Georgia Tech Home PageIPST Home Page
Georgia Institute of TechnologyMaterials Science and EngineeringPaper Science and TechnologyGT MapGT DirectorySite MapWebsite Use InstructionsSearch Engine
  Main Menu

About Us
Group Publications and Presentations
P&P Related Corrosion Publications and Presentations
Fun Stuff


Corrosion Problems in Recovery Boilers

Effects of Mill Closure on Equipment Corrosion

Environment-Sensitive Fracture of Suction Rolls
Top of page
Top of page

Pulp and Paper Related Corrosion Publications and Presentations

Corrosion Problems in Recovery Boilers

The objective of corrosion research at IPST is to understand the causes of high temperature corrosion and stress corrosion cracking of the waterwall tubes in the lower furnace of the kraft recovery boiler. Better understanding of these causes is necessary to develop successful corrosion mitigation strategies. Different types of material-related problems on the fireside have been observed and reported by various mills, and corrosion has been implicated in several smelt-water explosions over the years. Obviously, safe operation of the boiler is of prime concern, so the objective of the research program at IPST is to develop a sound understanding of the general and specific corrosion problems of kraft recovery boilers.

  1. Monitoring of Corrosive Environments in Recovery Boilers
  2. Corrosion Kinetics Database
  3. Factors Affecting High Temperature Corrosion of Waterwall Tubes in Kraft Recovery Boilers (Experimental Study)
  4. Probe to Monitor Fireside Corrosion in Recovery Boilers
  1. Monitoring of Corrosive Environments in Recovery Boilers

    Corrosion of the kraft recovery boiler lower furnace is a continuing problem in the pulp and paper industry. Although, the use of stainless steel composite tubes has reduced corrosion rate in the lower furnace area, the problem of composite tube cracking in floor and near- floor waterwalls is forcing manufacturers and users to consider new construction materials or barrier coatings for waterwall tubes in the lower furnace areas. It is known that the fireside corrosion of waterwall tubes in the lower furnace is due to high temperature sulfidation. However, very little is known about the environments to which waterwall tubes are exposed in the lower furnace during recovery operation. The environment of importance to waterwall corrosion is not the bulk environment but the local environment at/near the tube surface. This environment can be very different from the bulk environment in the lower furnace and depends upon various factors like presence of smelt, on-wall liquor pyrolysis, other local reactions, and differences in local tube surface temperatures.

    Corrosion in any system depends upon two major factors material used and environmental conditions. Different materials may corrode at different rates in a given environment. However, waterwall tubes made of the same material may experience different corrosion rates in different areas of the lower furnace. Regular tube thickness inspections reveal that areas with consistantly very low or moderate rates may exist very close to an area with high corrosion rates. Local environments, which come in contact with the tube surface, may differ in different areas. Localized environment for the waterwall tube corrosion includes tube surface temperature, localized gas composition at the tube surface, and smelt composition at the tube surface. It must be reemphasized that even though bulk environmental parameters affect local environments at tube surface, it is the local environment next to the metal surface that controls the corrosion of waterwall tubes from the fireside.

    Variables of corrosive environments like local gas composition, tube surface temperature, and smelt composition are studied. Real-time bulk gas composition is analyzed in different areas of the boiler with a “Portable” gas chromatograph with an on-line gas sampler. Gas sampling ports were designed and installed in lower furnace areas of interest. The temperature of waterwall tubes is continuously measured in selected areas of the recovery boiler by installing wall-embedded thermocouples. Smelt samples are also collected from the tube surfaces in selected areas and will be analyzed. These studies have given us new insight into the corrosive environments inside the recovery boiler. This work will be continued in other recovery boilers with different designs and firing practices.

  2. Corrosion Kinetics Database

    As a part of the DOE funded project, this task involved designing and developing a corrosion kinetics database. The corrosion kinetics database was developed at IPST with two main objectives in mind:

    1. to collect published information on corrosion kinetics, relevant to recovery boilers, in a form of database,
    2. to develop a database which may be used with the indirect corrosion measuring techniques (e.g., thermocouples and/or gas sensors) for the prediction of corrosion in kraft recovery boilers.

    The database was designed using a commercial Microsoft Accessâ package. This database is an icon-driven, user-friendly, windows-operated system. Published corrosion kinetics data were collected from gas, coal gasification, pulp and paper, and other industries, as well as from other fundamental studies. An on-line database search as well as back-referencing were used to identify and collect the relevant published papers for the corrosivity database. The database already has over 12,000 data points, representing data from more than 300 materials and 400 environmental conditions relevant to the recovery boiler fireside environments. Two papers showing the capabilities and use of this database for corrosion prediction and material selection for recovery boilers have been submitted for publication. One paper was submitted for the International Recovery Boiler Conference and the other for the International Symposium on Corrosion Problems in the Pulp and Paper Industry.

  3. Factors Affecting High Temperature Corrosion of Waterwall Tubes in Kraft Recovery Boilers

    The published literature covers most of the relevant alloys and environmental conditions similar to kraft recovery boilers. However, there are a significant number of gaps in the published data on the environmental/material combinations, which are very important for this database. To fill in these gaps, experiments were conducted at IPST to generate this data. Three different types of tests were identified: like isothermal tests, tests with different smelt compositions, and cyclic temperature tests. Data generated from these tests were also entered into the database. Experimental results have provided us with a better understanding of the effects of various factors on the corrosion in the lower furnace of kraft recovery boilers.

    Smelt/Gas Interactions and their Effects on Waterwall Tube Corrosion
    In Phase II, preliminary work was started at IPST to generate an understanding of the effects of smelt on the gaseous corrosion of carbon steels. Results from this effort have already been discussed in the section on Task II of the member-dues funded project. These efforts will be extended into Phase III  (1997-98) of the DOE project so that we can predict the local environments at the interface by knowing the smelt and bulk gas compositions in a particular area of the recovery boiler.

    Effects of Temperature Excursions on Waterwall Tube Corrosion
    Waterwall tube surfaces in the kraft recovery boilers experience temperature excursions. These excursions can increase the corrosion rate in the localized areas of the recovery boiler. Occasional temperature excursions can influence the corrosion of the boiler tubes on fireside by one or a combination of the following mechanisms:

    1. increase in diffusion controlled reaction rates and overall sulfidation reaction rate,
    2. microcracking of the semi-protective sulfide scale due to thermal stresses and increased reaction kinetics,
    3. molten salt corrosion (if the molten smelt is in touch with the fireside tube).

    When the temperature excursions are more than the melting point of smelt then the molten salt corrosion will increase metal wastage significantly. However, excursions above the melting point of frozen smelt in the lower furnace areas have not been reported. In the present series of tests, SA-210 carbon steel samples were tested in 1% H2S gas mixture tested at 320°C spiked to  480°C at different time locations. The results from these tests have shown that once damaged by the thermal excursion, the sulfide scales do not effectively heal in this system during the post-spike exposure. Results from this study have very important implications from the recovery boiler point of view in that even a couple of infrequent spikes, which cause scale damage through microcracking, followed by a long post-spike exposure will lead a to very high corrosion rate. The corrosion of waterwall tubes due to infrequent thermal spikes may even be higher than the isothermal exposure at higher temperature of spike.

    General conclusions from this series of experiments are as follows:

    • Thermal excursions generally lead to higher sulfidation rates compared to equivalent tests without spikes.
    • Scale damage may depend upon scale thickness, history of exposure, and spike location.
    • Sulfide scale damage due to thermal excursion may not heal and may lead to very high corrosion rates in the post-spike exposure. Even a couple of infrequent spikes, which lead to scale damage, may lead to even higher corrosion rates than the isothermal tests at upper temperature of the spike.

  4. Development of Probe to Monitor Fireside Corrosion in Recovery Boilers

    Electrical resistance probes were developed, constructed, and tested at IPST. Probes were tested under different environmental conditions relevant to recovery boilers. The main objective of this study is to check the feasibility of this technology for corrosion measurement, and to check the parameters such as stability of signal, reproducibility, etc. SA210 coupons were also tested along with the probe to compare the probe output signal with the coupon weight loss results. The correlation between the results from the weight loss tests and the probe output signal was very good and the results were reproducible.
  Georgia Tech Home Page