Estimating VOC Emissions from Fixed-Roof Storage Tanks Using AP-42 Section 7.1: A Senior Chemical Process Engineer’s Technical Guide

Introduction: Why VOC Emission Estimation Matters

Volatile organic compound (VOC) emissions from liquid storage tanks represent a significant source of air pollution in the chemical, petroleum refining, pharmaceutical, and coatings industries. Uncontrolled emissions contribute to ground-level ozone formation, smog, and adverse human health effects—and are tightly regulated under national and international environmental frameworks, including the U.S. EPA’s National Emission Standards for Hazardous Air Pollutants (NESHAP), New Source Performance Standards (NSPS) Subpart Kb, and the EU Industrial Emissions Directive (IED). Accurate estimation is not merely a compliance checkbox: it informs capital decisions on emission controls (e.g., internal/external floating roofs, vapor recovery units, or nitrogen blankets), supports life-cycle cost analysis, enables credible greenhouse gas (GHG) and carbon footprint reporting, and underpins environmental risk assessments.

The U.S. Environmental Protection Agency’s Compilation of Air Pollutant Emission Factors (AP-42), specifically Chapter 7.1 — “Storage Tanks”, provides empirically derived, widely accepted methodologies for estimating VOC emissions from various tank configurations. While Sections 7.1.2–7.1.5 address breathing, working, and flashing losses for fixed-roof and floating-roof tanks, Section 7.1.1—titled “Emissions from Fixed-Roof Tanks”—introduces the foundational mass-transfer-based approach for estimating evaporation losses from the exposed liquid surface in non-floating-roof tanks. This guide focuses exclusively on that methodology, demystifying its theoretical basis, correct implementation, regulatory context, pitfalls, and practical application.

Theoretical Basis and Equation Derivation

AP-42 Section 7.1.1 does not prescribe a single universal equation. Instead, it presents two complementary approaches depending on whether the stored liquid is water-miscible (e.g., methanol, acetone, ethanol) or water-immiscible (e.g., benzene, toluene, xylene, gasoline). The core principle is mass transfer across the liquid–vapor interface, governed by Fick’s law and modified by vapor-phase resistance and equilibrium thermodynamics.

For Water-Immiscible Liquids (Most Common Case)

The primary emission rate equation cited in AP-42 Section 7.1.1 is:

$$ E = K \cdot A \cdot \frac{P_v}{P_{\text{atm}}} $$

Where:

• $E$ = VOC emission rate (kg/hr)
• $K$ = Tank emission factor, with units kg/(m²·hr) — not a dimensionless constant. This factor empirically encapsulates temperature-dependent vapor-phase mass transfer coefficients, tank geometry effects (e.g., height-to-diameter ratio), and ventilation characteristics (natural convection, wind exposure). Per AP-42 Table 7.1-1, typical values range from 0.1 to 1.0 kg/(m²·hr) for ambient-temperature, vented, fixed-roof tanks storing hydrocarbons. The default value of 0.5 kg/(m²·hr) in the tool reflects a mid-range conservative estimate for moderately ventilated tanks.
• $A$ = Surface area of the liquid (m²) — not tank cross-sectional area, but the actual projected area of the liquid–air interface. For cylindrical tanks, this equals $\pi \cdot (D/2)^2$, where $D$ is the tank diameter. Critical note: $A$ must be the average surface area over the operating level range if inventory fluctuates significantly.
• $P_v$ = True vapor pressure of the liquid at the average liquid temperature (kPa) — not Reid Vapor Pressure (RVP) or bubble point. True vapor pressure is the saturation pressure exerted by the pure component (or mixture) in equilibrium with its liquid phase. AP-42 explicitly states (Section 7.1, p. 7.1-2): “True vapor pressure should be used… RVP is not appropriate for emission estimation.”
• $P_{\text{atm}}$ = Local atmospheric pressure (kPa) — typically ~101.3 kPa at sea level. While often omitted in simplified versions (assuming $P_v \ll P_{\text{atm}}$), AP-42 retains it for rigor, especially at high elevations or for very volatile compounds.

This equation is essentially an empirical adaptation of the basic evaporation flux model $E = k_g \cdot (C^_v - C_v)$, where $k_g$ is the gas-phase mass transfer coefficient and $C^_v$ is the equilibrium vapor concentration ($= P_v / RT$). The factor $K$ subsumes $k_g$, $RT$, and geometric scaling.

For Water-Miscible Liquids (e.g., Alcohols, Ketones)

When the VOC partitions significantly into the aqueous phase (e.g., wastewater tanks containing dissolved solvents), AP-42 recommends using a Henry’s Law-based approach, recognizing that the effective driving force is reduced by dissolution. The emission rate is estimated as:

$$ E = \frac{K_H \cdot A \cdot H \cdot P_v}{H \cdot P_v + K_H \cdot P_{\text{atm}}} $$

However, AP-42 notes this form is rarely used in practice due to data scarcity and complexity. Instead, the document (Section 7.1.1, p. 7.1-3) endorses a simplified, widely adopted approximation:

$$ E = K \cdot A \cdot \frac{P_v}{P_{\text{atm}}} \cdot \left(1 + \frac{H}{K_H}\right)^{-1} $$

Where:

• $H$ = Henry’s Law constant for the chemical in water at the average liquid temperature (unit: m³/kmol) — this is the value input as henrys_law_constant. Note: AP-42 uses the dimensionless form $H^{cp}$ in some tables, but the tool’s unit (m³/kmol) aligns with the common definition $H = P / C_{\text{aq}}$, where $C_{\text{aq}}$ is aqueous-phase concentration (kmol/m³).
• $K_H$ = A characteristic constant representing the liquid-phase mass transfer resistance, with typical values of 100–1000 m³/kmol for well-mixed aqueous systems. The tool’s default of 100 m³/kmol assumes moderate mixing and low viscosity.

Crucially, AP-42 emphasizes that this correction is only meaningful when $H$ is small (i.e., high solubility, strong partitioning into water). For example, methanol ($H \approx 4.2$ m³/kmol) would see a large reduction, while benzene ($H \approx 2900$ m³/kmol) would be virtually unaffected.

The height_above_liquid input (average ullage height) is not part of the core AP-42 7.1.1 equations. Its inclusion in the tool reflects an engineering judgment refinement: taller ullage volumes dilute the vapor-phase concentration, reducing the effective driving force for diffusion. While not codified in AP-42, this is consistent with first-principles modeling (e.g., using the stagnant-film model with a defined diffusion path length) and is recommended in supplemental guidance like EPA’s TANKS software documentation.

Regulatory Context and Standard Requirements

AP-42 is not a regulation itself—it is a technical reference document published by the U.S. EPA to support emission inventory development and permitting. However, its methodologies are mandated or strongly endorsed in several regulatory contexts:

• 40 CFR Part 60, Subpart Kb (Standards of Performance for Volatile Organic Liquid Storage Vessels): Requires use of AP-42 Chapter 7.1 methods (or equivalent) for determining whether a tank is subject to control requirements based on calculated potential emissions.
• 40 CFR Part 63, Subpart GGG (National Emission Standards for Hazardous Air Pollutants for Group I Polymers and Resins) and similar MACT standards: Frequently require AP-42-compliant calculations for initial compliance demonstrations and periodic performance tests.
• State Implementation Plans (SIPs): Many state agencies (e.g., Texas Commission on Environmental Quality, California Air Resources Board) explicitly require AP-42 Section 7.1 for permit applications involving storage tanks.

Key clauses from AP-42 Chapter 7.1 (5th ed., updated Jan 2023) include:

• Section 7.1.1, p. 7.1-2: “The emission factor approach… is applicable to fixed-roof tanks storing liquids with true vapor pressures greater than 0.7 kPa (0.1 psi) at the maximum liquid storage temperature.”
• Section 7.1.1, p. 7.1-3: “For miscible liquids, the emission factor should be adjusted downward using Henry’s law constants… If reliable Henry’s law data are unavailable, assume no reduction (i.e., use the immiscible equation).”
• Section 7.1.1, p. 7.1-4: “The tank emission factor $K$ is highly dependent on local meteorological conditions and tank design… Site-specific measurement or modeling is preferred when $K$ is critical to compliance.”

Failure to adhere to these stipulations—e.g., using RVP instead of true vapor pressure, neglecting temperature dependence, or misapplying the Henry’s law correction—can invalidate a permit application or trigger enforcement action.

Common Mistakes and How to Avoid Them

1. Using Reid Vapor Pressure (RVP) Instead of True Vapor Pressure
   Mistake: Inputting RVP (a standardized 100°F test) for $P_v$.
   Consequence: Systematic overestimation (RVP ≈ 1.5–2× true $P_v$ at 20°C for many hydrocarbons).
   Fix: Use Antoine equation, NIST Chemistry WebBook, or process simulation software (Aspen HYSYS, CHEMCAD) to calculate $P_v$ at the actual average liquid temperature (e.g., 35°C for a Gulf Coast refinery tank).

2. Misinterpreting the Tank Emission Factor $K$
   Mistake: Assuming $K = 0.5$ is universally valid, or treating it as a fundamental property.
   Consequence: Errors of ±200% in $E$.
   Fix: Consult AP-42 Table 7.1-1 for tank-specific guidance (e.g., $K = 0.15$ for unvented, insulated tanks; $K = 0.8$ for poorly sealed, windy locations). Calibrate $K$ using site-specific monitoring data if emissions are high-risk.

3. Incorrect Henry’s Law Constant Application
   Mistake: Applying the Henry’s correction to immiscible liquids (e.g., crude oil) or using $H$ values at 25°C for a 60°C wastewater tank.
   Consequence: Gross underestimation (for immiscible) or overestimation (if $H$ decreases with temperature but a low-T value is used).
   Fix: Verify miscibility first. Use temperature-corrected $H$ (e.g., van’t Hoff equation) or source data at the actual operating temperature.

4. Neglecting Average Operating Conditions
   Mistake: Using maximum tank diameter for $A$ or summer-maximum $P_v$ without weighting by residence time.
   Consequence: Overly conservative (and costly) control decisions.
   Fix: Calculate $A$ and $P_v$ as time-weighted averages over the annual operating cycle.

5. Ignoring Ullage Height Effects
   Mistake: Omitting $h$ (height above liquid) in high-ullage scenarios (e.g., large spherical tanks operating at 20% capacity).
   Consequence: Up to 3–5× overestimation of $E$ due to vapor-phase dilution.
   Fix: Incorporate $h$ via the inverse-square relationship in advanced models or use the tool’s built-in adjustment logic.

Worked Example: Realistic Petrochemical Application

Scenario: A fixed-roof, carbon steel storage tank at a Houston refinery holds 50,000 bbl of unstabilized naphtha (true vapor pressure highly temperature-sensitive). Engineering data:

• Average liquid temperature: 42°C → $P_v = 50$ kPa (from lab measurement & Antoine fit)
• Tank diameter: 30 m → $A = \pi \cdot (15)^2 = 707$ m²
• Average ullage height: $h = 10$ m (tank height = 15 m, average liquid level = 5 m)
• Tank is vented, uninsulated, located in open field → $K = 0.65$ kg/(m²·hr) (conservative value from AP-42 Table 7.1-1, upper quartile)
• Naphtha is water-immiscible → Henry’s correction not applied ($H$ irrelevant)
• Local $P_{\text{atm}} = 100.8$ kPa (Houston elevation ~15 m)

Calculation: $$ E = K \cdot A \cdot \frac{P_v}{P_{\text{atm}}} = 0.65 \cdot 707 \cdot \frac{50}{100.8} $$ $$ E = 0.65 \cdot 707 \cdot 0.496 \approx 0.65 \cdot 350.7 \approx \mathbf{228.0} \text{ kg/hr} $$

Interpretation: This emission rate (~2,000 metric tons/year) exceeds NSPS Kb thresholds, triggering mandatory control (e.g., external floating roof or vapor recovery). A sensitivity analysis shows that reducing $K$ to 0.45 (via improved tank sealing) lowers $E$ to 159 kg/hr—a 30% reduction justifying maintenance investment.

Tool Validation: Inputting $K=0.65$, $P_v=50$, $A=707$, $H=\text{N/A}$, $h=10$ yields $E \approx 228.0$ kg/hr — confirming the tool’s core calculation aligns with AP-42 theory.

Conclusion

Estimating VOC emissions from fixed-roof tanks using AP-42 Section 7.1.1 is a deceptively simple yet profoundly consequential engineering task. It bridges fundamental physical chemistry (vapor–liquid equilibrium, mass transfer), empirical observation (the $K$ factor), and regulatory pragmatism. Success demands rigorous attention to data quality—especially true vapor pressure and tank-specific emission factors—and a clear understanding of the underlying assumptions. As industry shifts toward predictive emissions management and digital twin technologies, mastering these foundational methods remains essential. Always remember: AP-42 provides a starting point, not an endpoint. When emissions impact is high, supplement with direct measurement (EPA Method 21), dispersion modeling, or dynamic simulation to close the uncertainty gap.