US3870467A

US3870467A - Monitoring chemical recovery furnace

Info

Publication number: US3870467A
Application number: US353828A
Authority: US
Inventors: Hugh Wharton Nelson
Original assignee: Combustion Engineering Inc
Current assignee: Combustion Engineering Inc
Priority date: 1973-04-23
Filing date: 1973-04-23
Publication date: 1975-03-11
Anticipated expiration: 1992-03-11
Also published as: FR2226505A1; CA1013904A; SE407595B; FI54952B; FI54952C; JPS5146676B2; ES425583A1; FR2226505B1; JPS5013296A

Abstract

The production of sodium vapor in the char bed of a chemical recovery furnace is monitored as an indirect indication of the efficiency of the reduction of Na2SO4 to Na2S. The sodium vapor production is monitored by means of an optical device sensitive to the photon energy emitted by burning sodium and relatively insensitive to other photon energy from the burning char bed.

Description

United States Patent Nelson MONITORING CHEMICAL RECOVERY FURNACE Inventor: Hugh Wharton Nelson, West Hartford, Conn.

Combustion Engineering Inc., Windsor, Conn.

Filed: Apr. 23, 1973 App1. No.: 353,828

Assignee:

U.S. Cl. 23/230 R, 23/253 R, 356/87 Int. Cl. G0lj 3/30, G01n 31/12 Field of Search 23/230 R, 253 R, 230 PC,

References Cited UNITED STATES PATENTS 5/1927 Brace 23/253 R 5] Mar. 11,1975

3,512,956 5/1970 Holderreed et a1. 23/230 R X 3,565,538 2/1971 Kahn et a]. 356/87 X 3,607,083 9/1971 Chowdhry 23/230 R OTHER PUBLICATIONS Diamond et a1., Anal. Chem., 25, 1825 (1953).

Primary E.raminerRobert M. Reese Attorney, Agent, or Firm-Richard H. Berneike [57] ABSTRACT The production of sodium vapor in the char bed of a chemical recovery furnace is monitored as an indirect indication of the efficiency of the reduction of Na SO, to Na s. The sodium vapor production is monitored by means of an optical device sensitive to the photon energy emitted by burning sodium and relatively insensitive to other photon energy from the burning char bed.

4 Claims, 5 Drawing Figures Pmminm 1 H s SHEET 8 BF 3 v FIG. 3

PATENTEDHAR] 1 1975 3.870.467

um I!!! FIG. 4

MONITORING CHEMICAL RECOVERY FURNACE BACKGROUND OF THE INVENTION The operation and control of chemical recovery furnaces of the recovery of chemicals from kraft black liquor is a difficult problem from the standpoint of both maximizing the efficiency of the process, and minimiz ing air pollution and the explosion safety hazards. One factor which makes the operation and control difficult is that the composition (particularly solids content) of the black liquor varies from time to time. Another factor is the variety of chemical reactions which take place in the furnace. Several important changes in the black liquor take place in the chemical recovery furnace, the first of which is the vaporization of the water which has not been previously removed by direct and/or indirect evaporators upstream. The carbonaceous material in the black liquor then burns in a pile on the hearth while the inorganic materials fuse and form smelt in the furnace. Reducing conditions are maintained in the lower hearth region of the furnace while oxidizing conditions are maintained higher up in the furnace. Some of the primary reactions which take place in the chemical recovery furnace are as follows:

Na SO, 2C Na S 2C0,

a CO 4: Na O (solid) CO Nil- O (solid) C Na (gas) CO N21 (gas) l/20 Na O (fume) Na O SO, Na SO (fume) Na O CO Na CO (fume) The control of the operating conditions in the lower furnace controls the reduction of the sodium sulfate to the sodium sulfide and conversion of sodium carbonate to sodium. oxide and sodium oxide to sodium vapor.

The amount of sodium vapor formed by reduction in turn controls the concentration of sulfur oxides and the sodium sulfate and sodium carbonate fume emitted in flue gas from the furnace. Proper control of the reducing conditions is therefore necessary not only to regulate the efficiency of the production of sodium sulfide from sodium sulfate, but also to regulate furnace emissions. Increased steam production is an additional benefit of correct operation.

In general, a high bed burning temperature and a deep char bed in the furnace are desirable to maximize reduction efficiency and to decrease sulfur oxide and particulate (fume) emission. However, furnace operators are often reluctant to operate with a deep char bed because there is an increased possibility of severe dissolving tank explosions caused by chance rapid burnout of the bed which creates a flood of molten smelt run off which is difficult to shatter. An unshattered smelt stream can cause dangerous smelt-water explosions in the dissolving tank. Another disadvantage of operating with a deep char bed is that the air ports around the periphery of the furnace can become blocked, causing a black out when an undercut char bed falls over toward a furnace wall. A further problem with deep bed operation is that it is often difficult to determine by mere visual observation the exact height of the bed and, therefore, difficult to tell if the bed is getting too deep. For these reasons, many operators operate the furnace under conditions which will maintain a lower bed thereby impairing the reduction efficiency and increasing the carry over of char in the gas stream which forms low melting slag deposits on heat transfer SUI'ffiCCS downstream.

SUMMARY OF THE INVENTION The present invention involves a technique for monitoring the burning conditions within a chemical recovery furnace bed whereby pertinent variables may be controlled to optimize furnace operation. More specifically, it has been found that the reduction efficiency in a chemical recovery furnace as well as other factors such as the release of pollutants from the furnace and steam production are directly related to the formation of sodium vapor in the char bed. It has also been found that the production of sodium vapor can be monitored and furnace operating conditions varied in accordance with these monitored readings so as to maintain the production of sodium vapor at a desired level or within a desired range to optimize furnace operation.

Therefore, the present invention involves the use of means for continuously detecting the rate at which sodium vapor is being produced in the bed of a chemical recovery furnace so that the operator can control key operating parameters to maintain sodium vapor production at an optimum level or within an optimum ange. More specifically, the means for detecting so dium vapor production comprises an optical device that receives and measures the photon energy emitted by the sodium burning reaction in the furnace char bed while blocking the photon energy from other reactions.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic elevational view of a black liquor chemical recovery furnace incorporating the present invention.

FIG. 2 is a schematic cross sectional view taken along line 2-2 of FIG. 1.

FIG. 3 is a side elevation view of the detector.

FIG. 4 is a front elevation view of the detector.

FIG. 5 is a schematic diagram of the detection circurt.

BRIEF DESCRIPTION OF THE PREFERRED EMBODIMENT FIG. 1 illustrates a chemical recovery furnace 10 Y which is typical of chemical recovery furnaces used for the processing of black liquor. The walls of this furnace are lined with steam generating tubes 12 which form a part of the heat exchange surface of the chemical recovery unit with there being additional heat exchange surface identified generally at 14 in the upper region of the unit.

Black liquor obtained from the kraft pulping process and/or other sodium based pulping processes'which has been processed by evaporation to the desired solids content is introduced into the furnace 10 through the nozzles 16. The liquor thus sprayed into the. furnace-descends downwardly towards the furnace bottom passing through rising combustion gas such that a majority of the moisture in the liquor is immediately evaporated. The solid particles fall downwardly through this rising combustion gas stream and form a pile or char bed 18 on the hearth or furnace bottom 20. A portion of the combustibles are consumed during this descent through the furnace with remaining combustibles being consumed in the char bed 20. The noncombustibles,

inorganic chemicals, are smelted and decanted from the furnace through the discharge spout 22.

Combustion supporting air is introduced into the furnace at two locations. The primary air is introduced through nozzles or ports 24 spaced relatively close to the bottom while the secondary air is introduced through the nozzles or ports 26 located above the liquor nozzles 16.

In addition to the reduction of sodium sulfate in the bed according to reaction (I) l. Na- SO 2C Na S 2CO the sodium carbonate is thermally decomposed according to the reaction (2) 2. Na CO e: Na O (solid) C The reverse reaction is depressed due to the escape of the CO and also to the immediate use of Na O in reaction (3).

3. Na O C Na (vapor) CO Sodium oxide is a thermally stable, relatively nonvolatile solid with a boiling point of 2320F. This is well above bed temperatures which are normally about l500F to 2000F. The elemental sodium produced in reaction (3) by contrast, has a low boiling point of l6l8F and thus readily volatilizes from the bed. It is a very reactive substance and quickly burns according to reaction (4) just above the bed giving off substantial heat and a bright yellow light.

4. Na (vapor) 1/20 Na O (fume) AH Light This yellow light comes from the thermal excitation of sodium atoms in the extremely fine, hot solid particles of Na O fume. It has sharp peaks ofintensity in the 5890 Angstrom wave length region of the visible light spectrum. This wave length is characteristic of sodium either in the elemental or compound form. It cannot be confused with the light from other excited atoms present in the furnace. The spectral lines for burning carbon or sulfur, for example, are far removed from this characteristic sodium wave length. Since the char bed is usually 300"F to 500F cooler than the air-rich regions above the bed, the light emissions from less volatile sodium compounds in the bed is limited. In addition the light from in situ sodium compounds in the bed is screend by the opaque carbon which is present. Yellow light from the bed would thus be due primarily to the sodium vapor burning above the bed. A thermodynamic study of the two reduction reactions taking place in the char bed indicates that considerations like the high dilution of sodium sulfate with carbonate and the requirement of a molten state for efficient reduction place the two reactions in close temperature regimes to each other. This makes the formation and luminous burning of sodium a useful tracer reaction for the reduction of sodium sulfate.

Two types of dust are produced in a chemical recovery furnace which may be carried over in the flue gases leaving the furnace. One type is caused by mechanical carry over the black liquor particles when the black 1i quor is introduced in an excessively fine spray. These too fine char particles burn in flight and produce relatively large spheres of smelt having a low melting point. The other type of dust consists of submicron sized particles of high melting sodium sulfate and sodium carbonate fume formed by reactions of furnace gases with metallic sodium vapor produced in the bed. It has been found that if the bed brightness is at a very high level, the amount of fume in the flue gases may be excessive. Bed brightness, therefore, should be maintained within a range of high brightness values determined by prior calibration to optimize the reduction efficiency without creating the high fume levels. Maintenance of brightness within this range of values and maintenance of proper (large) black liquor spray size will permit the operator to sustain high reduction efficiency and minimize sulfur oxide concentration in the flue gas as well as to balance these factors against the dust loading in the exit flue gases.

The meter for use in the present invention to receive and indicate the photon energy emitted by the sodium burning reaction is substantially an optical device. A photodetector is used in the meter which is particularly responsive to the sodium spectral line. Photoresistors, photodiodes, photocells, phototransistors, prisms, diffraction gratings, and the like may be employed. Detectors such as photodetectors which are responsive to particular portions of the visible light spectrum are commercially available items and therefore will not be further described herein. It is only necessary to select a detector which is responsive to the sodium spectral line and relatively insensitive to other light coming from the furnace. In addition, optical filters that pass the sodium spectral line and attenuate the side bands are employed. This combination of photodetectors and filters results in a meter which is about responsive to the sodium spectral line and only about 5% responsive to the spectral line from the gas or oil flames from auxiliary burners or the hydrocarbons that are being burned in the furnace.

FIGS. 1 and 2 illustrate the meter 28 as applied to a chemical recovery furnace. FIG. 1' shows the angle of view of each meter in the vertical plane and FIG. 2 shows the angle of view in the horizontal plane. From these two figures it can be seen that by the use of four meters placed around the furnace substantially all of the char bed can be viewed. A lesser number of meters can, of course, be used with consequently less sensitive registration of bed brightness. The meters may be located so as to view through peep holes in the furnace walls or through unused black liquor gun ports if they are available. The meters are ideally mounted near the corners of the furnace and they are aimed down at about a 45 angle so that the field of vision is about one quarter of the char bed. Each meter monitors an area of the bed opposite the location of the meter.

In FIG. 1, two of the meters are illustrated as being connected to an averaging device 30. Although not shown, the other two meters would also be connected to this averaging device. The output of the averaging device is then fed to a recorder 32 in which case the recorder indicates the average reading of the multiple meters. It may be desirable in certain instances to record the separate outputs from each meter so that the furnace operator can monitor each portion of the char bed. Alarms can be provided to warn of very low or very high brightness readings either for the average reading or for each individual reading. The low brightness alarm would be a warning of impending problems such as an impending blackout or a water leak and the high brightness alarm would indicate a high fume load in the exit gases.

FIGS. 3 and 4 illustrate a preferred construction of the meter of the present invention. The photodetector 34 of the meter 28 is housed within a compartment 36 which also houses other electric components of the meter as will be described hereinafter. This compartment 36 is attached to a plate member 38 which supports the slide 40. This slide 40 contains two optical filters 42. The plate member 38 has aperture 44 therein through which the light shines on the photodetector 34. The slide 40 is arranged so that one of the optical filters 42 is covering the aperture 44 while the other optical filter 42 is exposed so that it can be cleaned. Filters of different pass bands and percentage transmittance may be used in the same slide 40 to permit the selection of the proper wave length and intensity of photon energy for the photodetector to keep the photodetectors at the optimum illumination levels. Attached to the front surface of this plate 38 is the tubular section 46 through which the light from the furnace passes. The connection 48 on the tubular section 46 is for the purpose of admitting purge gas into the tubular section 46 to pre vent blowback from the furnace which could cause black liquor and ash to deposit on the filter window. Air, steam or other inert gas might typically be used for this purge.

FIG. 5 is a schematic diagram ofthe meter circuit. This circuit is a conventional bridge circuit 50 with the photodetector 34 being a portion of one leg of the bridge. The output of the bridge feeds into the amplitier 52 with the output of the amplifier feeding the previously mentioned averaging device or separate recorders as desired. The potentiometer 54 is employed to adjust the bridge.

A number of furnace operation parameters affect the reduction efficiency and therefore the brightness of the sodium spectral line. One parameter is the concentration or percentage of solids in the black liquor. The higher the concentration of solids, the higher the brightness. The solids concentration of the black liquor is determined by several factors. One of these factors is the evaporation rate in the evaporator prior to the firing of the black liquor. This is normally, although not necessarily, a direct contact evaporator such as a cascade evaporator in which the black liquor is contacted with flue gas. The concentration of solids leaving this evaporator can be changed, for example, by varying the gas flow through the evaporator. Another factor determining the concentration of the black liquor fed to the furnace is the amount of water that leaks into the black liquor in the pump which supplies the black liquor guns. These pumps have water seals and a significant amount of water can leak into the black liquor stream if the gland seals are not adjusted properly.

Another furnace operating parameter which has a significant effect on the efficiency and brightness is the size of the droplets of black liquor being sprayed into the furnace. With a fine spray there is a high degree of mechanical carryover which decreases the depth of the char bed and the reduction efficiency and increases the ash fouling of heat transfer surfaces. With a coarse spray the mechanical carryover is reduced and the bed depth, brightness and reduction efficiency all increase. The droplet size of the black liquor is directly affected by the viscosity of the black liquor which is related to the black liquor temperature. A black liquor heater is conventionally employed to control this temperature and a change of only a degree or two has a significant effect on the particle size and bed brightness.

Reduction efficiency is governed by the intensity of the reducing condition. This intensity is influenced by the surface area of contact between molten smelt and carbonaceous char as well as the absence of oxygen and the temperature of the molten phase. Char bed depth is a good indicator of the potential for reduction and is affected by the parameters previously mentioned. A further parameter which affects both bed temperature and depth is the total air flow to the'furmace and especially the ratio of primary to secondary air flow. The air flow is in turn affected by the furnace draft which can be effected by factors such as the amount of ash collected on furnace surfaces. Therefore, the operation of the furnace soot blowers can affect the reduction efficiency and brightness. Also. the cleanliness of the primary air ports has a significant affeet.

The monitoring of the sodium spectral line indicates changes in furnace operating parameters prior to the time that these changes would be apparent to the operator either from a visual inspection or from other indications. For example, a drop in the level of char bed-of about 15 inches at the sides and 4 feet in thecenter produces a change in the meter reading. A desirable level for the char bed in the center is about 6 feet. The influx of pump gland seal water into the black liquor can lower the concentration of solids by 3% or 4% if not properly adjusted. This can affect the meter output by 25% to 30% and immediatelyalert the operator. As previously mentioned, changes in the black liquor temperature, and therefore the spray droplet size, greatly affect the brightness and can also be immediately detected.

The recorders which may be used in the practice of the present invention may be calibrated directly in terms of reduction efficiency. This is accomplished by conducting tests on the green liquor from the smelt dissolving tank corresponding to various brightness values to determine the reduction value (percent sodium sulfate converted to sodium sulfide) and then graphing brightness versus reduction efficiency.

Meters have been used in the past which measured total (white) light coming from the furnace. However, the total light is not indicative of reduction efficiency. For example the reading on a meter which indicates total light measures flammable gas burning and does not relate to the bed depth as contrasted to the present invention in which the meterreadings do relateto bed depth and reducing conditions. Also, thetotal light meter reading goes up when viewing an oil flame with white light meters while the reading on the meter used in the present invention goes-down. In addition the total light meter registers light from-flames burning around the sighting port caused by air-leakage into this air deficient portion of the furnace; the present meter does not. It is not mislead by such light, but'rather registers only sodium light emitted from the bed'area.

While the invention has been shown and described with reference to specific embodiments and techniques, it will be understood that such showing and de scription is merely illustrative and that changes may be made without departing from the scope of the *invention as claimed.

What is claimed is:

1. A method of monitoring the reduction efficiency in a chemical recovery furnace in which Na SO is reduced to Na- S in a char bed in said furnace and in which sodium gas is produced and burned to form Na O fume comprising the step of monitoring the intensity of only that portion of the visible light spectrum in the region of the wave length characteristic of excited sodium atoms coming from said char bed.

2. A method as recited in claim 1 wherein said wave length is in the region of 5890 Angstroms.

3. A method as recited in claim 2 and further including the step of controlling the operation of said furnace said characteristic wave length.

Claims

1. A METHOD OF MONITORING THE REDUCTION EFFICIENCY IN A CHEMICAL RECOVERY FURNACE IN WHICH NA2SO4 IS REDUCED TO NA2S IN A CHAR BED IN SAID FURNACE AND IN WHICH SODIUM GAS IS PRODUCED AND BURNED TO FORM NA2O FUME COMPRISING THE STEP OF MONITORING THE INTENSITY OF ONLY THAT PORTION OF THE VISIBLE LIGHT SPECTRUM IN THE REGION OF THE WAVE LENGTH CHARACTERISTIC OF EXCITED SODIUM ATOMS COMING FROM SAID CHAR BED.

3. A method as recited in claim 2 and further including the step of controlling the operation of said furnace so as to maintain said intensity within a desired range.