NACE Paper No.

MECCOCT18-12292

PROTECTIVE COATING FOR PROCESS IMPROVEMENT IN SHELL & TUBE CORROSION

HEAT EXCHANGERS

Yasir Idlibi Ed Curran

Curran International Curran International
4610 Vicksburg Lane 4610 Vicksburg Lane
Dickinson, TX 77539 Dickinson, TX 77539

Alexander Barre
Curran International
4610 Vicksburg Lane
Dickinson, TX 77539

ABSTRACT

Corrosion and fouling in shell and tube heat exchanger tube ID’s causes reduced heat transfer efficiency,
operational disruptions, and downtime associated with major maintenance costs. Such issues can be
resolved by implementing appropriate cleaning techniques and use of protective coating to eliminate both
corrosion and fouling. One company in particular saved an average $1 million per month in lost production
in a single alkylation unit by coating tube ID’s of their heat exchangers. By taking advantage of proper
coatings and cleaning processes, dramatically system performance and operational efficiency are
improved. This paper discusses advantages of applying protective coatings in heat exchangers’ tube ID
and associated process improvement.

INTRODUCTION

Corrosion and fouling in heat exchanger tube ID cause decreased heat transfer efficiency, increased
downtime and operational and maintenance costs. Maintenance costs of heat exchangers are estimated
to account for 0.25% of the world GDP. In many cases, the losses from inefficient heat-transfer
applications can amount to tens or even hundreds of thousands of dollars per day. Depending on the
size of the business, however, it is possible to save millions, even billions a year by deploying appropriate
maintenance techniques for tubular systems. One company in particular saved an average $1 million per
month in lost production in a single alkylation unit.
By taking advantage of proper coatings and cleaning processes, owners can dramatically enhance
system performance and operational efficiency. Good tubular maintenance leads to long-term reliability
and a host of additional benefits.

HISTORY

Coatings for tube ID have been first utilized in the 1950’s and were applied in specialized shops using
the “fill, drain, and rotate” method. Coatings protected cooling water systems from scaling and corrosion.

After experiencing extremely high fouling rates in condensers due to effluent discharge being pulled into
the cooling water inlet producing a rich broth of bacterial and mineral component in several power
stations, Italian engineers in the mid 1980’s were able to apply cured epoxy-phenolics in-situ into heat
exchangers and condensers tube ID’s using air-atomized spray techniques. Fouling had adversely
affected the efficiency and power generation capacity. The spray coating application resulted in excellent
and improved fouling and corrosion resistance to the main condensers, restoring actual normal operating
capacity

In North America, a Florida-based utility company experienced rapid-through wall leaks two years after a
re-tubing. They began searching for alternatives for re-tubing which led to a collaborative effort to
investigate tube ID coatings by the Electric Power Research Institute, EPRI. The EPRI project TR-107068
investigated coating material, heat transfer resistance, wear resistance, back pressure and other
quantitative variables. A trial application was performed in 1993 on 6000 tubes of 12,000 tube condenser.
Performance data were collected over three years and demonstrated better heat transfer performance of
coated condenser tubes compared to uncoated ones. This confirmed that coatings offer an effective
solution. The final EPRI report was published in 1997 with coated tubes still in operations. The results of
this study opened doors for evolution in cleaning methods, coating material, thermal conductivity and
application equipment.

TUBE SURFACE PREPARATION

Surface preparation is one of the key elements in coating application that ensures durability and longevity
of a coating system. Tubes surfaces were first prepared using typical sandblast nozzles held up against
the tube ends and directing the grit into and through the tube. The 11-meter (36-foot) tubes had to be
blasted from each side, consuming large amounts of grit as well as time and labor, but left the center
sections of the tubes insufficiently clean for the long-term. Applicators developed and experimented
several new grit-blasting methods such as lance blasting, radial-pattern blasting, swirl blasting and high-
velocity grit blasting (HVGB).

In 1993, an EPRI demonstration project successfully proved the efficacy of lance blasting. However, it
was cumbersome to manage the length of the lances, which ranged between 6 and 12 meters (20 and
40 feet). There were also safety issues, as the metal lances tended to wear through at the hose-to-lance
transition coupling after eight hours of blasting. Radial nozzles mounted on the end of lances also suffered
excessive wear and tear, not to mention mechanical damage from insertion into the tube ends. It was
clear that grit-blasting procedures needed further development and improvements.
A gas dynamics lab at a US university had previously researched the grit-blast nozzle design. They were
provided data to develop a new design for a grit-blasting nozzle specifically to clean tube IDs. Typical
sandblast nozzles attained an air speed of 340 m/s (1115 ft/s) and grit velocity of 134 m/s (440 ft/s), which
was inadequate to maintain the grit velocity needed to clean tubes more than 3-meters (10-feet) long.
The new design increased air speed to 686 m/s (2250 ft/s) and grit velocity to 242 m/s (796 f/s). This
increased the kinetic energy in the grit particles by 81%, and proved highly effective as a tube ID cleaning
method. Grit blasting using this special sandblast nozzle has since become the standard cleaning method
for tubes prior to coating. To date, millions of tubes have been cleaned to white metal for surface
preparation prior to coating.

This methodology has proven so effective that it is now widely used to clean tubes with tenacious, sticky
deposits that UHP (ultra-high pressure) water jetting could not remove. An entire industry has grown out
of the need to clean tubes prior to LOTIS, IRIS, RFET, and other NDT inspections.

Figure 1 shows difference between hydro jetted and grit blasted tube ID’s. The tube in Figure 1A was
cleaned three times at at three different pressures; 15,000, 16,000, and 17,000 psi, yet fouling was not
removed. Figure 1B shows the same tube after being grit blasted for 40 seconds only. Apparently grit
blasting is more effective in removing fouling and eliminate repeated work. Grit blasting also provides
accurate inspection data since the scale that interferes with inspection signals is completely removed
making the tube ready for inspection such as IRIS.

(a) (b)
Figure 1: Difference between hydro jetting vs grit blasting. – (a) Tube cleaned using hydro
jetting for 3 times 3 different pressures, yet fouling was not completely removed. (b) Tube
cleaned using grit blasting for 40 seconds only, removing all fouling.

Figure 2 was captured by an NDE company performing Remote Field Testing (RFT) evaluation of carbon
steel tubes for a Houston area chemical plant. The same tube was inspected after hydro jetting and after
grit blasting. It was noticed that there is an extremely noisy absolute channel due to scale and deposits
past hydro jetting. The grit blasted tube, on the other hand, which was cleaned from all deposits and
scales, had a clear signal.
 (a) (b)
Figure 2: FRT signal after – (a) hydro jetting; note the extremely noisy signal, and (b) after grit
blasting; note the extremely clear signal

While the grit cleaning method has been widely adapted for heat-exchanger tube cleaning, the question
of tube wall erosion due to the cleaning procedures has also been raised and studied by certain end
users and initially by EPRI using 90/10 Cu/Ni (copper-to-nickel) tubes to test the benchmark, the average
dwell time is 30 seconds for grit-cleaning tubes to SSPC SP5/NACE #1, which is the standard for white
metal cleanliness. Next, grit-blasting tubes six-feet in length in two repeated tests for three minutes each
yielded no measurable wall loss. This test of erosive potential from the cleaning method has been
repeated many times and up to 10 minutes dwell time with no measurable wall loss.
High-resolution video probes can also be sued to verify tube cleanliness. Some video probes can identify
extremely small surface imperfections, making it easier to spot residual scale and confirm surface
cleanliness.

COATING APPLICATION DEVELOPMENT

Earlier application methods by fill, drain and rotate worked well enough but could not be done in situ.
These also required high-solvent loading to control thickness and prevent coating set-up before solvent
flash-off. Since solvents are recognized as being hazardous environmental contaminants, such practices
are being phased out.
Application method refinements with the spray application approach apply a reliable, uniform layer of
coating. Current application equipment can actually coat up to four tubes at a time at speeds of 2 m/sec
(6-ft/sec). Spray methods also allow for application of different coatings and solids content to 100%.
Typically, there is a less than 37.5-microns (1.5-mils) differential in dry-film thickness for the tube ID
circumference. Coatings can be applied to a pinhole-free condition and verified with a full-length holiday
test designed specifically for tube ID. Figure 3 show a tube ID coating application.
 Figure 3: Application of Tube ID coating

COATING MATERIAL

To avoid impacting heat transfer, coatings for heat-transfer equipment are typically applied at less than
than 14 mils (355 µm) dry-film thickness. Coatings used in heat exchanger tube ID are baked phenolics,
phenolic epoxy, epoxy novolacs, fluoropolymers, and thermoplastics. Coatings are chosen according to
the service temperature as well as conditions in which they are to be applied; shop vs. field application.

100

90 Bare Tube Coated Tube

80

70
Heat Transfer Rate

60

50

40

30

20

10

Days in Service
Figure 4: Coating Effect on Heat Transfer
Table 1 provides a list of generic types of coatings used in heat exchanger tube ID. Coatings have always
suffered from the perception that heat transfer rate will be affected due to their lower thermal
conductivities vis-à-vis metallurgy. However, case histories has demonstrated just the opposite. Decades
of service history have shown that tube coatings can actually enhance heat transfer and overall
performance to a significant degree. While the thermal conductivity of the coating is much less than the
parent tube, its impact is offset by several factors. Research have indicated that impact to overall thermal
conductivity if nil or minor. Figure 4 shows that for a bare uncoated tube the heat transfer efficiency starts
at 100% but will decline over time due to fouling build up whereas a coated tube the heat transfer
efficiency starts a little less than 100% but stabilizes overtime since the coating prevents scale and fouling
build up.

Table 1

Generic Types of Coatings used in Heat Exchanger Tube ID

Coating Material Temperature Limitation Cure Type

Epoxy Phenolic 120°C Ambient Cure
Novolac Epoxy 120°C Ambient Cure
Baked Phenolics 150°C Heat cured at 200°C
Thermoplastics 200°C Heat cured at 325°C
Bisphenol A Epoxy 80°C Ambient cure

COATING EFFECT ON HEAT TRANSFER

While the thermal conductivity of the coating is much less than the parent tube, its impact is offset by
several factors.

The first factor covers normal design considerations. HEI and TEMA design condensers to operate at
85% capacity or with 15% of tubes plugged without affecting performance. Fouling factor are also built
into the equation to add additional performance hedging into the the design. Applying the coating either
totally eliminates the subsequent fouling or greatly reduces the accumulation of typical micro-/macro-
fouling, mitigating the initial design consideration.

The second major factor is the boundary layer-drag reduction. Fully 70 percent of total heat transfer
resistance (HTR) across a heat exchanger tube is the slow-moving fluid coming into contact with the tube
wall. Tube wall friction reduces this flow and creates an insulating barrier of low velocity fluid. Polymer
coatings reduce the surface tension at the tube wall substantially, by a factor of 30 to 40 dynes/cm2
compared to metallurgy (1200 dynes/cm2) in a non-oxidized or new condition. Reducing friction reduces
the boundary layer drag and substantially opens up the flow profile. Studies show flow rate improvements
of 80 to 100 percent with polymer coatings compared to new uncoated tubes in the same fluid train. This
increase in flow and low surface energy of the coating contributes to the improved overall thermal
efficiency of the heat exchanger in fluid service. This increase flow at the tube wall also inhibits nucleation
sites for micro and macro scaling deposition to begin. An added benefit and energy saving is reducing of
energy needed to power circulation pumps.

EPRI studied thermal conductivities of various polymer coatings resins and reported in Table 2.

Table 2

Thermal Conductivity of Various Polymer Coating Resins

Tube Condition
Thermal Conductivity w/mK
(21 mm brass tube x 23 mm wall)

Average scaled - <0.150 mm 14.9

Epoxy 100% Solids – applied at 75 µm average 2.737

Epoxy phenolic – applied at 75 µm average 8.82

Epoxy phenolic – proprietary pigment loading 14.764

COATING APPLICATION QUALITY CONTROL

Tube ID coatings can be inspected with methods similar to tank lining with some limitations. Blotter tests
or black light examination can confirm or eliminate the presence of hydrocarbons. Chloride testing is also
viable, but usually on tube sheets. Holiday detection testing can be accomplished using NACE TM0384
“Detection of Internal Tubular Coatings of Less Than 250 µm (10 mils) Dry Film Thickness” or NACE
TM0186 “Holiday Detection of Internal Tubular Coatings 355 µm to 760 µm (14 – 30 mils) Dry Film
Thickness”. The testing is adapted to reach all the way through the tube with the sponges/brushes sized
to fin into the thee tube ID. The most limiting QA method is the measurement of dry film thickness. The
current available instruments can only reach 1 meter into the tube-end to verify adherence to the
specifications. If additional verification is needed, sample tube can be coated, split, and measured for
verification of minimum DFT throughout the tube.

ADVANTAGES OF COATING TUBE ID

Elimination of under deposit corrosion – Copper alloyed tubes have been the mainstay of condenser
metallurgy because if their superior heat transfer and resistance to sea and lake water corrosion. Scale
accumulation, however, can be detrimental to the tube with resulting under deposit corrosion. Removal
of the scale and application of coating has eliminated this issue.

Prevention of fouling buildup – smooth polymer coatings both promote flow velocities at the tube wall to
eliminate fouling nucleation sites and act as a dielectric barrier to ionic bonding between the tube wall
and the mineral laden cooling water. Coated tubes have eliminated all cleaning cycles or greatly extended
cleaning cycles times. Any deposits are generally removed easily by brush or 200 bar water pressure.
Prevention of Copper Ion Discharge – coating of condenser tubes prevents the tube wall from oxidizing
and leaching Cu+ into cooling water.

Resistance to Chemical Attack – coatings are resistant to H2S and other organic acids that can quickly
through wall copper alloyed tubes.

Mitigation of MIC or Manganese Attack – polymer coatings preferentially wick into pitting by capillary
action. Case histories have shown penetration into MIC or manganese pitting and provided multi-year
protection against further tube leaks and corrosion.

Erosion Prevention – tubes operating in high suspended or high velocity fluids continually have the lightly
bonded protective oxides removed. Polymer coatings eliminate the oxidation process plus pigmenting
with ceramics creates a hard abrasion resistant barrier.

Prevention of De-Alloying – Aged Copper alloy tubes are subject to de-alloying from continual preferential
ion release into the cooling fluid. Polymer coatings act as a barrier to prevent the ion exchange.

TYPES OF COATING MATERIAL

Epoxy based coatings are ambient cure offering hydrothermal stability up to 200, high chemical
resistance and fouling resistance properties typically applied in one to two coats for quick return to service
and have demonstrated long service, +10 years, in cooling water service. Can be blended with PTFE for
additional anti-oxidation and foul release. Applied up to 356 µm (14 mils)

Ceramic based coatings are ambient cure offering high antifouling properties for high temperature crude
services. Can be applied to U tubes and other tubular geometries. Can be applied to the OD of existing
exchangers. Applied in thin films; less than 50 µm (2 mils).

Sol gels based coatings applied to exchangers in very low film thicknesses, improve the repellency of
steel substrate in a broad range of fouling services including crude and process water exchangers as
well as sea water service. Applied in thin film from 6 – 100 µm (0.24 – 2 mils) depending on service.
Some formulations are new with limited service history. Some formulations are heat cured up to 300c.
Some formulations can only be applied to straight tubes.

CASE STUDY FOR COST SAVINGS

At a petrochemical refinery, six heat exchangers in the catalytic cracker recovery unit’s refrigeration
section required maintenance. They were not operating efficiently, and upon examination, the diagnostics
was severe tube corrosion and pitting. Two of the six exchangers required complete retubing due to age
and damage over time. The remaining four exchangers had only been operating for three years, but still
had telltale water and tear, corrosion and pitting. The refinery management opted to apply coatings to
all six exchangers to prevent further damage and to decrease fouling from sulfate-reducing bacteria. By
coating all the equipment, preventative maintenance in the future would suffice to reduce stoppages, leak
repairs, replacement and the need for any retubing, and the unit would see better performance from the
equipment in the refrigeration area. That bundle previously lasted no longer than 6 months before it was
removed from service. After 24 months, inspection revealed that no fouling had occurred. Before cleaning
and coating the six sets of exchanger tubes, the ammonia coolant pressure was 230 psi. Afterwards, the
pressure dropped to 205 psi. Currently, the heat exchanger's pressure drop runs about 190-200 psi. In
the prior 3 years of service, with bare carbon steel tubes, these four exchangers experienced a pressure
drop increase of about 15 psi/year. For these four sets of exchanger tubes, the petrochemical company
expects to get a 10-year coating life on its tubes. Minor touchups on the tubesheet and gasket surfaces
will probably be required during maintenance periods. After 10 years, the tube bundle will probably need
to be blasted and coated again. With the new coating, petrochemical company expects a bundle life of
more than 20 years.

REFERENCES

1. Curran, E. L., “Solving Heat Exchanger Tube Problems with Thin Film Thermally Conductive Coating
Applications and Novel Tube and Pipe Cleaning as a Precursor to Coatings Application and NDT”,
Proceedings of International Conference on Heat Exchanger Fouling and Cleaning VIII, 2009 (Peer-
reviewed).

2. Gawlik, K., Sugama, T., Webster, R., and Reams, W, “Field Testing of Heat Exchanger Tube
Coatings”, Geothermal Resources Council 1998 Annual Meeting.

3. “In Situe Coating Application Expected to Extend Tube Life”, Oil & Gas Journal, June 1999