Publication: Pulp & Paper Magazine
Issue: March 1, 1996

Numerous Recovery Options Offer Solutions for Mill Effluent Closure


Despite a heavier load on the recovery island from fiberline changes, incremental capacity increases are available

When evaluating potential pulp and paper mill modifications to reduce chlorine use and/or recover organic residues from pulping and bleaching, process engineers must consider the impact of such changes on the recovery system, including evaporators, recovery boiler, recausticizing, and the lime kiln. There are many engineering solutions to the problems of recovery system bottlenecks which can be implemented at a moderate cost, without having to take downtime beyond normal scheduled maintenance shutdowns.

This article estimates the magnitude of changes in load on each component of the recovery system and reviews potential solutions from the point of view of a bleach plant engineer who may not be an expert in the recovery area. Most of the technology discussed is applicable to any recovery limited mill.

For the purpose of analysis, the kraft recovery cycle can be considered in four areas:

Evaporation of black liquor to firing concentration

Recovery boiler

Recausticizing (production of cooking liquor from recovery boiler smelt)

Lime kiln.

The effects on recovery cycle load by implementing the best-known process closure technologies are summarized in Table 1. The impacts on the lime kiln are proportional to those for the recausticizing area.

Although not addressed here, mills attempting to eliminate all planned effluent discharges will have to address many issues in addition to the capacity of the conventional recovery system, such as scaling, metals management, and the overall soda/sulfur balance. The more closely a mill approaches complete process closure, the more significant these issues will become.

Rules-of-thumb and generic estimates of changes in process values that may be encountered are used for this article. However, these are no substitute for detailed, site-specific, millwide mass and energy balances, which should be carried out using modern process simulation techniques as an essential part of determining the best approach to overcoming limitations in recovery capacity.

KRAFT PULPING/CHEMICAL RECOVERY. The recovery process is an integral part of the mill operation, so some consideration of the pulping and washing process is required when evaluating the effects of changes. Kocurek provides a detailed description of this process,1 and the classic unbleached kraft mill process cycles are shown in Figure 1.

The pulping chemicals, sodium, and sulfur in various forms cycle through the loops shown in Figure 1, and none are actually consumed in the process. In practice, both sodium and sulfur are lost from several points in the cycle, so makeup chemicals are required.

Normally, the most economical makeup chemical is sodium sulfate (saltcake), which is added to the cycle just upstream of the recovery boiler as a dry powder. As indicated by its chemical formula (Na2SO4), the ratio of sodium to sulfur in saltcake is fixed at 46/32 = 1.44.

In almost all mills, if the correct amount of saltcake is added to compensate for sulfur losses, the associated sodium will be insufficient. The usual solution is to purchase sodium hydroxide (caustic) solution. This is normally added to the cycle at the white liquor tank. In the relatively rare situation where the ratio of sodium to sulfur losses is below 1.44, it will be necessary to make up with a mixture of sulfur (or sodium hydrosulfide) and sodium hydroxide.

In today's mills, there is often more sodium sulfate available as a byproduct of onsite chlorine dioxide generation than is required. This material is added to the black liquor and processed in essentially the same way as purchased saltcake when it is added at the inlet to the recovery boiler, shown in Figure 1.

Operating costs mentioned in this article have been calculated by the author and include only direct costs for chemicals and energy. In some cases maintenance costs would increase, while in others it would decrease. The unit costs used are shown in Table 2.

INCREASING EVAPORATOR CAPACITY. It is technically feasible to increase the capacity of an evaporator up to at least 10% by implementing one or more of the following modifications:

Convert internal heat exchange surface to external surface

Add a pre-evaporator that uses waste heat, frequently from batch digesters

Add an evaporator body

Install a concentrator and run existing evaporators to lower strong black liquor concentration (this will also normally increase recovery boiler capacity)

Extend the surface condenser.

Some modifications to the mill process outside the evaporator area can increase evaporator capacity or reduce the quantity of water requiring evaporation. These can include improved brownstock washing, replacement of pump packing with mechanical seals to minimize entry of water, conversion of batch digesters to indirect steaming or repair of such systems, and replacement of hydraulic doctors on washers with air doctors. None of these options has been examined in detail in this article. In some mills, a significant increase in capacity can be obtained at modest cost by carrying out a detailed engineering and optimization study, which can lead to correcting minor defects and improving operating skills.

REDUCING BOILER LOAD. It may be more practical or economical to reduce the recovery boiler load, rather than increase the capacity of an existing installation. The key issue is the thermal load. Solids feed rate and steam production are rarely important. The list of possible measures below is not exhaustive, since there is extensive scope for engineering ingenuity in this field.

Black liquor oxidation. It is common practice to oxidize black liquor to convert sodium sulfide to sodium thiosulfate or sodium sulfate, thus reducing the formation of hydrogen sulfide in the direct contact evaporator of traditional recovery boiler designs. Most of the systems in operation use air as an oxidant, but several use oxygen. The latter alternative has the advantage that there are no atmospheric emissions from the oxidation system.

In responding to an EPA Recovery Furnace Survey in 1994, mills operating a total of 55 recovery boilers using black liquor oxidation systems reported on the drop in calorific value across the oxidizer. The data are summarized in Figure 2. Data on five boilers that showed much greater drops in the heating value are not shown, since the values were considered to be improbable. The average reduction in heating value of the black liquor was 5.6%.

These data are based on standard measurements of the higher heating value of the black liquor. The standard method involves combustion in an oxidizing atmosphere, which differs significantly from the conditions in a recovery boiler, where the conditions in the lower furnace are controlled to reduce sodium sulfate and sodium thiosulfate to sodium sulfide. This reaction is endothermic.

Assuming that the sulfide content of the black liquor is 1.5% and it is oxidized to thiosulfate, the oxidized black liquor will absorb about 0.5% of the higher heating value of the liquor to reduce the thiosulfate to sulfide in the recovery furnace. The effective drop in heating value of black liquor due to oxidation is therefore about 0.5% greater than shown in Figure 2.

An independent confirmation of the feasibility of reducing the thermal load on a recovery boiler by oxidizing black liquor was provided by Connagahan, who reported an increase in recovery boiler solids throughput of 5% when a conventional (air) black liquor oxidation system was installed at the Crestbrook mill in Canada.2

None of these systems was installed with the objective of lowering heating value. Some of the oxygen-based systems are preceded by air oxidation systems, so they do not show the maximum potential for reducing heat value. It is therefore reasonable to assume that where a system is designed and operated with the objective of reducing the heating value of the liquor, that it will be more effective than the average system shown in Figure 2.

Oxygen-based systems can also be designed to generate steam from the heat of the oxidizing reaction. When using oxygen to oxidize the black liquor, the heating value of the liquor can be reduced by up to 1.2 MJ/kg, or about 8%, according to oxidation system vendors.

Air-based oxidation systems are unlikely to be installed in the future due to atmospheric emission considerations. The principal operating cost for oxygen-based black liquor oxidation is the supply of oxygen. Calculations from the operating data published by Perkins showed that about 25 kg oxygen is used per ton of black liquor solids, equivalent to $0.72/ton black liquor solids.3 This is counterbalanced by the heat recovered, so that the operating cost is close to zero. The capital cost is typically a few hundred thousand dollars.

A few mills-including Stone-Consolidated at Trois Rivieres, Que., Federal Paper Board in Augusta, Ga., and Riocell in Brazil-use the patented THR black liquor oxidation process (Air Products), which uses industrial oxygen, and report increases in recovery boiler capacity. Mullen reported that installation of the THR process was shown in at least one mill to also provide an increase in evaporator capacity in addition to the above-mentioned benefits with respect to recovery boiler thermal loading.4 This is supported by contact with the mill staff.

AQ pulping. Anthraquinone has been widely used in the pulp and paper industry in Japan, and to some extent elsewhere, for more than 15 years. Addition of small amounts of anthraquinone (typically less than 1 kg/ton of wood) to the digester at the start of the cooking process will modify the pulping yield, the required cooking time, and the degree of delignification attained. These three variables are closely interrelated, and the results obtained can be controlled by the operator by selecting the time, temperature, and cooking chemical charge.

In the context of this article, the principal interest in anthraquinone is its potential to compensate for increases in load on the chemical recovery and liquor preparation systems by improving pulping yield and/or reducing the required quantity of cooking chemicals. Holton and Chapman, Ringley, and Goyal showed that anthraquinone could be used to improve mill pulping yields.5,6,7,8

Figure 3 shows the net costs for additions of sufficient anthraquinone to the digester for corresponding recovery boiler load reductions, assuming production is held constant. The effective alkali charge required is also reduced by the use of AQ, which generally compensates for the additional load imposed on the white liquor preparation equipment for supply of white liquor for oxidizing prior to use in an oxygen delignification stage.

The costs shown in Figure 3 are highly sensitive to wood costs. The graph is based on a wood cost of $125/ton. If the cost is $150/ton, then the operating cost of anthraquinone addition is essentially zero, while if the wood cost is $100/ton, the cost is approximately double that shown in Figure 3. The capital cost of the mill modifications to use anthraquinone is well under $100,000.

Remote recovery boilers. It is common practice for mills that lack adequate recovery boiler capacity to ship black liquor to another kraft mill for processing. In most cases, green liquor produced by dissolving the smelt produced through liquor incineration in the boiler is returned to the originating mill. Capital costs are minimal, and operating costs are in the range of several dollars per ton of pulp,9 so this is a practical solution for overloaded recovery boilers in many cases.

Extended O2 delignification. Parsad showed that it is feasible to cook bleachable grade softwood pulp to higher-than-conventional kappa numbers (40 to 60) and then delignify it to below 15 kappa with oxygen while maintaining pulp properties.10 An interesting feature of the process is that it results in an increase in pulping yield sufficient to avoid an increase in load on the recovery boiler. At least one U.S. mill and one overseas mill practice extended oxygen delignification.

Soap separation. If the soap is separated and sold, burned outside the recovery boiler, or converted to tall oil, the heating value of the black liquor is dropped by 4 to 8%. Hardwoods would be at the low end of the scale, and softwoods would be at the high end. Most mills processing woods with high soap yields already practice soap separation, so the scope for using this technique for further reduction in boiler load is probably limited to a few mills.

Soap has a very high calorific value, so in some cases it may be more cost-effective to ship soap to a remote boiler than to ship excess black liquor. It has been suggested that soap could be used effectively as lime kiln fuel, but we are not aware of any full-scale experience. Sodium content could be an obstacle.

Black liquor gasification. Black liquor gasification processes have been proposed many times as alternatives to the conventional recovery boiler. The only system currently available commercially is the Chemrec process developed by Kvaerner Pulping. This process oxidizes the organics in a side stream of concentrated black liquor with air at 900C, while simultaneously reducing the sulfur in the liquor to sodium sulfide. A low heating value gas is produced, which is burned in the mill's power boiler. The gas is not suitable for firing in a lime kiln.

The only currently operating full-scale installation of this process is at Frovifors in Sweden. The capital cost of a system with an equivalent pulp capacity of 130 tpd would be about $20 million. This is much greater than the incremental capacity likely to be required to compensate for pollution prevention, but a detailed evaluation may be merited when a mill wishes to increase capacity as well as compensate for increased boiler load due to chlorine reduction modifications.

UPGRADING EXISTING RECOVERY BOILERS. The most significant characteristic of the liquor fed to a recovery boiler is the thermal load (total heat input, including the calorific value of the fuel, air heaters, and any fossil fuel burned with the black liquor). Any increase in thermal load will require modifications to a boiler that is currently operating at maximum capacity.

There are several ways of increasing the capacity of most recovery boilers by up to about 10%.9,11,12 Many have been developed and applied in the past few years and can be implemented within normal scheduled boiler shutdowns. Operator know-how is always important, and the first and least expensive way of increasing recovery boiler throughput is to ensure that supervisors and operators are well trained to optimize the operation.

In most boilers, the limit on capacity is the firing rate at which the tubes in the boiler bank plug more rapidly than can be tolerated. The pluggage can always be cleared by water washing the boiler, but this requires a shutdown for about two days and normally requires curtailment of mill production unless water washes are limited to the regular annual or semi-annual maintenance shutdowns.

The most common upgrade to pre-1985 boilers is to improve the air and liquor delivery systems. This normally involves installing high-pressure fans for the tertiary air and improving turbulence in the boiler. This approach has become popular because it is possible to carry out measurements on an operating boiler and optimize the design to a degree that was not possible with the level of knowledge at the time of original design.

Firing arrangements can often be improved to make use of current technology. Replacement of direct steam heating for the liquor improves safety and can provide almost a 1% increase in capacity, as well as improving thermal efficiency.

CAPACITY ISSUES. Boiler capacity is normally specified by the boiler manufacturer in terms of the feed of liquor solids per day. This value is calculated by considering a number of aspects, including the heat release rate in the furnace ("heat release rate" is essentially the feed rate of the fuel fired to the boiler multiplied by its calorific value), gas flows, rate of steam generation, and the capability to reduce sodium sulfate in the feed liquor to sodium sulfide. Chamberlain and McCubbin discuss practical recovery boiler capacities.9,13

Nameplate capacity is only a rough indicator of actual boiler capacity. It is quite common for boiler capacity to be expressed in terms of solid feed rate, assuming a nominal heating value of the black liquor of 15.1 MJ/kg (6,500 Btu/lb). In most mills the heating value of the liquor is lower, providing a margin of capacity that is not immediately obvious to the layperson. The routine operating limit for some boilers can be 30% above its rated equivalent pulp production rate.14

Overloaded boilers usually have to shut down frequently (several times per year) to water wash accumulated saltcake and related material from the boiler bank (Figure 4). In the context of this article, a boiler is considered overloaded if it has to be shut down for water washing more frequently than normal scheduled maintenance and inspections would require. All boilers are shut down for inspection at least once a year, which requires water washing. Many mills schedule biannual boiler shutdowns.

A considerable number of factors can limit the capacity of any one recovery boiler, and the essence of any capacity upgrade is to determine the critical ones and rectify them. In some cases, it is a relatively simple matter of inadequate auxiliary equipment, such as liquor feed pumps, or inadequate maintenance or operator training. Readily corrected situations, such as these, are more common than they should be, and every effort should be made to correct them before embarking on an analysis of the more expensive modifications discussed below.

Fireside deposits in the upper boiler. By far the most common limitation of the quantity of black liquor a boiler can process is plugging in the upper boiler, particularly in the boiler bank (Figure 4). This is caused by accumulation of solid deposits on the tubes, which manifests itself by one or more of the following:

Increasing draft losses across convection banks

Increasing induced draft fan speed

Increasing gas temperatures through the convection area

Decreasing final steam temperatures.

The increases in draft losses and fan speed are the result of the reduction in free flow area. The higher gas temperatures and lower steam temperatures are the result of the insulating effects of the deposits that restrict heat transfer from the fluegas to the water and steam.

Plugging occurs as the cumulative effects of the initial deposit formation and subsequent accumulation and growth, and in the case of overloaded boilers can require frequent shutdowns to water wash the boiler. The rate of pluggage is a function of the quantity and quality of ash carryover, the fluegas temperature, and the fluegas velocities, all of which generally increase with the firing rate of black liquor solids and calorific value of these solids.

For most mills, it is recommended to limit the fluegas temperature entering the boiler bank to 700C. Where the chloride content of the black liquor is high, then 600C could be the maximum temperature that could be tolerated without the ash becoming sticky and causing premature plugging. One objective of many boiler upgrade projects is to lower the temperature of the gases at this point in the boiler, thus preventing an increase in the firing rate from raising the temperature above the point at which the boiler starts to plug.

Furnace size and potential for upgrading. There is a practical limit to the quantity of liquor that can be successfully processed in a boiler of any given size. This is best defined by the hearth heat release rate per unit area. It is impractical to increase the physical area of a recovery furnace without accepting a shutdown of at least one month in duration.

Figure 5 shows the hearth loading in most of the recovery boilers in the U.S., as reported in the Recovery Furnace Survey.15 It is generally accepted that most boilers can be fired at least up to 950,000 Btu/hour/ft2, assuming that auxiliary equipment is adequate.

Boilers in the lower range of hearth heat release rates shown in Figure 5 can almost certainly be upgraded by the improvements to air systems or liquor systems discussed below. Those operating at higher rates are probably poor candidates for upgrading, so the mills would have to use one of the techniques described previously for reducing recovery boiler load.

Air system upgrades. In most boilers built before 1985, the air system can be improved to increase capacity and lower stack emissions. One of the most productive ways of increasing recovery boiler capacity in the past few years has been modification of the air systems.

To some extent, more recent upgrades of older boilers are a continuation of long-term trends in recovery boiler design, but such work has been characterized by increased application of measurements on operating boilers and much more intensive engineering than was normal in the past. Verloop discussed this at some length and emphasized that modifications are usually site specific so that detailed measurements of boiler operating conditions are essential for the design of an optimized system.12

The most common deficiency in air supply systems is inadequate velocity through the secondary and tertiary ports. Other factors are low temperatures and inadequate air flow control systems. In a few cases, total air supply is a limiting factor, and this is relatively easily corrected by increasing fan capacity.

The prime objective of most air system improvements is to attain as complete combustion as possible in the lower part of the furnace so that carryover of unburned particles is substantially eliminated. In addition, the increased heat transfer to the lower water walls results in reduced temperatures in the upper boiler at the entrance to the relatively closely spaced tubes of the steam generating bank, where plugging is most likely to occur.

While the air systems in some recent boilers have been designed using sophisticated physical and/or mathematical modeling, previous practice was empirical, and designers rarely had access to adequate data to evaluate their previous designs. Thus, improvements were rather slow. This almost universal deficiency in pre-1985 air system design provides an opportunity for increases in capacity, which has been exploited increasingly since the mid-1980s by boiler manufacturers as well as third parties.

The exact location of the air ports is important, but the principal feature of the more recent capacity upgrades has been the use of high-pressure air through small ports to create much greater turbulence in the lower furnace than was previously normal.

It is often necessary to increase the delivery pressure of forced draft fans to attain the velocities necessary for adequate turbulence in the boiler. While manufacturer recommendations differ, it appears that in most cases a low-pressure fan should be used for primary air only, with a separate higher-pressure fan used for secondary and tertiary air. In some cases, it is best to have three independent fans.

Liquor delivery. Optimizing liquor delivery is an essential part of many recovery boiler upgrades and is often associated with air system upgrades discussed above. The current approach to improving the liquor delivery system is to use multiple stationary spray nozzles located on all four walls of the furnace. The concept is to spread the liquor firing out to obtain a better, more even mixing of the liquor with the combustion air admitted at the secondary level.

The liquor is heated in an indirect liquor heater and with this approach can be raised to higher temperatures than a direct contact type. The temperature can also be controlled to within 0.5C, which is desirable to control particle size in the combustion zone in the interests of optimizing combustion conditions.

This type of liquor temperature control is required to give the correct black liquor droplet size. The correct droplet size is one that will travel from the liquor spray nozzle down toward the bed and, in doing so, reach the combustion zone just as all of the water has been evaporated from the liquor. The droplet initially swells as the water vapor is evaporated, and then once it is dry it collapses to a particle ready for combustion and smelting.

Steaming rate limitations. In relatively rare cases, the limiting factor on the capacity of a recovery boiler can be on the steam side. Problems such as poor separation of steam from water or limitations on circulation within the tubes can place a ceiling on capacity.

Where steaming rate is the limiting factor, some improvement can be attained by reducing the boiler feedwater temperature and/or reducing the temperature of the combustion air where it is heated by the steam coil air heater. Either of these techniques reduces the steaming rate by several percent. Where the problem is a design defect, it can sometimes be corrected at modest cost during a scheduled boiler shutdown. Where it is impractical to overcome limitations in steaming rate, it would be necessary to use one of the techniques discussed previously for reducing thermal load on the boiler.

FIRING HIGH-CONCENTRATION LIQUOR. Black liquor can be fired at concentrations from about 58 to 80% dry solids. The low end of the scale is considered unsafe by many engineers. Boilers with direct contact evaporators (DCE), mostly installed before 1970, are mostly fired at between 64 and 68% concentration, as shown in Figure 6. Boilers that operate without direct contact evaporators (NDCE), mostly built since 1970, mostly operate with rather higher firing concentrations, as indicated in Figure 6.

The DCE boilers are limited to a maximum practical firing consistency of about 72%, since the viscosity of the liquor at higher concentrations is too high for the DCE equipment. The NDCE boilers operate at up to 80% firing consistency.

It is generally desirable to fire recovery boilers with high-consistency black liquor. This reduces the total gas flow through the boiler, reducing the tendency to plug and/or reach the limiting capacity of the fans. Higher-consistency liquor also raises the combustion zone temperatures, resulting in greater heat transfer to the water walls. High temperatures in the lower furnace generally tend to reduce emissions of TRS and SO2 and to improve the reduction efficiency of the recovery boiler, with a number of benefits to the mill's operation.

In some cases, the carryover of incompletely burned liquor causes blockage of the gas passages in the boiler and superheater bank. It is clear that as the liquor consistency increases, the quantity of water introduced to the boiler with the liquor declines. Thus, the gas flow will also drop so that some reduction in carryover could be expected. Some concerns have been expressed about the possibilities of higher temperatures causing excessive corrosion in the lower furnace, but no documented evidence of this was found.

Raising the firing concentration of the black liquor has been shown to be feasible and to raise the capacity of the boiler by many authors.14,16,17,18 Figure 6 includes five boilers in separate U.S. mills being fired with black liquor at concentrations between 75 and 79%. Ten more boilers report firing at between 72 and 74% concentration. Many evaporator installations since about 1989 have been designed to produce 80% concentration black liquor, but the author is not aware of any mills firing at such a high concentration.

Huber described a project that resulted in raising the concentration from 68 to 75% of the liquor fired in the boiler at Pope & Talbot's Halsey, Ore., mill.17 This increased the boiler capacity from 806 tpd of black liquor solids to 1,020 tpd (approximately 16% increase). Subsequent personal communication with one of the authors indicated that the boiler has been successfully fired at up to 1,180 tpd, representing an increase of 28% in capacity. This project also included the installation of a liquor heat treatment system, which reduces the calorific value of the liquor by about 5%. Some of the increase in capacity can be attributed to this.

Lefebvre stated that calculations using Combustion Engineering Inc. (ABB) boiler simulation software indicate that where the capacity of a boiler is limited by plugging, increased solids throughput could be achieved by increasing the liquor firing consistency.19 They did not mention the extent of the increase attainable in that paper, but personal communication indicates an increase of several percent in capacity due to reduced temperatures in the boiler bank and reduced gas flow.

The higher flame temperatures resulting from the high-concentration liquors increases heat transfer to the water walls in the lower furnace, which leads to lower temperatures in the critical boiler bank,14 thus reducing the tendency to plug the passages within the boiler.

Imelainen described modifications to a recovery boiler at the Kemi mill in Finland, where he attributed an increase of 15% in boiler capacity to raising liquor firing consistency from the low 60% to the low 70%.20 However, the project to increase liquor firing consistency included some unspecified modifications to the boiler's air system and improved computer control, and it is not possible to separate the improvements which depended solely on the increase in liquor consistency.

In some specific experiments, they found that raising consistency from 65 to 75% reduced SO2 emissions substantially and raised the thermal efficiency from 55 to 60% while improving reduction efficiency from 95 to 97%. The same change in firing conditions raised the dust emissions from the boiler itself by 30%, but resulted in lower stack particulate emissions due to an improvement in the electrostatic precipitator's performance resulting from the lower gas flows and temperatures.

Hyoty described three years' experience of operating a "super concentrator" to evaporate black liquor to 78 to 82% and fire it in a recovery boiler at a mill in Pori, Finland.16 The authors concluded that boiler operations had improved in a number of ways, with the only difficulty being the extra operator attention required by the high-solids evaporator. Personal communication with the authors indicated that the effective increase in boiler capacity was 15% when increasing firing consistency from 65 to 80%. The boiler is reportedly fired routinely with 78% consistency liquor, since this provides sufficient capacity for the mill's requirements.

Increasing the black liquor firing consistency generally involves installing an additional black liquor evaporator, usually known as a concentrator, to raise the concentration to the desired level, which may be up to 80% dry solids. In most cases, the viscosity of the thick black liquor is the limiting factor. This can be readily reduced by heat treatment, as discussed below, but the capital cost of the necessary equipment (typically a few million dollars) may be unacceptable.

Several mills use a heat treatment process to lower the viscosity of high-concentration black liquor. Ryham described a system where the black liquor was heated to 180C near the end of the evaporator set, permanently reducing viscosity by a factor of about 5.21 The absolute reduction was most pronounced at high-liquor concentrations (65 to 80%).

The equipment used to treat the liquor is effectively a multi-flash evaporator, and it raised the concentration of the liquor by about 5%. It also caused sufficient emission of reduced sulfur gases to replace 12% of the fuel burned in the mill's lime kiln. This corresponds to about a 4% drop in the calorific value of the black liquor, which in itself increases the boiler's capacity to burn black liquor solids. Ryham reported an improvement in reduction efficiency from 93 to 96%. This is marginally beneficial to recovery boiler capacity.

Huber described a comparable installation at the Halsey, Ore., mill.17 Personal communication with mill staff indicates that the heating value of the black liquor was lowered by 4 to 5% by the heat treatment, due to evolution of sulfur gases that are burned in the lime kiln.

The increase in boiler capacity was 28%, as mentioned above. This was attributed to a combination of lowered calorific value of the black liquor, increased liquor firing concentration (from 68 to 75%+), and a cleaner burning liquor.

RECAUSTICIZING AND LIME KILN UPGRADES. Some mills would have to increase the capacity of their recausticizing systems to provide the oxidized white liquor required for cost-effective oxygen delignification and/or ozone delignification. There are a number of proven ways of doing this, which consist essentially of identifying the departmental bottleneck and taking corrective action.

Most recausticizing plants can achieve a 7% increase in white liquor production by upgrading either the green liquor clarifier or the slaker. Alternatively, the relatively new approach of recovering undiluted white liquor from the lime mud leaving the white liquor clarifier/filter could be employed to provide the additional white liquor while maintaining existing green liquor flow rates.

Typical capital costs of the upgrades have been $2 million to $3 million for mills with fully loaded recausticizing systems that have installed oxygen delignification. Where a mill is adding both oxygen delignification and ozone delignification, the total increase in recausticizing department load would be about 12%. In some cases, major work may be required.

There are several proven methods of obtaining marginal increases in capacity of a lime kiln, including improved control, improved drying of lime mud feedstock, installing one or more dams in the kiln to increase retention time of lime, and oxygen enrichment of the combustion air. The latter is the most commonly used.

Oxygen enrichment is a well-proven technique and has been used for more than 20 years. The necessary technology is available from several competitive suppliers of industrial gases. Garrido lists nine U.S. and two Canadian kraft pulp mills using oxygen enrichment in the lime kiln, as well as several other installations outside the pulp industry.22

The equipment required is relatively simple, consisting of feed piping, an injection lance, and controls. The capital cost is negligible relative to the costs of other recausticizing plant upgrades being considered. Calculations by the author based on the above references indicates that there is no significant effect on lime production costs where oxygen enrichment is used. n


1. M.J. Kocurek (Editor), Alkaline Pulping, Pulp and Paper Manufacture, Vol. 5, TAPPI Press, 1989, Atlanta, Ga.

2. C.L. Connaghan and G.E. Fenner, Strong Black Liquor Oxidation, Its Effects on Mill Operations and H2S Emissions, Pulp & Paper Canada, Mar. 1976.

3. A.S. Perkins, D. Hornsey, W. Parker Pearman, G.F. Sharpe, and Ronald L. Smith, Mill Experience Using Oxygen to Oxidize Weak and Strong Black Liquor in Pipeline Reactors, Proceedings of the TAPPI Pulping Conference, Nov. 1993, Atlanta, Ga.

4. T. Mullen and L.G. Mayotte, Effects of Black Liquor Oxidation on Overloaded Evaporators, Pulp & Paper Canada, Vol. 89, No. 5, May 1988.

5. H.H. Holton and F.L. Chapman, Kraft Pulping with Anthraquinone, TAPPI Journal, 1977, Vol. 60, No. 11, p. 121.

6. H.H. Holton and T.J. Blain, Economics of Anthraquinone Pulping, Pulp & Paper Canada, 1983, Vol. 84, No. 6, p. 124-129.

7. M.B. Ringley, Westvaco Used Anthraquinone to Increase Alkaline Pulping Yields, American Papermaker, Apr. 1991.

8. G.C. Goyal, J. Powers, and M. Cronlund, Anthraquinone-Simple Approach for Extended Delignification in Conventional Kraft Pulping, Proceedings of the 1992 TAPPI Pulping Conference, Boston, Mass., Nov. 1992, p. 1047-1054.

9. N. McCubbin, Review of Technology for Overcoming Capacity Limitations in Kraft Recovery Boilers, Report prepared for Industry, Science and Technology Canada by N. McCubbin Consultants Inc., Ottawa, Canada, July 1990.

10. B. Parsad, J. Gratzl, A. Kirkman, H. Jameel, and T. Rost., Extended Delignification by Kraft Pulping Followed by Oxygen/Alkali Treatment: Technical and Economic Evaluation, Proceedings of the 1993 TAPPI Pulping Conference, Nov. 1993, Atlanta, Ga.

11. B. Harsent, J.S. Elliott, and T. Sonnichsen, The Mackenzie Recovery Boiler Upgrade, Proceedings of the CPPA Technical Section Spring Conference, Jasper, Alta., May 1990.

12. A. Verloop, J.H. Jansen, and D.E. Danroth, Capacity Upgrade of a 1965 Babcock and Wilcox Recovery Boiler through Installation of an Improved Over-bed Air System, Proceedings of the 1989 International Chemical Recovery Conference, Ottawa, Canada.

13. R.E. Chamberlain, 1981, An Approach to Establishing and Maintaining Safe Kraft Recovery Furnace Firing Rates, Pulp & Paper Canada, 1981, Vol. 82, No. 6, p. T196-T200.

14. American Forest & Paper Assn., Kraft Recovery Boiler Physical and Chemical Processes, 1988, Textbook sponsored and published by AF&PA.

15. EPA, NCASI. Survey of Recovery Furnaces in the Kraft Pulp Sector, carried out by NCASI in cooperation with EPA in 1994-95.

16. P.A. Hyoty and S.T. Ojala, High Solids Black Liquor Combustion, TAPPI Journal, Jan. 1988, p. 108-111.

17. E. Huber, P. Rank, J.W. Rauscher, and D.F. Williamson, High Black Liquor Solids Firing at Pope and Talbot, Proceedings of the 1995 International Chemical Recovery Pulping Conference, TAPPI Press, Toronto, Ont., Apr. 1995.

18. S. Ibach, Conversion to High Solids Firing, Proceedings of the International Chemical Recovery Conference, TAPPI Press, Toronto, Ont., Apr. 1995.

19. B.E. Lefebvre and G.M. Santyr, Effect of Operating Variables on Kraft Recovery Boiler Performance, Pulp & Paper Canada, Dec. 1989.

20. K. Imelainen, J. Ekholm, K. Kukkanen, and I. Pasola, A Computer Controlled Recovery Boiler Burning High Dry Solids, Pulp & Paper Canada, 1989, Vol. 90, No. 4.

21. R. Ryham and S. Nikkanen, Liquor Heat Treatment and its Impact on Chemical Recovery, Proceedings of the 1992 European Pulp & Paper Week, Bologna, Italy, May 1992.

22. G.F. Garrido, A.S. Perkins, and J.B. Ayton, Upgrading Lime Recovery with Oxygen Enrichment, Proceedings of the 1981 Annual Meeting of the CPPA Technical Section, Montreal, Que.



Search Home Subscribe to Pulp & Paper Magazine


Copyright 2000 Paperloop Inc. All rights reserved.