated. The various aspects and embodiments disclosed herein are for purposes of illustration and are not intended to be limiting, with the true scope and spirit being indicated by the following claims.

Claims

CLAIMS What is claimed is:

1 . A method, comprising:

introducing fuel and air into a combustion volume in a first ratio that is inside a first range defined by an upper stability limit and a lower stability limit of the fuel;

igniting the fuel;

producing a modified range defined by a modified upper stability limit and a modified lower stability limit of the fuel by applying an electric field across a flame supported by the fuel and air; and

after igniting the fuel, and after producing the modified range, introducing fuel and air into the combustion volume in a second ratio that falls within the modified range but does not fall within the first range.

2. The method of claim 1 , comprising:

prior to producing the modified range, monitoring one or more

characteristics of the flame supported by the fuel and air; and

detecting values of the monitored characteristics that indicate a requirement that the ratio of fuel and air be adjusted from the first ratio to the second ratio, wherein producing the modified range is performed in response to detecting values of the monitored characteristic.

3. The method of claim 1 wherein applying the electric field across the flame modifies an upper flammability limit or a lower flammability limit of the flame.

4. A combustion system, comprising:

a burner configured to support a combustion reaction by emitting fuel and oxidizer;

first and second electrodes positioned and configured to apply an electric field across a combustion reaction supported by the burner; and a voltage supply operatively coupled to the first and second electrodes and configured to supply voltage signals to the first and second electrodes; and a controller configured to detect operation of the combustion reaction at or near either of an upper stability limit and a lower stability limit of the fuel, and to control the voltage supply to supply voltage signals to the first and second electrodes sufficient to produce a modified upper stability limit and a modified lower stability limit of the fuel.

5. The system of claim 4, wherein the first electrode is a nozzle, or a portion of a nozzle, of the burner.

6. The system of claim 4, wherein the second electrode is a mesh electrode positioned above the burner

7. The system of claim 4 wherein the controller is configured to control the voltage supply to supply voltage signals to the first and second electrodes sufficient to produce a modified upper flammability limit and a modified lower flammability limit of the fuel.

8. A method for controlling a combustion reaction, comprising:

receiving, via a data interface, a command to establish a particular fuel stability limit;

reading data corresponding to a fuel parameter;

determining, as a function of the data, a particular voltage to be applied to a physical electrode operatively coupled to a combustion reaction;

applying the particular voltage to the physical electrode; and

causing the fuel to combust at the particular stability limit responsive to an electric field generated in part by application of the particular voltage to the physical electrode.

9. The method for controlling a combustion reaction of claim 8 wherein determining the particular voltage includes determining second data

corresponding to the particular voltage.

10. The method for controlling a combustion reaction of claim 9 wherein determining the particular voltage includes

converting the second data into the signal;

driving a voltage amplifier with the signal; and

outputting from the voltage amplifier and transmitting to the electrode, the particular voltage.

1 1 . The method for controlling a combustion reaction of claim 9, wherein determining the second data includes algorithmically calculating the second data.

12. The method for controlling a combustion reaction of claim 9, wherein determining the second data includes looking up the second data.

13. The method for controlling a combustion reaction of claim 8, wherein the fuel parameter includes ambient pressure.

14. The method for controlling a combustion reaction of claim 8, wherein the fuel parameter includes ignition temperature.

15. A low NOx burner, comprising:

a physical flame holder configured to receive a particular fuel and oxidant mixture at a particular condition; and

a first and second electrode configured to apply an electric field to the fuel and oxidant mixture, wherein the fuel and oxidant are characterized by a leaner mixture than would undergo the combustion reaction at the particular condition without being exposed to the electric field.

16. The burner of claim 15, wherein the particular condition is a temperature proximate to the physical flame holder.

17. The burner of claim 15, wherein the particular condition is atmospheric pressure proximate to the physical flame holder.

18. A method, comprising:

introducing fuel and air into a combustion volume in a first ratio that is inside a first range defined by an upper flannnnability limit and a lower flannnnability limit of the fuel;

igniting the fuel;

producing a modified range defined by a modified upper flannnnability limit and a modified lower flannnnability limit of the fuel by applying an electric field across a flame supported by the fuel and air; and

after igniting the fuel, and after producing the modified range, adjusting a value of the ratio of fuel and air from the first ratio to a second ratio that falls within the modified range but does not fall within the first range.

19. The method of claim 18, comprising:

prior to producing the modified range, monitoring one or more

characteristics of the flame supported by the fuel and air; and

detecting values of the monitored characteristics that indicate a

requirement that the ratio of fuel and air be adjusted from the first ratio to the second ratio, wherein producing the modified range is performed in response to detecting values of the monitored characteristic.

20. A combustion system, comprising:

first and second electrodes positioned and configured to apply an electric field across a combustion reaction supported by the burner; and a voltage supply operatively coupled to the first and second electrodes and configured to supply voltage signals to the first and second electrodes; and a controller configured to detect operation of the combustion reaction at or near either of an upper flamnnability limit and a lower flamnnability limit of of the fuel, and to control the voltage supply to supply voltage signals to the first and second electrodes sufficient to produce a modified upper flamnnability limit and a modified lower flamnnability limit of the fuel.

21 . A method for controlling a combustion reaction, comprising:

receiving, via a data interface, a command to establish a particular fuel flamnnability limit;

reading data corresponding to a fuel parameter;

applying the particular voltage to the physical electrode; and

causing the fuel to combust at the particular flamnnability limit responsive to an electric field generated in part by application of the particular voltage to the physical electrode.

22. The method for controlling a combustion reaction of claim 21 , wherein determining the particular voltage includes determining second data

corresponding to the particular voltage.

23. The method for controlling a combustion reaction of claim 22, wherein applying the particular voltage includes

converting the second data into the signal;

driving a voltage amplifier with the signal; and

24. The method for controlling a combustion reaction of claim 22, wherein determining the second data includes algorithnnically calculating the second data.

25. The method for controlling a combustion reaction of claim 22, wherein determining the second data includes looking up the second data.

26. The method for controlling a combustion reaction of claim 21 , wherein the fuel parameter includes ambient pressure.

27. The method for controlling a combustion reaction of claim 21 , wherein the fuel parameter includes ignition temperature.

28. A method, comprising:

introducing fuel and air into a combustion volume in a first ratio that is outside a range defined by an upper stability limit and a lower stability limit of the fuel;

producing a modified range defined by a modified upper stability limit and a modified lower stability limit of the fuel by applying an electric field across the fuel and air, the first ratio falling within the modified range; and

after producing the modified range, igniting the fuel.

29. The method of claim 28, comprising:

prior to producing the modified range, monitoring one or more

characteristics of the fuel and air;

prior to producing the modified range, detecting values of the monitored characteristics that indicate that the first ratio of the fuel and air is at or outside the upper stability limit or the lower stability limit; and

producing the modified range after the monitoring and the detecting.

30. The method of claim 28, wherein producing a modified range includes: determining a magnitude of the electric field to be applied, based in part on a value of the first ratio and a value of one of the upper stability limit or the lower stability limit; and

applying the electric field at the determined magnitude.

31 . The method of claim 29, wherein the producing a modified range includes: applying the electric field at a preselected magnitude;

repeating the monitoring and the detecting; and

increasing the magnitude of the applied electric field by a preselected increment.

32. A method, comprising:

introducing fuel and air into a combustion volume in a first ratio that is outside a range defined by an upper flannnnability limit and a lower flannnnability limit of the fuel;

producing a modified range defined by a modified upper flannnnability limit and a modified lower flannnnability limit of the fuel by applying an electric field across the fuel and air, the first ratio falling within the modified range; and

after producing the modified range, igniting the fuel.

33. The method of claim 32, comprising:

prior to producing the modified range, monitoring one or more

characteristics of the fuel and air;

prior to producing the modified range, detecting values of the monitored characteristics that indicate that the first ratio of the fuel and air is at or outside the upper flannnnability limit or the lower flannnnability limit; and

producing the modified range after the monitoring and the detecting.

34. The method of claim 33, wherein the producing a modified range includes: applying the electric field at a preselected magnitude;

repeating the monitoring and the detecting; increasing the magnitude of the applied electric field by a preselected increment.

35. The method of claim 32, wherein producing a modified range includes: determining a magnitude of the electric field to be applied, based in part on a value of the first ratio and a value of one of the upper flamnnability limit or the lower flamnnability limit; and

applying the electric field at the determined magnitude.