Emission Assessment

8/7/2019 Emission Assessment

1/17

1

Development of anEmissions Assessment

BMP for Storage TanksPTACAir Issues Forum

CalgarySeptember 30, 2009By D. Colley


2/17


3/17


4/17

Potential Emissions Flashing, breathing and working losses.

Flashing losses occur primarily at production facilities: In-tank boil-off of dissolved gas absorbed by the liquid

product during contact with natural gas at elevatedpressures in separators and other vessels.

Breathing and working losses are most important inapplications involving storage of stabilized orweathered product: Breathing losses are attributed to diurnal temperature

changes that cause expansion and contraction of the gasvolume in the vapor space.

Working losses result from physical displacement ofaccumulated evaporation and evaporation of productresidue on the tank walls above the liquid or floatingroof.


5/17

Potential Emissions Unintentional gas carry-through to storage

tanks: Leakage past the seats of drain and dump valves. Malfunctioning level controllers.

Inefficient upstream gas/liquid separation.

Operational changes resulting in unstabilized liquid product.

Malfunctioning vapor control or recoverysystems: Faulty blanket gas regulators or pressure controllers.

Leaking pressure-vacuum valves and thief hatches.

Undersized of systems due to neglect of diurnal temperature effects. Fouled vapor collection lines.

Damaged floating roofs or seals.

Inconsistently applied Directed Inspection & Maintenance (DI&M).


6/17

Storage Tank Emissions

Field measurement results for 9 gas plants.


7/17

Needs

A protocol for assessing flashing losses fromproduction tanks. Significantly different results depending on the

assumptions made when applying E&P TANKS or processsimulators.

TVP weathers down to values less than local atmosphericpressure but flash calculations are often only taken toatmospheric pressure.

Lack of metering to show liquid flow rates from eachseparator and scrubber at a facility:

Assuming all liquids come from the lowest pressuresource may underestimate emissions.

Assuming all liquids come from the highest pressuresource may overstate emissions.


8/17

Needs

Instantaneous emissions rates, not average values,

to assess potential air quality impacts. US EPA TANKS is intended for predicting annual average

emissions for inventorying purposes, not instantaneousemissions for VRU sizing or air quality predictions.

A database to predict emissions of traceconstituents, such as reduced sulphur compounds,CH4 and other compounds, that do not show up in

typical liquid analyses but are of an air quality oremissions concern. Generally increasing levels of H2S and mercaptans in crude

oil streams are resulting in concerns for odour issues.


9/17

Needs

A model that accounts for continuous convective

breathing losses that occur during cold weatherperiods when the ambient air is colder than thetemperature of the stored product and the ullagespace (not currently accounted for in US EPA

TANKS) : Stratification in the vapor space tends to inhibit

evaporation losses during summer period.

Continuous convective mixing with ambient air duringthe winter significantly enhances breathing losses: Dense (cold) ambient air continuously flows into the

ullage spaced displacing the lighter (warm) air-vapormixture and this action promotes further evaporation dueto the dilution effect.


10/17

Needs

Guidance on measurement techniques that can be

used to verify predicted emission estimates. Methods and related guidance to regularly monitor

tanks for emissions upsets or deterioratedemissions control systems (especially for floating

roof tanks): Downwind air quality monitoring tends to identify

problems after the fact and does not necessary pinpointthe source.

Headspace monitoring not widely used. DIAL is used in Europe but at infrequent intervals due to

the high expensive (e.g., once every 5 years).

Roof-top vapor sensors are usually used only after issuesarise.


11/17


12/17

Work Done to Date Initial Literature Review

Estimation and measurement

techniques. Wind effects.

Initiated sampling program: Crude Oil:

Physical properties (density, RVP). Detailed liquid compositions (C1 to C30+

and sulphur content).

Vapors: C1 to C12+ and reduced sulphur

compounds (GC/TCD, GC/FID, GC/MS

and GC/SCD). Measurement of vertical concentration

profiles in the ullage space to examinestratification and mixing.


13/17

Tank Fluid StratificationTank Fluid E xam ple 1 (Liquid Analyses )

0

20000

40000

60000

80000

100000

120000

140000

160000

Input

Bottom

Middle

Top

Conce

ntration(ppm)

Isobutane (iC4)

Butane (nC4)

Isope ntane (iC5)

Pentane (nC5)

Hexane (C6)

Heptane (C7)

Methylcyc lohexane (C8)

Octane (C8)

Nonane (C9)


14/17

Ullage Space StratificationUllageSpaceExample1(GasAnalyses)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

Halfwaytothetop Topoffloatingroof

ppm

Hexane(C6)

Cyclohexane(C6)

Heptane(C7)

Methylcyclohexane

(C8)

Toluene(C8)

Nonane(C9)

oXylene(C9)


15/17

Emissions During Roof FloatingVOC Data

0

25

50

75

100

125

150

175

200

8/15/2009 12:43 8/15/2009 13:12 8/15/2009 13:40 8/15/2009 14:09 8/15/2009 14:38 8/15/2009 15:07 8/15/2009 15:36

VOCasIso

butleneConcent

ration

m

Min (ppm)

Avg (ppm)

Max (ppm)


16/17

Next Steps Continue testing program to examine:

Flashing loss issues.Thermally enhanced emission issues.

Floating roof emission behavior.

Floating roof landing and re-floatingemission characteristics and issues roofs.

Development and testing ofmonitoring systems.


17/17

Emission Assessment

Documents

Transcript of Emission Assessment