Proceedings of International Conference on Heat Exchanger Fouling and Cleaning - 2013 (Peer-reviewed)

June 09 - 14, 2013, Budapest, Hungary

Editors: M.R. Malayeri, H. Müller-Steinhagen and A.P. Watkinson
Published online
www.heatexchanger-fouling.com

MINIMIZING CRUDE OIL FOULING BY MODIFYING THE SURFACE OF HEAT

EXCHANGERS WITH A FLEXIBLE CERAMIC COATING

O. Santos1, J. Anehamre1, C. Wictor2, A. Tornqvist3 and M. Nilsson2

1
Materials and Chemistry Center, Alfa Laval Lund AB, P.O. Box 74, SE-22100 Lund, Sweden
E-mail: [email protected]
2
Product Center – Compact Heat Exchangers, Alfa Laval Lund AB, P.O. Box 74, SE-22100 Lund, Sweden
3
Parts and Service Process Division, Alfa Laval Lund AB, P.O. Box 74, SE-22100 Lund, Sweden

ABSTRACT wall of the heat exchanger. The WAT depends on the crude
An oil repellent organic inorganic hybrid coating was oil composition and also on the determination method.
applied on titanium plate heat exchanger plates and field Elsharkawy et al. (2000) used differential scanning
tested in an offshore crude oil cooler. The first trial calorimetry to determine the WAT for different crude oils
consisted in a partially coated plate heat exchanger (PHE) and reported values ranging from 32 to 38°C. The
where after, due to promising results, a fully coated unit was composition of the wax deposit is time dependent starting
installed. After several months in operation the coated plates with a porous structure which gradually becomes denser as
were inspected both visually and by the help of a light more wax molecules diffuses into the porous structure and
optical microscope. The following parameters were oil diffuses out (Singh et al., 2000). The denser the deposit
assessed: Fouling amount and strength of adhesion to the layer the higher the impairment on the heat transfer due to
plate, coating condition and repellency. The coating was the insulating effect of the wax layer (thermal conductivity
found to be successful in minimizing the amount of fouling of 0.073 W / m K, Torresola, 1998) at the surface and the
and reducing its adhesion to the plate, where the fouled plate more challenging the cleaning becomes. Several approaches
could be completely cleaned by a high pressure water jet, to mitigate oil fouling are reported in the literature including
thereby avoiding the use of cleaning chemicals. Only minor heating, chemical treatment (inhibitors), paraffin
areas of the coating presented defects and in major parts of hydrodynamics and surface treatment (Merino-Garcia and
the plate the coating repellency was still intact. Correra, 2008; Paso et al., 2009). In this work we follow the
As opposed to the need of sending the plate heat later approach i.e. alteration of the surface properties of the
exchanger onshore for reconditioning once or twice a year, metal substrate.
depending on the season, the fully coated unit maintained its A literature search regarding the use of coatings for
thermal performance for around 2 years. This implies a fouling reduction in crude oil cooling plate heat exchangers
huge benefit for the end user not only in terms of increased was unfruitful. Some coatings have however been tested in
process efficiency but also lower cleaning related costs crude oil. Fernandez and Rothan (2009) compared the
(time, energy and chemicals). performance of coated and uncoated shell and tube
reboilers. The coating, a phenolic epoxy polymer, was
applied on the tube side where the crude oil flows and was
INTRODUCTION heated with steam on the shell side. No fouling reduction
Fouling of heat exchangers in the oil industry results in was achieved on the coated reboilers and the coating
a loss of performance which leads to production stops for diminished the heat transfer due to its low thermal
cleaning and therefore increased operation costs conductivity. The effect of different coatings on paraffin
(Deshannavar, et al., 2010). Crude oil dehydration process wax fouling of cold steel surfaces have also been
consists of produced water being removed from the wet investigated (Johnsen et al., 2009). Three coatings (no
crude oil by a series of separators where the heat transfer details were reported on two of the coatings while the third
between the media is carried out by plate heat exchangers. consisted of an organic inorganic hybrid coating based on
One of these heat exchangers, the crude oil cooler, cools aminopropylsilane) were tested in a batch baffle stirred
down the crude oil to the required level by circulating sea reactor with a binary test fluid consisting of a paraffinic
water in the adjacent channel. The crude oil will then be solvent (n-decane) and paraffin wax solute (n-tetracosane) at
stored and transported. The cooling is necessary to minimize different flow velocities. Two of the coatings reduced the
evaporation of light hydrocarbons that would otherwise amount of wax deposit independent of the flow velocity.
create an explosive atmosphere in tanks. Fouling occurs due Several fouling reducing coatings (SiF3+ and MoS22+ ion
to the fact that the temperature at and near the wall in the oil implantation, Diamond-like Carbon (DLC) sputtering, DLC-
channel can be lower than the wax appearance temperature Si-O and SiOx Plasma Enhanced Chemical Vapour
(WAT) leading to paraffin wax crystallization at the plate Deposition (PECVD), autocatalytic Ni-P-PTFE and silica)

79
 Santos et al. / Minimizing crude oil fouling by modifying the surface of …

have been applied to stainless steel and its surface properties

characterized (Santos et al., 2004). Some of these coatings
were then tested both at laboratory (Santos et al., 2006) and
pilot scale heat exchangers used in the food industry
(Rosmaninho et al., 2006; Balasubramanian and Puri, 2009;
Patel et al., 2013; Barish and Goddard, 2013). While some
improvements in fouling and cleaning were observed for
some of the coatings, there is still, to the author’s
knowledge, no commercially available coated plate heat
exchanger. This is due to the complexity of both the fouling
mechanism and the coating properties. Besides being able to
reduce foulant adhesion a coating, to be suitable to use in a
heat exchanger, needs to have a decent thermal conductivity,
good wear and abrasion resistance, good temperature and
chemical resistance and a good adhesion to the metal
substrate.
In this full-scale industrial case study, an organic
inorganic hybrid coating has been field tested in a crude oil
rig with promising results. The coating was applied on heat Fig. 1 Diagram of the synthesis of the repellent flexible
exchanger plates and the coated plates assembled on a M20- ceramic coating.
MFD unit. The duty was cooling of crude oil from 70 to 40
°C using sea water as a cooling media. The duty is highly
fouling with wax solidifying on the plate wall together with The properties of the coating relative to the reference
crystalline substances. The operating time of an uncoated titanium surface are presented in Table 1 and 2, where the
heat exchanger varies between 6 to 9 months depending on contact angle and surface energy values were provided by
the season where thereafter the increase in the pressure drop the coating manufacturer. The arithmetic mean surface
to unacceptable values together with the inability to cool the roughness (Ra) was measured by a Perthometer Perthen
crude oil to the required temperature requires the heat M4P profilometer on 10 different locations. The sampling
exchanger to be open and the plate pack send for points were ridges on the heat transfer surface and the
reconditioning. During the operating time the heat measuring length 4.8 mm (Fig. 2). The coating thickness
exchanger is cleaned bimonthly by the so-called hot runs varies from 5-10 µm as can be seen in a cross-section of a
(heating up the oil side by shutting off the cooling water representative coated plate imaged with a scanning electron
flow resulting in wax being partially dissolved). microscope (Inspect S50) (Fig. 3).

EXPERIMENTAL

Coated Surfaces

The oil repellent organic inorganic hybrid coating

(ORC) used in the field test was manufactured and applied
by an external company. The synthesis of the coating is
schematically presented in Fig. 1.

Fig. 2 Roughness measurement with a Perthometer at the

ridges on the heat transfer area.

www.heatexchanger-fouling.com 80
 Heat Exchanger Fouling and Cleaning – 2013

Table 1. Surface properties of the reference and coated

substrates.

Surface Water Formide Surface

contact contact α- roughness
angle angle (˚) Bromo (µm)
(˚) naphtha
lene
Titanium 80 (4s) 60 ± 1 36 ± 2 1,03 ± 0,04
40 (180
s)a
ORC 105 ± 1 92 ± 1 66 ± 2 0,52 ± 0,14
a
Two time dependent angles are provided since the contact angle is not
Fig. 4 – Schematic picture of a plate heat exchanger
static.
showing the locations of the coated plates.
Table 2. Surface energy of the reference and coated
substrates.
Fully coated plate pack. The next step was to field test
two PHE’s consisting of a fully coated plate pack. One of
Surface Time γTOT γLW γ+ γ-
the heat exchangers was in operation for 5 months before
(mN/m) (mN/m) (mN/m) (mN/m)
being disassembled and the coating performance
investigated. The second heat exchanger was in operation
Titanium 4s 39 36 0.2 5.8
for 2 years.
180 s 47 36 1.4 20
The coating condition was assessed by visual inspection
ORC static 21 21 0 1.2
γTOT Total surface energy and with the aid of a LOM. Cross sections of different areas
γLW Lifshitz-van der Waals component of the surface energy of the plate (distribution area and heat transfer area) were
γ+ acid component of the surface energy prepared to measure the coating thickness. The coating
γ – base component of the surface energy repellent properties were investigated by the permanent
marker test and the fouling adhesion to the coating by
cleaning with a cloth and high pressure water jet.

RESULTS

Partially Coated Plate Pack

The coated plates had a lower amount of fouling, a
reduction of 65% was determined (Table 3, Fig. 5). The
growth of the wax layer was found to depend on the
substrate. The uncoated titanium surfaces were covered with
a homogeneous layer whereas on the coated surfaces only
small islands of wax were seen. Adhesion of the wax layer
to the surface was also significantly reduced compared to
the uncoated plates, where it was possible to clean the
coated plates only with high pressure water jet.
Fig. 3 SEM image of a cross section of a coated plate.

Full-Scale Industrial Field Test Table 3. Amount of fouling determined on both the titanium
and coated plate. The amount of fouling was determined
Partially coated plate pack. The first field test was run from 5 plates placed at the end of the plate pack. sd standard
with a partially coated plate pack in a M20-MFD plate heat deviation
exchanger. From the total 349 titanium plates 15 were
coated and 5 each placed in the beginning, middle and end
Surface Average fouling Fouling reduction
of the plate pack (Fig. 4). The unit was in operation for 8
(g) ± sd (%)
months with hot runs being the only cleaning performed.
The amount of fouling on the coated plates compared to Titanium 585 ± 125 -
the uncoated reference plates was assessed visually and by ORC 203 ± 48 65
weighting the plates before and after the field test. Fouling
adhesion was roughly estimated by removal the fouling with
a cloth and by high pressure water jet.

www.heatexchanger-fouling.com 81
 Santos et al. / Minimizing crude oil fouling by modifying the surface of …

Fig. 5 – Fouling appearance on the reference titanium plate

(left) and on the ORC plate (right).

Fig. 7 – Area of a coated plate cleaned with a dry cloth

Fouling was also observed to depend on the relative showing that the easy to clean and repellent properties of the
position of the coated plates in the PHE, where less fouling coating remained intact after 5 months in operation.
was seen on the plates positioned in the beginning. This can
be related to a slightly higher flow rate in this position of the
heat exchanger. Minor coating defects and a reduction in The permanent market test gives an indication
coating adhesion were observed after 8 months operation. regarding the repellent and easy to clean properties of a
However, the coating still showed an intact oleophobicity. coating. In some areas of the coated plate no repellency was
observed but the easy to clean properties were unaltered. In
other areas/other plates the repellent properties were still
Fully Coated Plate Pack evident (Fig. 7).
Fig. 6 shows that the amount of fouling on the coated The plates were almost completely cleaned after high
plates was minimal and, as observed on the partially coated pressure water jet where only some minor deposits remained
plate pack, the wax deposited as dispersed islands. The (Fig. 8).
adhesion of the fouling layer to the plates was very weak
where after cleaning with a dry cloth almost no fouling
remained (Fig. 7).

Fig.8 – Coated plate after high pressure water jet cleaning.

Fig. 6 – Amount and appearance of the fouling on a coated
plate. Coating degradation was observed in some plates mainly
located at the edges adjacent to the long side of the plate
(Fig. 9).

www.heatexchanger-fouling.com 82
 Heat Exchanger Fouling and Cleaning – 2013

groups have a surface energy of about 30 mN/m and the side

–CH2- groups of about 36 mN/m (Wang and Ober, 1997). It
is believed that the paraffin crystals are oriented with the
end groups contacting the surface to minimize the total free
energy of the system. From Table 2 it is seen that the ORC
has a surface energy lower than the end groups of the
paraffin molecules while the titanium surface has a higher
surface energy.
The low adhesion forces between the coated surface
and wax are also responsible for the different wax layer
growth observed depending on the substrate. The weaker
adhesive forces on the coated surfaces make it difficult for
Fig. 9 – Coating degradation mainly located near the long the wax molecules to attach in a coherent layer. However,
sides of the plate. Degraded area is encircle in red (matt once some wax molecules have already attached subsequent
appearance). deposition will proceed on the deposited wax forming small
islands on the surfaces. The same behavior was observed by
Johnsen et al. (2011) when studying wax deposition on cold
The dry film thickness of the coating from different uncoated and coated steel.
areas of the plate (heat transfer and distribution area – see The amount of fouling was found to depend slightly
fig. 6 where the locations of the cross sections are indicated on the location of the plate in the plate pack where less
by the yellow squares) varied from 4 to 11 µm (Fig. 10). fouling was observed in the first plates (near the frame
plate). Kukulka and Leising (2010) while evaluating
compact heat exchanger coatings determined different
amounts of fouling depending on the position of the plates
in the plate pack. For both large and small PHE a lower
amount of deposit was also calculated in the beginning of
the plate pack. This is related to the flow velocity which is
somewhat higher on this location.
In general the coating presented a good stability after
150 days in the crude oil, however some coating degradation
was observed in some of the plates. The location of these
failures was mainly at the edges of the long sides of the plate
Fig. 10 - LOM photo of a cross section taken from the heat
(Fig. 9). The explanation could be related to the flow
transfer (left) and distribution area (right) of a coated plate
velocity and to the mechanical stresses which are larger at
(see Fig. 6) where measurements of the dry film thickness
these locations. The thickness of the coating measured at
gave values from 5 to 11 µm.
different locations on one plate corresponded to the original
thickness after coating application. This means that in the
areas where the coating was present it was in a good
condition.
DISCUSSION
Another aspect one needs to consider regarding
coatings for heat exchangers is the influence the coating will
Adhesion of a substance to a substrate depends on
have on the thermal performance of the PHE. The coating
surface properties such as roughness and surface free
will act as a fouling layer in respect to the heat transfer,
energy. It is observed, in general, that low surface free
however due to the low thickness and reasonable thermal
energies lead to minimal adhesion forces between the
conductivity (~ 2 W / m K) of the present coating the
surfaces and deposit. An increase in the roughness of
reduction in heat transfer was estimated to be around 4%. It
stainless steel surfaces was found to result in a larger wax
should also be noted that an uncoated heat exchanger with a
deposition probably due to an increase of heterogeneous
thick wax deposit (low heat conductivity) will result in a
nucleation sites (Jorda, 1966). Therefore, a coating with
lower thermal performance as compared to a coated heat
both lower surface energy and lower roughness compared to
exchanger where only a thin wax layer deposits.
the metal surface should reduce the amount of wax
The stability/repellency of the coating after the 2
deposition. This was in fact observed in the present work
years field test with the fully coated heat exchanger could
where the ORC, having the surface energy and roughness
not be evaluated so far but the lack of the need for
reduced by half compared to the titanium surface (Table 2),
reconditioning points to a good coating performance.
decreased the amount of fouling (Fig. 5). In order to
minimize adhesion and therefore fouling the surface energy
of the metal surface should be lower than the surface tension CONCLUSIONS
of the foulant. The surface free energy of paraffin molecules The conclusions of this study are as follows.
in an orthorhombic lattice is low, where the end -CH3

www.heatexchanger-fouling.com 83
 Santos et al. / Minimizing crude oil fouling by modifying the surface of …

1. The coated plates in the partially coated plate pack Kukulka, D. J., and Leising, P., 2010, Evaluation of
showed a 65% decrease in the amount of fouling after 8 heat exchanger surface coatings, Applied Thermal Eng.,
months operation. Vol. 30, pp. 2333-2338.
2. Fouling adhesion to the coated plates was also reduced Merino-Garcia, D., and Correra, S., 2008, Cold flow: A
where the plates could be cleaned using only a high review of a technology to avoid wax deposition, Pet. Sci.
pressure water jet. Technol., Vol. 26, pp. 446-459.
3. Only minor coating failures were observed on the fully
coated unit after 5 months operation. Patel, J. S., Bansal, B., Jones, M. I., and Hyland, M.,
4. No stop for reconditioning was necessary with the fully 2013, Fouling behavior of milk and whey protein isolate
coated plate pack during the 2 years field test, leading solution on doped diamond-like carbon modified surfaces, J.
to an increase of operating time between maintenance Food Eng., Vol. 116, pp. 413-421.
of more than double. Paso, K., Kompalla, T., Aske, N., Rønningsen, H. P.,
5. The ORC proved to be successful in reducing fouling Øye, G., and Sjöblom, J., 2009, Novel surfaces with
and improving cleanability in an offshore crude oil applicability for preventing wax deposition: A review, J.
cooler heat exchanger. Dispersion Sci. Technology, Vol. 30, pp. 757-781.
Rosmaninho, R., Santos, O., Nylander, T., Paulsson, P.,
Beuf, M., Benezech, T., Yiantsios, S., Andritsos, N.,
NOMENCLATURE Karabelas, A., Rizzo, G., 2006, Modified stainless steel
surfaces targeted to reduce fouling – Evaluation of fouling
PDMS Polydimethylsiloxane by milk components, J. Food Eng., Vol. 80, pp.1176-1187.
ORC Oil repellent coating Santos, O., Nylander, T., Rosmaninho, R., Rizzo, G.,
SEM Scanning electron microscopy Yiantsios, S., Andritsos, N., Karabelas, A., Muller-
LOM Light optical microscopy Steinhagen, H., Melo, L., and Boulang-Petermann, L., 2004,
PHE Plate heat exchanger Modified stainless steel surfaces targeted to reduce fouling –
WAT Wax appearance temperature surface characterization, J. Food Eng., Vol. 64, pp. 63-79.
sd Standard deviation Santos, O., Nylander, T., Shillén, K., Paulsson, M., and
γTOT Total surface energy Trägårdh, C., 2006, Effect of surface and bulk solution
γLW Lifshitz-van der Waals component of the surface energy properties on the adsorption of whey protein onto steel
γ+ acid component of the surface energy surfaces at high temperature, J. Food Eng., Vol. 73, pp.
γ – base component of the surface energy 174-189.
Singh, P., Venkatesan, R., Fogler, H. S., and Nagarajan,
N., 2000, Formation and aging of incipient thin film wax-oil
gels, AIChE J., Vol. 46, pp. 1059-1074.
REFERENCES Torresola, J., Solidification properties of certain waxed
Balasubramanian, S., and Puri, V. M., 2009, Thermal and paraffins, 1998, Master’s thesis, Massachusetts Institute
energy savings in pilot-scale plate heat exchangers system of Technology, Department of Mechanical Engineering, p.
during product processing using modified surfaces, J. Food 51.
Eng., Vol. 91, pp. 608-611. Wang, J., and Ober, C. K., 1997, Self-organizing
Barish, J. A., and Goddard, J., 2013, Anti-fouling materials with low surface energy: The synthesis and solid-
surface modified stainless steel for food processing, Food state properties of semifluorinated side-chain ionenes,
and Bioproducts processing, In press. Macromolecules, Vol. 30, pp. 7560-7567.
Deshannavar, U. B., Rafeen, M. S., Ramasamy, M., and
Subbarao, D., 2010, Crude oil fouling: A Review, J. Applied
Sci., Vol. 10, pp. 3167-3174.
Elsharkawy, A. M., Al-Sahhaf, T.A., and Fahim, M. A.,
2000, Wax deposition from Middle East crudes, Fuel, Vol.
79, pp. 1047-1055.
Fernandez, G. T., and Rothan, T. A., 2009,
Effectiveness of coatings to reduce fouling in Shell and
Tube heat exchangers, Proc. AIChE The 2009 Spring
National Meeting, Tampa, Florida, 104c.
Johnsen, S.G., Hanetho, S. M., Tetlie, P., Johansen, S.
T:, Einarsrud, M. A., Kaus, I. and Simon, C. R., Studies of
paraffin wax deposition on coated and non-coated steel
surfaces, Proc. Int. Conference on Heat Exchanger Fouling
and Cleaning 2011, Crete Island, Greece, pp. 251-258.
Jorda, R. M., 1966, Paraffin deposition and prevention
in oil wells, J. Pet. Technol., Vol. 18, pp. 1605-1612.

www.heatexchanger-fouling.com 84