Hydrocarbon leaks into cooling water don’t just foul exchangers — they strip into the air, burn biocide, and can push towers toward their lower explosive limit. Case studies show online analyzers cut losses by 90%+ and trigger fixes within minutes.
Refinery cooling towers are designed to shed heat, not hydrocarbons. Yet once process fluid seeps across a failed tube, the water circuit becomes a conveyor for volatile organic compounds (VOCs, gas‑forming organics) and food for microbes. Biofouling — microbial slime that can be up to 4× more insulating than mineral scale — rapidly blankets exchanger surfaces (Chemical Engineering), forcing higher dosing and driving down heat‑transfer. Worse, a refinery case found that ~99% of leaked VOCs in circulating cooling water were stripped into the atmosphere (Hydrocarbon Engineering), and towers in extreme events have reached their LEL (lower explosive limit, concentration at which a vapor can ignite) and exploded (Hydrocarbon Engineering).
Product loss is the quieter headline. One 330,000 bpd refinery, running without VOC monitors, was losing ~1,600 lb/day of hydrocarbons (≈700 ppbw, parts per billion by weight) in its cooling loop — roughly 250 metric tons/year. After installing monitoring, levels fell to ≤84 ppbw, a >90% reduction (Hydrocarbon Engineering). On the compliance side, any hydrocarbon present in blowdown inflates COD (chemical oxygen demand) and BODC (biodegradable oxygen demand) — risking limits under national standards such as Indonesia’s Permen LH No. 5/2014 (peraturan.bpk.go.id) — while light fractions vent as VOC/HAPs (hazardous air pollutants) (ResearchGate).
Operating indicators and early warning
Front‑line signals in the basin or return headers remain surprisingly telling: an oily sheen or rainbow interference film, a hydrocarbon odor, gas bubbling from dissolved light ends, a sudden pH drop (often linked to H₂S generation), and a precipitous fall in free‑chlorine residual (Chemical Engineering) (Chemical Engineering). Heavy oil ingress drives turbidity (cloudiness) sharply higher — >300 NTU (nephelometric turbidity units) has been observed — and bacterial counts spike, with SRB (sulfate‑reducing bacteria) at 10³–10⁴ cells/mL and total bacteria >10⁵/mL (Chemical Engineering) (Chemical Engineering).
Digital instrumentation tightens the net. ORP (oxidation–reduction potential, a millivolt indicator of oxidizing power) dips are a red flag; values below ~450 mV at return headers correlate with foul‑up and potential leaks (Chemical Engineering).
Online hydrocarbon analyzers in practice
Modern plants deploy continuous oil/VOC‑in‑water analyzers based on UV‑induced fluorescence or infrared spectroscopy. Laser‑induced fluorescence monitors emit UV and detect hydrocarbon‑specific fluorescence (Boqi Instrument), differentiating oil from turbidity or natural organics. Commercial units — e.g., Turner Designs TD‑120 or Keco 204 PermaStream — strip a sample through a membrane and sensor or use fiber optics, reporting oil in water down to parts‑per‑billion.
These online monitors outperform infrequent grab tests that can take days — long enough for serious fouling to develop — while continuous devices alarm within minutes of a few‑ppbw leak (Benchmark Measurement Solutions). Fluorescence‑based systems are notably insensitive to cooling‑water turbidity, maintaining reliable low‑level detection (Benchmark Measurement Solutions).
Alarm thresholds and action levels
Chemical and permit programs set alarms far below safety limits. One Texas permit calibrated monitoring to ~0.015 ppmw (15 ppbw) for specific aromatics (ResearchGate), with daily sampling for benzene at 0.013 ppmw (13 ppbw) detection (ResearchGate). Under that permit, any confirmed benzene >13 ppbw for five days triggered repairs within 45 days (ResearchGate) (ResearchGate). Upstream operations often target <5 ppbw benzene or total VOCs in recirculating water (Hydrocarbon Engineering). Trending ppbw data against unit operations helps localize a leaking exchanger train.
Emergency isolation and safety measures
When an analyzer or lab result hits an action limit, the operating response is immediate: isolate the suspected exchanger or loop by bypassing or valving off the unit, even if it forces a partial process slowdown. This stops further contamination and reduces fire risk (Chemical Engineering). Notify operations and HSE, secure ignition sources near the tower, and ensure drift curtains or VOC abatement systems are active. If combustible vapor is feeding the tower, safety assessments can support an intentional blowdown or train shutdown to avoid reaching LEL (Hydrocarbon Engineering).
Adjusting the chemical treatment program is part of the first wave of actions, including biocide control and dispersant dosing; utilities teams typically frame this under a defined cooling tower program (cooling tower chemicals).
Leak localization and confirmation
Diagnosis proceeds via side‑by‑side readings: compare sump samples and header measurements on the supply and return of each unit to pinpoint the offender. A sharp ORP drop — >50–100 mV below normal — at an exchanger exit is a strong indicator (Chemical Engineering). Confirm with grab samples or quick GC (gas chromatography) tests for the specific hydrocarbons, then schedule immediate repair or tube replacement.
Chemical controls during containment
While mechanics mobilize, chemistry buys time:
- Boost oxidizing biocide (e.g., chlorine or chlorine dioxide) above normal to suppress biofouling; under leak conditions, chlorine residuals have fallen to zero within days (Chemical Engineering). Stronger oxidants like bromine or ozone are also used (Chemical Engineering). Targeted programs often source dedicated biocides.
- Apply a non‑oxidizing biocide “shock” (e.g., glutaraldehyde, quaternary ammonium compounds) to attack sessile biofilm organisms that resist oxidizers.
- Add a biodispersant (biosurfactant) at high concentration to emulsify hydrocarbon fouling and detach biofilms; dispersants “faster disengage organisms from the surface so that biocides act effectively” (Chemical Engineering). Many suppliers recommend ≈1000+ ppm in an event versus ~200 ppm baseline.
- Execute a controlled blowdown or sump overflow to purge dissolved oil, biomass, and froth; slowly draining to waste “eliminate[s] oil, biomass and froth from the sump” (Chemical Engineering). Balance blowdown with makeup to maintain circulation.
Maintaining consistent feeds under upset conditions benefits from precise metering; utilities commonly rely on accurate injection hardware to control rates, such as an automated dosing pump. When dispersing biofilms and hydrocarbon load, specialty agents are typically sourced as dispersant chemicals.
Repair, restart, and documentation
Mechanical repair is expedited — industry targets “as soon as practical, not later than ~45 days” after leak identification (ResearchGate). Pressure‑test and tightness‑test the exchanger, then clean or replace filters. A chemical/acid flush may be considered for the affected loop before refill and re‑chlorination. Many operators bring in a focused cooling tower cleaning service to remove residual deposits.
Throughout, document sensor readings, actions, and timing. If blowdown is routed to wastewater, ensure treatment or segregation meets regulations; in Indonesia, industrial cooling‑water discharges must comply with Permen LH No. 5/2014 on effluent quality (peraturan.bpk.go.id). Incidents are integrated into HSE reporting (e.g., a spill or UPDL report), and preventive maintenance is adjusted — including replacing flanged exchangers with welded types where feasible, as some permits now require (ResearchGate).
Return on monitoring and response
Real‑time analyzers deliver measurable ROI by limiting losses and downtime. In one case study, continuous monitoring capped VOCs at ~84 ppbw versus ~700 ppbw when unmonitored, cutting leak volumes by roughly an order of magnitude (Hydrocarbon Engineering). Operators reacted within 30 minutes of a 2 ppbw benzene spike, isolated and repaired the leak, and saw levels return to baseline in ~1.5 hours (Hydrocarbon Engineering).
Vendors report such systems “pay for themselves in saved product and labor” (Hydrocarbon Engineering) (Benchmark Measurement Solutions). Continuous trends also verify stable post‑repair conditions; for example, no residual tail leak at 0.5 ppbw was noted after fixes (Hydrocarbon Engineering).
Summary of key actions
- Monitor continuously: install online oil‑in‑water or VOC analyzers on recirculation returns — fluorescence/IR monitors are recommended (Benchmark Measurement Solutions) (Boqi Instrument).
- Set low thresholds: alarm at a few ppbw of total oil or target VOC to catch leaks early (e.g., 10–15 ppbw, comparable to strict permit LODs) (ResearchGate) (ResearchGate).
- Isolate leaks: on any positive alarm, immediately valve off the affected exchanger(s) (Chemical Engineering).
- Adjust treatment: increase biocides and shock‑dose biodispersant/biocide to counter oil‑fuelled biofouling (Chemical Engineering); perform controlled blowdown/flush to remove contaminants (Chemical Engineering). Where on‑spec feeds matter, teams lean on a dedicated dosing pump.
- Fix equipment, verify, document: expedite mechanical repair, pressure‑test, clean, and maintain enhanced monitoring; document and report per permit requirements. When scouring residual deposits post‑event, operators often call in a cooling tower cleaning service.
Sources and regulatory context
Authoritative industry and regulatory references inform each step above. For example, one refinery study reports a 1600 lb/day VOC leak (~700 ppbw) without monitoring versus ~84 ppbw with monitoring (Hydrocarbon Engineering). Government and expert guidance (e.g., US permit case studies) set action levels at ~13–15 ppbw of benzene (ResearchGate) (ResearchGate). Chemical‑treatment literature outlines the efficacy of oxidizers, biocides, and biosurfactants (“biodispersants”) in combating hydrocarbon‑induced fouling (Chemical Engineering) (Chemical Engineering). These and other studies provide the quantitative basis for the thresholds, timelines, and mitigation tactics recommended above.
Source notes and metadata
- Analytical Systems International. “Hydrocarbon analysers.” Hydrocarbon Engineering (Feb 2015) (Hydrocarbon Engineering).
- Ghosal, S. (Indian Oil Corp.). “Caring for Cooling Water Systems.” Chemical Engineering (Feb 15 2008) (Chemical Engineering) (Chemical Engineering).
- Hile, A.C. III, Lai, L., Kolmetz, K., Walker, J. “Cooling Tower Monitoring and Environmental Compliance.” Paper, AIChE Spring 2000 (Atlanta) (ResearchGate) (ResearchGate).
- Benchmark Measurement Solutions (Turner Designs). “Online Monitors for Oil Leaks in Cooling Water.” Industry Application Note (Feb 26 2021) (Benchmark Measurement Solutions).
- Boqi Instrument. “Real‑Time Oil in Water Monitoring for Industrial Cooling Towers.” (web article, Feb 19 2024) (Boqi Instrument).
- Kementerian Lingkungan Hidup RI. Peraturan Menteri LH No. 5 Tahun 2014 tentang Baku Mutu Air Limbah. (Jakarta, Oct 15 2014) (peraturan.bpk.go.id).
