The hottest upgrades in cracking: furnaces and heat recovery that slash fuel

Cracking furnaces are the energy sink of the refinery — and that’s exactly why they’re the richest target. Modern fired heaters can approach ~90–95% thermal efficiency in theory; in practice, new furnaces with air preheating and heat recovery often reach ~90–92% (see Ipieca; Heatflux). Older designs hover around ~80%.

The delta is money. One revamp of a 45 MW heater raised overall efficiency from ~80% to >90% by adding economizer and superheater coils — cutting stack temperature from ~350 °C to <185 °C (Ipieca). Since flue‑gas heat dominates furnace losses, limiting stack losses — keeping flue‑gas temperature as low as safely possible above the acid dew point (the temperature where acidic species condense and attack equipment) — is the lever (Heatflux).

Best practice is brutally specific: tight combustion control (excess O₂ ≈2–3% by volume, i.e., oxygen above stoichiometric needs) and advanced burners to minimize excess air. Air preheating is a proven retrofit; using low‑grade heat (e.g., steam at 150–200 °C or hot quench water) to warm combustion air cuts firing rates. By 2020, industry reports counted over 300 ethylene (steam‑cracking) furnaces in China with combustion‑air preheaters, citing “economic benefit [and] emission reduction” (AIChE). In short, optimized burner and furnace designs can recover ~5–10% of fuel by themselves, while theoretically achievable efficiency tops out in the mid‑90% range (Heatflux; Ipieca).

Advanced furnace and burner design

Combustion‑air preheating is low drama, high impact: retrofits preheating burners with 150–200 °C steam or hot condensate have been widely applied (AIChE), reducing firing rates without extensive equipment changes. For projects that warm boiler feed or process water via flue‑gas economizers, operators typically coordinate supporting gear such as water‑treatment ancillaries within the steam/water loop.

Recuparation and recovery are the other front. Modern designs use self‑recuperative burners or in‑furnace air preheaters (e.g., tubular air preheaters atop convection sections) to approach 90%+ efficiency (Heatflux). Multiple convection passes and surface enhancements (fins/studs on tubes) drive heat transfer. Low‑NOx and staged burners (staged secondary air, flue‑gas recirculation) improve mixing and limit excess O₂, pushing efficiency higher; with optimized burners and controls, stack outlet energy can be driven to just above the acid dew point blind of ~10–15 °C, permitting ≈90–94% thermal efficiency (Heatflux; Ipieca).

Maintenance and operation are non‑negotiable. Frequent decoking and cleaning matter; fouling (e.g., from residual fuel oil) strips heat‑transfer performance. Periodic “swooping and swinging” of convection coils (soot‑blowing or acid‑cleaning) restores percentage points of efficiency. Automatic O₂ and draft controls keep the air/fuel ratio on target during load changes — and chemical cleaning routines often rely on metered delivery via a dosing pump to manage solvents safely.

The gains are not theoretical. After an air‑preheat retrofit, one operator reported a 35% reduction in operating fuel cost (with around 48% total energy reduction) by avoiding direct firing for heat (Springer). A blend of preheated air, low excess O₂, and enhanced convection (economizers) can cut cracking furnace fuel use by a few tens of percent in practice (Ipieca; Springer).

Heat‑recovery systems in cracking units

Recovering flue‑gas heat is the workhorse move. Many cracking units already deploy waste‑heat boilers (WHBs, boilers placed in hot flue streams) in the regenerator duct. AITESA describes an FCC (fluid catalytic cracking) regenerator WHB producing 32 ton/hr of steam at 430 °C (≈25 MW) — “the new boiler has a much higher efficiency” and it “reduced the emission … by not requiring [additional] steam generation…via fossil fuels” (AITESA; AITESA). In effect, virtually all coke‑burning heat is captured in steam or sent to the stack at lower temperature.

Furnace convection sections can be upgraded, too. Ipieca cites a case where adding extra tube coils (superheater/economizer rows) jumped a heater’s thermal efficiency from ~80% to >90%, dropping stack temperature from ≈350 °C to ~185 °C (Ipieca). Flue‑gas economizers (preheating boiler feed or process water) and air preheaters are commonly used techniques in fired heaters (Ipieca) and alone can recover ~5–15% of input fuel by lifting base‑fluid temperature.

Compact heat exchangers broaden the playbook. Spiral or welded‑plate exchangers resist fouling and handle tough services (e.g., crude preflash, heavy oil coolers). Alfa Laval reports their plate/spiral exchangers in refineries save ~54 TWh per year (13 MtCO₂) globally, and claims that if every refinery switched to high‑efficiency exchangers, basin‑wide heat use could drop ≈23% (Alfa Laval; Alfa Laval). Based on industry experience, ~50 plate exchangers are now installed in FCC slurry and fractionator‑overhead services (Refining Community). In one case study, replacing four shell‑and‑tube exchangers with a single plate exchanger trimmed 7 MW fuel‑gas firing and cut 14,600 t/yr CO₂ on a CDU — a concrete outcome applicable to hot cracked‑gas cooling as well (Alfa Laval).

Overall, advanced heat‑recovery hardware can halve external heating demand in cracking. Jarullah et al. (2019) found an FCC heat‑integration scheme cut cooling requirements ~48% compared to the unfired case (Springer), implying similar fuel savings. Practically, installing economizers and steam coils to capture regenerator flue or hot fractionator vapors often pays back in a few years on fuel alone. One design study showed ~35% cost saving and ~48% drop in CO₂ after full heat integration (Springer). Energy consultants echo that furnace efficiencies above 90% and typical opportunities >10% are reachable, subject to capital trade‑offs (Ipieca; Heatflux).

Cross‑unit integration and waste‑heat use

Cracking furnaces can tap waste heat from elsewhere in the refinery. Any hot stream — overhead vapors, regenerated solvent, flue from other units — can preheat the feed or generate power. One approach uses FCU (fluid catalytic unit) flue to preheat FCC feed; industry also routes regenerator heat to steam cycles. On preheat networks, adding an ORC (Organic Rankine Cycle, a power cycle using organic working fluids) yielded a payback of <2 years in one study (ACS Publications). Downstream, converting hot fractionator or vacuum tower vapors into district heating or power via ORC showed positive economics, with payback depending on scale (IJTech).

Integrated sites make this tangible. PETRONAS’ Pengerang refinery installed two 2,000‑ton modular waste‑heat boilers on its RFCC (resid FCC) unit to serve the downstream steam cracker (PETRONAS). System‑wide, pinch analysis (a method for matching hot and cold streams to minimize utility use) ties hydrocrackers, hydrogen reformers, LPG plants, and steam systems together. By exchanging heat across units, fuel in the cracking furnace can often be cut 20–50%; one FCC retrofit analysis demonstrated ~48% reduction in fuel (and CO₂) demand via internal heat‑exchange loops (Springer).

Outcome, cost, and documentation

The impact shows up fast in OPEX. Industry surveys consistently find process heaters dominate refinery energy use, and the cited case studies report double‑digit percentage drops in fuel use with corresponding CO₂ reductions (Springer; ACS Publications). Even for Indonesia, where regulators track processing costs closely, improved energy efficiency directly cuts “cracking cost” — the $10–13/bbl processing premium (ESDM).

Sources for the figures above include peer‑reviewed studies, industry case reports, and regulatory analyses. For example, Jarullah & Awad (2019) modeled an FCC unit and quantified ~48% energy savings via heat integration (Springer). Industry briefs (Alfa Laval, Ipieca) provide global energy figures and case outcomes (Alfa Laval; Ipieca). Conference papers and press releases document real installations — e.g., 308 steam‑cracking furnaces with air‑preheat retrofits in China (AIChE), a ~25 MW FCC waste‑heat boiler (AITESA), and the RFCC tie‑in at Pengerang (PETRONAS). All statements above are supported by the cited sources.