Fouling in Refineries

By James G. Speight

Fouling in Refineries is an important and ongoing problem that directly affects energy efficiency resulting in increased costs, production losses, and even unit shutdown, requiring costly expenditures to clean up equipment and return capacity to positive levels.

This text addresses this common challenge for the hydrocarbon processing community within each unit of the refinery. As refineries today face a greater challenge of accepting harder to process heavier crudes and the ongoing flow of the lighter shale oil feedstocks, resulting in bigger challenges to balance product stability within their process equipment, this text seeks to inform all relative refinery personnel on how to monitor fouling, characterize the deposits, and follow all available treatments.

With basic modeling and chemistry of fouling and each unit covered, users will learn how to operate at maximum production rates and elongate the efficiency of their refinery’s capacity.

• Presents an understanding of the breakdown of fouling per refinery unit, including distillation and coking units
• Provides all the factors, crude types, and refining blends that cause fouling, especially the unconventional feedstocks and high acid crudes used today
• Helps users develop an analysis-based treatment and control strategy that empowers them to operate refinery equipment at a level that prevents fouling from occurring

• Language: English
• Publisher: Elsevier Science
• Release date: May 14, 2015
• ISBN: 9780128011454

James G. Speight

Dr. Speight is currently editor of the journal Petroleum Science and Technology (formerly Fuel Science and Technology International) and editor of the journal Energy Sources. He is recognized as a world leader in the areas of fuels characterization and development. Dr. Speight is also Adjunct Professor of Chemical and Fuels Engineering at the University of Utah. James Speight is also a Consultant, Author and Lecturer on energy and environmental issues. He has a B.Sc. degree in Chemistry and a Ph.D. in Organic Chemistry, both from University of Manchester. James has worked for various corporations and research facilities including Exxon, Alberta Research Council and the University of Manchester. With more than 45 years of experience, he has authored more than 400 publications--including over 50 books--reports and presentations, taught more than 70 courses, and is the Editor on many journals including the Founding Editor of Petroleum Science and Technology.

Book preview

2006.

Chapter 1

The Concept of Fouling

Abstract

Fouling as it pertains to petroleum is deposit formation, encrustation, deposition, scaling, scale formation, slagging, and sludge formation which has an adverse effect on operations. It is the accumulation of unwanted material within a processing unit or on the on solid surfaces of the unit to the detriment of function. For example, when it does occur during refinery operations, the major effects include (1) loss of heat transfer as indicated by charge outlet temperature decrease and pressure drop increase, (2) blocked process pipes, (3) under-deposit corrosion and pollution, and (4) localized hot spots in reactors, all of which culminate in production losses and increased maintenance costs.

It is the purpose of this chapter to introduce the reader to fouling in the petroleum industry and to describe the different types of fouling that can occur.

Keywords

Fouling

Mechanisms

Rate of fouling

Fouling factor

Fouling potential

1.1 Introduction

Fouling, as it pertains to petroleum refineries (Speight, 2000; Speight and Ozum, 2002; Parkash, 2003; Hsu and Robinson, 2006; Gary et al., 2007; Speight, 2014a–e), is deposit formation, encrustation, deposition, scaling, scale formation, slagging, and sludge formation which has an adverse effect on operations. It is the accumulation of unwanted material within a processing unit or on the on solid surfaces of the unit to the detriment of function. For example, when it does occur during refinery operations, the major effects include (1) loss of heat transfer as indicated by charge outlet temperature decrease and pressure drop increase, (2) blocked process pipes, (3) under-deposit corrosion and pollution, and (4) localized hot spots in reactors, all of which culminate in production losses and increased maintenance costs. In addition, the term macrofouling if often used to generally describe the blockage of tubes and pipes while, on the other hand microfouling is generally iced to describe scaling on the walls of the tubes and pipes. Again, the outcome is a loss of efficiency and output to the refinery.

Fouling during production or transportation or refining can occur in a variety of processes, either inadvertently when the separation is detrimental to the process or by intent (such as in the deasphalting process or in the dewaxing process). Thus, separation of solids occurs whenever the solvent characteristics of the liquid phase are no longer adequate to maintain polar and/or high molecular weight constituents in solution. Examples of such occurrences are: (1) separation of asphaltene constituents, which occurs when the paraffin nature of the liquid medium increases, (2) wax separation which occurs when there is a drop in temperature or the aromaticity of the liquid medium increases, and (3) sludge/sediment formation in a reactor which occurs when the solvent characteristics of the liquid medium change so that asphaltic or wax materials separate, coke formation which occurs at high temperatures and commences when the solvent power of the liquid phase is not sufficient to maintain the coke precursors in solution, and sludge/sediment formation in fuel products which occurs because of the interplay of several chemical and physical factors.

Typically, the fouling material consists of organic and/or inorganic materials deposited by the feedstock that is deposited by the occurrence of instability or incompatibility of the feedstock (one crude oil) with another during and shortly after a blending operation (Speight, 2014a).

Blending is one of the typical operations that a refinery must pursue not only to prepare a product to meet sales specifications, but also to blend the different crudes and heavy feedstocks to produce a refinery feedstocks (Speight, 2000; Speight and Ozum, 2002; Parkash, 2003; Hsu and Robinson, 2006; Gary et al., 2007; Speight, 2014a–e). Although simple in principle, the blending operation must be performed with care and diligence with the regular acceptance by refineries of heavy feedstocks as part of the refinery slate. Lack of attention to the properties of the individual feedstocks prior to the blending operations can lead to asphaltene precipitation or phase separation (fouling) due to incompatibility of the different components of the blend (Schermer et al., 2004; Speight, 2014a–e). This would result in the occurrence of fouling deposits in heat transfer equipment and reactors as a substantial energy cost to the refinery (Stark and Asomaning, 2003; Van den Berg et al., 2003). Therefore, it is advisable for the refiner to be able to predict the potential for incompatibility by determining not only the appropriate components for the blend, but also the ration of individual crude oils and heavy feedstocks in the blend.

The compatibility of crude oils is generally evaluated by colloidal stability based on bulk composition or asphaltene precipitation (Mushrush and Speight, 1995, 1998; Asomaning and Watkinson, 2000). Typically, the test methods are performed under used to evaluate oil stability at ambient conditions, but applying the data to the potential for fouling under the actual parameters used in heat transfer equipment must be done with caution. Fouling is dependent upon not only the conditions of asphaltene separation fluid and the stability of the crude oil/heavy feedstock system (Chapter 4), but also on flow conditions and other parameters (Asomaning and Watkinson, 2000; Saleh et al., 2005; Stark and Asomaning, 2003; Derakhshesh et al., 2013). Fouling is concerned with not only asphaltene precipitation (Srinivasan and Watkinson, 2003). In addition, fouling can also be a consequence of corrosion in a unit when deposits of inorganic solids become evident (Speight, 2014b). With the influx of opportunity crudes, high-acid crudes, heavier crude oils, extra heavy crude oils, and tar sand bitumen into refineries (Chapter 2) fouling phenomena are more common and diverse (Speight, 2005, 2008, 2009, 2013a–c, 2014a).

In the petroleum industry, the components that may be subject to fouling and the corresponding effects of fouling are (1) the production zone of crude oil reservoirs and oil wells, which is reflected by a decrease in production with time though the formation of plugs which can lead to complete cessation of flow, (2) pipes and flow channels which results in reduced flow, increased pressure drop, increased upstream pressure, slugging in two-phase flow, and flow blockage, (3) heat exchangers surfaces, which results in a reduction in thermal efficiency along with decreased heat flux, increased temperature on the hot side, decreased temperature on the cold side, and under-deposit corrosion, (4) injection/spray nozzles (e.g., a nozzle spraying a fuel into a furnace or a reactor, in which the incorrect amount of feedstock is injected), and (5) within a reactor due to uncontrollable chemical and/or physical reaction. In addition, there is macrofouling and microfouling.

Macrofouling is caused by a coarse matter from either organic or biological or inorganic origin. Such substances foul the surfaces of heat exchangers and may cause deterioration of the relevant heat transfer coefficient as well as flow blockages. Microfouling is somewhat more complex and several distinctive events can be identified (Bott, 1995): (1) particulate fouling, which is the accumulation of particles on a surface, (2) chemical reaction fouling, such as decomposition of organic matter on heating surfaces, (3) solidification fouling, which occurs when components of a flowing fluid with a high-melting point freeze onto a subcooled surface, (4) corrosion fouling, which is caused by corrosion, (5) biofouling, which can often ensure after biocorrosion (Muthukumar et al., 2003), is due to the action of bacteria or algae, and (6) composite fouling, whereby fouling involves more than one foulant or fouling mechanism.

Fouling caused by the presence of particulate matter in the fluid is a common form of fouling, can be defined as the process in which particles in the process stream deposit onto heat exchanger surfaces (Al-haj et al., 2005; Müller-Steinhagen et al., 2002; Al-haj et al., 2006). These particles include particles originally carried by the feedstock before entering the heat exchanger and particles formed in the heat exchanger itself as a result of various reactions, aggregation, and flocculation. Particulate fouling increases with particle concentration, and typically particles greater than 1 μm size lead to significant fouling problems.

Fouling of a surface through the formation of deposits, does not always develop steadily with time. There may be an induction period when the surface is new or very clean and the foulant does not accumulate immediately. After the induction period, the fouling rate increases. On the other hand, there is also negative fouling which occurs when relatively small amounts of deposit can improve heat transfer, relative to clean surface, and give an appearance of a negative fouling rate and total amount of the foulant. After the initial period of surface roughness control or surface roughness adjustment, the fouling rate may become positive.

In asymptotic fouling, the fouling rate decreases with time, until it finally reaches zero and at this point, the deposit thickness remains constant with time. This often occurs when the deposits are relatively soft or poorly adherent deposits in areas of fast flow or turbulent flow and is usually assigned to the point as the deposition rate equals the deposit removal rate. However, accelerating fouling is almost the opposite since the fouling rate increases with time and the rate of deposit buildup accelerates with time until it becomes transport limited. This type of fouling can develop when fouling increases the surface roughness, or when the deposit surface exhibits higher chemical propensity to fouling than the pure underlying metal.

1.2 Fouling

The occurrence of fouling can (for the purpose of this text) be conveniently sub-divided into two categories: (1) fouling on surfaces such as on heat exchangers surfaces and (2) fouling in a reactor, such as the appearance of sediment in visbroken products or the deposition of carbonaceous sediment on to catalyst during catalytic process.

1.2.1 Fouling on Surfaces

One area where fouling occurs with some regularity is in heat exchanger systems (heat transfer systems) when feedstocks are being heated prior to entry into a reactor. Typically, the foulant will deposit as solid particle or as a semisolid gum-like material which after, contact with the hot surface, will bake and eventually solidify. This results in a loss of heat transfer capability with a subsequent decrease in the outlet temperature because of the lower thermal conductivity of the foulant. As a result, the thermal resistance to heat transfer from a hot fluid to a cooler fluid is increased and the thermal efficiency of heat exchangers are markedly reduced (Müller-Steinhagen et al., 2002). This is reflected in a cooler-than-desired feedstock (at a lower outlet temperature) passing into a distillation unit or reactor and the expected physical and chemical events do not occur. Instead, partial events occur and the desired products are not produced.

To combat such an adverse effect, the outlet temperature of the feedstock (from the heat exchanger) should be monitored closely and brought to the desired temperature by increasing the temperature of the furnace, with the associate increase in consumption of (and cost of) the furnace fuel. As an alternate means of outlet temperature correction, the heat exchanger surface area may have to be increased though installation of additional contact area (tubes, pipes, and flow channels), again at a cost while the foulant continues to buildup and reduce the efficiency of the unit. In fact, as the thickness of the foulant increases, the cross-sectional area of the flow channels is reduced (equivalent to arteriosclerosis in the human body). In addition, increased roughness of the surface of the foul will increase frictional resistance to flow which, in turn, leads to a more drastic pressure drop across the unit and thence to flow blockage.

The answer to such an event has been, on some occasions, to dilute the feedstock with an appropriate solvent (again, using an anatomical analogy, analogous to blood thinning) but use of a thinning solvent may only increase the inevitable and, at some time, the unit will have to be shutdown and cleaned. In addition, monitoring the pressure drop is not always as good an indicator of the onset of fouling as heat transfer data. In situations where blends change and the constituents of one blend may contain more heavy oil (or extra heavy oil or tar sand bitumen) than another blend significant differences flow rates (with the accompanying differences in pressure data) are to be anticipated are experienced. To combat such anticipated changes, it will be necessary to apply flow corrections (if the data for standard operations have been defined and are available) to both pressure drop and to heat transfer calculations to normalize the abnormal feedstock data to a standard flow.

1.3 Parameters Affecting Fouling

The fouling process is a dynamic and variable process insofar as several operational and design variables have well-defined effects. These parameters include the fluid flow velocity, the fluid properties, the surface temperature, the surface geometry, the surface material, the surface roughness, and the suspended particles concentration and properties.

Parameters of importance are related to operating conditions and equipment design, such as (1) fluid flow velocity, (2) surface temperature, (3) surface materials, (4) surface roughness, and last but certainly not least (5) fluid properties. All these and other factors that may affect fouling need to be considered and taken into account in order to be able to prevent fouling if possible or to predict the rate of fouling or fouling factor prior to taking the necessary steps for fouling mitigation, control, and removal.

The fouling factor is a means of measuring the performance of a heat exchanger which, in turn is a way of measuring the performance as it deteriorates with time (Gudmundsson, 2009). The fouling may be due to the accumulation of organic material, mineral deposits, rust, or the presence of microorganism on the heat transfer surfaces. These deposits increase the resistance of heat transfer and cause a decrease in the efficiency of the unit. The resistance is usually represented by a fouling factor, Rf, which measures the thermal resistance introduced by the action of the foulant. The development of fouling depends on number of things, major groups of fouling dependents are (1) composition of the fluids, (2) operating conditions in the heat exchanger, (3) type and characteristics of the heat exchanger, (4) location of the fouling deposit, and (5) presence of microorganism (Bansal and Chen, 2005; Bohnet, 2005; Cengel and Turner, 2005; Rizzo et al., 2005). However, the occurrence of an induction period before a noticeable amount of mineral deposits has formed so the overall heat transfer coefficient changes noticeably and thence the rate of fouling increases during the fouling period. In addition to the parameters mentioned above, it is necessary to consider the effect of fouling during the design of heat exchangers so that the units can withstand the effect of fouling up to a certain point without becoming harmful for the intended process (Gudmundsson, 2009).

1.3.1 Fluid Flow Velocity

The flow velocity has a strong effect on the fouling rate since it affects both the deposition and removal rates through the hydrodynamic effects at the surface of heat exchangers.

In the refinery, the shell and tube heat exchanger is a commonly used design class of heat exchanger (Kakaç and Liu, 2002) and is aptly suited for high-pressure applications. This type of heat exchanger consists of a large pressure vessel (the shell) with a bundle of tubes inside (Figure 1.1). One fluid runs through the tubes, and another fluid flows through the shell and over the tubes and heat is transferred between the two fluids. The conventional segment baffle geometry is largely responsible for higher fouling rates. Uneven velocity profiles, back-flows, and flow effects generated on the shell side of a segmentally baffled heat exchanger results in higher fouling and shorter run lengths between periodic cleaning and maintenance of tube bundles.

Figure 1.1 Schematic representation of a straight tube heat exchanger.

On the other hand, the flow velocity has indirect effects on deposit strength, the mass-transfer coefficient, and the adherence of the foulant to the surface. Increasing the flow velocity tends to increase the thermal performance of the exchanger and decrease the fouling rate—a uniform flow of process fluids past the heat transfer surface favors less fouling. Foulants suspended in the process fluids will deposit in low-velocity regions (such as pipe elbows unless the flow is turbulent), particularly where the velocity changes quickly, as in heat exchanger water boxes and on the shell side.

1.3.2 Surface Temperature

Generally, the rate of fouling is temperature dependent with different rates of fouling between the feed inlet and outlet sides of the heat exchanger and fouling will increase with an increase in temperature. This is due to a baking on effect, scaling tendencies, increased corrosion rate, faster reactions, crystal formation and polymerization, and loss in activity by some anti-foulants. Lower temperatures produce slower fouling buildup, and usually deposits that are easily removable. However, for some process fluids, low surface temperature promotes crystallization and solidification fouling. As expected, biological fouling is strongly dependent on temperature—there is a temperature below which reproduction and growth rate are arrested and a temperature above which the organism becomes damaged or killed. If, however, the temperature rises to an even higher level, some heat sensitive cells may die (Mukherjee, 1996).

1.3.3 Surface Material

The selection of surface material is significant to deal with corrosion fouling—carbon steel is corrosive, but least expensive while copper exhibits biocidal effects in water and its use is limited in certain applications. Noncorrosive materials such as titanium and nickel will prevent corrosion, but they are expensive and have no biocidal effects. Glass, graphite, and Teflon tubes often resist fouling and/or improve cleaning, but they have low thermal conductivity. Although the construction material is more important to resist fouling, surface treatment by plastics, vitreous enamel, glass, and some polymers will minimize the accumulation of deposits.

1.3.4 Surface Roughness

Surface roughness has been noted to have an enhancement on fouling insofar as the rough surface provides sites that enhance laying down the initial deposits of foulant. Rough surfaces encourage particulate deposition and provide a good chance for deposit sticking. After the initiation of fouling, the persistence of the roughness effects will be more a function of the deposit itself. A less rough surface finish has been shown to influence the delay of fouling and ease cleaning. Similarly, nonwetting surfaces delay fouling. However, smooth surfaces may become rough in due course due to scale formation, formation of corrosion products, or erosion.

1.3.5 Fluid Properties

In terms of fluid properties, there are also the oft-forgotten chemical and physicochemical aspects of fouling. For example, the structure of the deposit, usually dictated by the chemical species that form the deposit, can lead to different effects, such as localized fouling, under-deposit corrosion of the substrate material, deposit tubercles, and sludge piles. The factors that are most likely to influence deposit structure (and the ensuing effects) include deposit composition and its porosity and permeability, which are all related to feedstock composition. Even minor components of the deposits can sometimes cause severe corrosion of the underlying metal such as the hot corrosion caused by vanadium in the deposits of fired boilers (Herro, 1989).

However, in addition to feedstock composition, the factors that govern fouling on surface are, in fact, changers are multi-faceted and varied. As already noted, some factors are related to the feedstock properties such as its chemical constituents, API gravity, viscosity, diffusivity, pour point, interfacial properties, and feedstock stability. The propensity of feedstocks to encourage fouling depends on properties such as viscosity and density. Viscosity can play an important role for the sublayer thickness where the deposition process is taking place. On the other side the viscosity and density (usually monitored as API gravity) have a strong effect on the sheer stress which is the key element in the foulant removal process.

Indeed, the chemical constituents (and their individual or collective behavior) are a particularly important factor that affects the rate and extent of fouling (Chapters 2, 6, and 7). Indeed, the presence in the feed of unsaturated and unstable compounds, inorganic salts, trace elements such as sulfur, nitrogen, and oxygen, as well as the storage conditions (e.g., exposure to oxygen during storage) will also affect nature of the foulant and the rate of fouling.

Moreover, refinery fluids are seldom pure—the intrusion of minute amounts of impurities can initiate or substantially increase fouling and these impurities can either deposit as a fouling layer or acts as catalysts to the fouling processes. For example, chemical reaction fouling may be due to the presence of oxygen and/or trace elements such as nickel, vanadium, and molybdenum. In crystallization fouling, the presence of small particles of impurities may initiate the deposition process by seeding. In addition, impurities such as sand or other suspended particles in the fluid may have a scouring action, which will reduce or remove deposits (Bott, 1990). Suspended solids promote particulate fouling by sedimentation or settling under gravitation onto the heat transfer surfaces. Since particulate fouling is velocity dependent, prevention is achieved if stagnant areas are avoided. For water, high velocities (above 1 m/s) help prevent particulate fouling. Often it is economical to install an upstream filtration.

In a heat exchanger, the fluid velocity is generally lower on the shell side than on the tube side, less uniform throughout the bundle, and limited by flow-induced vibration. Zero-or low-velocity regions on the shell side serve as ideal locations for the accumulation of foulants. If fouling is expected on the shell side, attention should be paid to the selection of baffle design—segmental baffles have the tendency for poor flow distribution if spacing or baffle cut ratio is not in correct proportions. Too low or too high a ratio results in an unfavorable flow regime that favors fouling.

1.4 Fouling Mechanisms

Fouling can be caused by a number of different mechanisms, which include (1) particles in the feedstock, (2) particle formation, (3) corrosion fouling, (4) coking, (5) aggregation and flocculation, (6) phase separation, (7) particle deposition, (8) deposit growth, aging, and hardening, and (9) auto-retardation and erosion or removal. In addition to these stages, the rate of fouling and the prediction of fouling factor must also be considered.

1.4.1 Particles in the Feedstock

Particles in the crude oil feedstocks originate from poor desalter performance (Chapter 8) or have been entrained during the distillation process (Chapter 8). These particles are, for the most part, insoluble inorganic particles such as dirt, silt, and sand particles, and other inorganic salts such as sodium chloride, calcium chloride, and magnesium chloride (which arise from poor desalter performance), corrosion products (iron sulfide and rust), catalyst particles, or fines. There may also contain some organic particles that may have been formed during their storage or transport if, for example, the fetlock was exposed to oxygen. In particular, streams from such refinery process units as vacuum distillation, visbreaking, and cokers may have more particulates and metals than straight-run products due to the heavier feedstocks that are processed in these units. Feedstocks purchased from other refiners may also be suspected because of the increased transportation time and exposure to oxygen leading to higher levels of particulate matter as a result of various chemical reactions in addition to the higher potential for contract with corroded system and/or the potential from the feedstock itself to cause corrosion (Speight, 2014c).

These particles, wherever the source, can be sub-categorized into the following classes: (1) basic sediment and (2) filterable solids. The amount of filterable solids in the feedstock (reported in pounds per barrel % w/w) may be determined by filtration of the feedstock and it is possible to evaluate the potential for fouling by indicating the type of materials that could contribute to fouling if allowed to pass into the heat exchanger. Such particles in the feedstock at amounts in excess of 1 ptb (1 pound per thousand barrels) lead to significant fouling problems. The effect of the particles on fouling can be dished considerably (in not avoided) avoided by solid-liquid filtration, sedimentation, and centrifugation or by any of various fluid cleaning devices. However, particles that need to be considered as serious in terms of foulant production are those that are not filterable (such as the micron-sized clay minerals in crude oil) and which are likely to pass through the desalter and proceed into the heat exchanger prior to the distillation unit.

By way of explanation for the context of this text, sedimentation is the accumulation of solids that are deposited in low-velocity areas in process equipment. The equipment can include heat exchangers, tower distributors, distillation trays, random packing, and structured packing. If the feedstock contains suspended solids—such as salts, metal oxides, catalyst fines, asphaltene particles, and coke fines—sedimentation can occur on the mass-transfer surface (sedimentation fouling). Precipitation and crystallization of dissolved salts can occur when process conditions become super saturated, especially at mass-transfer surfaces. Ammonia salt deposition resulting from both water vaporization and direct solid deposition from the gas phase is a common refining problem.

In some cases, the deposit may adhere strongly to the surface and are self-limiting insofar as thicker a deposit becomes, the more likely is to be removed by the fluid flow and thus attain some asymptotic average value over time. Sedimentation fouling is strongly affected by fluid velocity and less so by temperature; however, a deposit can bake on a surface and become very difficult to remove.

Certain salts such as calcium sulfate are less soluble in warm water than cold and if such a stream encounters a surface at a temperature above that corresponding to saturation for the dissolved salt, the salt will crystallize on the surface. Typically, crystallization will begin at specially active points (nucleation sites) such as area where corrosion has occurred and after a considerable induction period will spread to cover the entire surface. The buildup of the foulant will continue as long as the surface in contact with the fluid has a temperature above saturation. In addition, solidification fouling can occur due to cooling below the solidification temperature of a dissolved component, such as solidification and separation of wax from crude oil.

1.4.2 Particle Formation

Particles are usually formed after the exposure of the feedstock to oxidative or thermal environments. The presence of oxidative conditions during feedstock storage to exposure to heat in a heat exchanger or in a refinery unit is the usual mechanisms of chemical particle formation. Trace contaminants present in the feedstock can have a significant effect on the fouling encountered in certain chemical processes. Such contaminants may include oxygen, nitrogen, ammonia (NH3), hydrogen sulfide (H2S), hydrogen cyanide (HCN), olefins, organic sulfides, organic chlorides, organometallic compounds, and high molecular weight compounds such as paraffin wax (Chapter 7) as well as resin constituents and asphaltene constituents (Chapter 6). Individual metals, which may exist as metal salts in the feedstock, can catalyze differently a variety of reactions. The concentrations of such metals are typically very low, not exceeding few parts per million, but small concentrations of certain metals can have a significant effect on the catalysis of fouling-related reactions.

In addition, particle formation by biological means (biofouling) may occur in sea water systems and in types of waste treatment systems. Biofouling may be of two kinds: microbial fouling, due to microorganisms (bacteria, algae, and fungi) and their products, and microbial fouling, due to the growth of macroorganisms such as barnacles, sponges, seaweeds, or mussels. On contact with heat transfer surfaces, these organisms can attach and breed, thereby reducing both flow and heat transfer and sometimes completely disrupting fluid flow (Bott, 1990). Such organisms may also trap silt or other suspended solids and give rise to deposit corrosion.

1.4.3 Corrosion Fouling

Corrosion fouling is fouling deposit formation as a result of the corrosion of the substrate metal of heat transfer surfaces or reactor surfaces. In this type of fouling, corrosion occurs first and initiates fouling whereas under-deposit corrosion occurs as an after-effect of fouling. Corrosion fouling is dependent on several factors such as thermal resistance, surface roughness, composition of the substrate, and composition of the feedstock. In particular, impurities present in the feedstock stream can greatly contribute to the onset of corrosion—examples of these impurities are hydrogen sulfide, ammonia, and hydrogen chloride.

In many crude oils (especially the heavy feedstocks such as heavy oil, extra heavy oil, tar sand bitumen, and the refinery-produced residua), sulfur-containing and nitrogen-containing compounds are common contaminants which are decomposed during refining to hydrogen sulfide and ammonia, respectively. Chlorides which may be found in feedstocks (and originate from the formation brine which is co-produced with crude oil) converted to hydrogen chloride in, for example, the distillation section of the refinery:

In addition to entering the refinery from the production wells, chlorides may also be derived from various chemicals used during oil production—as, for example, recovery enhancement chemicals and solvents used to clean tankers, barges, trucks, and pipelines. As the crude oil is processed, some of these chemicals and solvents, which are thermally stable and not soluble in water, pass overhead in the main tower of the atmospheric distillation unit along with the naphtha fraction (boiling range 0-200 °C; 32-390 °F). Further processing of the naphtha (a gasoline blend stock or solvent precursor) causes generation and release of the hydrogen chloride.

In a hydrogen sulfide environment, the sulfur reacts with any exposed iron to form iron sulfide compounds—this can occur in both the hot and cooler sections of a refinery unit. Although corrosion is typically expected, the iron sulfide can form a complex protective scale or lattice on the base metal, which inhibits further corrosion (Speight, 2014c). In such a case, and the corrosion rate would be minimal if no other impurities were present in the system to interact with the sulfide lattice (which is then unable to remain in equilibrium with the unit environment) and corrosion not only continues, but is accelerated.

Although often ignored in favor of discussion the sulfur-containing and nitrogen-containing contaminants and their contribution to fouling, hydrogen chloride is also an important contributor to fouling especially corrosion fouling. By way of clarification, hydrogen chloride itself is a much lesser problem and typically does not foul equipment or corrode the carbon steel. However, chloride corrosion and the ensuing fouling take place when hydrogen chloride, ammonia, and water all interact in the colder sections of a unit and cause serious damage—the extent of the damage depends on the concentration (of the chloride ions) and is directly dependent on pH, with the corrosion rate increasing rapidly with pH decrease to a more acidic environment. Hydrogen chloride is corrosive when it comes in contact with free water, that is, water that is not in the vapor phase. Hydrogen chloride is highly soluble in water, and in a free water environment, any hydrogen chloride present in the vapor or hydrocarbon liquid will be quickly absorbed by the water, thus decreasing the pH down.

If the iron sulfide lattice is intact this chloride competes with the bisulfide ion (HS−) for the iron ions in the lattice:

When there is a high concentration of hydrogen chloride, the reaction shifts to the right and as more and more bisulfide is released from the sulfide lattice, the metal is exposed to corrosive effects. On the other hand, in water the chloride ions react directly with any exposed iron to form iron chloride (FeCl2):

As the chloride concentration in water is reduced by removing the source, diluting with additional water or neutralizing with a base, the pH will increase at which point hydrogen sulfide will begin to react with the exposed iron a new protective sulfide layer will be formed and the rate of corrosion approaches a minimum.

Hydrogen chloride will also cause serious fouling problems if ammonia is present in the system—the ammonia reacts with hydrogen chloride to form ammonium chloride which may cause fouling and plugging problems.

The ammonium chloride condenses in the cooler parts of the unit and forms a solid deposit on the walls of the unit—the ammonium chloride can also be detached from the walls and be carried downstream to eventually deposit elsewhere. However, if free water is present in the system, ammonium chloride will be absorbed directly from the vapor phase into the water and no solid salts will form on the equipment.

Another problem associated with ammonium chloride salt deposits is under-deposit corrosion (under-deposit pitting), because of the hygroscopic nature of ammonium chloride, which result in a water-containing environment at the wall under the deposit. The chloride ions then react with the iron to form iron chloride causing serious localized corrosion, which is accelerated by the presence of hydrogen sulfide. Furthermore, the sulfide ion (as part of an ammonium sulfide salt) will react with the iron chloride to form iron sulfide, thus releasing the chloride ion to continue the process and increase the extent and rate of corrosion. Thus, the iron content in the deposits may be an indication of fouling by corrosion.

1.4.4 Aggregation and Flocculation

High molecular weight constituents of feedstocks (such as resin and asphaltene constituents, but particularly the asphaltene constituents) will separate from the liquid phase into aggregates (large particles) which has an adverse effect on the stability of the feedstock system (Chapter 4). Separation and flocculation of asphaltene constituents in highly paraffin environment is irreversible in that same environment. Once separation and deposition has occurred with adsorption on to a sold surface (especially the surface if a catalyst), the deposition process can be irreversible. Segments of the separated asphaltene constituents which contain nitrogen, sulfur, and/or hydrogen bonds could also start to flocculate and as a result produce the irreversible disposition of high molecular weight constituents (especially reacted asphaltene constituents) which may be insoluble in solvents that typically are solvents for the asphaltene constituents (Chapter 6). In addition, inorganic particles may also act as nuclei on which agglomeration of organic particles proceed until the particles become eventually large enough to drop out. This principle was the driving force for the older CANMET process in which a carbonaceous or iron-containing scavenger was used to remove coke-formers from the reaction mix (Chapter 13) (Speight and Ozum, 2002; Speight, 2014a).

1.4.5 Phase Separation

Phase separation (called sedimentation fouling when gravity is the controlling force) is the separation of solid particles from a feedstock stream and the eventual deposition of the particles on to a heat exchanger surface or within a reactor. This may be a result of a combination of many chemical and physical processes that can be assigned to the various constituents of feedstocks (Table 1.1), especially the asphaltene constituents which when thermally changed move into a region of instability and phase separate (Figure 1.2) (Chapters 6 and 7).

Table 1.1

The Constituents of Feedstocks That Can Promote or Cause Sediment Formation

Figure 1.2 Phase relationship between the various bulk components of a feedstock.

Suspended particles such as sand, silt, clay, and iron-oxide (which occur in many feedstocks) (Chapter 2) (Speight, 2014a) may, if the particles are beyond a limiting size (which is system dependent) separate from the feedstock. This phenomenon can often be prevented by pre-filtration or pre-sedimentation. Sedimentation fouling is strongly affected by fluid velocity, and suspended particles in the process fluids will deposit in low-velocity regions, particularly where the velocity changes quickly, as in pipe elbows (providing the elbow does not cause turbulent flow), heat exchanger water boxes, and on the shell side of the heat exchanger, or in the reactor.

On the other hand, precipitation fouling is dependent on the solubility of the material (usually salt) in the feedstock. The dependence of solubility on temperature is often the driving force for precipitation fouling, but temperature dependence is different for different salts. Salt solubility increases with increasing temperature so that different salts may cause fouling depending on the temperature. While for most salts solubility increases with increasing temperatures, there are salts such as calcium sulfate (CaSO4) which have retrograde solubility dependence and are therefore less soluble in warm (water-laden) feedstock streams (inverse solubility fouling). Such salts will crystallize on heat transfer surfaces if the feedstock stream contacts a surface at a temperature higher than the saturation temperature of these salts. The calcium sulfate scale is hard and adherent and usually requires vigorous mechanical cleaning (Stegelman and Renfftlen, 1983) or chemical treatment for removal and salts such as calcium carbonate and magnesium carbonate (as well as silica minerals) also cause scaling problems.

On the organic side, the solubility of certain high molecular weight constituents with high-melting points such as paraffin wax and polynuclear aromatic systems is highly dependent on temperature. If the temperature is decreased, these constituents (especially the wax constituents) may precipitate in the form of solid crystals. Deposition of paraffin wax in cooled heat exchanger tubes shows a limiting behavior due to decreasing heat flux and increasing shear stress (Bott and Gudmundsson, 1978). In addition, for the high molecular weight constituents of petroleum and heavy feedstocks their interactive effects largely determine their collective deposition especially when the interacting constituents are the constituents of the asphaltene fraction. In addition, changes in the nature of the liquid medium can lead to precipitation of resin constituents and asphaltene constituents (as practiced in the propane deasphalting process) (Chapter 9) because the dispersion (solubility) limits of the molecular species are limited in the changed medium. Thus, asphaltene separation which is a common cause of crude unit fouling is affected by many factors including the stability of the system, variations of temperature, pressure, composition, flow regime, and wall effects. This type of fouling (variously called precipitation fouling or solidification fouling) can also occur as a result of pressure changes, where the solubility (dispersability) decreases with decreasing pressure.

1.4.6 Particle Deposition

Deposition and attachment of solid particles to surfaces (such as heat exchanger surfaces) is a function of several different operating variables which include particle size and concentration, bulk fluid density and velocity through the heat exchanger (Müller-Steinhagen and Bloch, 1988; Müller-Steinhagen et al., 1988). Furthermore, the stickiness and attractive or repulsive forces between particles can significantly contribute to the deposition of particles. Organic deposits may also be the result of high molecular weight constituents bound to the metal surfaces by inorganic deposition. Deposition and subsequent attachment of the particles to a surface is also a function of the interfacial properties of the foulant and the quality (or roughness and wettability) of the surface. For example, smooth and nonwetting surfaces will, more than likely, delay fouling, while rough surfaces provide sites that encourage deposition of the initial foulant species so that smooth surfaces would tend to become uneven (rough) as particle deposition occurred after which surface unevenness (roughness) would then be a factor to be considered. On the other hand, some consideration would have to be given to the deposition of very fine particles on to uneven (rough) surfaces which could conceivably result in filling the surface cavities thereby creating a relatively smooth surface. But this later consideration would also have to take into account interactions between deposited and nondeposited particles.

1.4.7 Deposit Growth and Deposit Deterioration

After particle deposition, there follows (1) deposit expansion (increase) growth and consolidation of the foulant or alternatively (2) auto-retardation and erosion or removal may take place. The rate of deposition and increase in the deposited foulant on surfaces is a function of the nature of the fouling material, the composition of the feedstock from which the foulant is produced as well as variables such as temperature and fluid flow rate.

The strength of the foulant deposit may increase with time and there is a tendency for the deposit to harden through a variety of processes known collectively as aging. Some types of foulant particles can bake on the surface of the heat exchanger or reactor which makes them much manure difficult to remove if allowed to remain for prolonged periods. The character of the deposit may be further affected by the presence of resin and asphaltene constituents, which are highly polar compounds and are susceptible to inter-molecular interactions which may simulate the behavior of mastic. Such behavior is similar to the interactions between rock aggregate and an asphalt binder (Chapter 16). On the other hand, deposits of foulant that originate from biofouling mechanisms may deteriorate and weaken with time due to contamination from other organisms or sources.

The deterioration of foulant deposits is a consequence of the nature of the feedstock and the composition and properties of the foulant deposit. The process commences with a decline of the foulant (particle) deposition rate (auto-retardation) is a desirable but spontaneous process that is subject to any one or more of several mechanisms that may account for the progressive decrease in adherence of particles to the surface of the heat exchanger or reactor.

In fact, depending on the strength of the deposit, erosion can occur immediately after the first deposit has been laid down. The concept of saw-tooth fouling includes the possibility of alternate foulant deposition and deposit erosion. In this process, part of the deposit is detached after a residence time or once a critical thickness of the deposit has been reached, which is dependent upon the strength of the deposit and the adherence properties of the deposit which, in turn, are deposit/foulant dependent. In the next step of the process, the fouling layer then builds up only to erode once more. In some cases, impurities such as sand or other suspended particles in feedstock streams may have a scouring action, which will reduce or remove deposits (Gudmundsson, 1981).

1.5 Rate of Fouling and Fouling Factor

1.5.1 Rate of Fouling

The rate of fouling is determined from the average deposit surface loading per unit of surface area in a unit of time. The word avenge is used loosely here since the average value of any phenomenon is not always an accurate depiction of the phenomenon—a reason why caution is advocated in (2) the analysis of the foulant to determine the process and mechanics by which fouling occurred as well as (2) the development models to describe the fouling phenomenon. Deposit thickness (microns, μm) and porosity (% v/v) are also often used for description of the amount of fouling. Depending on the mechanism of fouling mechanism and conditions under which fouling occurred, the rate of fouling may be linear, falling, accelerating, asymptotic, or saw tooth. However, the development of a model may not address all fouling phenomena since parameter such as the average nature of the foulant, the avenge temperature, and the average rate of foulant deposition are just that—averages for which there is no definitive and finite value and the average composition of the feedstock may vary (especially if the feedstock was a blend) in time that the foulant was deposited.

Linear fouling is the type of fouling where the fouling rate can be steady with time with increasing fouling resistance and deposit thickness. This is perhaps the most common type of fouling and generally occurs where the temperature of the deposit in contact with the flowing fluid remains constant. On the other hand, falling fouling is the type of fouling where the fouling rate decreases with time, and the deposit thickness does not achieve a constant value, although the fouling rate never drops below a certain minimum value. This type of fouling is due to an increase of removal rate with time and the progress can often be described by two numbers: the initial fouling rate and the fouling rate after a prolonged period of time. Accelerating fouling is the type of fouling where the fouling rate increases with time. This type of fouling is the result of the formation of a hard and strongly adhering deposit where removal and aging of the foulant can be ignored and typically develops (1) when fouling increases the surface roughness or (2) when the deposit surface exhibits higher chemical propensity to fouling than the pure underlying metal.

The asymptotic fouling rate is used to describe a decrease in rate with time until the rate becomes negligible after a period of time when the deposition rate of the foulant becomes equal to the deposit removal rate, at which time the deposit thickness remains constant. Generally, this type of fouling occurs where the surface temperature of heat exchanger or reactor remains constant while the temperature of the flowing feedstock drops as a result of increased resistance of the deposited foulant to heat transfer. Asymptotic fouling may also be the result of soft or poorly adherent suspended solid deposits upon heat transfer surfaces in areas of fast (or turbulent) flow where the foulant does not adhere strongly to the surface and the result that the thicker the deposit becomes, the more likely it is to wash off and attain some average asymptotic value over a period of time. The asymptotic fouling resistance increases with increasing particle concentration as well as decreasing fluid bulk temperature, flow velocity, and particle diameter.

Saw-tooth fouling occurs where part of the deposit is detached after a critical residence time or once a critical deposit thickness has been reached. The fouling layer then builds up and breaks off again. This periodic variation could be due to pressure pulses, trapping of air inside the surface deposits during shutdowns or other reasons, and spalling (the creation of chips, fragment, or flakes due to corrosion). Saw-tooth fouling often occurs when there are (or have been) moments of system shutdown, start-up, or other operation influences and/or interruptions.

1.5.2 Fouling Factor

The outcome of fouling is the formation a solid or near-solid deposit of low thermal conductivity upon the metals surface of a heat exchanger or reactor. However, in the case of heat exchangers, since the thermal conductivity of the fouling layer and the thickness are not generally known and often connate be known until the system is shutdown, the only possible solution to define the heat transfer problem is by the use of a fouling factor. This factor should take into account the additional resistance to heat transfer and make possible the calculation of the overall heat transfer coefficient. In some cases, a fouling coefficient (the reciprocal value of the fouling factor) may also be sometimes specified. Again caution is advised in the use of such factors or coefficients since fouling is feedstock dependent upon several parameters and, more than likely, the fouling factor and the coefficient are also dependent upon several parameters and, therefore each facto or coefficient cannot be generally applied to different systems.

In fact, the influence of uncertainties inherent in fouling factors (an fouling coefficients) is generally greater than that of uncertainties in other design parameters such as fluid properties, flow rate, and temperatures (Riverol and Napolitano, 2002). A large fouling factor is sometimes adopted as a safety margin to cover uncertainties in fluid properties and even in process knowledge, but the use of an excessively large fouling factor will result in an oversized heat exchanger with two or three times more area than is really necessary. Thus, in addition to system parameters, acceptable evaluation of the effects of fouling needs to be judged and evaluated for each particular system and with the use of more specific information.

1.6 Determination of Fouling Potential

The focus of stability and incompatibility studies (leading to fouling) has usually been on the whole crude oil and specifically on the asphaltene fraction and the characteristics of this fraction (including the individual asphaltene constituents) depends on the crude oil source (Speight, 2014a). The problems with the asphaltene constituents have increased due to the need to accept even the heaviest feedstocks in refineries as well as the trend to extract large amounts of low-boiling fractions out of crude oil by visbreaking and cracking processes (Rhoe and de Blignieres, 1979; Radovanović and Speight, 2011; Speight, 2011a, 2014a) which disturbs the stability of the petroleum system and causes deposition of asphaltene constituents or reacted resin and/or asphaltene constituents. At the other end of the molecular weight scale and refining, the heteroatoms (particularly nitrogen, sulfur, and trace metals) (Speight, 2014a) that are present in petroleum and might be also expected to be present in liquid fuels and other products from petroleum. Perhaps the main factor is the location and nature of the heteroatom which, in turn, determines reactivity (Por, 1992; Mushrush and Speight, 1995, 1998).

When various crude oil feedstocks are blended at the refinery, fouling can occur due to the onset of acid-base catalyzed condensation reactions of the various organo-nitrogen compounds and organic acid constituents in the individual blending stocks. Furthermore, when a petroleum product is transferred to a storage tank or some other holding tank, incompatibility can occur by the free radical hydroperoxide-induced polymerization of active olefins. This is a relatively slow reaction, because the observed increase in hydroperoxide concentration is dependent on the dissolved oxygen content (Mushrush and Speight, 1995, 1998).

In relation to petroleum products, product complexity and the means by which the product is evaluated (Speight, 2014a, 2015) have made the industry unique among industries. But product complexity has also brought to the fore issues such as instability and incompatibility. Product complexity becomes even more disadvantageous when various fractions from different types of crude oil are blended or are allowed to remain under conditions of storage (prior to use) and a distinct phase separates from the bulk product. The adverse implications of this for refining the fractions to salable products increase (Batts and Fathoni, 1991; Por, 1992; Wiehe, 1993; Mushrush and Speight, 1995, 1998).

1.6.1 Definitions and Terminology

Although definitions of fouling (and related terminology) have been introduced elsewhere in this chapter, it is appropriate here to define some of the terms that are used as they relate to petroleum products to alleviate some potential for misunderstanding (which might be called verbal fouling). First and foremost, the terms instability and incompatibility can be classed as a subset of the general category of fouling.

Briefly, the term incompatibility refers to the formation of a precipitate (or sediment) or separate phase when two liquids are mixed. The term instability is often used in reference to the formation of color, sediment, or gum in the liquid over a period of time and is usually due to chemical reactions, such as oxidation, and is chemical rather than physical. The phenomenon of instability is often referred to as incompatibility, and more commonly known as sludge formation, and sediment formation, or deposit formation.

Gum formation (ASTM D525) is used to describe the formation of soluble organic material whereas sediment is the insoluble organic material. Storage stability (or storage instability) (ASTM D381; ASTM D4625) is a term used to describe the ability of the liquid to remain in storage over extended periods of time without appreciable deterioration as measured by gum formation and/or the formation sediment. Thermal stability is also defined as the ability of the liquid to withstand relatively high temperatures for short periods of time without the formation of sediment (i.e., carbonaceous deposits and/or coke) (Brinkman and White, 1981; Mushrush and Speight, 1995, 1998). Thermal oxidative stability is the ability of the liquid to withstand relatively high temperatures for short periods of time in the presence of oxidation and without the formation of sediment or deterioration of properties (ASTM D3241) and there is standard equipment for various oxidation tests (ASTM D4871). Stability is also as the ability of the liquid to withstand long periods at temperatures up to 100 °C (212 °F) without degradation. Determination of the reaction threshold temperature for various liquid and solid materials might be beneficial (ASTM D2883).

Existent-gum is the name given to the nonvolatile residue present in the fuel as received for test (ASTM D381). Potential gum (ASTM D873) is the term used to predict the potential for gum formation during prolonged storage that alludes to the oxidative stability of the product (ASTM D942; ASTM D2272; ASTM D2274). Dry sludge is defined as the material separated from petroleum and petroleum products by filtration and which is insoluble in heptane. Existent dry sludge is the dry sludge in the original sample as received and is distinguished from the accelerated dry sludge obtained after aging the sample by chemical addition or heat. The existent dry sludge is operationally defined as the material separated from the bulk of a crude oil or crude oil product by filtration and which is insoluble in heptane, which is used as an indicator of process operability and as a measure of potential downstream fouling.

Thus, petroleum constituents and petroleum products are incompatible when sludge, semisolid, or solid particles (for convenience here, these are termed secondary products to distinguish them from the actual petroleum product) are formed during and after blending. If the secondary products are marginally soluble in the blended petroleum product, use might detract from solubility of the secondary products and they will appear as sludge or sediment that can be separated by filtration or by extraction (ASTM D4310).

Fouling through the formation of sediments and deposits also originate from the inorganic constituents of petroleum. These may be formed from the inherent components of the crude oil (i.e., the metallo-porphyrin constituents) or from the ingestion of contaminants by the crude oil during the initial processing operations. For example, crude oil is known to pick up iron and other metal contaminants from contact with pipelines and pumps.

1.6.2 General Chemistry

The chemistry and physics of fouling can be elucidated, but many aspects still remain unknown (Por, 1992; Mushrush and Speight, 1995, 1998; Speight, 2014a). In addition to the chemical aspects, there are also aspects such as the attractive force differences, such as: (1) specific interactions between like/unlike molecules, such as hydrogen bonding and electron donor-acceptor phenomena, (2) field interactions such as dispersion forces and dipole-dipole interactions, and (3) any effects imposed on the system by the size and shape of the interacting molecular species.

1.6.3 Test Methods

Fouling by crude oil and crude oil products is manifested in the formation of sludge, sediment, and general darkening in color of the liquid (ASTM D1500). Sludge (or sediment) formation takes one of the following forms: (1) material dissolved in the liquid; (2) precipitated material; and (3) material emulsified in the liquid. Under favorable conditions, sludge or sediment will dissolve in the crude oil or product with the potential of increasing the viscosity. The sludge or sediment foulant, which is not soluble in the crude oil (ASTM D96; ASTM D473; ASTM D1796; ASTM D2273; ASTM D4007; ASTM D4807; ASTM D4870), may either settle at the bottom of the storage tanks or remain in the crude oil as an emulsion.

1.6.4 Determination of Fouling Potential

A number of experimental methods are available for estimation of the factors that influence fouling, which have been explored and attempts made to estimate the character of the fuel or product with varied results. It is the purpose of the section to note the methods that are used—a fuller description of the methods and their respective uses are presented elsewhere (Chapter 5).

1.6.4.1 Elemental Analysis

Of the data that are available, the proportions of the elements in petroleum vary only slightly over narrow limits: carbon: 83.0-87.0% w/w, hydrogen: 10.0-14.0% w/w, nitrogen: 0.10-2.0% w/w, oxygen: 0.05-1.5% w/w, and sulfur: 0.05-6.0% w/w (Chapter 2). And yet, there is a wide variation in physical properties from the lighter more mobile crude oils at one extreme to the extra heavy crude oil and tar sand bitumen at the other extreme (Chapter 2) (Speight, 1987; Dolbear, 1998; Speight, and Ozum, 2002; Hsu and Robinson, 2006; Gary et al., 2007; Cher et al., 2014; Speight, 2014a).

In terms of the instability and incompatibility leading to fouling in petroleum and petroleum products, the heteroatom content appears to represent the greatest influence. In fact, it is not only the sulfur and nitrogen content of crude oil are important parameters in respect of the processing methods which have to be used in order to produce fuels of specification sulfur concentrations, but also the type of sulfur and nitrogen species in the oil. The tendency of these heteroatom-containing constituents to foul catalysts is real, indicating a relationship between nitrogen and sulfur content types in crude oil and crude oil products and, in addition, higher nitrogen and sulfur in crude oil and crude oil products are indicative of higher sludge-forming tendencies (Mushrush and Speight, 1995,

Reviews for Fouling in Refineries

[Mohammad Matar · Rating: 5 out of 5 stars]

This book is well writting and it is a very good referance.