Feb-2019

Avoiding unplanned reactor shutdowns

In hydroprocessing operations, prevention is better than just dealing with the effects of corrosion and fouling problems.

Maria J L Perez, Roberta Cenni, Zeina Maroun and Kenneth G Myntman
Haldor Topsoe

Article Summary

Refineries worldwide are showing a growing interest in processing a wide range of renewable feedstocks and opportunity crudes. These can provide good commercial opportunities, but unfortunately also often feature more impurities, impurities of different, less familiar types and high total acid numbers. This in turn usually means having to deal with larger amounts of corrosion debris and other contaminants from upstream equipment and fittings.

In the chemically aggressive operating environments prevalent in refineries, older installations wear and corrode faster and more easily because they were not designed to cope with these many, newer kinds of more corrosive feedstocks. Such corrosion problems will only increase and accelerate, because it is an inescapable fact that the majority of refineries were built more than 40 years ago, and we are unlikely to see new refineries built in economies reducing their reliance on fossil fuels.

Corrosion and fouling impact on hydroprocessing
Hydroprocessing units are critical in effective refinery operations, and such corrosion – along with other forms of contamination that affect what is going on in the reactor – is one of the major causes of unplanned shutdowns here.

Any such unscheduled stoppages usually end up costing millions of dollars, with knock-on effects and penalties that reverberate throughout the supply chain. This makes them the one thing refineries want to avoid.

Pressure drop is disruptive and expensive
Corrosion particles and other contaminants entering the hydroprocessing reactor cause build-up of pressure drop across the hydroprocessing reactor by blocking the feedstock’s passage through the catalyst bed.

Two scenarios are common, usually depending upon how the scale particles are distributed in size (see Figures 1 and 2). The first is when the many corrosion particles and contaminants lodge in the voids between the catalyst pellets and accumulate more or less homogeneously throughout the grading, and sometimes even down to the active catalyst layer. The pressure drop build-up across the hydroprocessing reactor is then fairly continuous and relatively steady.

The second is when the contaminants get trapped in a relatively narrow layer between the different grading layers. In such cases, the pressure drop builds up exponentially and is very unpredictable. In these cases, the hydraulic or mechanical limitation of the reactor is often reached rapidly and it is therefore necessary to deal with it urgently.

Catalyst bed plugging is directly related to the total bed void available for contaminant deposition. As the available catalyst bed void is reduced over time, pressure drop develops across the reactor. The details are different for each catalyst type and shape as well as specific loading, but the overall pressure drop pattern is similar for all of them. The critical point is reached when the bed void volume is reduced to 20-25% of the original void available. From this point on, the pressure drop development curve is exponential. This is the point of no return as far as reactor shutdown is concerned.

Two realistic solutions
Once pressure drop has started escalating, the refinery is normally left with only two realistic solutions.

If the planned turnaround is fairly imminent and the refinery wants to keep the plant running until then, it is possible to reduce throughput and the recycle gas rate until the unit can be shut down for catalyst replacement, and accept the loss of revenue this would involve (see Case 1).
On the other hand, if it would cost less to take the unit out of operation completely for around 15 days, it is possible to ‘take the top off the contaminant load’ by skimming the catalyst bed.

Managing and coping with pressure drop build-up and its many effects are multi-faceted endeavours for any refinery. The aim is to ensure the maximum productive on-stream duration at the design throughput of the specific unit with the specific catalyst loading and grading configuration, up to the pressure drop limitation.

Case 1: Cost of operating at reduced capacity
A 27000 b/d hydrotreating unit operated for 470 days undisturbed. After 470 days, the reactor started experiencing pressure drop build-up and, consequently, the refinery chose to start reducing the feed rate in order to delay the shut-down. The unit was able to run for another 73 days.

In those 73 days, the loss of production was 61300 m3, corresponding to 386000 bbl. At a €6/bbl diesel spread, this corresponds to a total cost of €2.3 million for the reduced production (see Figure 3).

Reducing risk and uncertainty
Unscheduled interruption of refinery operations involves uncertainty and risk – which are always bad for business.
Different kinds of temporary solutions and traditional preventive measures enable refineries to reduce risk and uncertainty, and improve processing efficiency by ensuring better control of what is happening in the reactor at any given time.
 
Partial solutions are common
There are several traditional ways to try to prevent and deal with difficulties with pressure drop build-up across hydroprocessing reactors.

Feed filters
Topsoe recommends always having effective feed filter systems installed upstream of the reactor pump, in order to prevent the influx of carryover contaminants from transport and/or upstream processing.

Effective filtration requires a filter medium specifically configured to allow the right amounts of liquid to pass through, while holding back solid particles exactly as required. An effective, well-functioning feed filter normally removes impurities with particle sizes down to 25 microns.

Selection of the most appropriate filter type depends on feed type and quality, temperature and fouling level. The actual location of the feed filter also plays an important role in maximising filtration effectiveness, because the viscosity of the feed has to be low enough to ensure smooth filtration of impurities and to extend the filtration cycle.

Grading solutions
Conventional grading solutions are effective and satisfactory in many reactor operations. However, as soon as a refinery has fouling problems due to the incoming corrosion products and other particles (such as coke fines, salts or silicon), special care should be given to the design of the grading layer to ensure cost-effective operation.

The principle behind pressure drop build-up prevention is to use grading materials of differing sizes, shapes and specifications on top of the bulk catalyst in order to create a filtering effect that traps the largest inorganic feedstock contaminants in the upper layers and the smaller particles in the lower layers.