WO1996004508A1

WO1996004508A1 - REDUCTION OF NOx EMISSIONS IN FLUIDIZED BED INCINERATION

Info

Publication number: WO1996004508A1
Application number: PCT/US1995/000438
Authority: WO
Inventors: Lewis W. Clark; Richard L. Samu
Original assignee: Dorr Oliver Inc
Current assignee: Dorr Oliver Inc
Priority date: 1994-08-03
Filing date: 1995-01-17
Publication date: 1996-02-15
Anticipated expiration: 1997-02-03

Abstract

In a process for incinerating a waste water sludge (24) in a fluidized bed reactor (10) of the bubbling bed type, urea in granule form or in solution is injected directly into the bubbling fluidized bed of the reactor (10) through conduit (41) thereby effecting a reduction of NOx emissions by at least 25 %.

Description

DESCRIPTION

REDUCTION OF NO_x EMISSIONS IN FLUIDIZED BED INCINERATION

Technical Field

The invention is directed to the reduction of NO_x levels in the flue gas from fluidized bed incinerators, particularly fluidized bed incinerators of the bubbling bed type.

Background Art

Fluidized bed reactors are well-known means for generating heat and, in various forms, can carry out the processes of drying, roasting, calcining, incineration and heat treatment of solids in the chemical, metallurgical and other material processing fields. They are also used for the generation of hot gases, including steam, for use in driving electric power generation equipment, for process heat, for space heating, or for other purposes.

Fluidized bed reactors typically comprise a vessel having a substantially horizontal air distributor or constriction plate, which supports a bed of particulate solids in the reaction chamber and separates the reaction chamber from a windbox below the air distributor. Combustion air is introduced into the windbox and passes through the air distributor in sufficient volume to achieve a gas velocity that expands or fluidizes the solids bed, suspending the particulate solids of the bed in the flowing air stream and imparting to the individual particles a continuous random motion. A fluidized bed in appearance and properties resembles a boiling liquid. Conducting a combustion reaction in a fluidized bed has important advantages which include attainment of a substantially uniform bed temperature, combustion at relatively low temperatures, say 1350°-l 700°F, and a high heat transfer rate. Nitrogen oxides are generated when fuel is burned, both from thermal fixation of nitrogen in the air and from conversion of nitrogen present in the fuel. The former reaction is favored at high temperatures (above 1800°F) while the latter occurs at all temperatures. There is much concern with minimizing NO_x emissions in combustion systems due to their poisonous effect at low exposure levels and to their involvement in the formation of photochemical fog. The problem of NO_x reduction in flue gas has hitherto been addressed by injection of ammonia into the effluent stream with and without catalysts. U.S. Patent No. 3,900,554 suggests non-catalytic removal of NO_x from flue gases by injecting ammonia into a gas stream at a temperature in the range 1600°-2000°F. Reduction of NO_x in flue gases using ammonia with various catalysts is suggested in U.S. Patent Nos. 3,887,683; 4,056,600; 4,010,238; 4,049,777 and 4,070,440. The injection of ammonia or an ammonia-producing precursor directly into the fluidized bed combustion region of a furnace is suggested in U.S. Patent No. 4,181,705. U.S. Patent No. 4,208,386 is directed to a process in which an effluent stream is contacted with urea (as a solid or in solution) at a temperature in the range 1600°-2000°F to reduce NO_x.

U.S. Patent 4,719,092 calls for reducing the concentration of NO_x in oxygen-rich combustion effluents by injecting an aqueous solution of urea and an oxygenated hydrocarbon into the effluent at a temperature above 1600°F.

U.S. Patent No. 4,756,890 suggests injection of ammonia or an ammonia precursor into the gas stream within a high temperature cyclone separator at a location where there is a strong vortex region to obtain an efficient mixing of the NO_x reducing agent and the combustion product flue gas. This arrangement is particularly suitable for those fluidized bed reactors of the circulating bed type where large volumes of gas and solids are returned to the combustion reactor through a cyclone. It is not suitable for fluidized bed reactors of the bubbling bed type serving as incinerators because the incinerators generally do not employ cyclone separators. There is continuing interest in providing more convenient and effective methods for introducing NO_x reducing compounds into incinerators.

Disclosure of the Invention

It has been found in accordance with this invention that injection of urea as dry particulate matter or in solution directly into, the bubbling bed region of a fluidized bed incinerator substantially reduces NO_x emissions in the flue gas of the reactor.

This result is contrary to the effect produced by the introduction of urea into the sludge feed before that feed is forwarded to the fluidized bed region of the incinerator. Brief Description of the Drawings

Figure 1 is an elevational view, partially in section, of a fluidized bed incinerator of the bubbling bed type in which the process of the invention can be carried out. Figure 2 is a graph in which the ppm of NO_x generated is plotted as the flow of urea granules or prill in pounds per minute is increased over time.

Figure 3 is a graph in which ppm of NO_x generated is plotted as the flow of urea solution in gallons per minute is increased over time.

Modes for Carrying Out the Invention

In a series of NO_x emissions tests on an existing bubbling bed fluidized bed incinerator, parameters known to influence NO_x emissions were evaluated. These parameters included temperature, feed moisture content, excess air levels (% oxygen in the exhaust) and the form of nitrogen in the feed as nitrate/nitrite, i.e. calcium nitrate, Ca(NC>3)2 and as an organic, i.e. urea, CO(NH2)2- Additions of urea and calcium nitrate granules were made directly into the waste water sludge at a feed pump and sufficient time was allowed for blending and dissolving the additions into the sludge. The waste water sludge has a consistency resembling a wet peat moss or damp soil and pumping of this semi-solid (about 12% by weight solids) is accomplished using a progressive cavity type pump. The additions of urea and calcium nitrate were sprinkled into the sludge as it reached the pump. The treated sludge moved to the incinerator through a pipe about 100 feet long, taking about one hour in travel time to reach the incinerator. As indicated in Table 1, both additions, (i.e., of urea and calcium nitrate), added in this manner, greatly increased NO_x emissions from the incinerator.

It is now believed that the introduction of urea into the sludge feed, followed by transportation of the sludge mixture over a substantial distance to the incinerator through a conduit which consumed an hour or more and provided conditions in which the urea was well-blended with the sludge, created, upon injection into the bubbling bed fluidized incinerator, a localized reducing atmosphere within the bubbling bed which acted to inhibit the effectiveness of urea in reducing NO_x emissions.

It was thought that results similar to those set forth in Table 1 would be obtained by the convenient method of injecting the dry compounds (urea and calcium nitrate) directly into the fluidized bed combustion region of the incinerator using an eductor to draw dry granules from a bag. The results obtained by direct injection of the dry granules are shown in Table 2, and it is seen that, surprisingly, NO_x reductions of 30% and higher were achieved with injection of dry urea, whereas injection of calcium nitrate increased the NO_x emissions in accordance with expectations. It should be noted that where a 50-50 blend by weight of urea and calcium nitrate was injected directly into the bubbling bed region of the fluidized bed incinerator, the NO_x emissions increased, but to a significantly less extent than the increase when calcium nitrate was alone injected.

ln confirmation of the above favorable results obtained in NO_x reduction of incinerator emissions, laboratory tests are made on a paper mill de-inking sludge subjected to incineration in a fluidized bed test reactor. The average freeboard temperature in the reactor is 1600°F to 1750°F and the NO_x content of the emissions before treatment amounts to 167 ppm and higher. Urea in dry granule form is injected directly into the bubbling bed of the fluidized bed incinerator. The results of three test runs made are set forth in the following Table 3.

TABLE 3: BUBBLING BED UREA ADDITION TO DE-INKING

SLUDGE

NOTE: PPM NOχ REPORTED ON A DRY VOLUME BASIS CORRECTED TO 7% 0 .

ADDITION TYPE RATE K urea PPM NO_v %NOy.

(g/min) g N feed) Before After Reduction

UREA DRY 3.75 0.4 167 117 30

UREA DRY 3.75 0.4 246 155

³⁷

UREA DRY 1.94 0.2 167 137 18 |

As can be seen from the above table, very substantial reductions in NO_x emissions are obtained in these confirmatory tests. It should be noted that in the third test run of this series, the stoichiometry of the urea/nitrogen feed reaction is very closely approached with less than optimum NO_x reduction achieved. In general, some excess urea is made available for reaction to obtain maximum reduction of NO_x.

While initially it was concluded that the quite significant reduction in NO_x emissions achieved as described above was due to the form of the urea addition (dry rather than wet urea), still further surprising results awaited subsequent experiments in which solutions of urea were injected directly into the bubbling bed of the fluidized bed reactor. Thus, it was determined that, in employing urea to reduce NO_x emissions from fluidized bed reactors of the bubbling bed type, it is, above all, essential to inject the urea whether in the dry form or in the form of a solution, directly into the bubbling bed.

In Bed Injection of Urea

In-bed injection processes employing both dry urea and solutions of urea were further objects of this study. The objective was to determine the effectiveness of the two methods by quantifying NO_x reduction levels as a function of urea feed requirements (i.e. NSR values). NSR (normal stoichiometric ratio) is defined as the actual mole ratio of urea to NO_x formed divided by the theoretical stoichiometric ratio, which is 0.5 for the reaction between urea and NO_x. Therefore, an NSR of 1 means that the actual mole ratio of urea to NO_x equals the theoretical stoichiometric ratio of 0.5 (0% excess urea). An NSR of 2 means that urea is being fed at 100% excess, NSR of 3 implies 200% excess urea, etc.

Injection of Dry Urea Further dry urea in-bed testing was performed. A baseline NO_x and % O2 were established prior to injection of urea and then dry urea was injected directly into the bubbling bed via an oil gun. Straight urea prill was discharged from a screw feeder and drawn into the bed via an eductor. A single feed point independent of the sludge feed point was used. The results using natural gas as auxiliary fuel are presented in Figure 2.

The dry urea was continuously introduced into the bubbling bed during the test runs. Baseline NO_x levels were established at the beginning and end of the complete test run. Figure 2 indicates that as the rate of urea introduction into the bed (Ib/min) increased, the ppm NO_x levels decreased. Table 4 summarizes the results of the dry urea in-bed testing. The data indicate that as the level of dry urea introduced into the bed increased (i.e. Ib/min, NSR) the % NO_x reduction increased.

TABLE 4: NO_x REDUCTION USING DRY UREA IN-BED INJECTION PROCESS

NO_x NO_x

Run Dry FB Temp. Bed Temp. %0₂ (ppm) (ppm) %NO_x No. Urea (Deg. F) (Deg. F) (Dry) Before After Reduction NSR

(Ib/min)

1 0.6 1540 1500 1 1.9 220 173 21 5.7

2 1.2 1552 1510 1 1.8 220 141 36 11.3

3 2.2 1569 1507 11.4 220 102 54 20.8

4 3.1 1571 1499 1 1.4 220 72 67 29.2

5 4.0 1567 1483 1 1.7 220 63 71 37.7

NOTES: 1. All NO_x values presented are on dry volume basis and have been corrected to 7% 02.

2. For all Runs natural gas was used as auxiliary fuel. The tests were repeated using No. 2 oil as auxiliary fuel without significantly different results.

Direct Injection of Urea Solution

A 15% (by wt) solution of urea in water was prepared and introduced into a bed-directed oil gun by a positive displacement chemical metering pump. A single feed point independent of the sludge feed point was used. The sludge feed was a primary/secondary blend (61/39 dry weight basis) . In general, primary sludge differs from secondary sludge in that primary sludge is essentially a raw sludge whereas secondary sludge is a waste-activated sludge. Tests were conducted using either natural gas or No. 2 oil as auxiliary fuel. Once a baseline

NO_x and % O2 were established, the 15% urea solution was introduced into the bed at approximately 1.0 gpm for approximately 20 minutes. Urea feed to the combustor then ceased momentarily to adjust the urea flowrate to 1.5 gpm and re-establish the baseline NO_x. After pumping at 1.5 gpm, the urea flowrate was increased to 2.0 gpm, and then shut-off to once again return to baseline. The results of the urea solution tests using natural gas as auxiliary fuel are presented in

Figure 3. In both cases, as the flowrate of urea injection increased from 1.0 to 2.0 gpm, the ppm NO_x levels decreased.

A 30% (by wt) solution of urea in water was also prepared and injected into the bed. The conditions were the same as the 15% solution test runs except decreasing flowrates of 0.5, 1.0, and 1.5 gpm were used. The results obtained using natural gas as auxiliary fuel are presented in Table 5. As the flowrate decreased from 1.5 to 0.5 gpm, the ppm NO_x levels increased.

Results obtained using No. 2 oil as auxiliary fuel were not significantly different from those obtained with natural gas.

Table 5 below summarizes the results of the urea solution in-bed testing. The data indicate that as the amount of urea introduced into the bed increased (i.e. gpm/NSR), the % NO_x reduction increased. In addition, the data suggest that the same NO_x reduction levels can be attained (at the same NSR levels) when using either a 15% or 30% urea solution. However, a 30% solution of urea in water is recommended since it can be introduced at half the flowrate (gpm) of a 15% solution thereby resulting in less bed quench. TABLE 5: NO_x REDUCTION USING UREA SOLUTION IN-BED INJECTION PROCESS

GPM GPM Bed

(15% (30% FB Temp. NO_x NO_x % NO_x NS

Run Urea Urea Temp. (Deg. % 02 (ppm) 0 (ppm) Reduction

No. Solution) Solution) (Deg. F) (DRY) Before | After

F)

1 1.0 - 1632 1477 7.8 156 120 23 1 1.

2 1.5 - 1619 1468 8.2 151 85 44 18.3

3 2.0 - 1619 1460 8.1 147 67 54 23.2

4 - 1.5 1617 1478 8.5 167 39 77 34.0

5 - 1.0 1621 1473 8 177 68 62 22.5

6 - 0.5 1617 1480 8 149 103 31 ²,⁹

NOTES:

1. All NO_x values presented are on dry volume basis and have been corrected to 7% 02.

2. For all Runs, natural gas was used as auxiliary fuel.

In our opinion, injection of aqueous urea solution directly into the bubbling bed of a fluidized bed incinerator, at a separate feed point independent of the sludge feed point, provides an oxidizing environment for the addition which enhances urea's effectiveness in reducing NO_x. The data indicates that there is no major advantage of urea in dry form over urea solutions with respect to NO_x reduction (i.e., similar NO_x reductions can be achieved at equivalent NSR values). However, it is usually more convenient to supply urea in the form of an aqueous solution. Description of the Preferred Embodiments In Figure 1 , a fluidized bed reactor 10 is shown having a vessel wall 11 which comprises a steel shell 13 and a refractory lining 14. The reactor 10 is supported by grillage beams 36 which rest on a concrete pad or foundation 38. The reaction chamber 16 in the main portion of the fluid bed reactor 10 is separated from the windbox 17 in the lower portion of the fluid bed reactor by a perforated steel constriction plate or the perforated refractory dome 21 as shown. It will be understood that a steel constriction plate is suitable for cold windbox operation while the perforated refractory dome is used when the incoming air is heated; that is, for "hot windbox" operation. The freeboard region of the reaction chamber is designated by the reference character 16. The refractory dome 21 is provided with a number of tuyeres 22 for providing communication between the windbox 17 and the reaction chamber 16. A conduit 27 is provided for introducing feed stock into the reaction chamber 16. A fluidized bed 24 is illustrated in the reaction chamber 16 and a conduit 26 is provided for draining the bed material or removal of solid particulate products if required. A conduit 29 is provided for supplying air to the windbox 17. A duct 34 is provided for the off-gases emanating from the reaction chamber 16. A conduit 28 may be provided for introducing secondary air into the reaction chamber 16.

The desired reactions which take place in the fluidized bed region 24 in the case of NO_x reduction are:

1500°F-1700°F

A) CO(NH₂)2 + 2NO + V.0₂ > 2N + C0₂ + 2H₂0

and/or

1500°F-1700°F

B) 2CO(NH )2 + 2Nθ2 + 0₂ > 3N₂ + 2CO2 + 4H2O

It will be understood that reaction A above dominates to the extent that 90- 95% of the urea is involved in this reaction. In operation, the fluidized bed reactor 10 has within the reaction chamber

16 a body of particulate material 24 which is supported by the perforated dome or constriction element 21. Air supplied by a blower (not shown) through conduit 29 to the windbox 17 moves upward through the tuyeres 22 of the constriction element 21 into the bed material 24 and expands that bed to a substantial height within the combustion chamber 16, forming a fluidized bed of the bubbling bed type. Above the expanded bed 24 is the freeboard region of the combustion chamber 16 into which gas and fine particles are ejected from the bed 24. Secondary air may be introduced into the freeboard region just above the expanded bed 24 through conduit 28. Combustion takes place primarily in the expanded bed 24 at temperatures of 1350°F to 1600°F and at the level of secondary air introduction and there above in reaction chamber 16 where gases and suspended fine solid particles are burned at temperatures of 1350°F to 1700°F. The sludge to be incinerated is supplied directly into the fluidized bed through conduit 27. The urea is separately injected directly into the fluidized bed through supply conduit or hose 41. The amount of injected urea used is substantially in excess of that required to complete the desired reactions.

A fluidized bed reactor about 30 feet high and having a maximum diameter at the reaction chamber of about 27 feet is provided with a refractory combustion dome separating the reaction chamber from the windbox below.

A bed comprising inert material (i.e. silica sand) is charged into the reaction chamber. After preheating, it is fluidized in the combustion zone of the reaction chamber by introducing a flow of air from the windbox sufficient to establish an air flow of from 2.0 ft/sec to 4 ft/sec. This air flow expands the charge in the reaction chamber and forms a fluidized bed of the bubbling bed type. Waste material containing organics, i.e., waste water sludge, is fed into the reaction chamber and oxidized. The urea in granule form or as a solution is separately injected directly into the fluidized bed. Combustion of the organic material in the charge takes place at a temperature of from 1350°F to 1600°F. There is excess air present in the process so that emissions from the reactor contain about 4-10% by volume of oxygen.

Above the bubbling bed, secondary air can be admitted into the reaction chamber to complete combustion of the gases rising from the bubbling bed and the fine solids suspended in the gas. The secondary combustion occurring at the level of introduction of secondary air is minor and usually does not increase the gas temperature.

The percent NO_x reduction of which this process of urea injection is capable ranges from 25% up to about 70%.

There has thus been disclosed a simple and effective method for reducing the nitrogen oxide levels produced in the combustion of wastes in bubbling bed fluidized bed incinerators.

Thus, having described the invention, what is claimed is:

Claims

1. A method for selectively reducing NO_x emissions from a fluidized bed incineration process wherein the feed undergoing incineration is a nitrogen- bearing waste comprising: injecting dry urea directly into the fluidized bed region of said incineration process in an amount of about 0.1 to 10 moles per mole of NO_x generated in said fluidized bed, and the temperature of said fluidized bed being in the range from about 1350°F to

1600°F and the NO_x present in emissions from the process being within acceptable limits.

2. The method of claim 1 wherein an excess of oxygen is present in the incineration process.

3. The method of claim 1 wherein urea is introduced in an amount of about 0.5 to 2 moles per mole of NO_x generated in said fluidized bed.

4. The method of claim 1 wherein the temperature of the fluidized bed is in the range from 1500°F to 1700°F.

5. The method of claim 2 wherein the oxygen present in emissions from the process is at least 4-10%, by volume.

6. The method of claim 2 wherein the dry urea is in the form of granules.

7. The method of claim 6 wherein NO_x emissions from the incineration process are reduced from 25% up to about 70%.

8. A method for selectively reducing NO_x emissions from a bubbling bed fluidized bed incineration process wherein the feed undergoing incineration is a nitrogen-bearing waste comprising: injecting urea directly into the bubbling bed region of said incineration process in an amount of about 0.1 to 10 moles per mole of NO_x generated in said fluidized bed, and the temperature of said fluidized bed being in the range from about 1350°F to 1600°F and the NO_x present in emissions from the process being within acceptable limits.

9. The method of claim 8 wherein the urea is injected in the form of dry urea prills or granules..

10. The method of claim 8 wherein the urea is injected in the form of a aqueous urea solution.

11. The method of claim 10 wherein an excess of oxygen is present in the incineration process.

12. The method of claim 10 wherein urea is introduced in an amount of about 0.5 to 2 moles per mole of NO_x generated in said fluidized bed.

13. The method of claim 10 wherein the temperature of the fluidized bed is in the range from 1500°F to 1700°F.

14. The method of claim 10 wherein the oxygen present in emissions from the process is at least 4-10%, by volume.

15. The method of claim 1 1 wherein NO_x emmissions from the incineration process are reduced from 25% up to about 70%.