Calculator

Glycol dehydrator emissions

Methane and VOC from a TEG dehydration unit, estimated from a validated process model rather than a single still-vent factor. Enter what your unit actually runs at and see how much of the gas ends up in each vent.

Model last reviewed

The numbers, in one place

Quantity Value Source
Circulation ratio drives emissions 9x Vented methane from 1 to 7 gal/lb, this model
Routing flash gas to control 95% cut Same unit, flash tank not vented, this model
Gas-assisted pump share of dehydrator emissions 90.25% Mdigo et al. 2026, uncontrolled flash tanks, AMI 2024 field data
Gas-assisted pump share of site emissions 63.1% Mdigo et al. 2026, same units
Units running gas-driven pumps 31% 54 units reported under AMI 2024
Model accuracy, methane 0.00091 RMSE Against held-out process simulations, Table 2

Your dehydrator

calibrated 400 to 1200
calibrated 77 to 122
calibrated 1 to 7, EPA Natural Gas STAR recommends 3 to 5

Where the gas goes

Wet gas in 2,756 kg CH4/h
Dehydrator contactor, flash tank, reboiler
Dry gas to sales 2,755 product
Flash tank 0.344 vented
Still vent 0.019 vented

Methane, in kilograms an hour. Every stream is mass, not volume.

Methane vented 0.363 kg/h
Methane a year 3.18 tonnes
VOC vented 5.07 kg/h
Total hydrocarbon 5.79 kg/h

By species

Methane is the smaller part by mass. Heavier hydrocarbons are far more soluble in glycol, so a dehydrator vents more propane and heavier than it does methane.

Species Flash tank Still vent Vented t/yr vented
Methane 0.344 0.019 0.363 3.18
Ethane 0.293 0.065 0.358 3.13
Propane 0.293 0.133 0.426 3.73
Isobutane 0.058 0.036 0.094 0.823
n-Butane 0.166 0.162 0.328 2.88
Isopentane 0.046 0.073 0.119 1.04
n-Pentane 0.048 0.093 0.141 1.24
Hexane 0.066 0.238 0.305 2.67
Heptane plus 0.182 1.33 1.51 13.2
Octane plus 0.096 1.23 1.32 11.6
Nonane plus 0.033 0.785 0.818 7.16
Carbon dioxide 0.182 0.134 0.316 2.76
Nitrogen 0.013 0.00017 0.013 0.117
Hydrogen sulfide 0.0018 0.0068 0.0086 0.075

Kilograms an hour unless stated.

What this does not cover

Gas-assisted glycol pumps are not in this model
A gas-driven pump uses wet gas as its motive force and vents it. That gas is not absorbed in the contactor, so the absorption regressions do not see it. In the field study, pump emissions were 90.25 percent of total dehydrator emissions at units with uncontrolled flash tanks, and 63.10 percent of site-level emissions. If your unit has a gas-assisted pump, the number here is a floor and not a total.
Outside the calibrated range the model extrapolates
The regressions were fitted for circulation ratios of 1 to 7 gal/lb, 77 to 122 degrees F and 400 to 1200 psia. The paper reports facilities running at 18, 31 and 71 gal/lb, well outside that range, where the model underpredicts because absorption becomes increasingly non-linear at high solvent flow.
It was fitted on one gas composition
The regressions were characterised against a single inlet composition, held constant so the effects of circulation, pressure and temperature could be isolated. Heavier hydrocarbon and acid gas content change absorption, and the authors name composition sensitivity as a limitation and a direction for further work.
Reboiler combustion is excluded
The model does not cover carbon dioxide from burning fuel in the reboiler, or methane slip from that burner. Those are real emissions from a dehydrator and are not counted here.
Steady state only
Failure modes and upset conditions are not modelled. The 2024 field campaign observed a dehydrator failure emitting 138 kg an hour, which no steady-state model predicts.

Why a dehydrator emits anything at all

Wet gas meets lean triethylene glycol in the contactor tower, and the glycol absorbs the water. It also absorbs hydrocarbons, which is the part nobody wants. That rich glycol then goes to a reboiler to be dried out and reused, and everything it absorbed comes back out on the way, either in a flash tank if the unit has one or up the still column with the steam.

So the emission is not a leak and not a malfunction. It is dissolved gas being released exactly where the process is designed to release it, and the size of it depends on how much glycol you circulate, at what pressure, and at what temperature. Those are the three variables this model is fitted across.

What each outlet is

Dry gas to sales
The dehydrated product stream leaving the top of the contactor tower.
Flash tank
Gas released from the rich glycol when it is depressurised, typically to 40 to 100 psig. A flash tank can recover up to 90 percent of the absorbed gas, which can be routed to a control device or reused as reboiler fuel instead of vented.
Still vent
Hydrocarbons released with the steam in the still column as the rich glycol is heated and the water is boiled off.

Circulation rate is the lever

Everything else held constant, this is what circulation ratio does to a 5 MMscf/d unit at 800 psia and 95 degrees F, with an uncontrolled flash tank.

Circulation ratio, gal/lb Methane vented, kg/h Methane, t/yr Total hydrocarbon, kg/h
1 0.102 0.895 0.739
3 0.363 3.18 5.79
5 0.624 5.47 10.8
7 0.884 7.75 15.6

Deliberately a few rows rather than a dense grid. Use the calculator above for your own operating point.

Over-circulation buys nothing. Past the point where the glycol is already drying the gas to specification, extra circulation absorbs extra hydrocarbons and then vents them. EPA Natural Gas STAR recommends 3 to 5 gal/lb. The field study behind this model found 40.7 percent of gas-assisted pump units running above 7 gal/lb, and some above 80, generally because a pump sized for peak production stayed in place as throughput declined.

The part this model does not see

A gas-assisted glycol pump uses high pressure wet gas as its motive force and vents that gas with the rich glycol. None of it goes through the contactor, so no absorption model predicts it. In the field deployment behind this work, that single omission was the difference between a simulation that badly undercounted a site and one that matched reported emissions.

Across the units reported under the AMI 2024 project, 31 percent used gas-driven pumps. For the 6 of those with uncontrolled flash tanks, pump emissions were 90.25 percent of total dehydrator emissions, ranging from 83.2 to 92.4 percent, and 63.1 percent of everything the site emitted.

That is why the calculator asks about your pump, and why it will tell you its own answer is a floor if you say the pump is gas-driven without giving it the numbers. Replacing a gas-assisted pump with an electric or instrument-air pump removes the source outright.

How the model was built and checked

Second-order polynomial regressions fitted to 117 ProMax scenarios: 13 glycol circulation ratios by 3 wet gas temperatures by 3 wet gas pressures, with the contactor modelled as three theoretical equilibrium stages. Species-level transfer regressions predict the share of each gas component that leaves by each outlet, and a second set predicts how the inlet gas volume splits across those outlets. Seventy percent of the simulation set was used for fitting and the remaining thirty percent held back for validation.

Model output RMSE against held-out simulations
Methane transfer ratio 0.00091
Worst light hydrocarbon (n-pentane) 0.02935
Dry gas flow fraction 0.002273
Flash tank flow fraction 0.001411
Still vent flow fraction 0.002056

Root mean square error against held-out ProMax results. Thirty percent of the simulation set was reserved from fitting and used for validation.

Averaged across every simulated case, 99.9557 percent of the inlet gas leaves as dry gas to sales, with 0.0286 percent to the flash tank and 0.0145 percent to the still vent. Those are small fractions of a large stream, which is exactly why they are easy to overlook and worth calculating.

This calculator runs the model on our server rather than in your browser. The fitted coefficients are not published, so they are not shipped to the page. Nothing you enter is stored.

Questions

How do you calculate glycol dehydrator emissions?

Emissions come from the hydrocarbons the glycol absorbs in the contactor and then releases downstream, at the flash tank and the still vent. That absorption depends on the glycol circulation ratio, the wet gas pressure and the wet gas temperature, so a single emission factor cannot capture it. This tool evaluates regressions fitted to process simulations across all three variables and splits the result across each outlet.

What is the biggest driver of dehydrator emissions?

Circulation rate. Holding everything else constant, going from 1 to 7 gal of glycol per lb of water raises vented methane by about 9 times. EPA Natural Gas STAR recommends 3 to 5 gal/lb, and over-circulation absorbs hydrocarbons that are then vented for no gain in water removal.

Does a flash tank reduce dehydrator emissions?

Only if the flash gas goes somewhere. The flash tank separates most of the absorbed gas out of the rich glycol before the reboiler, but if that gas is vented it is still an emission. Routing it to a control device or back to the sales line cuts vented methane by about 95 percent in the default case here.

Why does my dehydrator vent more VOC than methane?

Hydrocarbon solubility in triethylene glycol rises with molecular weight, so propane, butane and heavier absorb far more readily than methane does. By mass, the heavier fraction usually dominates what a dehydrator vents, which is why dehydrators matter for VOC and hazardous air pollutant permitting as much as for methane.

Does this include gas-assisted glycol pump emissions?

Not unless you enter your pump's figures. A gas-driven pump vents its motive gas, which the absorption model does not see. In the field study behind this model, pump emissions were 90.25 percent of dehydrator emissions at units with uncontrolled flash tanks. If you have a gas-assisted pump and leave those fields blank, treat the result as a floor.

Source

Mechanistic Modeling of TEG Dehydrator Emissions in Oil and Gas Industry
Mdigo, J.; Santos, A.; Duggan, G.; Vora, P.; Shonkwiler, K.; Zimmerle, D. Fuels 2026, 7(2), 21. Open access under CC BY. doi:10.3390/fuels7020021
The Energy Institute, Colorado State University.

Model metadata, operating envelope and validation figures in machine-readable form: /data/dehydrator-model.json

Modelling every unit on a site

This is one dehydrator with one gas. MAES runs the same model across every piece of equipment at a facility, with the composition changing as the gas moves through it, which is what OGMP 2.0 Level 4 and 5 reporting asks for.

See how MAES models a site