• What are the likely causes of and solutions for severe fouling in a vacuum unit’s preheat section?



  • Sam Lordo, Nalco Champion, salordo@comcast.net

    As mentioned in another response, this is an uncommon design, however they do exist and are prone to fouling from several things such as:

    - Low flow in the exchanger.
    - Compatibility of the feedstock.
    - Caustic.
    - Crude solids.
    - Poor desalting.
    - Coke fines from upstream operations.
    - Corrosion byproducts.
    - Oxygen contamination.


  • Marco Roncato, CHIMEC SpA, mroncato@chimec.it

    The fouling affecting a vacuum distillation unit (VDU) HEX train depends a lot on the VDU’s position in the process stream:

    1) Downstream of a crude unit
    2) Downstream of a cracking unit (for instance a visbreaking)

    Case 1 (downstream a crude unit)

    In this case, the fouling content of the VDU feed can be both:

    • Inorganic (everything not removed at desalter stage will end up – concentrated – in the atmospheric residue, that is VDU feed)
    - sediment deposits
    - sand
    - corrosion products
    - salts
    - caustic injected in the crude oil after the desalter
    • Organic
    - slops
    - sludges
    - destabilised asphaltenes
    - heavy paraffins

    Which of the two will be predominant depends on the intrinsic nature of the crude diet (more or less clean, more or less stable) and on desalter performance.

    Cleaning of VDU feed

    For the inorganic part, the first measure is to maximise desalting efficiency: with lower basic sediment and water, and lower caustic injection in the desalted crude, their concentration in the VDU feed will be lower as will their role in reducing the heat exchangers’ performance.

    For more than four decades, Chimec has developed a programme – based on demulsifiers together with analytical and software tools – aimed at helping refineries in improving the desalter’s performance.

    But after this first step, despite good desalting efficiency, a further improvement can be achieved by injecting a caustic replacer in the desalted crude.

    Caustic injection downstream of the desalter is recognised as a cheap and effective method to reduce overhead corrosion, but unfortunately at the same time NaOH can be detrimental because it can contaminate the bottom streams affecting the downstream units.

    Caustic is itself a foulant agent; moreover, sodium is a well-known dehydrogenation catalyst, hence it is a coke promoter – it increases the coking rate in the downstream unit (for instance VDU) HEX train and furnaces.

    Together with this, it is worth mentioning that caustic is also responsible for catalyst poisoning in downstream catalytic plants (FCC unit, hydrocracking unit, residue desulphurisation and, so on) and for the production of low-quality fuel oil (fouling problems in the burners, for instance in the power station or in the fuel oil furnaces).

    In order to manage all issues, Chimec has developed a caustic replacer, Chimec 3034, to substitute completely or partially the injection of NaOH downstream of the desalter; it is completely organic and metal free which means it has no impact on coke promotion and catalyst deactivation.

    The overall effect is the reduction of caustic and sodium content in the atmospheric residue.

    This implies no risk of NaOH induced fouling in the preheat trains.

    Managing fouling precursors

    After ‘cleaning’ the VDU feed (by maximising the desalter’s efficiency and reducing NaOH injection), the following action is to inject a suitable dispersant chemical in the VDU feed able to provide protection from both organic fouling (mainly due to asphaltenes deposition) and the inorganic variety.

    Case 2 (downstream a cracking unit)

    In this case, together with inorganic and organic fouling, there is also the strong presence of a further fouling portion, which usually becomes predominant – coke deriving from cracking processes and heavy gums due to olefinic polymerisation reactions.

    For such units, usually the situation is more severe, because of the presence of coke and the higher amount of asphaltenes in the feed and generally lower stability.

    Therefore, the dispersant must be more specific and stronger compared to the one injected into the feed of VDUs processing atmospheric residue.

    Moreover, it is necessary to tackle the polymerisation reactions through the injection of a stabiliser, so the olefins cannot form heavy gums.

    In such a situation, if possible the chemicals have to be injected directly into the cracking unit’s MF bottom so that they can act from the very beginning, before coke and asphaltenes can settle and polymers are formed, fouling downstream equipment.


  • Celso Pajaro, Sulzer Chemtech, celso.pajaro@sulzer.com

    The question is not so clear about the nature of the atmospheric residue and/or the streams that exchange heat with it. These are some potential causes:

    • Low velocity of the atmospheric residue in the tubes (it should be above 6 ft/s)
    • Leakage in the heat exchanger that allows the hot stream into the residue creating compatibility problems
    • Mixing different types of atmospheric residues (paraffinic and non-paraffinic sources) that have compatibility problems leading to asphaltene precipitation.
    • Paraffinic residues stored at low temperatures which allow the precipitation of solid wax.


  • Mel Larson, Becht, mlarson@becht.com

    This is a bit of an unusual question as most vacuum columns are fed directly from the atmospheric column. A ‘pre-heat’ section would indicate a standalone vacuum column before the furnace. On the feed side of a pre-heat system would be potential oxygen contamination and interactions of compounds that result in precipitation either in a tank or as the solution is heated with an increased propensity for precipitation. The challenge on the passing effluent stream in preheat service is the temperature difference between the passing fluids and the resulting tube wall temperatures. On the bottoms side of heat exchange, the tube wall engagement will be ‘cooler’ rising to a higher potential of ‘adherence’ to a surface and or cooling rapidly enough so as to enhance a precipitate dropping out (some call this asphaltenes, however it can be any maltene structure). The challenge in these services is to accurately identify the cause. Considering this may be an interaction that causes some type of precipitate formation, find an appropriate solvent that can keep the system flowing and in solution. In very severe services it is justifiable to install spare standby exchangers with rotation of cleaning and service to minimise the lost revenue from outages. Others have found that some chemical additives work, but careful consideration is required as to the impact on any downstream systems.


  • Chris Claesen, Nalco Water, Chris.Claesen@ecolab.com

    There can be multiple reasons for this. If the preheat is continuously fouling it may be a design issue related to exchanger design or configuration. The fouling can also be related to crude blending, type of crudes processed, impurities or contaminants content such as solids and salts, desalter operation, and caustic addition. The use of intermediate storage of atmospheric residue and blending of other streams or imported materials can have a big impact and needs to be evaluated. Key is to have a good model of the heat exchanger train that can track the performance of the whole train by NFIT and of each individual exchanger such as Nalco Water’s Monitor. This can help to correlate fouling events with feed composition and operational conditions. It is also important to have a good analytical programme in place to trend feed stability, composition and contaminant content. This, together with analyses of fouling material, can help determine the root cause and develop a mitigation strategy. The mitigation can be by elimination of problem crudes, improved contaminant removal in the desalters, and controlled blending for increased stability. This can be combined with the use of a properly selected antifoulant, such as Nalco Water’s Thermogain which deals with the identified fouling mechanism.


  • Berthold Otzisk, Kurita Europe, berthold.otzisk@kurita-water.com

    The vacuum unit separates the heavier crude oil fractions to decrease the boiling temperatures with a lowering of the pressure. The boiling points are so high that thermal cracking would occur under atmospheric distillation.

    Many experienced specialists are working on vacuum heater designs and revamps to improve the efficiency with less coke fouling. Vacuum heaters are typically cabin or box type heaters with four or six passes in a single radiant cell. Maximising the convection section duty will decrease the radiant section duty, which reduces the coke formation potential. The localised oil film temperature and oil residence time in the radiant section depend on the heat flux, tube mass flux rates, and bulk oil temperature. They are variables to minimise the rate of coking.

    Some crude oils are less stable than others and have a poor thermal stability. Asphaltenes, maltenes, waxes, and resins are components that change the behaviour and structure with temperature, pressure, and the composition of crude oil. Asphaltenes are very sensitive to shearing forces and electrostatic interactions, which is why vacuum units are often affected by asphaltene fouling and coking. Crude oils with poor thermal stability begin to generate coke and gas at relatively low temperatures. Crude unit heaters with a high outlet temperature and high residence time in the crude column bottom decrease the oil stability with rapid coke and gas formation. Minimising the oil film temperature and oil residence time will help to decrease the potential for coke formation, which improves the run length of the heater.


  • Jérémy Provost, Technip Energies, Jeremy.Provost@technipenergies.com

    There are few severe fouling root causes in a vacuum unit’s preheat section. Their relative impacts depend mainly on the refinery’s strategy and objective (processing Middle East crude oils, heavy Canadian blendings, capture opportunity crudes on the market, and so on). These choices help determine the operation, (required flexibility) of the preheat train and its design. Nevertheless, the main causes which lead to severe fouling are identified below.

    Severe fouling in a vacuum unit’s preheat section is generally linked to asphaltene precipitation and even coking. This phenomenon is strongly governed by product temperature. Too high a wall temperature in a heat exchanger of the preheat train will be prone to fouling development. This can later spread through the whole train by reducing heat transfer. For instance, too cold a desalter inlet temperature, resulting from a poor performing and fouled cold train part, will jeopardise the desalter performance and result in a higher corrosion risk in the hot train part. A similar performance cut in the hot train part will impact consumption of the fired heater prior to the distillation column as well.

    In the early 2010s, Technip Energies, in partnership with Total and CEA, co-developed a dedicated test loop named BEECH (Fouling Exchanger Test Loop in French). Several measurement campaigns and studies have been performed showing the importance of wall temperature, shear stress, and heat exchanger design on the fouling mechanisms in a preheat train. The main outcomes to limit fouling development are:

    • Reduce the wall temperatures as low as possible by promoting a high heat transfer coefficient on the cold fluid side.

    • Maximise shear stresses at the wall on both exchanger sides to limit deposits and improve heat transfer coefficients. This can be done by combining high regime flow rates (for instance, above 2.0 m/s on the tube side) with well-chosen and proven turbulence promoter heat transfer solutions (grooves or corrugations on walls, inserts).
    • Pressure drop should not be a ‘target’ but a ‘consequence’: Designs with higher energy consumption, using a more powerful pump will be paid back by a limited fouling development and hence a longer production lifetime.
    • Manage pressure drop on the whole train and not item by item; pressure drop must be expended where it is needed.
    • Avoid excess area applying standard fouling resistance design method and prefer a margin based method where a limited heat transfer area is allocated to manage fouling. When using resistance based design, up to 85% of the heat transfer area can be dedicated, from the design phase, only to manage future fouling. In general, this results in very low velocities, leading to lower heat transfer coefficients and higher tube wall temperatures, accelerating fouling development.
    • Whatever the selected heat exchanger technology is, always take care of flow distribution in all the parts and avoid dead zone areas and bypass flows. These are starting points for fouling development. For instance, when designing a shell and tube heat exchanger, careful attention is given to the shell side to limit bypass and dead zone areas with relevant baffle type selection and arrangement.


  • Simon Calverley, KBC (A Yokogawa Company), Simon.Calverley@KBC.global

    Fouling in vacuum unit preheat trains is often associated with the higher temperature section of the preheat train though not exclusively so. There are numerous potential causes of fouling. Some of these are:

    • Low velocities can lead to fouling. These are often addressed by design changes to either the exchanger internals or by changing the arrangement of the exchanger (e.g. from parallel to series).
    • Corrosion deposits can cause fouling.
    • Heavy cracked slop oil can end up in the vacuum unit, particularly if there is a risk of oxygen contamination (for instance being stored in un-blanketed tanks). These are often better processed in a unit that has cracked streams such as a visbreaker or FCC. Alternatively, blanketing the tanks they are stored in can reduce the risk of fouling. Some refiners have had success with oxygen scavengers.
    • Asphaltene deposition can occur in the vacuum unit preheat train.
    • Inadequate desalting can lead to fouling, often because solids are not effectively removed. If the vacuum units are fed only with atmospheric residue from the refinery’s crude unit, this is more likely to be problematic in the crude unit. If the vacuum unit is fed with imported residue then either consider desalting it or, if it is desalted, then review the desalter operation.

    Analysis of the deposits helps in identifying the problem and the solutions to alleviate them. Antifoulants and/or corrosion inhibitors may help and there are commercially available programmes, such as KBC’s HX monitor, for heat exchanger monitoring, advice on which exchangers to clean, and estimating the cost of the fouling (such as increased energy consumption and CO2 emissions, yield reduction, and throughput reduction). All these will help decide how much investigation is warranted and in justifying remedial action.