Waste Heat Boilers in Hazardous Waste Incineration Service


IT3/HWC Conference 2010 #6

 

Prepared By Rick Ullrich

 

WastePro Engineering, Inc.

P.O. Box 74

Kennett Square, PA 19348


and Elizabeth Drake

 

Compliance Strategies & Solutions, Inc.

1301 Regents Park Drive, Suite 100

Houston, Texas  77058

Phone (281) 286-1861

Fax (281) 286-0325

edrake@cs2inc.com

 

 

ABSTRACT

 

The goal to extract useful energy from waste combustion – most often as steam – is a positive one, made even more important by rising energy costs and focus on reducing fossil fuel usage.  Installing a boiler in incineration service may be worthwhile, but maintenance challenges and on-stream time penalties can harm the economics of such equipment.  Experience world-wide, both positive and negative, will be reviewed and a matrix of key parameters will be offered for review of retrofit potential.

 

INTRODUCTION

 

Boilers are used throughout industry to generate steam, almost always from the combustion of fossil fuels.  Industrial furnaces utilize heat release to add energy to a process stream, and sometimes with direct combustion to oxidize a process material.  USEPA promulgated Boiler & Industrial Furnace (BIF) regulations for hazardous waste combustors (HWC) equipped with or retrofit into boilers many years ago, and has updated them in a maximum achievable control technology (MACT) regulatory package that has been legally challenged.  EPA has remanded these regulations, leaving compliance control balanced between the BIF and MACT until resolution of the various challenges (1).  A small set of combustors captured under the interim HWC MACT are equipped with boilers, but have always been treated permit-wise as incinerators.

 

Over eighty facilities underwent the original BIF permitting, including a number of cement kilns and other industrial furnaces that will not be discussed here.  Cement kilns, with enormous energy appetites, have continued to develop waste-sourced energy input, as a means to reduce fuel costs and now even to reduce fossil-fuel sourced CO2 emissions.  Most of this continuing effort is directed at non-hazardous waste sources.  Sulfuric acid regeneration furnaces are a BIF subset that began burning waste to offset natural gas consumption in direct combustion furnaces.  Other process-specific industrial furnaces include halogen recovery, generally as acid, and several large process tubular heaters converted to clean liquid waste as the primary fuel in indirect process heating.

 

Many of the constraints on boilers discussed herein are applicable to such furnaces, as are the positive aspects, but our focus will be on steam generation.

 

The economics of steam production from waste center on the trade-off in operating costs between the value of steam and the maintenance costs of the waste-fired boiler.  Fouling and/or corrosion of the boiler tubes can cripple the unit, and repair costs and downtime can be costly.  Several US incinerators, in fact, removed their waste heat boilers when the interim HWC MACT standard was implemented, due in part to borderline economics of the unit operation.  Replacement with a rapid quench changed the process, in some respects for the better, and eliminated dioxin “reformation” in the intermediate flue gas temperatures experienced in the boiler.

 

Environmental issues with a waste heat boiler are most often led by the reaction kinetics allowing dioxins and furans to form (and remain) in the combustion gases as they pass through a temperature range present in some portion of the boiler.  This window is generally described as 400-700oF (2).  Where chlorine is present, even in low concentrations, dioxin can be found at measurable levels when the detection (and emissions) limit is in the part per trillion range.

 

Many BIF boilers were retrofit, where a conventional fossil fuel boiler was equipped with waste-burning equipment, and these boilers often operate with little or no gas cleaning equipment.  Any waste burned in these retrofit boilers, therefore, needs to be increasingly clean (C, H, O and small amounts of N) to meet emissions regulations.  State and other air permits often add emissions constraints – such as NOx and SO2 – in areas that the HWC MACT has never addressed.

 

Boiler operating problems – measurable as efficiency, on-stream time, and even maintenance costs – relate to fouling of the heat transfer areas in the boiler and corrosion.  Steam is a basic utility in most manufacturing plants, and reliability of supply is paramount.  Backup boilers, firing natural gas or other fossil fuel only, often supplement the waste-fired boiler to ensure that reliability when contaminants in the waste put the boiler at risk.  Co-firing (waste and fossil fuel) is common with waste making up a percentage – often directly constrained in the permit – of the total fuel consumption.  The avoided fuel cost must be substantial to justify the capital and maintenance costs associated with the complexity added to steam system operations.

 

In new process design, installation of a boiler at a combustor exit allows downstream gas cleaning equipment, and cooling towers, to be smaller.  The resulting savings - added to the very real value of process steam - can help justify the capital cost of a boiler.

 

WASTE AS FUEL

 

Boilers in hazardous waste service have been a standard in Europe for many years.  Energy recovery is essentially a permit requirement for waste combustion – commercial or “captive” - installations.  To accommodate the contaminants and challenges of such service, the boilers deployed are generally quite large.  Notable are large inlet radiant sections to cool flue gases, and freeze any slag, before the gases enter convection and superheater sections with much more intimate contact between the gases and steaming surface area.  This at least minimizes fouling of the tubes.  Activated carbon injection is nearly standard, to capture both dioxins and mercury, in a gas cleaning train that is as complex as anything to be found on a US incinerator.  Catalytic (and sometimes non-catalytic) NOx control is common in these plants as well.

 

At the peak of the regulatory impact, over seventy boiler sites in the US carried BIF permits.  As many as twenty have shut down for various reasons, including the complete shutdown of manufacturing on a site.  Costs of compliance and more often maintenance (vs. fuel savings generated) have had a role in some of the shut downs of waste-fired boilers.  This trend may reverse with the combination of fuel cost instability and the potential for CO2 emissions regulation.  Recouping process steam from the heat released by proper destruction of a waste is arguably avoidance of fossil-fuel based CO2 generation.  Reflecting this in whatever regulatory structure emerges will be a miniscule – but meaningful – reflection of the debate between common sense and environmental zeal.

 

EPA has promulgated a full suite of tighter HWC regulations that include boilers (separated as solid- vs. liquid-fueled, and by existing vs. new/reconstructed).  These regulations build on the interim HWC MACT regulations by combining all of the combustors into a common, and presumably consistent, set of regulations.

 

The boiler limits are summarized in Table 1.  Some parameters are applicable only to a major emissions source, and others are bi-modal with lb/MMBTU or other derivative limits.  NA means not applicable in the table, though the solid-fuel boilers dioxin/furan (D/F) emissions are considered to be controlled by other organics limitations.  Semi-volatile metals (SVM; lead and cadmium) and low volatility metals (LVM; arsenic, beryllium and chromium) are the same metals grouping deployed in the interim HWC standard, although, the liquid-fueled boiler regulation only considers chromium as needing measurement.

 

Table 1. MACT Emissions Standards for Hazardous-Waste Fired Boilers

HWC Boiler MACT Limits

Solid-

Fueled

Liquid-

Fueled

 

Existing

New

Existing

New

D/F  (ng TEQ/dscm)

NA

NA

0.40

0.40

Mercury  (ug/dscm)

11

11

19

6.8

Particulate Matter (PM; gr/dscf)

0.03

0.015

0.035

.0087

SVM  (ug/dscm)

180

180

150

78

LVM (ug/dscm)

380

190

370

12

Total Chlorine  (ppmv)

440

73

31

31

 

The standards also include a CO limit of 100 ppmv, a total hydrocarbon limit of 10 ppmv, and a destruction and removal efficiency (DRE) minimum of 99.99%. 99.9999% DRE is mandated for Toxic Substances Control Act waste feeds.  Demonstrating this suite of compliance parameters is just as complicated for a boiler as for an incinerator, and the testing cost can be a factor in steam economics.

 

The MACT limits establish levels of contaminants in the waste feed to the boiler, directly in some cases, where a pound per million BTU of fuel limit is offered as an alternative to the concentration limit tabulated above.  Many boilers are not equipped with air pollution control equipment, so content of the waste feed of any inorganic constituent constrained by an emission limit is the only means of compliance.  The low removal efficiency of early air pollution control devices like cyclones and dry electrostatic precipitators can also result in tight feed limits on some constituents.

 

From a waste chemistry standpoint, then, a wide variety of contaminants are of interest for compliance reasons.  Adding maintenance concerns to this, elements of concern are tabulated below.  In a sense the entire periodic table is of interest.  H, O, and C are rarely so, though soot can be a PM problem.  N, to the extent that it yields NOx in combustion, can also be problematic.  Inert compounds form ash and PM from essentially every other element.  Particular concerns are highlighted below.

 

Table 2. Minor Waste Constituents and Specific Concerns

Acid Formers

Cl

HAP Emission

 

Cl, F, Br, I, S, P

Acid Dew Point

Ash

Every element except noted above

PM Emission

Metals

Hg, As, Be, Cr, Cd, Pb

MACT Emissions

 

Ni, Sb, Ba, Ag, Ti, Zn

Risk Drivers

 

V, Na, K

Slag Formers

 

With the key material of construction in all boilers being carbon steel, the corrosion issue can take an equal role in consideration of waste-firing, to that of compliance.

 

BOILER DESIGN

 

Many BIFs were pressed into waste combustion service as an economy measure, to reduce disposal costs for waste and purchased fuel costs at the same time.  An existing boiler was modified to allow co-combustion or outright fuel replacement from a new supply.  The same issues discussed in this section apply to a retrofit, as would apply to selection and detailed design of a new boiler.  With retrofit, however, whichever constraints emerge become limits on the boiler’s capability, not design stipulations.  The original boiler fuel – coal, oil (#6 or #2), or coal – defines the degree to which waste firing fits in most cases.

 

Boilers as a rule are fired with these cleaner fuels.  Natural gas is preferred, oil (#6 down to #2) is generally a back-up, and coal is of course an inexpensive alternative that has been pushed out of many industrial applications.  Where hazardous waste displaces coal as a fuel, in fact, emissions as a whole tend to be reduced.

 

The choice between fire-tube and water-tube design is rarely more than one of size and fabrication capability.  Fire-tube design can limit process gas corrosion issues a bit, though the steel tubes are still exposed to these gases.  Water-tube boilers are much more scalable and widespread, especially in Europe.  The radiant section of these boilers is considered a luxury in many cases, but can make the difference with respect to slag.  An alternative to this sizing issue might be a modest pre-quench on the boiler inlet to freeze slag and minimize sticking on the steaming surfaces.  Superheat and convection sections of the boiler are otherwise designed for optimal output steam pressure and temperature.

 

The key point in boiler design may be that early stage cooling of the incinerator exit gases, either in the boiler itself or in a pre-cooling stage, must provide the combination of cooling and time to freeze any slag particles in the gas.  This will preclude those particles sticking to the tubes, which would result in lost heat transfer and corrosion.

 

Another key point is that the flue gas acid dew point will be lower, and more variable, than that experienced in a typical fossil-fuel fired boiler.  Avoiding cold spots - on boiler tubes, the shell, and even exit ductwork - is important.  Both eutectics and dew point will be more variable when waste feed are involved, than with the much more consistent fossil fuels.

 

The actual burners of course must be designed for service with the quantity, heat content, and even viscosity of the waste to be burned.  Combustion air and atomization fluid are crucial to ensure adequate combustion.  Most standard gas and oil burner firebox designs are quite compact, and the uncooled residence time quite small (less than 0.1 second).  This may limit the waste throughput in a retrofit, or need to be expanded for a new boiler design.  The two second rule of thumb for residence time is excessive for many wastes, but is a de facto standard for many permitters.  Careful evaluation is possible with both historical data and advanced design tools to optimize this for capital cost vs. capability.  Destruction and removal efficiency calculation tools (3) can be utilized even in design to assure performance of the burner.  Once operating, burner condition – down to the many details of air/fuel mixing – becomes important as carbon soot will add to particulate emissions making it possible for total PM emissions to actually exceed the ash content of the waste feed.

 

Operating pressure of the boiler can be driven by needs in the steam consumer that the boiler will supply.  In that pressure and operating (especially exit) temperature of the boiler process-side are inter-related, this variable can otherwise be manipulated to minimize corrosion and sometimes D/F formation.  A low exit temperature will optimize heat recovery, often using an economizer at the end of the boiler, but this can lead to dew point corrosion.  A high exit temperature might reduce D/F formation, but this leaves much available heat unrecovered and is unlikely to completely avoid compliance issues.

 

Cold spots in the boiler are problematic.  Acid gas dew point – depending on the contaminants in the waste – can be devastating.  Acid condensation on the steaming surfaces in general is especially problematic, but this can be a localized problem as a function of fouling too.  Economizers are often not recommended for this very reason.  Even the boiler shell – if insulation is compromised or inadequate – is at risk from acid exposure.  Attention to detail, even flow regimes inside the unit, is crucial to minimize this problem.

 

Soot blowing can be complicated by slag issues, as frozen slag is amorphous and hard to remove.  Waste feed as discussed previously can minimize this problem.  This then becomes a design detail that can add to the efficiency and life of the boiler.  Some boilers resort to scheduled shutdowns where the tubes are manually cleaned, even with a fire hose.  Here again, clean fuel is ideal and ash contaminants add to costs and downtime.  Additives (to the waste or directly into the flue gas) can be utilized to modify the eutectic point of ash/slag, but must be managed carefully for cost and chemistry reasons.

 

EXPERIENCE

 

The specific technology of sulfuric acid regeneration may be a special case, but one that has become a force in the commercial waste market, at least for waste liquid incinerables.  This technology requires a boiler to minimize water evaporated into the process gas even without waste firing into the process gas.  The high SO2 process gas is converted to H2SO4, and excess water tends to dilute the net acid strength a plant is capable of delivering (with a target of essentially 100% acid).  Rapid quench would so overload the flue gas with water as to make the process non-functional.

 

Corrosion in a sulfuric acid boiler is a concern – SO3 dew point made early designs with economizers problematic.  Even before hazardous waste became a fuel in the acid furnaces, localized deposition and dew point (mostly SO3) in cold spots would generate corrosion quite quickly.  The acid regeneration business has standardized on a modular boiler design that allows for rapid maintenance, both preventative and in response to an actual tube failure.  With the preventative approach, tube failures are minimized.  Modular tube bundles are kept as spares, and even shared amongst plants, for both planned outage deployment and even emergency repairs.  While this option is not helpful to a single boiler, the quick turnaround partial retubing potential of the modular design can be helpful in maximizing boiler on-stream time.  If this is a key issue for the plant in question, the extra cost of this design should be considered.

 

In the late 1980’s, extensive ash modification testing was conducted in a laboratory, and then confirmed in sulfuric acid regeneration waste heat boilers.  There were two aspects of waste ash content studied, the percentage of ash generated and the characteristic of the ash.  These aspects vary based on the constituents of the ash and their concentrations.  The standard waste ash analysis reports the percentage of ash that will be generated during combustion.  The ash characteristic developed in this laboratory program is a visual parameter examined following the completion of an ash analysis.  The visual observation includes the ash color and the consistency of the ash, which can range from glazed or adherent, to friable or granular, to light or fluffy.

 

Table 3. Qualitative Ash Categories

Glazed/Adherent

Shiny, sticks to porcelain

Na, K especially

Friable/Granular

Hard, must be scraped off

V especially

Light/Fluffy

Not dense, loose

Si for example

 

The visual observation of ash is an inexpensive indicator of the composition and the qualities of that ash.  Wastes having sodium and potassium generate a glazed/adherent ash characteristic due to the fact that these oxides significantly lower the melting point of ash.  The resulting molten ash aggressively attacks refractory and boiler tubes.  Wastes having greater than 1,000 ppm sodium and/or potassium will yield glazed/adherent ash that is extremely corrosive to boilers.  Wastes having vanadium at a concentration of 1,000 ppm or more can also lower the melting point of ash.  These three elements are the most common slag formers as they relate to boiler operation.

 

Lowered eutectic ash can be modified with the addition of aluminum, magnesium, manganese, or a combination of these metals.  These ash modification additives elevate the melting point of the ash.  Different additives produce varying percentages of ash and varying ash characteristics.  The laboratory testing revealed that the amount of the additives necessary to yield light or fluffy ash ranged from 0.1 to 1.0% by weight.  The goal of the laboratory testing was to develop a repeatable procedure to select an additive or combination of additives that would generate a minimum percentage of ash, but having a light and fluffy ash characteristic.

 

This procedure was also used on wastes containing phosphate.  During the combustion of phosphate containing wastes, “metaphos”, a mixture of phosphorous oxides, is formed in the gas phase.  Metaphos condenses on the surface of the boiler tubes and is extremely corrosive.  The operating experience of the sulfuric acid regeneration plants revealed that the combustion of wastes containing greater than 1,000 ppm of phosphate caused severe corrosion and boiler tube failures within a matter of hours due to this corrosion.

 

Research was conducted in the laboratory to experiment with ash modification additives to bind the phosphate in an ash as a means to eliminate metaphos from the gas phase.  Success was achieved with a combination of magnesium and manganese to yield that target: a minimum percentage of ash that was light and fluffy.  The laboratory testing was confirmed in the sulfuric acid regeneration furnaces followed by waste heat boilers.  The additives were fed to the furnace and the dew point was measured constantly in the waste heat boiler.  The combination of magnesium and manganese at a concentration of 0.1 to 0.2 % by weight enabled the combustion of a waste stream containing 1 to 2 % phosphate with no lowering of the dew point in the boiler and no corrosion of the boiler tubes.

 

In summary, the chemistry of the waste is a major consideration to take into account prior to its combustion in a boiler.  It is possible to alter the chemistry to avoid serious corrosion issues and slagging issues that affect the economics and the operations of boilers.  This experience is still used today in assessing the viability of new wastes fired in permitted boilers.

 

In boilers where coal is the original (and co-burned) fuel, much experience has been that the greater the fraction of heat release from waste, the lower the emissions.  Even #6 fuel oil has contaminants that make many waste liquids an attractive alternate fuel.  With these two fuels especially, soot formation in the flame zone can easily add to stack PM levels.

 

Enhanced mechanical maintenance in the burner will minimize soot (and PM), leaving the allowable emissions as a limitation on the waste feed alone.

 

Ultimately, MACT regulations may drive some waste-fired boilers into the addition of gas cleaning equipment.  The baghouse would appear to be state-of-the-art for any such upgrades.  Gas temperature regime is in the right range, and additional items appended to this technology can control acid gases, dioxin, and even mercury emissions.  Furthermore, the creation of a modest quantity of dry waste is generally less environmentally negative than a blowdown water stream that must be treated further.

 

THE FUTURE OF WASTE AS BOILER FUEL

 

Greenhouse gas emissions regulation, focused on CO2 for the most part, will tend to make waste-to-energy, even hazardous waste-to-energy, more attractive.  A process that requires steam will utilize fossil fuel to produce that steam.  Even co-generation, which can be modestly more efficient than simple boiler operation, will be a significant source of CO2 to an operating facility.  If that same process yields hazardous waste that by regulation must be incinerated, a second CO2 source will be quantified for the process at some stage of the operation, even if the waste is shipped off-site.  Using the waste as boiler fuel will reduce the first source of CO2 with no change in the second.

 

Combustion with energy recovery is clearly superior to simple combustion in that regard.  Tracking CO2 emissions alone will quantify this potential, and perhaps even push some development of energy recovery from waste.  Further regulation - with financial impact - either through direct taxation or cap-and-trade, will add that incentive to push further development here. 

 

The existing “fleet” of hazardous-waste-fired boilers in the US is estimated to recover over 10 million pounds per hour of steam, the equivalent of 1,250 MW - one large power plant.  Expanding this output will be an economic positive and provide CO2 emissions reductions.  Shutting down this capability would incrementally increase our national CO2 production and is clearly a step in the wrong direction.

 

Gasification has often been offered up as a potential improved technology.  The syngas produced by partial combustion can be cleaned in a smaller gas cleaning train than would be required for full combustion.  This allows a large fraction of the energy content of that waste stream to be burned in a gas turbine before the boiler, in a boiler without gas cleaning, or even to be used in a process as the syngas directly.  These technologies are approaching technical feasibility, but are challenged economically with high capital costs.  “New” capital is even more challenged by the retrofit redeploying old capital at modest cost.  Permit questions will abound even if capital issues are not a constraint on deployment of this technology. 

 

MACT regulations are challenged in many technology areas, especially boilers.  The standards in the waste combustor MACT, as they relate to boilers, contrast with the incinerator emission limits as one extreme and the proposed industrial boiler MACT standard as the other.  The focus is consistent on the art of EPA, but the boiler MACT does not include Dioxin, most metals, and DRE at all - fairly logical with units burning conventional fuels.

 

In that EPA has been legally challenged by both industry and the environmental community on boiler MACT regulations, the current focus on CO2 emissions may be a welcome relief to the regulators.  To the extent that the focus is under a myriad of challenges, it is not clear where the regulations will go.  What is clear is that energy recovery from waste sources - hazardous in this case - brings both economic and environmental benefits when managed properly.  The use of boilers in this effort can be a positive if attention is paid to the fundamentals of chemistry and design.

 

CONCLUSIONS

 

Waste combustion with energy recovery, then, should take a larger role in manufacturing.  Steam value, by a combination of rising fuel costs and restraints on CO2 emissions, should trend upwards.  Maintenance costs on a waste-fired boiler, on the other hand, can be reduced by application of equal measures of common sense and technology, as discussed here.  The environmental challenges of D/F and PM emissions can be addressed with proven, cost-effective technologies.  Ultimately, the CO2 emissions issue may dominate the decision but even now economics can be a positive in any assessment of hazardous waste to energy opportunities.

 

REFERENCES

 

1.      www.epa.gov/osw/hazard/tsd/td/combust/finalmact/; and associated links

 

2.      Vehlow, Jürgen; “Dioxin in Waste Combustion - Conclusions from 20 Years of Research”; proceedings, Bioenergy Australia; Melbourne, Australia; 12/2005

 

3.      Lamb, Charles et al; “Theoretical DRE Calculations for Incinerators and Thermal Oxidizers”; proceedings, Incineration and Thermal treatment Technologies conference; New Orleans, LA, USA; 5/2002