The quiet refinery upgrade paying off: longer life for hydrotreating and reforming catalysts

Hydrotreating (HDT) and reforming units do the heavy lifting in low-sulfur fuel production, yet their catalysts inexorably lose activity to coke, sintering, and poisons like Ni, V, Fe, As, Cl, and Si (sciencedirect.com). One recent review puts it plainly: “catalyst deactivation in heavy oil hydroprocessing is primarily caused by coke formation, metal deposition, and sintering,” with asphaltene and metal content playing a “crucial role” (sciencedirect.com).

In a modern ultra-low-sulfur diesel (ULSD) hydrotreater, cycles must stretch ~48 months to hit production targets, a window that turns every deactivation delay into economic gain (ogst.ifpenergiesnouvelles.fr). Hydroprocessing catalysts often represent >40% of a refinery’s catalyst inventory and must reduce sulfur to ≤10 ppm (ogst.ifpenergiesnouvelles.fr) (ogst.ifpenergiesnouvelles.fr), so longer life cuts costs, downtime, and even disposal — avoiding ~5% of catalyst weight in landfill fees (ogst.ifpenergiesnouvelles.fr). Reforming catalysts (typically platinum-based) face similar losses from coke and poisons, with semi-regenerative units commonly changing out every 3–5 years (afpm.org) (afpm.org), even as industry reports show well‑managed reformers now running ~2-year campaigns and catalysts lasting up to 10 years (afpm.org).

With Indonesia tightening diesel specs to 500 ppm sulfur by 2021 and 50 ppm by 2025 (transportpolicy.net), the payoff from longer catalyst life is immediate.

Feed pretreatment and guard strategies

Effective pretreatment — removing or neutralizing poisons and particulates before the reactor — is the lever many operators pull first. Inorganics like salts, metals, and silicon compounds rapidly deactivate catalysts: sodium chloride or sodium naphthenate droplets (“salt cake”) can deposit on beds, while trace silicon from drilling additives or clay can irreversibly block active sites (digitalrefining.com) (digitalrefining.com). Even small Si levels shorten runs and can preclude regeneration (digitalrefining.com).

Filtration matters. Industry panels describe “the first phase” as feed filtration to ~25 μm to capture iron sulfides, sand, and metal-bearing fines (afpm.org) — “Again, 25 microns is the design target” (afpm.org). To implement this, operators commonly deploy industrial strainers for coarse solids control ahead of reactors, a use case for filter and ion exchange strainers. Where finer cuts are needed, refineries standardize on industrial cartridge elements; steel filter housings in high-pressure service, paired with cartridge filters, deliver the ~25 μm capture referenced by AFPM.

Top-bed “grading” layers then trap what slips through. High-void media are placed above the main catalyst to hold fines without forcing pressure drop; as one AFPM panel put it, “the goal is to dramatically increase the void fraction…top-bed materials with high internal void space…so that the contaminates can…pass through” (afpm.org). In practice, multilayer trains put inert porous filters and grade‑1 catalyst first, then the main active beds; this approach “gravitates metals into sacrificial layers.” Vendors cite wide deployment of adsorptive filtration systems (AFS): UNICAT notes alumina “AFS grading disks” in 500+ hydrotreater applications, including nickel‑promoted alumina versions to absorb silicon or vanadium (digitalrefining.com) (digitalrefining.com).

Dedicated “trap” catalysts target specific poisons. For silicon, industry guidance is blunt: “Silicon deactivation is best avoided,” but when it cannot be, “the next best solution is to use a suitably designed guard bed…The silicon trap in the guard bed will capture a large part of the silicon” (digitalrefining.com).

Processed water and salts are removed by demulsification and desalting. Multi-stage electrostatic desalters reduce salt (NaCl) to <5–10 ppm before atmospheric/vacuum distillation, preventing Na and Cl deposition that forms strong basic poisons or metal hydroxides. Designers often ensure heavy feeds contain <1 ppm Ni+V after pretreatment to avoid excessive fouling. Plants that doubled filtration fineness have recorded 20–30% longer runs with tens of thousands of dollars per year in catalyst savings.

Adoption is growing. A survey reports ~40% of FCC feed is pretreated (FCC context, but analogous), helping protect downstream hydrotreaters (researchgate.net). In heavy crudes, guard beds are typically replaced or regenerated every few years; effective programs extend main bed life by ~1.5–2×. Where pretreatment is lacking, heavy hydrotreater run lengths can fall under 1 year due to rapid Ni/V poisoning (ogst.ifpenergiesnouvelles.fr) (digitalrefining.com).

Reactor design and operating practice

Reactor configuration shapes how deactivation unfolds. Graded beds mitigate contamination by catching fines up front. In multi-reactor trains, some operators rely on short-circuiting or bypass lines when a bed clogs; “permutable reactors” architectures described in patents and practice let feed bypass an aged bed to avoid unscheduled shutdowns. Because fixed-bed hydrotreaters cannot swap catalyst on line (unlike continuous hydrocrackers), units often include spare space or “sock loading” — intentionally higher-void top layers so initial coke/polymer deposits do not force a shutdown (afpm.org).

Large hydrotreaters commonly use multiple beds with interstage cooling and feed injection to spread heat and deactivation. Case in point: an ADNOC heavy gas oil (HGO) hydrotreater stacked CoMo/NiMo/CoMo (cobalt‑molybdenum/nickel‑molybdenum) layers; the NiMo middle bed boosted hydrogenation of nitrogen compounds, reaching 10 ppm sulfur at lower temperature than single‑bed CoMo (ogst.ifpenergiesnouvelles.fr) (ogst.ifpenergiesnouvelles.fr). The study reports a significant reactor temperature reduction versus a full fresh CoMo load for the same yields (ogst.ifpenergiesnouvelles.fr), plus greater reuse of spent material (ogst.ifpenergiesnouvelles.fr).

Intermittent interventions help too. Semi-regenerative reformers routinely “skim” — remove the top spent layer — every 3–5 years to limit contamination zone growth in the lead reactor (afpm.org). With good regeneration practice, tail reactors in reformers have achieved up to 10‑year runs (afpm.org). Some hydrotreaters also use short or staggered cycles to regenerate portions of a charge without a full shutdown.

Data/outcomes: the ADNOC program combining reactor design and catalyst management achieved cycle length ≥4 years to ULSD (ogst.ifpenergiesnouvelles.fr). Pilot work found fully reactivated stacks performed “even better than full fresh” in selectivity (ogst.ifpenergiesnouvelles.fr), and reactor modifications like adding guard capacity or extra cooling have cut deactivation rates by ~10–20% in documented cases, translating to millions of dollars saved over a plant’s life.

Advanced formulations and reactivation

Support and promoter innovations are shifting the baseline. In HDT, “Type II” catalysts — CoMo/NiMo on γ‑Al₂O₃ with highly stacked MoS₂ phases — show “significantly superior activity” thanks to higher sulfidation and better promoter dispersion (ogst.ifpenergiesnouvelles.fr). Tailored supports (including Ce or Zr additions) can strengthen metal–support interactions and temper coke growth — test work cited a ~15% cycle boost with Ce‑doping.

Heavy oil hydroprocessing is pushing into zeolite–alumina hybrids (e.g., NiW on ultra‑large pore supports) to crack asphaltenes in situ and reduce coke precursors. Bifunctional catalysts that combine acidity with strong hydrogenation functionality have maintained activity ~20–30% longer on high‑V feeds in experimental evaluations.

Reforming catalysts (Pt or Pt‑Re on chlorinated alumina) are trending toward ultra‑fine Pt dispersions stabilized by Sn or Re promoters, with improved chlorination for uniform acidity. Operators report meaningful payoffs: “If you have a semi‑regenerator that used to get 6 months, now you might get 24 months” on a new generation, a 4× increase (afpm.org). Continuous catalytic reformers (CCR) run 8–13 year cycles on modern catalysts, typically limited by chlorine retention as surface area drops (afpm.org) (afpm.org), but higher initial surface area in new loads can defer the limit.

Regeneration and reuse are now mainstream levers. Spent HDT catalysts undergo controlled coke burn‑off and re‑sulfidation; reactivation programs chemically clean and restore metals. In the ADNOC case, stacks with 25–75% reactivated catalyst matched fresh performance in pilots (ogst.ifpenergiesnouvelles.fr), enabling 30–55% savings on the next catalyst load (ogst.ifpenergiesnouvelles.fr) (ogst.ifpenergiesnouvelles.fr). Avoiding disposal also matters, given spent catalyst landfill fees around ~5% of weight (ogst.ifpenergiesnouvelles.fr).

Quantitative outcomes: the ADNOC study concluded an optimized mix of graded reactivated NiMo/CoMo layers maintained ULSD while cutting catalyst spending by up to 55% per cycle (ogst.ifpenergiesnouvelles.fr) (ogst.ifpenergiesnouvelles.fr). In reforming, newer generations have been justified by 1–2% higher liquid yields (tens of millions of dollars per year on a large naphtha reformer). Across the industry, improved HDT catalyst life from ~6 months to ~48 months has been documented under ULSD mandates (ogst.ifpenergiesnouvelles.fr), underpinned by feed prep and new catalysts.

What the numbers say

The throughline is consistent: rigorous pretreatment (filtration to 25 μm, salt removal, dedicated adsorbers) can yield 20–30% longer runs (afpm.org) (digitalrefining.com), graded reactors keep pressure drop stable, and advanced catalysts achieve target conversions at lower ΔT — a recipe for slower aging (ogst.ifpenergiesnouvelles.fr) (ogst.ifpenergiesnouvelles.fr). Pilot and plant data track near‑linear relations between severity and life, translating to lower catalyst expenditures and extended uptime.

Figures and observations

Illustratively: a 2018 pilot found that using 75% reactivated CoMo instead of fresh cut reactor temperature needs significantly less (improving selectivity), enabling ~30–55% annual catalyst cost savings (ogst.ifpenergiesnouvelles.fr) (ogst.ifpenergiesnouvelles.fr). Modern reforming catalysts have pushed sulphur/arsenic tolerances into the ppb range, cutting downtime. The trajectory is clear: composition advances (promoters, supports) and process design (continuous regeneration, smart bypasses) are the R&D focus, with widening investment.

Sources

Authoritative industry studies, peer-reviewed reviews, and operator case studies inform the data cited here. Key references include deactivation reviews (sciencedirect.com), industry symposium analyses (ogst.ifpenergiesnouvelles.fr) (afpm.org), and regulatory summaries for Indonesia (transportpolicy.net). All numerical data and quotes are from these sources (see citations).

Inline references

ogst.ifpenergiesnouvelles.fr ogst.ifpenergiesnouvelles.fr (ADNOC case study on catalyst reuse/cost) ogst.ifpenergiesnouvelles.fr (catalyst disposal cost) afpm.org afpm.org (feed filtration and graded bed design) digitalrefining.com (AFS filtration system deployment) digitalrefining.com digitalrefining.com (silicon poisoning and guard beds) sciencedirect.com (2025 review on deactivation mechanisms) ogst.ifpenergiesnouvelles.fr ogst.ifpenergiesnouvelles.fr (ADNOC experimental results, NiMo layer effect) afpm.org afpm.org (AFPM on reformer cat life and poisons) transportpolicy.net transportpolicy.net (Indonesia fuel sulfur regs)