Jun-2017

Building robust hydroprocessing operations

Advances in catalyst technology enable refiners to maximise profitability by processing opportunity crudes and running heavier feeds.

AMIT KELKAR
Criterion Catalysts & Technologies

Article Summary

Unplanned downtime is one of the most critical factors that influence refinery profitability. In terms of hydroprocessing catalyst, minimising unplanned downtime equates to ensuring the catalyst system is designed to process the most profitable feeds while meeting the target cycle length. An unscheduled pit stop to replace catalyst early, either due to rapid activity loss or pressure drop growth, has severe financial consequences. Catalyst deactivation is caused either by contamination from feed poisons or by deposition of coke. Typical poisons include silicon (Si), nickel (Ni) and vanadium (V), iron (Fe) and arsenic (As). The mechanism of interaction with the catalyst and deactivation severity varies from one poison to the next. For example, for the same spent catalyst content, arsenic is a more severe poison than silicon, iron, nickel and vanadium. However, since these latter poisons often exist at higher feed concentrations than arsenic, they can still cause severe catalyst deactivation.

Strategies to combat the impact of arsenic and silicon poisoning will be discussed first, followed by nickel and vanadium and high Conradson carbon residue feeds. Commercial examples highlighting the concepts will be shared for each application.

Arsenic
Arsenic is indigenous to many types of crude, particularly heavy oil and/or oil sands derived crudes from Canada and South America (see Figure 1). It is also present in coker products and occurs over a wide range of boiling points, naphtha to heavy gasoil.

Arsenic in feed reacts with the nickel present in hydroprocessing catalysts to form nickel arsenide. A relatively low level of arsenic has a severe impact on catalyst activity which can easily be reduced to half with 1 wt% arsenic deposition (see Figure 2).

Since arsenic deactivates catalyst by attaching to active nickel sites, an effective trap catalyst is designed to take advantage of this affinity by using nickel to absorb arsenic while simultaneously providing hydrogenation activity. As Figure 3 shows, arsenic trap catalyst is more stable than primary conversion catalysts. It is loaded upstream of the main catalyst to protect it from arsenic induced aging.

A new generation MaxTrap[As] was introduced in 2014 with more available and highly dispersed nickel to increase arsenic pick-up. Test basket results (see Figure 4) show that new generation MaxTrap[As] can pick up 50% more arsenic compared to the previous generation and up to three to four times higher than standard NiMo hydrotreating catalysts.

Example
The benefit of the current MaxTrap[As] is validated in commercial operation (see Figure 5). It compares the normalised HDN activity for a bulk naphtha/kero hydrotreater for the last two cycles. Cycle 1 used first generation arsenic trap followed by a high activity NiMo catalyst. For Cycle 2, arsenic trap was switched to current MaxTrap[As]. The higher arsenic pick-up capacity of the new MaxTrap[As] makes the overall system more stable. Consequently, the deactivation rate is lower, resulting in a 30% increase in cycle length compared to Cycle 1.

Silicon
Coker derived streams are primary sources of silicon due to use of silicon based anti-foam agents in delayed cokers. Silicon is also present in a wide variety of crudes including Maya, Canadian syncrudes, Petrozuata, Cerro Negro, and Venezuelan. It may also show up in straight run naphtha, distillate, and gasoil streams due to injection of silicon based additives in upstream crude production and pipeline operations.

Unlike arsenic, silicon is removed by interaction with the alumina surface without involvement of the active metals present on the catalyst. The removal mechanism is adsorption of silicon by the hydroxyl groups present on the alumina carrier. The net result is accumulation of a layer of silicon on the catalyst surface area which restricts access to active sites, thus reducing catalyst activity.

Silicon deposition is greatly influenced by feed distillation and the pore geometry of the catalyst. Naphtha and distillate feeds consist of smaller molecules that are not diffusion limited and can easily penetrate the catalyst pellet. Consequently, silicon pick-up is primarily governed by catalyst surface area; more surface area results in higher capacity (see Figure 6a).

Heavier streams such as VGO include larger silicon containing molecules which are unable to fully penetrate the catalyst pellet, thus accessing only a small portion of the available surface area (see Figure 6b).

Catalysts with larger pore diameters are recommended for VGO service to mitigate diffusion limitations and increase utilisation of available catalyst surface area (see Figure 6c).

Silicon is a milder catalyst poison compared to arsenic; main catalysts are typically able to tolerate  up to 10 wt% silicon deposition before seeing significant impact on activity. HDN activity is more sensitive to silicon poisoning particularly in naphtha service. Hence, design of catalyst systems for most NHTs requires a balance of silicon capacity and HDS/HDN activity.

Criterion has recently launched a new dual function catalyst (DN-240) for naphtha and distillate service. It has the same silicon pick-up as the previous generation dual function catalyst (DN-140) and 25-30% higher HDS and HDN activity. Higher activity is a result of improved metals impregnation chemistry (see Figure 7).

Criterion’s MaxTrap[Si] is a specially designed Silicon trap catalyst which has very high uptake capacity and enough activity to saturate olefins and treat the easy sulphur and nitrogen molecules. The features of DN-240 (high silicon uptake and high activity) mean it can be combined with MaxTrap[Si] and our high activity main bed catalysts in a number of different combinations to satisfy the silicon capacity and HDS/HDN activity requirements of the unit. In addition, DN-240 is ideally suited for the top bed in DHT service to provide silicon pick-up as well as sufficient activity for easy HDS reactions.

Example
The following example highlights the importance of having a balanced system with both silicon uptake capacity and HDS/HDN activity to achieve overall unit stability and minimise unexpected downtime. A unit processing coker naphtha was limited by silicon uptake capacity. The unit was loaded with MaxTrap[Si] in Rx 1 and DN-140 in Rx 2. In an attempt to increase silicon capacity, the refiner switched to a higher surface area, inert silicon trap catalyst in Rx 1 and a high activity NiMo catalyst in Rx 2. As with most coker naphtha units, the two reactors are highly heat integrated. As Table 1 and Figure 8 show, the minimal activity of the inert silicon trap resulted in a cycle which processed less than half the total feed barrels due to a lack of exotherm in R1 and lower overall activity of the system. Spent catalyst analysis confirmed that silicon pick-up was less than half the previous system. The unit switched back to the original catalyst system in the next cycle; with the balanced system the total amount of silicon processed was close to the original cycle even though the silicon content of the feed had increased 30%.