Sizing a scavenger vessel is a math problem before it is a purchasing decision. Get it wrong in one direction and the bed breaks through early, forcing frequent changeouts and putting operators at risk of H2S exposure during service. Get it wrong in the other direction and you have paid for steel, media, and footprint the well or gathering line never needed.
The procedure below follows a Cornell University Dairy Environmental Systems Program design guide, which draws its sizing equations from Arnold and Stewart’s Surface Production Operations, Volume 2, a standard reference in gas processing engineering (Cornell, Part 3B, 2017).
One caveat on transferring it: Cornell wrote the guide for biogas systems, which typically operate below 15 psi, and notes that the compressibility factor Z can usually be taken as 1 at those pressures. Wellhead and gathering-line applications often run well above that. The equations themselves come from a gas processing source and scale fine, but pull Z from an actual gas analysis rather than assuming it once pressure, CO2, or H2S content climbs.
What loading capacity actually means
Granular iron oxide and iron hydroxide media remove H2S through a chemical reaction, not simple physical adsorption. Cornell writes the iron sponge reaction in its hydrated form:
2Fe2O3·H2O + 6H2S → 2Fe2S3·H2O + 6H2O
Water is a product of the reaction, and the incoming gas stream is often already water-saturated, which reduces the need to add moisture to the bed (Cornell, Part 3A, 2017).
Because this is a chemical conversion, every batch of media has a finite capacity before it is spent, and that capacity is a function of how much reactive iron the media contains. Classic wood-chip iron sponge is sold on ferric oxide content, commonly around 15 pounds of Fe2O3 per bushel (Cornell, Part 3A, 2017). Engineered granular media differ, so use the published figures for the specific product you are sizing around. Ole Red Dog publishes ours on our product specifications sheet.
Rated capacity only materializes if the vessel gives the gas enough contact time to use it. A University of Maryland Extension case study makes the point concretely. A farm-built scrubber, a 55-gallon plastic drum charged with scrap iron and steel filings, averaged 3.0% H2S removal across the monitoring period. Residence time ranged from roughly 15 seconds in summer to 6 minutes in winter, against the 1 to 15 minutes the study notes commercially prepared iron oxide media require. The authors attribute the poor result to a combination of unsuitable media and inadequate contact time, concluding that media prepared for desulfurization plus a vessel sized for more than a minute of residence time would have substantially improved performance (University of Maryland Extension, 2018). That is a biogas application rather than a sour gas well, but the lesson carries directly.
The inputs that drive vessel size
Six numbers determine everything else:
- Gas flow rate (Qg, MMscfd)
- Inlet H2S concentration (ppmv)
- Operating temperature (°R)
- Operating pressure (psia)
- Compressibility factor (Z)
- Target interval between media changeouts (days)
Get any one wrong and every downstream calculation skews with it. Inlet H2S concentration is the usual culprit, so work from a current gas analysis rather than an old one.
The sizing procedure
Minimum diameter, velocity limit. Gas velocity through the bed should not exceed 10 ft/min, or contact with the media suffers.
d(min) = √(360 × Qg × T × Z / P)
where d is in inches, T in degrees Rankine, and P in psia.
Minimum diameter, deposition rate limit. H2S should not deposit onto the media faster than about 15 grains per minute per square foot of bed cross-section.
d(min) = √(5.34 × 10⁶ × Qg × MF)
where MF is the H2S mole fraction (ppm ÷ 1,000,000). Take whichever of the two minimum diameters is larger.
Maximum diameter, channeling limit. Too low a velocity lets gas bypass the media instead of contacting it.
d(max) = √(1,800 × Qg × T × Z / P)
Minimum bed height. For at least 60 seconds of contact time:
H ≥ 3,600 × Qg × T × Z / (P × d²)
with a 5-foot floor regardless of what the formula returns.
Changeout interval. This is where loading capacity enters the arithmetic. Building the vessel larger than the minimums extends the time between media changes:
tc = 3.14 × 10⁻⁸ × Fe × d² × H × e / (Qg × MF)
where tc is in days, Fe is pounds of Fe2O3 per bushel of media, and e is reaction efficiency, typically 0.6 to 0.8. Media volume follows from:
Bushels = 4.4 × 10⁻³ × d² × H
All five equations are from Cornell, Part 3B, 2017.
A worked example
Take 2 MMscfd at 800 ppm H2S, 80°F (539.67°R), and 50 psig (64.7 psia), assuming Z ≈ 1.
Velocity-limited minimum diameter: 77.5 in
Deposition-limited minimum diameter: 92.4 in, so this one governs
Maximum diameter before channeling: 173.3 in
A 96-inch (8-foot) vessel sits comfortably in that window. At that diameter, the minimum bed height for 60 seconds of contact is 6.5 feet, which most operators round up for margin and a workable length-to-diameter ratio. Call it 7 feet.
That vessel holds about 284 bushels of media. At 15 lb Fe2O3 per bushel and a reaction efficiency of 0.7, the changeout interval works out to roughly 13 days.
What the example is telling you
Two numbers in that example are worth sitting with: an 8-foot-diameter vessel, and a media change every two weeks.
That is the economic ceiling on granular media showing up as arithmetic rather than as a rule of thumb. As inlet concentration and flow rate climb, the diameter equations push toward vessels that cost more in steel and media, and the changeout equation simultaneously pushes the service interval down. Both trends work against you at once.
Changeouts are also the part operators like least. Spent media has to be removed manually and drummed, each drum has to be recorded on a manifest kept in company records, and the work itself is dirty and unpleasant (Bowers and Cash, 2014). Cornell notes the same pressure from the other direction: media replacement frequency and cost have led most operators of large-scale digesters to choose other technologies for primary treatment, while iron sponge remains a good option for secondary or polishing duty (Cornell, Part 3A, 2017).
None of that makes granular media the wrong answer. Solid absorbent beds have handled low-concentration H2S for roughly a century, and they remain a low-complexity way to hit a pipeline spec, which is commonly set around 4 ppm, at the wellhead (Bowers and Cash, 2014). Liquid scavenging has its own ceiling, with most triazine-based systems unable to treat above about 2,000 ppm economically (Bowers and Cash, 2014). The point is that the crossover is specific to your gas, and the equations above are how you find it rather than guess at it.
Design details that matter beyond the math
Run more than one vessel.
Because a reactor has to come offline for media replacement, standard practice is to install two or more in parallel so gas can be routed through one while another is serviced (Cornell, Part 3B, 2017).
Keep spent media wet.
The regeneration reaction is exothermic and can generate enough heat to cause spontaneous combustion, which is why the media is kept moist in service and why material removed from the vessel must be kept wet to prevent auto-ignition (Cornell, Part 3A, 2017). Vessels are confined spaces and should not be entered without proper training and equipment.
Monitor two things, not one.
Rising outlet H2S indicates the reactive iron is being consumed. Rising pressure across the bed indicates the media may be fouling with solid iron sulfide (Cornell, Part 3A, 2017).
Watch moisture and temperature.
For classic wood-chip iron sponge, Cornell recommends holding temperature between 50°F and 130°F, noting that above 130°F the water of crystallization can be driven off and reduce reactivity, and maintaining trickle-water pH near 8 (Cornell, Part 3A, 2017). Engineered granular media carry their own moisture, temperature, and velocity parameters; check the specification sheet for the product you are running.
The takeaway
Running these five equations up front, and rerunning them whenever the gas analysis changes, is what keeps a scavenger vessel matched to the well it is actually treating instead of the well it was sized for on day one. If you want us to run them against your numbers, send us a current gas analysis and your target changeout interval and we will work it through with you.
Sources
- Shelford, T. and Gooch, C., “Hydrogen Sulfide Removal from Biogas, Part 3B: Iron Sponge Design Considerations: Vessel Sizing,” Cornell University Dairy Environmental Systems Program, November 2017. https://projects.sare.org/media/pdf/3/b/I/3b-Iron-Sponge-design-Considerations.pdf
- Shelford, T. and Gooch, C., “Hydrogen Sulfide Removal from Biogas, Part 3A: Iron Sponge Basics,” Cornell University Dairy Environmental Systems Program, November 2017. https://projects.sare.org/media/pdf/3/a/I/3a-Iron-Sponge-basics.pdf
- Hines, M., Choudhury, A., Felton, G. and Lansing, S., “Hydrogen Sulfide (H2S) Removal at a Northeastern Dairy Farm Digester using Iron Oxide: Case Study,” University of Maryland Extension, May 2018. https://enst.umd.edu/sites/enst.umd.edu/files/files/documents/Extension/Case-Study_H2S-Removal-Iron-Oxide_UMD.pdf
- Bowers, B. and Cash, E., “Maximize cost and safety benefits of scavenging H2S at the wellhead,” Gas Processing & LNG, December 2014. https://gasprocessingnews.com/articles/2014/12/maximize-cost-and-safety-benefits-of-scavenging-h2s-at-the-wellhead/
- Arnold, K. and Stewart, M., Surface Production Operations: Design of Gas-Handling Systems and Facilities, Volume 2, 2nd edition, 2008 (cited by source 1).
The information provided in this blog post is for general informational purposes only and does not constitute legal, financial, operational, safety, or professional advice of any kind. Some content on this site may be created or assisted by artificial intelligence tools; while we review this content for quality, AI-generated information can contain errors, outdated information, or inaccuracies, and should not be relied upon as a sole source of truth. While we aim for accuracy, we make no guarantees about the completeness, reliability, or timeliness of the information presented, and it may not reflect the most current developments or apply to your specific situation. Before acting on any information found here, you should consult a qualified professional. Ole Red assumes no liability for actions taken based on the content of this post. Customers remain solely responsible for safe operation of their systems and should always follow their own internal safety policies, operating procedures, and applicable laws, regulations, and governmental requirements over any technical guidance or recommendations provided by Ole Red.