C4 Fouling of Heat Exchanger Surfaces

C4 Fouling of Heat Exchanger Surfaces

Hans Müller-Steinhagen
Universität Stuttgart, Stuttgart, Germany

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 3.4.2 Allowable Pressure Drop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

1.1 The Fouling Resistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 3.4.3 Longitudinal Baffles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
1.2 Mechanisms of Heat Exchanger 3.5 Example for Fouling Mitigation
Fouling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 by Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
1.3 Sequential Events of Fouling. . . . . . . . . . . . . . . . . . . . . . . . . . 83 3.5.1 Original Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
1.4 Approximate Influence of Operating 3.5.2 Operation History with Original
Conditions on Fouling in Industrial Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Heat Exchangers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 3.5.3 Root Cause Analysis of Performance . . . . . . . . . . . . . . . . . 93
1.5 Costs due to Heat Exchanger Fouling . . . . . . . . . . . . . . . . 85 3.5.4 New Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
1.5.1 Capital Expenditure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 3.5.5 Final Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
1.5.2 Fuel Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
1.5.3 Maintenance Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 4 Online Mitigation Methods . . . . . . . . . . . . . . . . . . . . . . . . . 94
1.5.4 Costs due to Production Loss. . . . . . . . . . . . . . . . . . . . . . . . . 86 4.1 Start-Up Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.2 Chemical Fouling Mitigation Methods . . . . . . . . . . . . . . . 94
2 Consideration of Fouling in the Design of Heat 4.2.1 Scale Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Exchangers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 4.2.2 Particulate Fouling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
2.1 Preliminary Remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 4.2.3 Chemical Reaction Fouling . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
2.2 Selection of Heat Exchanger Type . . . . . . . . . . . . . . . . . . . . 87 4.2.4 Biofouling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
2.3 Material Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 4.2.5 Corrosion Fouling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
2.4 Shell and Tube Heat Exchangers . . . . . . . . . . . . . . . . . . . . . . 87 4.2.6 Gas-Side Fouling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
2.5 Other Heat Exchanger Types. . . . . . . . . . . . . . . . . . . . . . . . . . 89 4.3 Mechanical Fouling Mitigation Methods . . . . . . . . . . . . . 97
2.5.1 Plate and Frame Heat Exchangers . . . . . . . . . . . . . . . . . . . . 89 4.3.1 Liquid Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
2.5.2 Plate-Fin Heat Exchangers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 4.3.2 Gas Flow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
2.5.3 Printed Circuit Heat Exchanger . . . . . . . . . . . . . . . . . . . . . . 90 4.3.3 Other Devices for Fouling Mitigation . . . . . . . . . . . . . . . . 99
2.5.4 Polymer Compact Heat Exchangers . . . . . . . . . . . . . . . . . . 90
2.6 Effect of Fouling on Pressure Drop . . . . . . . . . . . . . . . . . . . 90 5 Cleaning of Heat Exchangers. . . . . . . . . . . . . . . . . . . . . . . 100
2.6.1 Tube-Side Pressure Drop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 5.1 Chemical Cleaning Methods . . . . . . . . . . . . . . . . . . . . . . . . . 100
2.6.2 Shell-Side Pressure Drop. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.1.1 The Basic Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
5.1.2 Cleaning Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3 HTRI Fouling Mitigation by Design 5.1.3 Cleaning Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.1.4 OnStream Chemical Cleaning . . . . . . . . . . . . . . . . . . . . . . . 102
3.1 Crude Oil Best Practice Operating 5.1.5 Problems Associated with Chemical
Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Cleaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
3.2 Cooling Tower Water Best Practice Operating 5.2 Mechanical Cleaning Methods. . . . . . . . . . . . . . . . . . . . . . . 102
Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.3 Excess Surface/Coefficient Adjustments . . . . . . . . . . . . . . 92 6 Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
3.4 Design Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.4.1 Shell-Side Bundle Geometry Exit/Entrance 7 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

1 Introduction on the heat transfer surfaces, i.e., fouling. Since the thermal
conductivity of such deposits is low, see Table 1, their resistance
In most industrial processes, heat exchanging fluids contain to heat transfer may well exceed that of the clean fluids, resulting
certain amounts of dissolved or suspended material or provide in significantly reduced heat exchanger performance.
conditions favorable for the growth of biological organisms. As a result, substantial safety margins in the design, pre-
Design and operation of heat exchangers are still to a major treatment of hot/cold fluids and regular cleaning of equipment
extent determined by the process-related formation of deposits may be required.

VDI-GVC (ed.), VDI Heat Atlas, DOI 10.1007/978-3-540-77877-6_7,

# Springer-Verlag Berlin Heidelberg 2010
80 C4 Fouling of Heat Exchanger Surfaces

C4. Table 1. Thermal conductivity of various deposits C4. Table 2. Fouling of heat exchangers in various industries [3]

Sodium aluminum silicate 0.2–0.4 W/mK Extent of

Milk components 0.5–0.7 W/mK Industry group Type of fouling problem
Hematite (boiler deposit) 0.6 W/mK Food and kindred products Chemical Major
Biofilm 0.7 W/mK reaction
Calcium sulfate (boiler) 0.8–2.2 W/mK Crystallization Major
Calcite (boiler deposit) 0.9 W/mK Biological Medium
Serpentine (boiler deposit) 1.0 W/mK Particulate Minor/Major
Gypsum (boiler deposit) 1.3 W/mK Corrosion Minor
Calcium sulfate 2.3 W/mK Textile mill products Particulate
Magnesium phosphate 2.3 W/mK Biological
Calcium phosphate 2.6 W/mK Wood and paper products Crystallization Major
Calcium carbonate 2.9 W/mK Particulate Minor
Magnetite iron oxide 2.9 W/mK Biological Minor
Chemical Minor
reaction
Surveys [1–3] have indicated that more than 90% of heat
Corrosion Medium
exchangers suffer from fouling problems. Table 2 identifies the kind
Chemical and allied Crystallization Medium
and typical extent of fouling for fluids from various industries [39].
industries Particulate Minor/Medium
More information about research and industrial fouling pro-
blems may be found under www.heatexchanger-fouling.com. Biological Medium
Chemical Minor/Major
reaction
1.1 The Fouling Resistance Corrosion Medium
Petroleum refineries Chemical Major
The possibility of deposition on heat transfer surfaces is gener-
reaction
ally considered in the design of heat exchangers by using the so-
Crystallization Medium
called fouling resistances in the calculation of the overall heat
transfer coefficient k. Particulate Minor/Medium
  Biological Medium
1 1 A2 1
¼ þ Rf ;1 þ Rwall þ þ Rf ;2 ð1Þ Corrosion Medium
k a1 A1 a2
Tone, glass, concrete Particulate Minor/Major
In Eq. (1), a, A, and Rf are the heat transfer coefficients, the Electricity generation Biological Major
heat transfer areas and the fouling resistances, respectively, for Crystallization Medium
the two heat exchanging fluids; Rwall is the thermal resistance of
Particulate Major
the separating wall. It is obvious that the frequently used expres-
Freezing Major
sion ‘‘fouling factor’’ is incorrect, as the effect of fouling is to create
an additional thermal resistance. The fouling resistance reduces Corrosion Minor
the overall heat transfer coefficient k, and hence leads to the
reduction of heat duty of an existing heat exchanger or to addi- in Tables 3–5. Although the TEMA tables were originally con-
tional surface requirement in the design of new heat exchangers. sidered to be only a rough guideline for shell-and-tube heat
In the utility industry, it is common to use the cleanliness exchanger design, they are unfortunately often treated as accu-
factor CF rate values. This may cause considerable errors because
kf
CF ¼ ð2Þ 1. The values were developed in 1949 and were based on hand
kc
calculation procedures for heat exchanger design in use at
instead of the fouling resistance Rf, where CF is a function of tube the time. Even though calculation methods have evolved,
material, flow velocity, and fouling propensity of the cooling the values presented by TEMA have not been altered since its
water. Typically, cleanliness factors are adding less excess surface publication.
and are, hence, more realistic than fouling resistances. However, 2. Fouling resistances are included for a number of fluids that
this purely empirical ratio of overall heat transfer coefficients for are known not to foul, such as refrigerants, demineralized
fouled and dirty conditions does not provide any access to the water, liquefied natural gas (LNG) and non-polymerizing
understanding of the mechanisms of deposit formation and hence (diolefin-free) condensing gases, or any other streams which
the potential effects of operating conditions. do not foul within the operating conditions of the heat
The main source of publicly available fouling resistances are exchanger.
the approximately 100 values suggested by the Tubular Exchanger 3. Rf values in the tables tend to be used without adequate
Manufacturers Association (TEMA) [4], which are reproduced engineering, resulting in exchangers that are much larger
 Fouling of Heat Exchanger Surfaces C4 81

C4. Table 3. TEMA fouling resistances for water (m2K/kW) [4] C4. Table 4. TEMA fouling resistances for oil refinery streams in
shell-and-tube exchangers, (m2K/kW) [4]
Temperature of heating
 
medium Up to 115 C 115–200 C Crude and vacuum unit gases and vapors

Temperature of water 50  C Over 50  C Atmospheric tower overhead vapors 0.18

Water velocity Water velocity Light naphthas 0.18

m/s m/s Vacuum overhead vapors 0.35
1.0 and Over 1.0 and Over Crude and vacuum liquids
less 1.0 less 1.0 Crude oil
 
Cooling tower and artificial 0–120 C 120–175 C
spray pond Velocity (m/s) Velocity (m/s)
Treated make up 0.088 0.088 0.17 0.17 <0.6 0.6–1.2 >1.2 <0.6 0.6–1.2 >1.2
Untreated 0.35 0.17 0.53 0.35 Dry 0.53 0.35 0.35 0.53 0.35 0.35
City or well water 0.17 0.17 0.35 0.35 Salta 0.53 0.35 0.35 0.88 0.70 0.70
River water 175–230 C 230 C and over
Minimum 0.35 0.17 0.53 0.35 Velocity (m/s) Velocity (m/s)
Average 0.53 0.35 0.70 0.53 <0.6 0.6–1.2 >1.2 <0.6 0.6–1.2 >1.2
Muddy or silty 0.53 0.35 0.70 0.53 Dry 0.70 0.53 0.53 0.88 0.70 0.70
Hard (over 250 ppm) 0.53 0.53 0.88 0.88 Salta 1.1 0.88 0.88 1.2 1.1 1.1
Engine jacket 0.17 0.17 0.17 0.17 Assumes desalting at approximately 120 C
a

Distilled or closed cycle Gasoline 0.35

Condensate 0.088 0.088 0.088 0.088 Naphtha and light distillates 0.35–0.53
Treated boiler feedwater 0.17 0.088 0.17 0.17 Kerosene 0.35–0.53
Boiler blowdown 0.35 0.35 0.35 0.35 Light gas oil 0.35–0.53
Seawater 0.09 0.09 0.18 0.18 Heavy gas oil 0.53–0.88
Brakish water 0.35 0.18 0.54 0.35 Heavy fuel oils 0.88–1.2
Asphalt and residuum
Vacuum tower bottoms 1.8
than required. For example, Fig. 1 includes a summary of Atmosphere tower bottoms 1.2
the percent overdesign of all cases submitted to HTRI in
2006. As can be seen from the figure, most exchangers are equation, the percentage excess surface area for a fixed heat duty
50–500% too large, based primarily on the assumed fouling increases with increasing clean heat transfer coefficient.
resistances. This results in unnecessary capital expense and Af
exchangers that foul due to poor design. ¼ 1 þ k c Rf ð3Þ
Ac
4. Rf values are only available for a limited number of process
fluids and process conditions. Obviously, the impact of the fouling resistance is more severe
5. Tabulated Rf values provide only limited information about for heat exchangers with high overall heat transfer coefficients.
the influence of process parameters such as flow velocity, To account for unreliable design procedures and operational
fluid, and wall temperature on the fouling resistance. These problems, heat exchangers are typically overdesigned by 70–80%,
parameters have a considerable influence on the deposition from which 30–50% is attributed to fouling [8]. While the instal-
of foulants matter onto the heat transfer surface. lation of excess heat transfer surface may extend the operation
6. Using constant fouling resistances, the transient character of time of heat exchangers, it provides no remedy against the
the fouling process is ignored. Conditions in the initially deposition of dirt. Fluid pretreatment, antifouling installations,
overdesigned heat exchanger often promote deposition, and regular cleaning will still be required in most cases. Indus-
thus making fouling a self-fulfilling prophecy. trial practice and state of the art of such fouling mitigation and
7. The TEMA values do not apply for heat exchanger types of heat exchanger cleaning are outlined in this chapter.
other than conventional shell and tube heat exchangers, e.g.,
plate heat exchangers, compact heat exchangers or finned 1.2 Mechanisms of Heat Exchanger Fouling
tubes. Some fouling resistances for other heat exchanger
types are given in Tables 6–8. Because of the great variety of fouling mechanisms it is useful to
divide fouling according to the key physical/chemical processes
To demonstrate the significance of the selected fouling resistance
into five major categories:
on the sizing of heat exchangers, Table 9 shows the excess heat
transfer surface required for several heat exchanger types, if a 1. Crystallization Fouling
typical TEMA fouling resistance of 0.18 m2K/kW is used for each ● Precipitation and deposition of dissolved salts, which at
of the two heat exchanging fluids. As shown in the following process conditions become supersaturated at the heat
82 C4 Fouling of Heat Exchanger Surfaces

C4. Table 5. Fouling resistances for various processing streams in (d) Mixing of streams with different composition
shell-and-tube exchangers, (m2K/kW) (e) Variation of pH which affects the solubility of CO2
in water
Hydrocarbons Fuel oil no.2 0.35
(f) Solidification fouling due to cooling below the
Fuel oil no.6 0.88
solidification temperature of a dissolved compo-
Transformer oil 0.18 nent (e.g., solidification of wax from crude oil
Lube oil 0.18 streams)
Hydraulic oil 0.18 2. Particulate Fouling
Quench oil 0.7 ● Deposition of small suspended particles such as clay, silt,
Pitch 0.8 or iron oxide on heat transfer surfaces of any orientation
Tar 0.9 ● Gravitational settling of relatively large particles onto
horizontal surfaces
Vegetable oil 0.53
3. Chemical Reaction Fouling
Gases and Steam (oil free) 0.0
Deposit formation at the heat transfer surface by a
vapors Steam (oil contaminated) 0.18 chemical reaction in which the surface material itself does
Refrigerant vapor (oil 0.35 not participate (polymerization, food processing).
contaminated) 0.0 4. Corrosion Fouling
Alcohol vapor 0.09 The thermal resistance of corrosion layers is usually low
Organic vapor 0.18 because of the relatively high thermal conductivity of oxides.
Ammonia 0.35 However, the increased surface roughness may promote
Carbon dioxide 0.18 fouling due to other fouling mechanisms.
Combustion gas (coal) 0.18–0.35 5. Biological Fouling
Biological fouling refers to the development and
Combustion gas (natural gas) 1.8
deposition of organic films consisting of microorganisms
Diesel exhaust gas 1.8
and their products such as bacteria (microbial or micro-
Synthesis gas 0.18–0.35 biofouling) and the attachment and growth of macro-
Compressed air 0.18 organisms such as mussels, algae, etc. (macro-biofouling)
Natural gas 0.18–0.35 on the heat transfer surfaces. Microbial fouling always
Nitrogen 0.18–0.35 precedes fouling by macro-organisms, with the microorgan-
Stabil column overhead product isms acting as the nutrient source for the macro-organisms.
Liquids Refrigerant 0.10–0.18 Suspensions of seaweed and other organic fibres often cause
fouling. Many types of bacteria will deposit slime on the
Organic heat transfer liquid 0.18–0.35
heat transfer surfaces and other types of foulants can adhere
Ammonia (oil free) 0.18
to these deposits. Larger growth restricts the fluid flow and
Ammonia (oil contaminated) 0.53 often causes pitting of the metal. This type of fouling is
Methanol solution 0.35 common in untreated water such as sea, river, or lake
Ethanol solution 0.35 water. Microbiological fouling is a particularly serious prob-
Glycol solution 0.35 lem as the microbes may be introduced into the water cycle
LPG, LNG 0.18–0.35 not only by the fluid, but also from the ambient air in the
MEA- and DEA solution 0.35 cooling tower. Temperatures between 15 C and 50 C in
cooling towers are ideal for microbial growth. Both dead
DEG- and TEG solution 0.35
and alive microorganisms adhere to the heat transfer sur-
Stable column side stream 0.18–0.35
faces and form a layer of slime with a thermal conductivity
Stable column bottom stream 0.18–0.35 similar to that of water. As a consequence of the filtering
Caustic solutions 0.35 effect of biological layers, more suspended particles amass in
Black liquor 0.7–1.4 the deposit.
6. Mixed Fouling
Fouling mechanisms within each category may be de-
transfer surface. Supersaturation may be caused by the
scribed with similar models. Generally, several fouling
following processes:
mechanisms occur at the same time, nearly always being
(a) Evaporation of solvent
mutually reinforcing. Exceptions are the combination of
(b) Cooling below solubility limit for solution with
crystallization and particulate fouling, where particles of
normal solubility, e.g., increasing solubility with
the crystallizing matter accelerate fouling, whereas particles
increasing temperature
from other material may lead to reduced fouling due to a
(c) Heating above the solubility limit for solutions with
weakening of the deposit structure [9].
inverse solubility such as CaCO3, CaSO4,
Ca3(PO4)2, CaSiO3, Ca(OH)2, Mg(OH)2, MgSiO3, Figures 2 and 3 show typical effects of surface temperature
Na2SO4, Li2SO4, and Li2CO3 in water and flow velocity on most of the above fouling mechanisms.
 Fouling of Heat Exchanger Surfaces C4 83

C4. Fig. 1. Impact of fouling resistance on 2000 shell and tube heat exchangers designed from 2003–2008

1.3 Sequential Events of Fouling C4. Table 6. Fouling resistances in plate and frame heat exchan-
gers, (m2K/W) [5]
The above fouling mechanisms generally occur in five consecu-
Fluid Fouling resistance (m2 K/kW)
tive steps:
Water
(i) Initiation Period or Delay Period
Demineralized or distilled 0.009
When the new or cleaned heat exchanger has been taken
into operation, the initially high heat transfer coefficients may Soft 0.017
remain unchanged for a certain time. During this time, nuclei Hard 0.043
for crystallization are formed or nutrients for biological Treated cooling tower water 0.034
growth are deposited. This delay period may last anytime Coastal sea water 0.043
from few seconds to several days. According to [10], no Ocean sea water 0.026
delay period occurs for particulate fouling. For crystallization River water 0.043
fouling and for chemical reaction fouling, the initiation period
Engine jacket 0.052
decreases with increasing surface temperature, as supersatura-
Lube oil 0.017–0.043
tion and/or reaction rate increase [11]. Generally, it is reported
that the delay time, before deposition starts, decreases with Vegetable oil 0.017–0.052
increasing roughness of the heat transfer surface. Organic solvents 0.009–0.026
(ii) Mass Transport Steam 0.009
To form a deposit at the heat transfer surface it is necessary General process fluids 0.009–0.052
that at least one key component is transported from the
fluid bulk to the heat transfer surface. In most cases, this
(v) Aging
occurs by diffusion. For the transport of particles to the
Every deposit is subjected to aging. Aging may increase the
wall, inertia effects and thermophoretic forces have to be
strength of the deposit by polymerization, recrystallization,
considered, as well.
dehydration, etc. Biological deposits get poisoned by metal
(iii) Formation of Deposit
ions and may be washed away by the bulk flow. Aging is the
After the foulant has been transported to the heat transfer
least investigated and understood step and is usually ig-
surface, it must stick to the surface (for particulate fouling)
nored in modelling attempts.
or react to the deposit forming substance (e.g., CaCO3).
Depending on the process parameters and the dominant
(iv) Removal or Auto-Retardation
fouling mechanism, the fouling rate can be either constant
Depending on the strength of the deposit, erosion occurs
or decreasing with time (see Fig. 4).
immediately after the first deposit has been laid down.
Furthermore, several mechanisms exist, which cause For hard, adherent deposits such as silicates and some polymeri-
auto-retardation of the deposition process. For the thermal zation products, steps iv and v may be ignored and the growth rate
boundary condition of constant temperature difference of deposits is constant or continuously decreasing with time. For
between heated and cooled fluid, the growth of deposit weaker deposits (e.g., particles >1 mm), the fouling resistance
causes a reduction of the driving temperature difference approaches a constant (or asymptotic) value which may or may
between heat transfer surface and fluid. not allow acceptable operation of the process.
84 C4 Fouling of Heat Exchanger Surfaces

C4. Table 7. Fouling resistances for combustion gases on finned

surfaces (m2K/W) [6]

Fouling resistance Flow velocity

Fuel (m2 K/kW) m/s
Natural gas 0.09–0.53 30–40
Propane 0.18–0.53 25–30
Butane 0.18–0.53 18–24
Clean turbine gas 0.18 15–21
Moderately clean 0.27–0.5
turbine gas
Light oil 0.36–0.7
Diesel oil 0.53
C4. Fig. 2. Effect of flow velocity on the fouling resistance for flow
Heavy oil 0.53–1.24 of water
Crude oil 0.7–2.7
Coal 0.89–8.85

C4. Table 8. Fouling resistances in evaporators, (m2K/W) [7]

Fouling resistance
(m2 K/kW)
Boiling medium
Hydrocarbon C1–C4 0–0.18
Higher hydrocarbons 0.18–0.5
Olefins and polymerizing hydrocarbons 0.5–0.9
Heating medium
Condensing steam 0–0.09
Condensing organic vapor 0.09–0.18
C4. Fig. 3. Effect of surface temperature on the fouling resistance
Organic liquid 0.09–0.052 for flow of water

C4. Table 9. Excess surface area for various heat exchanger applica-
tions, Rf = 0.36 m2K/kW

Clean overall Excess

Application coefficient area
Gas/gas shell & tube heat 50 W/m2K 1.8%
exchanger
Liquid/gas shell & tube heat 150 W/m2K 5.4%
exchanger
Liquid/liquid shell & tube heat 1000 W/m2K 36%
exchanger
Liquid/liquid plate & frame heat 3000 W/m2K 108%
exchanger
Water-cooled shell & tube steam 4500 W/m2K 162%
condenser
C4. Fig. 4. Possible fouling resistance versus time curves
 Fouling of Heat Exchanger Surfaces C4 85

C4. Fig. 6. Total annual costs of a double pipe heat exchanger

arrangement as a function of the flow velocity

C4. Fig. 5. Fouling resistance in shell and tube heat exchangers as a

function of flow velocity and water quality [13]. Water quality
multiple double pipe heat exchangers as a function of the flow
decreasing from 1 to 4
velocity in the pipes. One curve has been calculated according to
Martin [14] for a constant fouling resistance, the second is for
the case where the fouling resistance is velocity dependent as
suggested above. The optimum flow velocity shifts from 0.8 to
1.4 Approximate Influence of Operating about 1.3 m/s and the total annual costs are reduced by 10%
Conditions on Fouling in Industrial despite the higher friction losses.
Heat Exchangers

Many correlations have been recommended for the prediction

1.5 Costs due to Heat Exchanger Fouling
of individual fouling mechanisms [12]. However, these correla-
tions are generally not applicable to industrial conditions where
Despite the enormous costs associated with heat exchanger
a combination of fouling mechanisms and foulants occurs.
fouling, only very limited research has been done to determine
Comparing fouling data from a range of industries, the follow-
accurately the economic penalties due to fouling and to attri-
ing approximate influence of process parameters on industrial
bute these costs to the various aspects of heat exchanger design
fouling has been found:
and operation. However, reliable knowledge of fouling econom-
(a) Fouling usually increases linearly with increasing foulant ics is desirable to evaluate the cost-efficiency of various mitiga-
concentration in the fluid bulk. tion strategies. The total fouling related costs consist of
(b) The fouling resistance nearly always decreases with increas-
ing wall shear stress due to increased removal forces. As an
average, it was found that the fouling resistance is propor- 1.5.1 Capital Expenditure
tional to the flow velocity to the power of 1.5. Figure 5
shows the effect of dirt content and of flow velocity on the According to Thackery [15], total capital cost due to fouling in
fouling resistance for cooling water [13]. England added up to £100 million in 1978, which corresponds
(c) For many fouling mechanisms, the fouling resistance to US$190 million. For the United States, Garrett-Price et al.
increases with increasing surface temperature (see e.g., found that capital costs excluding costs for antifouling equip-
Fig. 3). For crystallization and chemical reaction fouling, ment were US$960–280 million in 1982 [3].
this trend frequently follows an Arrhenius relationship
dRf E (a) Excess Heat Transfer Surface Area
¼ Ke =RTS ð4Þ Thackery [15] found that design excess surface area for fouling
dt
varies between 10–500%, with an average around 30%. This
For biological fouling, a maximum is observed for tem-
result was confirmed by Garrett-Price et al. [3], who obtained a
peratures around 35 C.
similar value for the USA and by Steinhagen et al. for New
(d) Fouling was found to increase with increasing roughness of
Zealand [2]. Investigations by Heat Transfer Research Inc.
the heat transfer surface.
and TEMA among the major North American heat exchanger
To date, not even these simple rules are considered in the design manufacturers showed excess surface areas between 11% and
of heat exchangers, even though they could significantly im- 67% [8].
prove some heat exchanger optimization procedures. This is Excess area of 30–40% may correspond to 25% additional
demonstrated in Fig. 6, which shows the sum of annual capital cost. To estimate the absolute costs of excess heat transfer
operating and capital service costs of an arrangement of area in the United Kingdom, Pritchard [16] took the value of
86 C4 Fouling of Heat Exchanger Surfaces

the process plant built in the UK. As much as 6.5% of the 1.5.4 Costs due to Production Loss
process plant hardware consisted of heat exchangers. If each
heat exchanger has 30–40% extra surface area to allow for Because of planned and unplanned plant shutdowns due to
fouling, he concluded that additional costs were £5 million in fouling in heat exchangers, large production losses are possible.
1968 and £20 million in 1977. The corresponding American These costs are often considered to be the main cost of fouling.
figure is $US320 million per year for 1982 [3]. For example: Sart and Eimer [18] state that the loss of produc-
tion for 1 day shutdown of a 1300 MW power plant is about
(b) Transport and Installation Costs $US500,000; Taborek [13] estimates that shutdown losses of a
As a result of additional surface area, heat exchangers become large oil refinery are about $US1.5 million per day. In addition
bigger and heavier. Therefore costs for stronger foundations, to production losses during plant shutdown and start-up,
provisions for extra space, increased transport and installation penalties for not keeping to a deadline and the loss of customers
costs must be considered. Woods et al. [17] assume that instal- must be considered. Garrett-Price et al. [3] suggested that an
lation costs tend to increase with the size of the heat exchanger upper limit for loss of production costs may be estimated by
and are usually 2–3 times the delivered costs. Adding these extra assuming that the loss of production has to be less than the cost
costs to the costs for excess heat transfer surface may inc- of providing redundant exchangers. For the US in 1984, this
rease the costs for oversized equipment to $640–960 million would be $US200 million. According to Thackery, [15] 1978
per year [3]. costs due to production losses in the UK are about £100 million
($US190 million).
(c) Capital Costs for Antifouling Equipment The above fouling-related costs will have to be inflated to
These costs include expenses for online and off-line cleaning current prizes. For crude oil heat exchangers, more recent infor-
equipment, extra cost for providing non-fouling heat exchan- mation is available in [19].
gers such as scraped surface or fluidized bed heat exchangers,
pretreatment plants, cleaning-in-place equipment, dosing
pumps and tanks for antifouling chemicals. 2 Consideration of Fouling in the Design
of Heat Exchangers

1.5.2 Fuel Costs 2.1 Preliminary Remarks

Costs for extra fuel only occur if fouling leads to extra fuel burning Not all heat exchangers have serious problems with fouling;
in furnaces or boilers or if more secondary energy such as electric- many of them operate satisfactorily for long periods of time
ity or process steam is needed to overcome the effects of fouling. without being cleaned. If fouling is anticipated, however, some
Thackery [15] estimated additional UK fuel costs in 1978 as allowance must be provided in the design of a heat exchanger.
£100–200 million ($US290–480 million). Garrett-Price et al. [3] Regardless of the approach, the selection of appropriate values
calculate that 1–5% of the energy consumed by the industrial still relies more on engineering judgement from past experience
sector is used to overcome fouling. The result leads to fouling than on the application of results from experimental and theo-
related fuel costs between $US700 and $US3,500 million. retical research. It is important to keep in mind that heat
exchanger fouling can be effectively mitigated at the design
stage of the heat exchanger. To design for reliable operation,
1.5.3 Maintenance Costs
(i) Select a suitable heat exchanger type
(ii) Try to avoid operating conditions which promote fouling
Maintenance costs are costs for removing fouling deposits and
(iii) Attempt an optimum design with adequate velocities in
costs for chemicals or other operating costs of antifouling
the heat exchanger and which avoids hot spots, bypass flow
devices. According to Pritchard [16] and Thackery [15], about
or dead zones
15% of the maintenance costs of process plant could be attrib-
(iv) Design for easy cleaning
uted to heat exchangers and boilers and of that 50% was proba-
bly due to fouling. Garrett-Price et al. [3] quote a figure of Additional guidance is included in Sect. 3. Due to their frequent
$US2,000 million for annual sales of companies supplying occurrence and economic importance, detailed best practice
heat exchanger online and off-line cleaning equipment, chemi- guidelines have been prepared by ESDU [19–21] for fouling in
cals, and cleaning services in the USA for 1982. crude oil preheat exchangers, and for seawater and fresh water as

C4. Table 10. Reboiler selection guide [8]

Anticipated fouling Kettle or internal boiler Horizontal shell-side thermosyphon Vertical tube-side thermosyphon Forced flow
No fouling Good Good Good Expensive
Moderate Risky Good Best Expensive
Heavy Poor Risky Best Good
Very heavy Poor Poor Risky Best
 Fouling of Heat Exchanger Surfaces C4 87

cooling media. These reports present the state-of-the-art of surface area. As a result of this difference, stronger adhesion
fundamental aspects and industrial practice. should occur on rough surfaces. This is confirmed by measure-
ments with Kraft black liquor in electropolished tubes shown in
Fig. 8 [28].
2.2 Selection of Heat Exchanger Type Surface coatings for reducing the adhesion of deposits on
heat transfer surfaces have attracted increased interest in recent
If fouling will be significant, it may well control the selection years. For example, organic materials such as PTFE and Säka-
of the type of heat exchanger and its size. This is very phen have indeed been shown to reduce fouling from various
clearly demonstrated in Table 10, which recommends different fluids, for example during seawater evaporation and heat trans-
reboiler types depending on the severity of fouling [7]. Other fer to Kraft black liquor. The main reason why such materials/
examples are: coatings are not more widely used is that they are poor heat
conductors and form an additional resistance to heat transfer
● Shell and tube heat exchangers are not particularly suitable
which is comparable to the TEMA fouling resistance for cooling
for fouling conditions; however, good design practices
water. If very thin coatings were used, the resistance against
[22–26] and special baffle and tube design may be applied
erosion or other mechanical stress would be greatly diminished.
to reduce fouling.
These problems may be avoided with several novel coating
● Plate and frame heat exchangers may be attractive as they
methods, such as Ion Beam Implantation, Magnetron Sputter-
can be disassembled for cleaning and sterilizing.
ing, Multi-Arc Ion Plating, Filtered Cathodic Vacuum Arc
● Since there are no local low velocity regions in spiral plate
Plating or electroless Ni-P-PTFE plating which have been inves-
heat exchangers, these heat exchangers perform well for
tigated in recent years [29–33]. These thin and stable coatings
fluids with a high concentration of suspended solids.
have been found to reduce scale formation during convective
● Scraped heat exchangers improve the heat transfer by con-
and boiling heat transfer as well as the adhesion of bacteria.
tinuously removing deposit from the heat transfer surfaces
with rotating blades.
● Fluidized bed heat exchangers can be used where the flui-
dized particles remove deposit from the embedded tubes. 2.4 Shell and Tube Heat Exchangers
● Direct contact heat transfer may be a suitable alternative.
● Highly compact heat exchangers are normally avoided for Experience has shown that higher flow velocities and lower tube
severe fouling conditions, as they are difficult to clean. surface temperatures generally tend to reduce fouling. There-
fore, arrangements that eliminate stagnant or low velocity
regions have less overall fouling. As a general rule, the more
fouling and more corroding fluid should be placed on the tube-
2.3 Material Selection
side. The inside of tubes can be cleaned much easier than the
outside and the tubes can be made from exotic alloys at lower
The second most important point is the proper selection of the
cost than the shell.
heat exchanger material, as already minor corrosion may con-
The orientation of a heat exchanger influences the ease by
siderably increase other fouling mechanisms. In addition, the
which it can be cleaned and can effect particulate fouling. If
pipe material itself can also have an effect on fouling. For
particulate fouling is anticipated, a vertical down-flow orienta-
example, biofouling is reduced in brass tubing, as shown in
tion will permit the solids to move through the exchanger. If a
Fig. 7 for seawater fouling [27], and crude oil fouling can be
horizontal orientation is unavoidable, place the slurry on the
minimized by careful material selection [26].
Surface roughness increases the contact surface area such
that the true contact area is much larger than the apparent

C4. Fig. 8. Reduction of heat transfer coefficient during the evap-

C4. Fig. 7. Influence of pipe material on biofouling [27] oration of Kraft black liquor in pulp mills [28]
88 C4 Fouling of Heat Exchanger Surfaces

C4. Table 11. Cooling water flow velocities in condensers [29]

Pipe material Recommended velocity Minimum velocity

Arsenical copper <1.5 m/s 1.0 m/s
Admiralty 1.4–2.0 m/s 1.0 m/s
Aluminum brass 1.8–2.2 m/s 1.0 m/s
9010 cupro-nickel 1.8–2.5 m/s 1.5 m/s
90/30 cupro/nickel 2.4–3.5 m/s 1.8 m/s
Cu 1.5–2.0 m/s 1.0 m/s
Steel 2.0–4.0 m/s 1.0 m/s

C4. Fig. 10. Helixchanger Baffles (courtesy ABB Lummus)

C4. Fig. 11. Twisted Tubes (courtesy Brown Fintube Company)

C4. Fig. 9. Sketch of heat exchanger geometry and observed de-

posit formation [8]
stagnant flow zones [36]. Reduced fouling has also been
reported for the EM-baffle design developed by Shell Global
tube-side and ensure down-flow for multiple tube pass designs. Solutions [37].
While placing a slurry on the shell-side of a horizontal heat Successful installations of heat exchangers with twisted
exchanger is not recommended, some success has been reported tubes (Fig. 11) have been reported which may reduce deposit
for vertical-cut double-segmental baffles because they allow the formation both, on the shell and on the tube-side [38].
sediment to travel through the exchanger. It is often assumed that finned tubes tend more to fouling
While heat exchangers used to be designed for tube-side because of low flow velocity zones at the base of the fins. While
flow velocities around 1 m/s, modern design velocities are these problems may occur for fouling mechanisms which de-
about 2.0 m/s. Table 11 shows optimum and minimum cooling pend strongly on the flow velocity (such as biological fouling
water velocities for condensers of various pipe materials [34]. and particulate fouling), there are a number of applications
Based on 20 years of experience in the design of heat where fouling was even reduced by the use of finned tubes.
exchangers, Gilmour [35] states that ‘‘for most applications, This effect is explained by the nonuniform thermal expansion
only negligible fouling occurs if the heat exchanger is well of finned tubes due do the temperature profile along the fins,
designed. It is obvious that the majority of poorly performing which may reduce the strength of hard and adherent deposits. In
shell and tube heat exchangers were caused by mistakes in the [39] numerous investigations on fouling on finned tubes are
design of the shell-side flow path.’’ He especially emphasizes that compared with respect to the effects of fin geometry on deposit
zones with low flow velocity and bypass flows should be avoided formation. When selecting finned tubes for fouling duties it
under any circumstances. If half-moon baffles are used, the should always be considered that mechanical cleaning of finned
baffle cut should not exceed 20% of the shell inside diameter. surfaces may be difficult or even impossible.
Vertical baffles should only be used for condensation or eva- A fairly recent solution for heat transfer involving severely
poration duties but not for sensible heat transfer situations, fouling liquids is the fluid bed heat exchanger, which has been
because they allow a stratification of flow and hence a sedimen- described by Klaren [40]. Small solid particles (glass, ceramic,
tation of suspended particles. Figure 9 [8] shows qualitatively metal) are fluidized inside parallel tubes by the upward flow of
the deposit formation in two shell and tube heat exchangers for liquid. The solid particles regularly break through the viscous
identical heat duty. The smaller heat exchanger is designed with boundary layer, so that good heat transfer is achieved in spite of
appropriate baffle spacing and baffle cut, and hence has higher relatively low flow velocities. More importantly, the solid parti-
heat transfer coefficients and less fouling. cles have a slightly abrasive effect on the wall of the heat
Helical flow baffles, as shown in Fig. 10, have been used exchanger tubes, thus removing most deposits at an early
successfully because they avoid both, flow stratification and stage. Fluid bed heat exchangers have been installed in water
 Fouling of Heat Exchanger Surfaces C4 89

treatment plants, paper mills, food and dairy plants, geothermal 2.5.1 Plate and Frame Heat Exchangers
plants and in various chemical plants. In all cases, a substantial
reduction in fouling has been achieved. The application of plate heat exchangers in the chemical process
industry is increasing rapidly, where they begin to replace tubu-
lar heat exchangers in several traditional applications. Cooper
2.5 Other Heat Exchanger Types [36] investigated cooling water fouling using an APV model
R405 plate heat exchanger. The water was chemically treated
The widespread installation of compact heat exchangers has before entering the heat exchangers. As shown in Fig. 12, the
been hindered by the perception that the small passages are fouling resistance in the plate and frame heat exchanger is
stronger affected by the formation of deposits. Obviously, com- significantly lower than in the shell and tube heat exchanger,
pact heat exchangers are unsuitable for fluids containing large despite the typically lower flow velocities. If the flow velocity is
particulate material or debris. However, several investigations increased, the fouling resistance decreases similarly as it is found
demonstrated that the high shear forces, low wall superheat and for shell and tube heat exchangers. Novak [42] studied the
homogeneous flow distribution typical for compact heat fouling behavior of Rhine River water near Mannheim (Ger-
exchangers reduce the formation and adhesion of deposits on many), and of Öresund seawater in Sweden. For both waters,
the heat transfer surfaces. The use of more corrosion resistant mainly biological fouling was observed. The fouling resistances
materials with smoother heat transfer surfaces further reduces increased almost linearly over the observed period of time.
the formation of deposits. Most compact heat exchangers have Table 12 summarizes the measured effect of flow velocity on
to be cleaned chemically. Unfortunately, there is very little the fouling rate.
published information about fouling and cleaning of compact Typical values for fouling resistances in plate heat exchan-
heat exchanger types. gers are given in Table 6. Most manufacturers of plate and frame
As indicated by Eq. (3), the excess heat transfer surface area heat exchangers recommend that the excess surface should not
increases with increasing clean heat transfer coefficient for a exceed 25% of the heat transfer surface area calculated for the
constant heat duty. This places a heavy penalty on compact clean duty. Due to the nonuniformity of flow distribution and
heat exchanger types, such as plate and frame heat exchangers, deposit formation, measured pressure drop increases are signif-
if, because of ignorance or because of cautiousness, the TEMA icantly higher than values predicted using an average deposit
fouling resistances for shell and tube heat exchangers are used. thickness calculated from the fouling resistance.
Typical clean overall heat transfer coefficients for plate and The actual plate geometry (angle, amplitude, and wave-
frame heat exchangers are about 3000 W/m2K, for shell and length of corrugations) affects the formation of deposits [43].
tube heat exchangers about 1000 W/m2K. A design fouling Delplace et al. found that deposition from whey protein solu-
resistances of 0.3 m2K/kW then corresponds to 30% overdesign tions on chevron plates is only half of that of straight corruga-
for a shell and tube heat exchanger and to 90% overdesign for a tions, for otherwise identical conditions [44].
plate and frame heat exchanger.

2.5.2 Plate-Fin Heat Exchangers

Plate-fin heat exchangers are brazed/welded compact heat

exchangers with a heat transfer surface density of about ten
times that of tubular heat exchangers. Typical applications are
cryogenic, chemical/petrochemical, and hydrocarbon offshore
installations. Molecular sieves and 100 mm filters are used in
cryogenic installations to remove particulate matter or compo-
nents that may freeze-out on the heat transfer surfaces. System-
atic investigations have been performed on particulate fouling
[45] and on river water fouling [46].

C4. Table 12. Fouling rates of Rhine river water for a surface
temperature of 25 C [33]

Type u, m/s t, Pa dRf/dt, 104 m2K/kWh

Plate heat exchanger 0.13 6.7 7.4
Plate heat exchanger 0.19 14.5 4.3

C4. Fig. 12. Comparison of fouling in plate and frame, and in shell Plate heat exchanger 0.77 190.0 0.6
and tube heat exchangers [41] Spiral plate exchanger 0.43 7.5 5.0
90 C4 Fouling of Heat Exchanger Surfaces

For 3 mm ferric oxide particles suspended in water, no 2.6 Effect of Fouling on Pressure Drop
blockage of plain fin or wavy fin channels was observed. Wavy
fin channels fouled more than plain fin channels. All experi- The formation of deposits on the heat transfer surfaces causes
ments showed asymptotic behavior. Higher deposition rates an increase of the frictional pressure drop due to increased
were obtained for non isothermal conditions and at higher surface roughness and restricted cross-sectional flow area.
bulk temperatures. Maximum deposition occurred at a Rey- According to Chenoweth [8], more heat exchangers are taken
nolds number of about 1500 [45]. out of service because of excessive pressure drop than because of
Fibrous and biological material partially block the inlet of reduced heat transfer.
the aluminum plate-fin test sections when used with river water,
which was filtered through a 1 mm mesh. Some deposition was
found at locations where corrosion of the aluminum had oc-
2.6.1 Tube-Side Pressure Drop
curred. In the wavy fin test section, a thin, uniform deposit of
fine mud was observed. Pressure drop for the plain finning
A rough estimate of the tube-side pressure drop can be made if
increased linearly with time, whereas asymptotic behavior was
it is assumed that the deposit is distributed evenly at the tube
found for the wavy finning. The initial slope of the relative
inside. The frictional pressure drop in cylindrical tubes is calcu-
pressure drop vs. time curves was 5.8·108s1 for the plain
lated from Eq. (5):
fins and 1.71·107s1 for the wavy fins. For the latter, an initial
deposition rate of 4.8·1012 s m2K/W and an asymptotic  Dp r u2
fouling resistance of 6·106m2K/W was measured [46]. ¼x ð5Þ
DL 2di

the friction factor for smooth tubes

2.5.3 Printed Circuit Heat Exchanger
x ¼ 0:0056 þ 0:5 Re0;32 ð6Þ
The passages in Printed Circuit Heat Exchangers (PCHEs) are and for rough tubes
typically between 0.3 mm and 1.5 mm deep. The specific design
x ¼ 0:014 þ 1:056 Re0;42 ð7Þ
leads to volumetric heat transfer areas of 500–2,500 m2/m3,
which is an order of magnitude higher than shell and tube If the fouling resistance and the thermal conductivity of the
heat exchangers. Experiments are described in [47] to compare deposit are known, the inside diameter of the fouled pipe can be
the fouling related drop in performance of a PCHE and of a determined by Eqs. (8) and (9):
double pipe heat exchanger (DPHE). The cooling water treated  
di di
against corrosion, scale formation and biofouling, and a Rf ¼ ln ð8Þ
2ld df
0.5/1.0 mm strainers was installed to reduce particulate fouling.  
For operating times of 500–660 h, no change in thermal effec- 2ld Rf
df ¼ di exp ð9Þ
tiveness was observed for the PCHE, but the pressure di
drop increased by up to 55% due to the deposition of particulate
material. The addition of a stainless steel mesh insert for The pressure drop for the fouled tube is obtained by using
the removal of fibrous material significantly reduced the increase Eq. (5) in conjunction with Eqs. (7) and (9). It is generally
in pressure drop. No deposition was observed in the parallel found that the above equations under-predict the effect of
DPHE. fouling on pressure drop, since they assume a uniform distribu-
PCHEs have been used for gas cooling using seawater [47]. tion of deposit over the total heat transfer surface.
200 mm strainers have been installed upstream of the heat
exchanger and chlorine was added to counter biofouling. No
operational problems have been reported. Another application
involved the heating of tail gas in a nitric acid plant using C4. Table 13. Ratio of fouled to clean shell-side pressure drop [48]
condensing steam. After 18 months of operation, no indication Shell diameter/baffle
of channel blockage could be detected. spacing
Deposit heat transfer coefficient 1.0 2.0 5.0

2.5.4 Polymer Compact Heat Exchangers Laminar flow

1/Rf = 6000 W/m2K 1.06 1.20 1.28
Polymer heat exchangers are used for low pressure operations 1/Rf = 2000 W/m2K 1.19 1.44 1.55
involving corrosive gases or liquids. The low surface energy and 1/Rf < 1000 W/m K2
1.32 1.99 2.38
the smooth surface of their construction materials (polypropyl- Turbulent flow
ene, fluoropolymer etc.) reduce the stickability of most deposits. 1/Rf = 6000 W/m2K 1.12 1.38 1.55
Since clean heat transfer coefficients are already low (150–250 2
1/Rf = 2000 W/m K 1.37 2.31 2.96
W/m2K), these heat exchangers react less sensitively to an addi-
1/Rf < 1000 W/m2K 1.64 3.44 4.77
tional fouling resistance than metallic heat exchangers.
 Fouling of Heat Exchanger Surfaces C4 91

2.6.2 Shell-Side Pressure Drop ● Heavy particulate matter (e.g., catalyst fines) absent
● Reasonable salt content (no desalter malfunctions)
The effect of fouling on the shell-side pressure drop can be
a) Minimum Liquid Velocity
estimated using Table 13 from Coulson et al. [48].
● Tube-side velocity of 2 m/s. This velocity limit is applicable
for tubes with outside diameters of 19.05 mm and 25.4 mm.
Increase velocity to 2.2 m/s for tube diameters of 31.75 mm
3 HTRI Fouling Mitigation by Design and 38.1 mm to maintain shear stress.
Method ● Shell-side cross flow stream should be at least 0.6 m/s. If the
shell-side flow is fully longitudinal, the minimum shell-side
Based on almost 50 years of experience as some of the world’s velocity should be 1.2 m/s. For longitudinal flow bundles,
leading heat exchanger design experts, Heat Transfer Research tubes removed for entrance/exit considerations at the shell
Incorporated (HTRI) [22–26] have developed a design method- nozzles should be replaced in the bundle proper with
ology that yields smaller, more cost-effective shell and tube plugged dummy tubes or rods of the same diameter to
heat exchangers with extended run times between cleanings. maintain a uniform flow field and minimize bypass streams.
While this methodology has, so far, only been validated for The bundle-to-shell diameter ratio is to be made as close to
crude oil processing, its rigorous approach can be taken as an 1.0 as practical under TEMA [4] clearance rules.
example for other fluids and heat exchangers types. These tech-
niques were developed by HTRI with help from industry through b) Maximum Temperature
the HTRI Exchanger Design Margin Task Force (EDMTF). The ● The maximum tube wall temperature should be 300 C.
goal of the EDMTF is to develop the design philosophy for adding Shell-side design with cross-flow baffles
margins to heat exchangers to allow for process uncertainties ● The B-stream fraction according to TEMA nomenclature
and fouling. [4] should be at least 0.65.
Experience has shown that fouling may be mitigated for ● Single-segmental baffles should be selected. If the shell-side
many services through proper heat exchanger design and oper- pressure drop is prohibitively high, double-segmental, heli-
ation. For the experienced designer, fouling resistances are not cal, EMbaffle, rod baffle, squared, or no-tube-in-window
used when operating data for identical or similar services are (NTIW) baffle configurations may be considered.
available. In these cases, designing with the proper attention to ● Baffle cut orientation should normally be horizontal for
velocity (or shear stress) and wall temperature can prevent TEMA type E and J shells. Baffle cut orientation for TEMA
significant fouling whereas ‘‘the mere use of a high fouling type F and G shells should be vertical. If slurry must be
resistance will generally engender a high degree of fouling.’’ placed on the shell-side of a horizontal heat exchanger,
A small design margin may be added to the design to address consider vertical cut double-segmental baffles to allow the
design uncertainties. Rarely is this margin in excess of 30%. More sediment to exit the shell.
than 30% excess margin calls for a root cause analysis of the ● Baffle cut for single-segmental designs should be 20–25% of
problem followed by a fouling (or design) mitigation strategy. the shell inside diameter, where 20% is preferred. It may be
Except for rare cases of intentional high variability in throughput, increased up to 25% to reduce leakage streams.
more than 30% excess margin in a heat exchanger design indi- ● The ratio of window velocity to cross-flow velocity (includ-
cates the presence of unresolved engineering issues and can often ing leakage streams) should be less than 2.0 for designs with
be a significant source of hidden cost to the owner. tubes in the window (1.0–1.5 is preferred). For no tubes in
It is good practice to design for an allowable pressure drop window designs, the ratio of window velocity to cross flow
derived by reducing the maximum available pressure drop in velocity should be less than 3.0 (1.5–2.0 is preferred). Re-
the clean condition by the amount of excess margin anticipated. finement of this guideline is an area of research.
This permits any excess margin to be applied in such a way that
design shear rates and wall temperatures are not reduced. The
maximum available pressure drop in the clean condition is
estimated as the maximum available pressure drop divided by 3.2 Cooling Tower Water Best Practice
the fractional pressure drop increase when the exchanger is Operating Conditions
operated in the fouled condition.
For the case of those cooling water streams which are closely
regulated in the plant for velocity control and are kept reason-
ably clean with a water maintenance program, fouling mitiga-
3.1 Crude Oil Best Practice Operating
tion strategies apply. The cooling water temperatures should be
Conditions
designed and operated to not exceed a maximum bulk temper-
ature of 50 C or a maximum wall temperature of 60 C.
The fluid scope for this design methodology is:
In addition, there must be sufficient velocity to maintain
● Medium-to-high boiling point liquid hydrocarbon mixture any particulate in suspension as well as to produce enough wall
with API gravity less than 45 shear to stabilize any fouling which does occur. There are many
92 C4 Fouling of Heat Exchanger Surfaces

sources for information on minimum cooling water velocities drop (and hence the flow velocity) is too low, fouling may
for design, such as those given in Table 11. In reality, the exact become inevitable and fouling mitigation impractical.
minimum value for any cooling water system is so dependent
upon the contaminants dissolved in the water that one single
value for this purpose can only be regarded as an approxima- 3.4.3 Longitudinal Baffles
tion. In the final analysis, the judgment as to minimum design
water velocities while adhering to prudent water temperature If a longitudinal baffle is used in heavy fouling service where
limitations must be made by those knowledgeable about the shell-side pressure drop in one shell exceeds 35 kPa (70 kPa with
water used. a ‘‘kempchen’’ style of seal), the baffle shall be welded to the
shell. Note that welding the longitudinal baffle to the
shell requires a shell inside diameter of at least 0.7 m and, for
3.3 Excess Surface/Coefficient Adjustments the bundle to be removable, U-tubes must be used with the
U-bends in the horizontal plane (normally two or more tube
● If both fluids are within the scope outlined above, approxi- passes per shell pass). The designer should investigate differen-
mately 20–25% excess surface should be provided instead of tial thermal stresses across the shell. In general, a welded
applying fouling resistances. This design margin may be longitudinal baffle is probably acceptable where the shell-side
reduced when the designer has confidence in the fluid temperature difference across one shell does not exceed 90 C.
properties, predictive methods, and successful mitigation For leaf seal construction, shell-side operating temperature dif-
of fouling (usually based on prior experience for a similar ferentials of 195 C across one shell have been accommodated
service). with proper mechanical design. These rules are for 6.7 m
● If only one fluid is operating under the best practice condi- straight-length tube bundles and will vary with bundle length.
tions, a fouling resistance should be selected for the fluid Bundle slide rails in both top and bottom portions of the bundle
outside scope. For non-fouling fluids outside scope, a foul- need to be provided.
ing resistance of 0.000088 m2K/W is recommended to com- The following segmental-baffle construction features may be
pensate for heat transfer surface changes during start-up. considered to improve shell-side performance:
For the fluid within scope, the heat transfer coefficient is ● American Petroleum Institute Standard 660 requires that
multiplied by 0.83 and no fouling resistance is used. As a seal device (dummy tubes, rods, or strips) be implemented
above, the design margin may be reduced based upon from 25–75 mm from the baffle tips, and for every 5–7 tube
operating experience. pitches thereafter. The number of seals may have to be
increased to limit the bundle and pass lane leak streams.
● Where the tube-to-baffle diametral tolerance (as specified by
3.4 Design Recommendations TEMA [4]) is 0.8 mm, the tolerance may be reduced to 0.4
mm if required to reduce the leak stream between the tube
3.4.1 Shell-Side Bundle Geometry and baffle hole.
Exit/Entrance Constraints ● The TEMA [4] baffle-to-shell diametral clearance may be
reduced to limit the baffle-to-shell leakage stream. A clear-
● Where impingement protection is required, use impinge- ance of 0.0035–0.004 times the shell diameter is achievable
ment rods. One row of rods is acceptable for 90 tube for shells rolled from plate, but use this extra tight clearance
layouts, two rows for staggered pitch. Impingement plates only if necessary, as it is difficult to guarantee compliance.
should be avoided. Extra tight clearance is not recommended for shells made
● Large baffle end spaces and correspondingly low local veloc- from pipe (typically NPS 24 and smaller).
ity sometimes occur due to geometry constraints. When the ● Baffled TEMA [4] F and G shells may be considered to
end baffle space is greater than or equal to 1.5 times the increase shell-side velocity, reduce the number of shells in
space between baffles, the area back from the first baffle to series, and/or improve the baffle-spacing-to-shell-diameter
1.5 times the baffle spacing is to be considered 65% effective aspect ratio.
regardless of baffle type or orientation. All remaining area to
the tube-sheet is to be considered ineffective for heat trans-
fer. Additional area should be provided in the bundle proper
3.5 Example for Fouling Mitigation by Design
as compensation. An annular distributor may be considered
if the affected surface area is large.
To demonstrate the potential savings of the design methods
outlined in this section, the following example will be used.
This example is for the last shell-and-tube heat exchanger in
3.4.2 Allowable Pressure Drop the crude oil preheat train prior to the fired heater. Selection of
material is important, and stainless steel or high chrome steel
Pressure drop should be provided as required to meet the should be selected for the design. The process conditions are
minimum critical velocities noted in Sect. 3.1. If the pressure summarized in Table 14a.
 Fouling of Heat Exchanger Surfaces C4 93

3.5.1 Original Design from adhering to the outside of the tubes and allow it to be
carried through the exchanger without deposition. How-
The original heat exchanger is a TEMA type AES consisting of ever, to do this required an appropriate redesign of the heat
two shells. The design and performance parameters for the exchanger.
exchanger unit are given in Table 14b.

3.5.4 New Design

3.5.2 Operation History with Original Design
Using the procedures outlined in this section, a replace-
This service was a consistent high fouling problem and would ment heat exchanger was designed for this service as given
lose about 57% of its performance capacity within the first 6 in Table 14c.
months after cleaning, resulting in excess energy (fired heater The new design consists of a horizontal welded long baffle
fuel) and related costs of about €62.000 per month averaged on the shell-side, shell nozzles located at the rear of the shell
over a 2 year turnaround cycle. beyond the bundle U-bends so that there are no erosion con-
cerns, and horizontal U-bends with a vertical channel pass
3.5.3 Root Cause Analysis of Performance partition to produce a removable, two tube pass bundle in a
welded long baffle F-shell. Both the original and new designs
At the next turnaround, the bundles were pulled for cleaning. employed segmental baffles on the shell-side.
Prior to cleaning, visual bundle inspection revealed a ‘‘fuzzy’’
looking coating on the outside of the tubes consisting of oil and
coking fines from the vacuum unit. The coating was slightly 3.5.5 Final Outcome
sticky but did not solidly adhere to the surface and could be
easily wiped away. The tube-side fouling appeared to be slight. The new design used no preset fouling resistances but instead
In addition, analyzing the fouling performance monitor over used the allowable pressure drop to produce the higher shear
the course of the 2-year turnaround cycle indicated a perfor- rate necessary to inhibit deposition. A minimum of 15% excess
mance plateau at about 6 months after which the fouling surface was thought prudent to handle design uncertainties in
seemed to stabilize for the next 1.5 years. These data strongly the new configuration. The new exchanger performed at or
indicated shear rate controlled fouling at the shell-side. above expected design over the 2-year turnaround cycle pro-
From the above data, it was estimated that a shell-side ducing an average €68.000 per month cost savings over the
velocity slightly greater than 1.5 m/s should keep the fouling original heat exchanger.

C4. Table 14a. Process conditions for design example

Stream Fluid Flow rate Temperature in Temperature out Allowable DP

 
Location Designation kg/s C C kPa
Shell-side Heavy Vacuum Gas Oil 71 366 338 100
Tube-side Crude Preheat 88 289 311 70

C4. Table 14b. Original heat exchanger design

Number of shells TEMA Number of passes Fouling resistance Velocity m/s DP design kPa Uc/Uf
Size Type Shell Tube Shell Tube Shell Tube Shell Tube
2 AES 1 2 0.00123 0.0007 0.37 1.22 8.5 32 1.95
1016 mm  6.1 m

C4. Table 14c. Improved heat exchanger design

Number of shells TEMA Number of passes Fouling resistance Velocity m/s DP design kPa Uc/Uf
Size Type Shell Tube Shell Tube Shell Tube Shell Tube
1 AFU 2 2 – – 1.85 2.38 91 58 1.16
610 mm  6.1 m
94 C4 Fouling of Heat Exchanger Surfaces

C4. Fig. 14. Recirculation of cooling water

4.2 Chemical Fouling Mitigation Methods

Since about 1920, a number of companies have specialized in

the mitigation of fouling and corrosion, mainly for the flow of
cooling water and hydrocarbons. These companies have gained
considerable expertise and have developed a wide range of
C4. Fig. 13. Fouling resistance as a function of flow velocity and
additives and equipment. Services include the supply of chemi-
surface temperature during start-up of a new or cleaned heat
cals as well as the analysis of cooling water, the evaluation of
exchanger
potential fouling and corrosion problems, and complete treat-
ment programs including continuous monitoring of the system.
In what follows in this subsection, only a small selection of
4 Online Mitigation Methods methods to reduce fouling by chemical means is discussed.
For the final selection of the treatment as well as for the dosage
The following section provides an overview of the broad cate- of treatment chemicals, specialists should be consulted. More
gories of mitigation methods, and describes some general details may be found in [49] provided by industrial fouling
approaches. For more detailed information see [49]. mitigation companies.
Commercial antifoulants usually contain a number of com-
ponents. These polyfunctional antifoulants are more versatile
4.1 Start-Up Procedures and effective since they can be designed to combat various types
of fouling that can be present in any given system. Antifoulants
The use of constant fouling resistances in the design of heat are designed to prevent equipment surfaces from fouling, but
exchangers leads to initially oversized equipment. Heat duties in they are not designed to clean up existing deposits. Therefore,
new or cleaned heat exchangers can, therefore, be considerably antifoulant addition should be started immediately after equip-
higher than the design specifications. In most chemical pro- ment is cleaned.
cesses, however, product inlet and outlet temperatures, product
flow rate and cooling water inlet temperature are specified. If
4.2.1 Scale Formation
this is the case, the heat exchanger is usually controlled via the
flow rate of the cooling water. To reduce the heat duty, the water
In general, there are three alternatives available to mitigate or to
flow velocity must be reduced. Figure 13 shows that this proce-
prevent scale deposition due to high concentration of scale-
dure may cause a considerable increase of fouling as compared
forming ions in aqueous solutions:
to fouling under design operating conditions.
Point ‘‘A’’ refers to the design values of flow velocity and heat
Removal of Scaling Species
transfer surface temperature. As the heat exchanger is initially
Scaling species may be removed by ion exchange and by chemi-
overdesigned, the cooling water flow velocity is throttled which
cal treatment. In the latter treatment, carbonic acid, and calci-
also causes an increase of the heat transfer surface temperature,
um hardness are removed by the addition of chemicals. If the
see point ‘‘B.’’ However, fouling at ‘‘B’’ is considerably worse and
lime treatment is used:
deposits created during this part of the operation may not be
removed completely, even if the flow velocity is increased, later. CaðHCO3 Þ2 þ CaðOH Þ2 ! 2 CaCO3 þ 2H2 O ð10Þ
Therefore, by specifying high fouling resistances, fouling may
become a self-fulfilling prophecy. If part of the cooling water is
CO2 þ CaðOH Þ2 ! CaCO3 þ H2 O ð11Þ
recirculated, as shown in Fig. 14, the flow velocity and the
cooling water inlet temperature can be increased to meet the During slow decarbonization (1–3 h reactor residence time), the
required heat duty. The anticipated fouling (‘‘C’’) will be similar calcium carbonate precipitates as silt, during fast decarboniza-
to the design value (‘‘A’’), but a price is to be paid to provide the tion (5–10 min reactor residence time) it precipitates in the
higher flow velocity. form of particles. With the exception of installations with high
 Fouling of Heat Exchanger Surfaces C4 95

calcium hardness or large throughput, chemical removal of maintain CaCO3 in solution. Lists of additives to reduce crys-
scaling species is not used anymore. tallization from hard waters have been compiled by Harris and
Instead, ion exchangers are used in which the ‘‘harmful’’ Marshall [52] and by Krisher [53].
scaling species in the fluid are replaced by ‘‘harmless’’ ions (for
example Ca++ or Mg++ by Na+). Ion exchangers are usually
manufactured from styrene based polymers. The so-called cat- 4.2.2 Particulate Fouling
ionic exchangers contain weak and strong acids; the anionic
exchangers contain weak and strong alkaline groups. With Particulate fouling is usually mitigated by the addition of sur-
these two variations, all cations and anions can be removed factants or dispersants. If the surface tension is reduced, large
from the fluid. Ion exchangers have to be regenerated regularly particle agglomerates can break down into smaller particles,
with the appropriate salt solution. According to [50], chemical which tend less to sedimentation. Dispersants impart like
decarbonization leaves a residual hardness of 17–30 ppm as charges to both the heat transfer surface and the particles and
CaCO3, ion exchange can reduce the hardness down to 2 ppm reduce deposition. For cooling water applications polyacrylates
as CaCO3. Both methods of fluid treatment have high capital or polysulfonates are used with molecular weights between 2000
and operating costs. and 3000 g/mol. According to [53], the addition of polypho-
In the oil industry, desalters are installed at the beginning of sphates to reduce scaling may cause a slight reduction of the
the crude oil heat exchanger train to replace salty water, which dispersion of particulates.
may otherwise cause scale formation at higher temperatures.

pH Control 4.2.3 Chemical Reaction Fouling

The solubility of scale-forming constituents increases with de-
creasing pH. Many treatment programs, therefore, involve the Chemical reaction fouling increases exponentially with increas-
addition of acid (usually H2SO4) to the system to maintain a pH ing heat transfer surface temperature according to an Arrhenius
in the region of 6.5–7.5. If the system contains corrosion-resis- term, see Eq. (4). As activation energies E are fairly high for
tant materials, a pH may be selected at which no scaling will chemical processes, even a modest reduction of the heat transfer
occur. The Langelier Saturation Index or the Ryznar Stability surface temperature due to process or design modifications may
Index [51] are commonly used to determine the value of pH to already cause a considerable reduction of fouling. Particles
be adjusted. suspended in the fluid (e.g., from upstream corrosion) can act
as catalysts. Reaction fouling may be mitigated by removing
Scale Inhibitors these particles.
Growth of crystals or the nucleation of crystals can be inhibited Especially for oil refining processes a number of chemical
by the addition of scale inhibitors. Many proprietary com- additives to reduce reaction fouling have been developed.
pounds are available for scale control. Most antifoulants have several functions. Generally they are
Chelating agents (for example EDTA) complex strongly oxygen scavengers, metal deactivators and dispersants [54].
with the scaling cations and hence inhibit their deposition at For autoxidation-induced fouling, antioxidants can be added
the heat transfer surface. Inhibitor and scalant must be available to consume oxygen or react with oxidation products in a way
in stoichiometric ratios. as to prevent the chain reaction of the autoxidation process,
Processes, which are based on physical rather than on chem- or metal deactivators are added to chelate metal ions thereby
ical reactions are those that stabilize supersaturated solutions by preventing their catalytic effect on the autoxidation process.
adsorption at the crystal nuclei (for example polyphosphates) or Once insolubles form by either autoxidation or thermal decom-
that modify or weaken the crystalline structure (for example position, dispersants can be added to minimize agglomeration
polycarboxylic acid). Table 15, which has been adopted from of small insoluble polymeric or coke-like particles into
Harris and Marshall [52], shows the ability of additives to larger particles or deposit, or sticking of particles to the tube
wall [55].

C4. Table 15. Ability of various additives to maintain CaCO3 in Antioxidants

solution [52] Even very small amounts of oxygen can cause or accelerate
% Inhibition at dose level
polymerization. Accordingly, antioxidant type antifoulants
have been developed to prevent oxygen from initiating poly-
Additive 2.5 ppm 5.0 ppm 7.5 ppm 10 ppm
merization. Antioxidants act as chain-stoppers by forming inert
Polyphosphate 98% 98% 99% 100% molecules with the oxidized free radical hydrocarbons.
Aminophosphonic acid 97% 96% 95% 94%
Acetodiphosphonic acid 83% 82% 83% 90% Metal Deactivators
Polyacrylate 30% 65% 84% 93% Traces of metals are invariably present in hydrocarbon streams,
which may catalyze polymerization reactions. For example tran-
Polymaleic acid 26% 35% 44% 56%
sition metal ions, such as Cu, Fe, Zn, and Mn are powerful
EDTA 15% 20% 20% 20%
hydroperoxide decomposers and provide a steady source of free
96 C4 Fouling of Heat Exchanger Surfaces

radicals for oxidation chain initiation. By complexing the metal stickability of suspended matter and hence promote further
ion, it can be prevented from participating. Thus chelating deposition. Biological growth is usually controlled by addition
compounds are used as metal deactivators [55]. of biocides. In recent years, chlorine has most widely been used,
which reacts with water to hydrochloric and hypochlorous acid:
Dispersants Cl2 þ H2 O ! HCl þ HOCl ð12Þ
Dispersants or stabilizers prevent insoluble polymers, coke, and
other particulate matter from agglomerating into large particles Hypochlorous acid is an extremely powerful oxidant that easily
which can settle out of the process stream and adhere to the diffuses through the cellular walls of microorganisms. It is
metal surfaces of process equipment. They also modify the assumed [64–66] that HClO oxidizes the active sites of certain
particle surface so that polymerization cannot readily take enzyme sulfhydryl groups, which constitute intermediate steps
place. Dispersants generally play an important role in antifou- in the production of adenosine triphosphate (ATP). The system
lant programs. The feedstock or hydrocarbon stream may al- ATP-ADP allows conversion of carbohydrates and hence the
ready contain polymerized materials which, if allowed to energy supply for living organisms. Contrariwise to other, non-
agglomerate, would deposit. In most applications it is not oxidizing biocides, chlorination also weakens the biofilm matrix
possible to fully eliminate oxygen- or metal-induced reactions allowing the removal of biofilms from the heat transfer surface
and dispersants are necessary to prevent the polymerized mate- by fluid shear forces. Continuous application of chlorine at
rials from agglomerating and depositing on heat transfer sur- concentrations between 0.1 and 0.5 ppm has shown to be a
faces [54, 56]. Dispersants are believed to function by absorbing reliable but costly method to avoid deposition. Cheaper but less
on the surface of materials which are insoluble in the organic effective is a dosage of 1–10 ppm for 15 min in intervals of 4 h.
fluid and converting them to stable colloidal suspensions. They However, it was found that biofilm growth is accelerated after a
generally contain polar groups which absorb on the particle shock chlorination, see Fig. 16.
surface and nonpolar hydrocarbon-soluble groups to affect dis- Biological fouling control with chlorination has the disad-
persion by their solubilization. Mayo et al. [57–62] argue that it vantage that chlorine has to be added continuously, since it does
is the solubilizing properties of dispersants which are most not only react with microbes but also with process contami-
important in reducing deposit formation. nants such as H2S or NH3. Chlorine concentration in water
Figure 15 shows the performance of the feed/effluent exceeding 0.5 ppm may give rise to corrosion problems, espe-
heat exchanger in an oil refinery with and without chemical cially for stainless steel equipment. Due to the biocidal action of
treatment. chlorine, there are increasing restrictions on the effluent chlo-
rine concentration. For these reasons, chlorine is increasingly
replaced by other chemicals such as methylene-thiocyanate or
4.2.4 Biofouling chlorophenoles, see Waite and Fagan [68].
Even though the addition of hypochlorous acid provides
The environment in cooling towers and cooling systems is par- an effective method against the growth of a wide range of
ticularly conducive to the growth of microorganisms in water and bacteria and algae, there are a number of species that
on surfaces of the system. Microorganisms attach and grow on can only be controlled by excessively high HOCl concentra-
surfaces, and produce polysaccharides, which increase the tions (>30 ppm). Since this may cause operational problems,
compounds have been developed to eliminate these species,
which can be added to the chlorinated water. Grade and
Thomas [69] discuss treatment programs, which are effective
against bacteria and algae. Generally, it is recommended to

C4. Fig. 16. Effect of shock chlorination on the growth of biological

C4. Fig. 15. Reformer feed/effluent exchanger performance [63] matter at heat transfer surfaces [67]
 Fouling of Heat Exchanger Surfaces C4 97

vary biocide treatment regularly to avoid immunization of slag. Magnesium oxide can minimize fouling of coal-fired
microorganisms. boilers by acting as a catalytic inhibitor to retard the formation
Because of the toxic effect of copper ions on biological of SO3 and reacting to MgSO4 which is inert and has a high
matter, another method to reduce bacterial growth is the use melting point. Often, the amount of MgO required, 0.4–3%
of piping with a copper content above 60% or the addition of of the fuel burned, makes the practice uneconomical. How-
copper sulfate to the water. For potable water, the copper ever, using fine, particle-size magnesia dispersions at rates
concentration must be below 1 ppm. of 0.005–0.015% (weight) of the fuel has reportedly reduced
fouling.
For oil-fired boilers, additives are used to control SO3
4.2.5 Corrosion Fouling related problems, high temperature fouling, high temperature
corrosion.
Generally, it is desirable to have a thin, passivating oxide layer
on the heat transfer pipes. Epstein [11] mentions that this oxide
layer is removed if the flow velocity exceeds 30 m/s. Excessive
corrosion can be controlled by the addition of corrosion inhi-
4.3 Mechanical Fouling Mitigation Methods
bitors (chromate or polyphosphate based) or by control of the
A number of mechanical mitigation techniques have been de-
pH. Chromate is a highly efficient and cost effective inhibitor.
veloped which generally are based on one of the following
However, the toxicity of chromates in the environment has
mechanisms:
restricted their use. This also holds for zinc based inhibitors.
Under some circumstances, corrosion inhibitors (such as (a) Short-time overheating of the heat transfer surfaces. The
phosphates) themselves can be the source of fouling in heat different thermal expansion of tubes and tube deposits may
exchangers as they increase the total salt content of the water. cause cracking of the deposit
However, this can be mitigated by careful control of parameters (b) Mechanical vibration of heat transfer surfaces
such as inhibitor concentration, flow velocity and surface (c) Acoustical vibration of heat transfer surfaces
temperature [70]. (d) Increased shear stress at fluid deposit interface.
(e) Reduced adhesion of deposits
4.2.6 Gas-Side Fouling [3]

Removal of contaminants which promote fouling, such as 4.3.1 Liquid Flow

sodium, sulfur, or vanadium, from fuels prior to combustion
and contaminant removal from combustion gases are two Most of the commonly used fouling mitigation techniques have
approaches to mitigate gas-side fouling. Water washing has been developed for the tube-side liquid in shell and tube heat
helped to overcome some of the fouling problems experienced exchangers. Even though attempts have been made to develop
with residual oils in marine applications by removing sodium mechanical online mitigation devices for non-tubular heat
and sediment. Inorganic sulfur can be removed from coal by exchangers, their installation has not penetrated the market.
gravity settling or by froth floatation if the mineral particles are
well above micron size. Reversal of Flow Direction
Electrostatic precipitators, mechanical collectors, fabric fil- Regular reversal of the flow direction in conjunction with a
ters or wet scrubbers can be used to remove particles from short-time increase of the flow velocity is sometimes used as a
combustion gas streams. Removal of gaseous constituents, method to mitigate the formation of weak deposits. Figure 17
which is considerably more difficult than particle removal, shows that this procedure reduces the fouling resistance, but
may involve limestone addition, wet scrubbing without sulfur
recovery, MgO systems with sulfur recovery or use of dry sor-
bent systems.
Under certain conditions, chemical fuel additives or
chemical flue gas additives can mitigate the effects of gas-side
fouling and corrosion. Many proprietary additives have been
marketed, with varying degrees of success in mitigating fouling.
Fuel additives to improve combustion efficiency in boilers
and to keep gas-side heat transfer surfaces clean by reducing
soot and slag formation have been developed by companies
such as Betz [71], Dearborn, Drew [72], and Nalco [73]. Addi-
tives that have been used to control gas-side fouling in boilers
include aluminum oxide, ammonium bicarbonate, magnesium
oxide, magnesium carbonate, silica, and zinc. For coal-fired
boilers, additives are used for fly ash conditioning for electro-
static precipitation, convective tube fouling and coal-ash C4. Fig. 17. Continuous cleaning by reversal of flow direction [74]
98 C4 Fouling of Heat Exchanger Surfaces

only for a short period of time. A much better performance efficiency is always accompanied by an increased pressure drop
could be achieved by operating at a higher flow velocity. per unit length; therefore, these inserts work best for flow in the
laminar or transitional flow regime. In combination with fur-
Gas Rumbling ther reduction of flow velocity (i.e., tube passes) design varia-
Deposits with moderate stickability to the heat transfer surfaces tions may be possible where significant improvements of heat
(e.g., particulate, and some biological deposits) can be dislo- transfer can be achieved with no or little increase in pressure
cated and washed out by increasing the fluid shear forces for a drop. Typical inserts are twisted tapes, coils (Fig. 18) and wire
short time, in regular time intervals. This can be achieved by matrix inserts (Fig. 19). The potential of some of these inserts
increasing the flow velocity, if enough pump capacity is avail- with respect to reducing deposit formation is reported in [49].
able. More effective is, however, to introduce compressed air or Figure 20 shows the effectiveness of the TURBOTAL system in
nitrogen into the liquid system. The resulting highly turbulent reducing fouling in a crude oil preheater [77].
gas-liquid two-phase flow can provide shear forces and pressure
fluctuations, which are substantially higher than for single-
phase flow. Gas rumbling is commonly used in cooling water
applications.

Ultrasound
On the laboratory scale, some success has been achieved in
removing/inhibiting deposits by ultrasonic vibrations. So far,
however, technical limitations have prevented the extrapolation
of these results into industrial practice.

Tube Inserts
Tube corrugations and tube inserts can increase the plain tube
heat transfer coefficient by a factor of 2–15 [49]. This is achieved
by reducing the average thermal boundary layer thickness. As
deposition rates for most fouling mechanisms are inversely C4. Fig. 20. Crude oil fouling mitigation with Turbotal inserts [77]
dependent on fluid wall shear stress and heat transfer surface
temperature, reduction of the viscous and thermal sublayer
thickness may also considerably reduce fouling. It must be
considered that, for constant mass flux, the increased thermal

C4. Fig. 21. Continuous cleaning with wire brush system

C4. Fig. 18. Spiral insert (SPIRELF system) [75]

C4. Fig. 19. Cal-Gavin wire insert to increase heat transfer for flow in
pipes [76] C4. Fig. 22. Typical layout of sponge ball cleaning system [79]
 Fouling of Heat Exchanger Surfaces C4 99

Continuous Transport of Cleaning Devices Through Tubes dislodge and frequently require shutdown for their removal.
These methods require major modifications of the flow system Jet soot blowers come in two types: (i) The fixed position
and are, therefore, best implemented in the design stage. How- rotating type is installed inside the heat exchanger and (ii)
ever, they have the advantage that exchangers may be kept clean the retractable type periodically passes an externally
over long periods of time. All systems work best if applied to an mounted nozzle through the heat exchanger. The fixed
initially clean heat exchanger. position soot blowers require little additional floor space,
A number of companies (MAN, Water Services of America, but they can usually not be used if the temperature exceeds
KALVO [78], ATCS) have developed continuous tube cleaning 1000 C. As more than 100 soot blowers may be installed in
systems using small nylon brushes which are inserted into large fired boilers, steam, and pressurized air consumption
each tube, see Fig. 21. These brushes are pushed through the may cause considerable costs.
tubes by the fluid flow. For continuous operation and optimum ● According to the manufacturers, installation, and opera-
cleaning efficiency, the flow direction has to be reversed about tion costs of sonic soot blowers are only 10% of those of jet
every 8 h. Life expectancy of the brushes is about 5 years. It is soot blowers. Sonic soot blowers perform best in the cooler
claimed that the time for amortization is between 8 and regions of furnaces or in other apparatus where glassy phases
16 months. of deposits are not encountered. They operate by emitting
There are many examples for the successful application of sound pressure waves that loosen the particulates and allow
the brush tube cleaning system. However, their most effective them to be carried away with the gas stream. Under normal
installation is in smaller, water-cooled heat exchangers, for operations, sonic horns need only sound for 15–30 s every
example, for the central air-conditioning systems of office 10–30 min. Horns are constructed of materials that can
buildings, hotels, or hospitals. withstand temperatures up to 1000 C. Sonic soot blowers
For large installations, more consistent results were obtained may not, however, be able to loosen the harder deposits that
with a system where sponge balls with a rough surface are can be removed by the high velocity steam, air or water jets.
circulated through the heat exchanger, see Fig. 22. The diameter Sonic soot blowers are available at sonic or infrasonic range.
of the sponge balls is slightly larger than the inside diameter of
For very sticky deposits or if jet soot blowing may cause the
the tubes and the system is designed such that each tube sees a
temperature to drop below the acid dew point, 5 mm diameter
sponge ball every 5–10 minutes. Since the diameter of the
cast iron spheres may be poured over the pipe arrangement. For
sponge balls decreases with time and because of inevitable ball
extremely severe gas-side fouling problems, fluidized bed tech-
losses through the screening system, the sponge balls have to be
nology should be considered as an alternative.
replaced regularly. For hard and adherent deposits, carborun-
The control of operating conditions is a very important
dum coated sponge balls can be used. According to the manu-
consideration in the prevention of gas-side fouling. Some of
facturers, application of sponge ball systems may reduce the
the most important controls are:
fouling resistance to close to 0 m2K/kW. The application of
sponge ball systems is limited to temperatures below 120 C. ● Maintain surface temperature above acid dew-point
Several companies, for example, TAPROGGE or CQM supply temperature
sponge ball systems and complete maintenance packages with ● Control amount of excess air, which governs the conversion
different levels of complexity, size, and cost. of SO2 to SO3 and hence the amount of H2SO4 formed
Online cleaning systems are not effective against stones, ● Control combustion parameters such as fuel injection pat-
clamshells, etc. and need upstream devices to remove debris tern, fuel injection schedule and fuel viscosity
and macroscopic organic matter from the incoming water. ● Use fuel/air premixing to eliminate soot production
● Quench hot flue gases to solidify molten and soft particles to
prevent attachment at cooler heat transfer surfaces
4.3.2 Gas Flow The control of combustion conditions is a difficult task due to
the great variability in the quality of fuel supplies. Variability of
Online mechanical techniques vary greatly, but soot blowers fuel characteristics is a particular problem for those industries
are the most popular for gas-side use [80]. Some of other that burn waste products.
techniques such as scrapers, rappers, and chains work well in
special applications but are not as readily available. Two com-
mon types of soot blowers are jet soot blowers and sonic soot
4.3.3 Other Devices for Fouling Mitigation
blowers.
● The jet type of soot blower operates by emitting pulses of When it comes to commercial mitigation of scale formation,
steam, air or water at programmed intervals directed at one of the most frequently and emotionally discussed topics are
the tubes and/or down tube lanes to dislodge the deposits devices, which claim to reduce scaling by magnetic, electronic,
and re-entrain them in the gas stream. These soot blowers or catalytic means. To-date, no conclusive scientific proof or
work best if used frequently, thus avoiding the build-up of theory for the mechanisms, which may be responsible for the
material. When build-up occurs, it insulates the surface beneficial effects of such technologies, has been found. A con-
from the coolant, allowing a temperature rise that can pro- siderable number of investigations have been reported in the
duce a glassy deposit. Glassy deposits are much harder to literature; many of them claim some sort of success with the
100 C4 Fouling of Heat Exchanger Surfaces

applied technology. Most of the research-related literature ori- 5.1 Chemical Cleaning Methods
ginates from the former Soviet Union, the UK and the USA,
while several systematic investigations have been performed by Chemical cleaning methods have a number of advantages over
public and private organizations in Austria, Germany, and mechanical methods, namely:
Switzerland. German Industry Standards (DIN) have been for-
(a) They are relatively quick.
mulated for performance evaluation of physical water condi-
(b) Surfaces do not experience mechanical damage.
tioners. Pilot plant and laboratory scale investigations have
(c) Chemical solutions reach normally inaccessible areas.
provided contradicting results. For example, [81, 82] report
(d) They are less labor intensive than mechanical cleaning.
that the installation of magnets considerably reduced cooling
(e) Cleaning can be performed in situ.
water fouling, whereas [83, 84] found no effect of the water
conditioner. Even the mechanisms of scale inhibition are highly
disputed. There are claims that clatherate formation or impact
on the nuclear spin of dissolved ions will reduce the chemical 5.1.1 The Basic Process
reaction on the surface; or that very high frequency current
favors bulk precipitation of scale-forming materials and hence Most chemical cleans consist of five distinct processes, each
weakly adhering particulate deposit rather than a strong crys- being monitored for results before proceeding to the next. The
talline layer. Other investigators believe that minute changes in five stages are:
local pH may affect the CaCO3 equilibrium in the solution.
1. The alkaline clean primarily aims to remove the organic
More recently, claims have been made that the key mechanism
portion of the deposit (oil, fat) in order to render the
is the release of iron ions or iron oxide into the water, which has
inorganic surface hydrophilic. This is necessary to make
an adverse effect on the growth behavior of crystals. Another
the following acid cleaning effective.
plausible explanation is that the electrical field, together with
2. Before and after each chemical step, high flow water flu-
dissolved impurities, changes the crystal form of CaCO3 from
shes are required to physically remove loose or softened
Calcite to Aragonite. Some agreement exists that magnetic or
material.
electromagnetic fields are effective for a relatively narrow range
3. Once the surface is hydrophilic, the deposit is softened and/
of flow velocities only [85].
or dissolved by application of the appropriate acid blend.
Manufacturers of such equipment have impressive reference
This blend usually contains an inhibitor, which prevents
lists of successful installations, where the formation of crystalline
corrosion of the base metal by the acid. The analysis of the
deposits has been substantially reduced or even avoided. Howev-
spent acid strength and the concentration of dissolved scale
er, it is also fair to say that there have been numerous cases where
species indicate whether the acid clean is completed.
no improvement has been achieved. Until the applications and
4. After the acid stage, water rinsing is required to remove
limitations of these installations have been clearly established,
loose debris, sludge, and residual acid. Water rinsing may
no general statement can be made about their economic evalua-
be accompanied by inert gas purging and sequestrant addi-
tion with respect to other available scale prevention methods.
tion, depending on the cleaning technique and the plant
While most suppliers of physical water treatment facilities rec-
configuration.
ommend installation of their devices at a relatively short distance
5. After the acid and rinse stages, the base metal which has
upstream of the heat exchanger which is to be protected, others
been exposed as a result of the cleaning operation is in a very
claim ‘‘memory effects’’ in the fluid of up to 6 months.
active state. If left and exposed to the atmosphere, the
surface would rapidly reoxidize in an uncontrolled fashion.
A passivation process is performed to form a tightly adher-
5 Cleaning of Heat Exchangers
ent, protective oxide film on the base metal.
Periodical cleaning of heat exchangers will be necessary, even if Particular applications may require modifications of the above
the heat exchanger is well-designed and the fluid treatment is sequence. The selection of the cleaning agent and the cleaning
effective. Additionally, conditions in the heat exchanger may procedure strongly depend on the type of deposit, the material
deviate from the design conditions due to changes in flow rates and configuration of the installation and on economical and
and temperatures, plant failures, ingress of air and bacteria, environmental considerations.
changes in the fluid composition or upstream corrosion,
which all may promote fouling. If a heat exchanger or pipe-
line suffers from deposit formation, this can be the start of a 5.1.2 Cleaning Procedure
whole series of problems. Corrosion processes may take place
under the deposit, fouling rates may be increased due to the Among the many possible choices available for cleaning proce-
surface roughness of the deposit or irregular behavior of the dures are ambient temperature treatments, high temperature
exchanger may be observed due to build-up and removal of treatments, fill-and-soak techniques, circulating techniques, on-
deposits. It is, therefore, advantageous to remove non-protec- stream techniques, vapor phase techniques, foam techniques
tive deposits soon after the onset of their formation. Heat and emulsion techniques. Soaking treatments are effective in
exchangers may be cleaned by chemical or mechanical methods many instances. Their application generally reduces equipment
or by a combination of both. costs while increasing chemical costs and downtime costs.
 Fouling of Heat Exchanger Surfaces C4 101

Obviously, it is advantageous to circulate the cleaning agent in acids. Although some metal loss is inevitable, the addition of
order to improve evenly mixing of the chemicals and to reduce inhibitors greatly reduces corrosion. It must be emphasized that
concentration profiles near the fouled surfaces. Circulation also inhibitors are only suitable for specific metal under specific
increases physical disintegration of the deposit by mechanical conditions. Personal danger and disposal problems have to be
scouring. As chemical reaction rates increase exponentially with
temperature, the cleaning process may be improved if the clean-
ing agent is heated. Foaming reduces the cleaning agent require-
ments and increases the effectiveness of cleaning. Also, the C4. Table 18. Common types of chemicals utilized for in-situ chem-
foaming treatment may be faster in some cases. It allows for ical cleaning [87]
good contact in large shell and tube heat exchanger. Complexing
Research on the mechanisms of chemical cleaning of heat Acids Alkalis agents
transfer surfaces is far less developed than research of fouling
Hydrochloric Caustic Soda EDTA
mechanisms. Nevertheless, some first modelling has been
Nitric Ammonia Gluconates
attempted, assuming that the cleaning process is an inverse
fouling process. Sulfuric Trisodium
phosphate
Hydrofluoric Sodium metasilicate
5.1.3 Cleaning Agents Citric Soda ash
Formic
Table 16 shows typical deposits that can be removed chemically
[86]. Deposits that can not be removed are given in Table 17. Sulphamic
Table 18 lists a number of chemicals used for cleaning. Oxidants Solvents Others
Sulfuric acid and hydrochloric acid are the most widely used Potassium Aromatic Biocides
chemical cleaning agents. When used properly, they are safe, permangangate
relatively low-cost materials. However, these mineral acids are
Sodium bromate Aliphatic Surfactants
highly ionized and strong, which may cause rapid corrosion
Sodium nitrite Chlorinated Inhibitors
if the solution is insufficiently inhibited. Therefore, weaker
organic acids and chelating agents are coming into wider use. Sodium hypochlorite Emulsifiers Antifoams
Generally, a mixture of several chemicals is used to attack Ammonium Dewatering Dispersants
complex deposits. Dispersants are added to disperse oils or fats persulfate formulations Hydrazine
and to allow better penetration of the deposit. Sometimes, the bifluoride
addition of small quantities of a second cleaning agent (e.g., the
addition of 0.25% ammonium bifluoride or 0.5% sodium bro-
mate to citric solutions) may considerably increase the effective- C4. Table 19. Scale and deposit removal [87]
ness of the cleanup [87].
Deposit Cleaning method
The type of cleaning agent to be chosen has a major effect on
the economics of the cleaning job. The selection of cleaning Calcium sulfate Boil with Na2CO3 solution, treat with inhibited
chemicals is not only depending on the type of deposit, but also acid
on the exchanger material and the cleaning conditions. In many Calcium Inhibited acid
cases, chemical cleaning of heat exchangers involves the use of carbonate
Magnesium Inhibited acid
hydroxide
Calcium Inhibited acid
C4. Table 16. Typical deposits removed by chemical cleaning [86]
phosphate sludge
Organic Magnesium Inhibited acid + small % of HF
Oil, grease, fat, waxes, soft carbon, tars, silt, vegetation, biological silicate
matter, polymers, resins, paints Sodium Hydrofluoric acid
Inorganic aluminum silicate
Rust, magnetite, calcium carbonate, calcium sulfate, magnesium Ferric oxide Inhibited acid or ammoniated citric acid
hydroxide, calcium phosphate, silica, magnesium silicate, copper, (Fe2O3)
copper oxides, alumina, nickel oxides Ferrous oxide Inhibited acid or ammoniated citric acid. Add
(Fe3O4) 0.25–0.5% stannous chloride for protection
against ferric ion attack. Ammoniated EDTA
Copper Ammoniated bromate, ammoniated EDTA
Organic material High velocity liquids or circulate 20% chromic
C4. Table 17. Intractable deposits [86]
acid heated to 95 C. Hot alkaline solutions
Glasses, ceramics, hard carbon, inert plastics, vulcanized rubber, useful for removing light oils, grease, or other
rubber latex sludge materials
102 C4 Fouling of Heat Exchanger Surfaces

considered, too. Table 19 shows some deposits and scales with 5.1.4 OnStream Chemical Cleaning
the recommended cleaning agents [87].
For plants in operation, the timing of cleaning operations is In most cases, chemical cleaning is done while the heat exchanger
of economical importance. Energy losses, production losses and is off-line. However, there have been several attempts to develop
safety aspects have to be compared with cleaning costs and chemicals that can be used for onstream cleaning to avoid expen-
losses due to shutdown. Table 20 shows typical application of sive plant shutdown. Examples for these chemicals are complex-
some of the most common cleaning agents. ing agents which can be added during operation of boilers.
Shock dosage of chlorine (>15 ppm) is often used to re-
move bio-deposits while the exchanger is onstream. However, it
was found that biofilms regrow at an accelerated rate after the
chlorine dosage. As chlorination has come under increasing
C4. Table 20. Chemical cleaning agents [87]
regulatory control, methods to use chlorine more efficiently
Agent Application and to minimize the amount of chlorine discharged to the
Hydrochloric General removal of rust and scale for water ways become increasingly important.
acid non-stainless steel piping
Inhibited HCL Same as above 5.1.5 Problems Associated with Chemical
Sulfuric acid General usage for stainless steel for removal of Cleaning
rust and scale. H2SO4 is dangerous to personnel
and must be used with caution Problems associated with chemical cleaning of heat exchangers
Inhibited H2SO4 Same as above are due to the danger of handling (burn, toxic), due to elevated
Hydroxyacetic General acidizing. Safer than H2SO4 application temperatures, due to the costs of cleaning agents,
acid due to the chemical attack on the heat exchanger material
Formic acid Often used as 1% formic, 2% hydroxy-acetic (overcleaning, uneven cleaning, corrosion) and due to disposal
acid solution for cleaning of supercritical ‘‘once- problems. Acids and alkalis must be neutralized, organic mate-
through’’ boilers where stainless steel is the rials may be burned and fluorides must be reacted to inactive
prevalent material of construction and chloride solid residues. Some of the organic acids, such as citric acid and
ions must be avoided. Also used for non-ferrous gluconic acid are biodegradable.
metals as a 1–2% solution. Must be heated to
95 C. Safter than mineral acids
Citric acid Used as 0.01 solution to chelate iron. Used at 5.2 Mechanical Cleaning Methods
ambient temperature or slightly warm
Ammoniated Very safe (corrosion and handling) cleaning For the following cleaning methods, heat exchangers have to be
citric acid agent. By changing pH and adding sodium taken off-line and dismantled. Some of the deposits may then be
nitrite, the same solution. may be used for removed manually, for example from the water box. Steam-
passivation. pH is adjusted by adding NH3 and blasting and hydro-blasting with pressures up to 1500 bar are
resulting solution is effective for copper probably the most common mechanical cleaning methods, see
removal [49]. They can be performed completely manually or semiau-
Hydrofluoric Used for new piping and for stainless steel tomatically in cleaning stations. If deposits are very tenacious,
acid where chloride ions are critical as a 3–5% sand can be added to the pressurized water to increase the
solution. Also used for silica deposit removal. cleaning efficiency. Both, steam- and hydro-blasting are labor
Dangerous! intensive and keep the exchanger off-line for a considerable
Sulphamic acid Fairly safe. Used at a 5–10% solution for removal time. Blasting may not completely eliminate all deposits and
of iron oxides. Must be maintained below 60 C some significant roughness can remain. The shell-side of tube
or decomposes to ammonium bisulfate bundles can only be cleaned completely if the tubes are arranged
Ammoniated Use for removal of copper. Does not attack scale in-line. The particular geometry of twisted tubes provides flow
bromate or iron oxides. For each kg of copper use lanes for pressurized water or steam which facilitates cleaning.
0.9 kg NaBrO3 Experienced maintenance crews are required and strict safety
1.4 kg (NH4)2CO3 regulations must be obeyed due to the danger of handling
4.5 l NH4OH 30% equipment at very high pressures.
Phosphoric acid Less aggressive than sulfuric acid and more
aggressive than sulphamic acid. Use for general
removal of oxides and scales, particularly for
stainless steel
Ammoniated Use for removal of iron oxides or copper, as a
EDTA 5–10% solution. Safe material!
Sodium salt Use for removal of water hardness scales as a
5–10% solution
C4. Fig. 23. CONCO cleaning device [89]
 Fouling of Heat Exchanger Surfaces C4 103

While blasting is the only available alternative for the shell- 4. TEMA (1978) Standards of the Tubular Exchanger Manufacturers Associa-
tion, 6th edn., New York
side of the tube bundle, several cleaning methods can be used
5. Marriott J (1971) Where and how to use Plate Heat Exchangers. Chem Eng
for the inside of straight tubes. The continuous cleaning sponge 78(8):127–134
ball system described in detail in Sect. 4.3 can also be used as a 6. Weierman RC (1982) Design of heat transfer equipment for gas-side fouling
transportable, off-line cleaning system, particularly if used with service. In: Marner WJ, Webb RL (eds) Workshop on an assesment of
corundum-coated sponge balls. Very dirty and plugged tubes Gas-Side Fouling in Fossil Fuel Exhaust Environments. Publication 82–67,
Jet Propulsion Laboratory, California Institute of Technology, Pasadena,
can be cleaned with drills equipped with drill bits, brushes, or
California
bit-brush combinations. 7. Palen JW (1983) Shell and tube re-boiler. Sect. 3.7.8, Heat Exchanger Design
To avoid damage of the heat transfer surfaces, cleaning must Handbook. Hemisphere Publ. Corp
be done carefully, thus increasing costs for labor and downtime. 8. Chenoweth JM (1987) General design of heat exchangers for fouling condi-
Using air- or hydropressure, rubber plugs or metal scrapers tions. In the NATO Advanced Study Institute on Advances in Fouling
Science and Technology, Alvor, Portugal
can be shot through the tubes. These techniques are consider-
9. Bansal B, Chen XD, Müller-Steinhagen H (1997) Effect of suspended parti-
ably faster than the above methods, cleaning up to 15,000 tubes cles on calcium sulphate fouling in plate heat exchangers. ASME J Heat
in 24 h. Rubber plugs fail for hard deposits. Shooting metal Transfer 119:568–574
scrapers through the tubes at a water pressure of 35 bar and a 10. Blöchl R, Müller-Steinhagen HM (1990) Influence of particle size and
scraper velocity of 3–6 m/s results in the removal of most particle/liquid combination on particulate fouling in heat exchangers. Can
J Chem Eng 68(4):585–591
deposits [88]. An example of those scrapers is shown in
11. Epstein N (1983) Fouling in heat exchangers. In: Taborek J, Hewitt G (eds)
Fig. 23. In general, water pressure systems are safer than air Heat exchanger theory and practice, McGraw-Hill, New York
pressure systems, due to the compressibility and subsequent 12. Müller-Steinhagen H (2000) Modellierung der Ablagerungsbildung in Wär-
rapid expansion of gases. meübertragern – Vom Laborversuch zur Produktionsanlage – Berichte zur
Most mechanical cleaning methods remove not only the Energie- und Verfahrenstechnik Vol 20.1, ISBN 3-931901-14-9
13. Taborek J (1987) Private communications
deposit but also the protective oxide layer. Under certain cir-
14. Martin H (1988) Wärmeübertrager. Georg Thieme, Verlag Stuttgart
cumstances, this may create a corrosion problem. On the other 15. Thackery PA (1980) The cost of fouling in heat exchanger plant. Effluent
hand, regular cleaning removes deposit and avoids flow condi- and Water Treatment J 20(3):112–115
tions, which promote corrosion due to chemical reaction or 16. Pritchard AM (1987) The economics of fouling. In: Melo LF, Bott TR,
stagnant flow. For very severe fouling problems, a combination Bernardo CA (eds) Fouling science and technology, NATO ASI Series E,
vol. 145. Kluwer, Amsterdam
of chemical and mechanical cleaning may be recommended.
17. Woods DR, Anderson SJ, Norman SL (1976) Evaluation of capital-cost
data – heat-exchangers. Can J Chem Eng 54(6):469–488
18. Sart P, Eimer K (1979) Control of scaling or fouling effects in cooling water
6 Symbols system for improvement of heat exchanger efficiency, Paper presented at
international meeting on industrial heat exchangers and heat recovery,
CF Cleanliness factor Liege, Belgium
E Activation energy (J/mol) 19. Hewitt G, Müller-Steinhagen H (2000) Heat exchanger fouling in crude oil
K Constant distillation units. ESDU Data Item 00016:1–80
20. Hewitt G, Müller-Steinhagen, H (2003) Fouling in cooling water systems
R Universal Gas Constant, 8.314 J/(mol K)
using seawater. ESDU Data Item 03004, pp 1–100, International Ltd.,
Re Reynolds Number London
Rf Fouling resistance (m2K/W) 21. Hewitt G, Müller-Steinhagen H (2007) Fouling in cooling systems using
s Deposit thickness (m) fresh water. ESDU Data Item 08002, International Ltd., London, pp 1–172
u Flow velocity (m/s) 22. Bennett CA, Kistler RS, Lestina TG, et al. (2007) Improving heat exchanger
designs. CEP 103(4):40–45
tw Wall shear stress (N/m2)
23. Aguirre FJ ( 2006) Reviewing the use of fouling factors in heat exchanger
x Friction factor design. http://www.ChemicalProcessing.com. Cited 15 February 2000
1,2 Fluid 1, fluid 2 24. Bennett CA (2005) Using a non-traditional approach to account for crude
oil fouling in heat exchangers. http://www.ChemicalProcessing.com. Cited
Subscript/Superscripts 30 September 2005
c Clean 25. Nesta JM, Bennett CA (2005) Fouling mitigation by design. The 6th inter-
d Deposit national conference on petroleum phase behavior and fouling, Amsterdam,
f Fouled The Netherlands
26. Nesta J, Bennett CA (2004) Reduce fouling in shell-and-tube heat exchan-
gers. Hydrocarb Process 83(7):77–82
27. Knudsen JG (1983) Fouling in heat exchangers. Heat Exchanger Design
7 Bibliography Handbook Hemisphere Publications Corporation, Washington, DC
28. TUBEC Tubes. AST, Avesta Sandvik Tube AB, Helmond, Holland
1. Steinhagen R, Müller-Steinhagen HM, Maani K (1993) Fouling 29. Bornhorst A, Zhao Q, Müller-Steinhagen H (1999) Reduction of scale
problems and fouling costs in New Zealand industries. Heat Transfer Eng formation by Ion implantation and magnetron sputtering on heat transfer
14(1):19–30 surfaces. Heat Transfer Eng 20(2):6–14
2. Steinhagen R, Müller-Steinhagen HM, MaaniK (1990) Heat exchanger 30. Müller-Steinhagen H, Zhao Q, Helalizadeh A, et al. (2000) The effect of
applications, fouling problems and fouling costs in New Zealand Industries. surface properties on CaSO4 scale formation during convective heat transfer
Ministry of Commerce Report RD8829 1–116 and subcooled flow boiling. Can J Chem Eng 78:12–20
3. Garrett-Price BA, et al. (1985) Fouling of heat exchangers – characteristics, 31. Förster M, Augustin W, Bohnet M (1999) Influence of the adhesion force
costs, prevention, control and removal. Noyes Publications, Park Ridge, crystal/heat exchanger surface on fouling mitigation. Chem Eng Process
New Jersey 38:449–461
104 C4 Fouling of Heat Exchanger Surfaces

32. Zhao Q, Müller-Steinhagen H (2001) Intermolecular and adhesion forces of 62. Mayo FR, Stavinoha LL, Lee GH (1988) Source of jet fuel thermal oxidation
deposits on modified heat transfer surfaces. In Proceedings of UEFC on heat tester deposits from an oxidized JP-8 fuel. Ind Eng Chem Res 27(2):362–363
exchanger fouling (Session II), Davos, Switzerland 63. Dugan CP, Van Nostrand WL, Jr. Haluska JL (1978) How antifoulants
33. Zhao Q, Liu Y, Müller-Steinhagen H, Liu G (2002) Graded Ni-P-PTFE reduce the energy consumption of refineries. Chem Eng Prog 74:53–57
coatings and their potential applications. Surface and Coatings Technology 64. Novak L (1982) Comparison of the Rhine river and the Öresund sea water
155:279–284 fouling and its removal by chlorination. J Heat Transfer 104:663–670
34. Paikert P (1983) Verschmutzung von Kondensatoren und Kühltürmen. GVC 65. Miller PC, Bott TR (1979) The removal of biological films using sodium
Weihnachtstagung 371–390 hypochloride. Internaional Chemical Engineering Conference on Fouling
35. Gilmour CH (1965) No fooling – no fouling. Chem Eng Prog 61(7):49–54 Science or Art? Surrey University, Guildford, England
36. Kral D, Stehlik P, van der Ploeg HJ, et al. (1996) Helical baffles in shell and 66. Birchall GA (1981) Achieving microbiolocal control in open recirculating
tube heat exchangers, Part I: Experimental verification. Heat Transfer Eng 17 cooling systems. Conference on progress in the prevention of fouling in
(1):93–101 industrial plant. University of Nottingham, Nottingham
37. http://www.embaffle.com/ 67. Grier JC, Christensen RJ (1975) Microbiological control in alkaline cooling
38. Butterworth D, Guy AR, Welkey JJ (1998) Design and application of twisted- water systems. Paper presented at the national association of corrosion
tube exchangers. Brown Fintube, London engineers annual meeting, Toronto, Canada
39. Müller-Steinhagen H (2001) Numerical investigation on plate heat exchanger 68. Waite TD, Fagan JR (1980) Summary of biofouling control alternatives.
design. UEF Conference on Compact Heat Exchangers, Davos, Switzerland Condenser Biofouling Control, Publisher J. Garey, Ann Arbor Science
40. Klaren DG (1987) Fluid bed heat exchanger. CPP Edition, Europe 69. Grade R, Thomas BM (1979) The influence and control of algae in industrial
41. Cooper A, Suitor JW, Usher JD (1980) Cooling water fouling in plate heat cooling systems. International Chemical Engineering Conference on Fouling
exchangers. Heat Transfer Eng 1(3):50–55 Science or Art? Surrey University, Guildford, England
42. Novak L (1982) Comparison of the Rhine river and the Öresund sea 70. Knudsen J (1991) Conquer cooling-water fouling. Chem Eng Prog 87(4):42–48
water fouling and its removal by chlorination. J Heat Transfer 104:663–670 71. Betz Laboratories (1976) Handbook of industrial water conditioning, 7th
43. Zettler HU, Weiss M, Zhao Q, et al. (2005) Influence of surface properties/ edn Trevose, PA, pp 24–29
characteristics on fouling in plate heat exchangers. Heat Transfer Eng 72. Drew Chemical Corporation (1977) Principles of industrial water treatment,
26:3–17 1st edn Boonton, NJ, pp 99–103
44. Delplace F, Leuliet JC, Bott TR (1997) Influence of plate geometry on 73. Nalco Chemical Comp (1979) Nalco Water Handbook, 1st edn. McGraw-
fouling of plate heat exchangers by whey protein solutions. In: Panchal Hill, New York
CB, Bott TR, Somerscales, FC, et al. (eds) Fouling mitigation of industrial 74. Knudsen JG, Jou HY, Herman KW (1985) Heat transfer characteristics of an
heat-exchange equipment. Begell House, Redding, pp 565–576 electrically heated annular test section for determining fouling resistances.
45. Masri MA, Cliffe KR (1997) Investigation into the fouling of a plate and DREW Ind. Div., Report CWI-TP-18
frame heat exchanger. In: Panchal CB, Bott TR, Somerscales FC, et al. (eds) 75. SPIRELF System (1987) American European consulting company.
Fouling mitigation of industrial heat-exchange equipment. Begell House, Houston, TX
Redding, pp 549–561 76. HEATEX Radial Mixing Element: A patented system developed by CAL
46. Pritchard AM, Clarke RH, de Block MX (1992) Fouling of small passages in GAVIN LTD., Birmingham, England
compact heat exchangers. In: Bott et al. (eds) Fouling mechanisms: Theo- 77. Haquet I (1994) TURBOTAL-system for reduced fouling in crude oil heat
retical and practical aspects. Eurotherm Semin 23:47–56 exchangers. Proceedings engineering foundation conference on heat ex-
47. Kew P (1991) An investigation into fouling of a printed circuit heat ex- changer fouling, Snells Beach, California
changer. Future Practice Report 13, Energy Efficiency Enquiries Bureau, 78. KALVO VOGLER GMBH: Automatisches Reinigungssystem für Kondensa-
Harwell, London toren und Röhrenwärmeaustauscher
48. Coulson JM, Richardson JF, Sinnott RK (1985) Chemical Engineering, 79. TAPROGGE Report 84–15 (1984) Test of TAPROGGE condenser tube
Volume 6. Pergamon Press cleaning system to prevent silica and calcium carbonate scaling. TAP-
49. Müller-Steinhagen H (2000) Heat exchanger fouling – mitigation and clean- ROGGE GmbH
ing technologies. Publico Publications, ISBN 0 85295 436(0):1–382 80. Simon S, Zachay R (2000) Clyde Bergemann cleaning technology for heat
50. DUBBEL (1974) Taschenbuch für den Maschinenbau, vol. 2, 13th edn., pp exchangers In: heat exchanger fouling – mitigation and cleaning technolo-
87–94 gies Publico Publications, ISBN 0 85295 436 0:1–382
51. Müller-Steinhagen H (1999) Cooling water fouling in heat exchangers. Adv 81. Parkinson G, Price W (1984) Getting the most out of cooling water. Chem
Heat Transfer 33:415–496 Eng 91(1):22–25
52. Harris A, Marshall A (1981) The evaluation of scale control additives. 82. Donaldson J, Grimes S (1988) Lifting the scale from our pipes. New Sci
Conference on progress in the prevention of fouling in industrial plant, 18:43–46
University of Nottingham, Nottingham 83. Hasson D, Bramson D (1985) Effectiveness of magnetic water treatment in
53. Krisher AS (1978) Raw water treatment in the CPI. Chem Eng 79–98 suppressing CaCO3 scale deposition. Ind Eng Chem Process Des Dev
54. Haluska JL (1976) Process fouling control by effective antifoulant selection. 24:588–592
Paper No. 153 presented at Corrosion/76, Houston 84. Söhnel O, Mullin J (1988) Some comments on the influence of a magnetic
55. Watkinson AP (1988) Critical review of organic fluid fouling: Final Report field on crystalline scale formation. Chem Ind 6:356–358
No. ANL/CNSV-TM-208 Argonne National Laboratory, III 85. Cho YI, Lee SH, Kim W, et al. (2003) Physical water treatment for the
56. Gillies WV (1979) Fouling and its control by chemical additives in hydro- mitigation of mineral fouling in cooling-tower water applications. In: Wat-
carbon streams. Proceedings of the Institute of Corrosion Science and kinson AP, Müller-Steinhagen H, Reza Malayeri M (eds) Heat exchanger
Technology, and Institute of Chemical Engineering Fouling Conference, fouling and cleaning: Fundamentals and applications, ECI Symposium
University of Surrey, Guildford, London Series, volume RP1. http://services.bepress.com/eci/heatexchanger/4
57. Mayo FR, Miller AA (1856) The oxidation of unsaturated compounds. The 86. French MA (1981) Chemical cleaning in practice. Conference on progress in
oxidation of styrene. J Am Chem Soc 78:1017–1022 the prevention of fouling in industrial plant. Universität Nottingham, Not-
58. Mayo FR, Miller AA (1956) The oxidation of unsaturated compounds. tingham
Reaction of styrene peroxide. J Am Chem Soc 78:1023–1034 87. Roebuck AH, Bennett CA (1977) Heat transfer payback is a key to chemical
59. Mayo FR, Miller AA, Russel GA (1958) The oxidation of unsaturated cleaning choice. Oil Gas J 9:93–96
compounds. J Am Chem Soc 80:2500–2507 88. Hovland AW (1978) Effective condenser cleaning improves system heat rate.
60. Mayo FR (1958) The oxidation of unsaturated compounds. The effect of Pow Eng 82:49–50
oxygen pressure on the oxidation of styrene. J Am Chem Soc 80:2465–2480 89. Conco Systems, Inc., 135 Sylvan St., Verona, PA 15147
61. Mayo FR (1986) Gum and deposit formation from jet turbine and diesel
fuels at 130 C. Ind Eng Chem Prod Res Dev 25:333–348