Ion Exchange for PFAS: Design Considerations and Operational Tips

When regulatory limits fall into the low parts-per-trillion range, ion exchange pfas removal is often the pragmatic choice for municipal systems and remediation sites. This guide gives engineers and operators the design and operational detail they need — resin selection, pretreatment, pilot testing, sizing calculations, regeneration and disposal options, monitoring protocols, and lifecycle cost drivers. Expect real-world trade-offs, a worked sizing example, and troubleshooting tips you can use during specification and commissioning.

Why choose ion exchange for PFAS removal in municipal and remediation contexts

Direct point: For sites where short-chain PFAS or mixed ionic PFAS dominate and footprint, speed of implementation, or predictable effluent are priorities, ion exchange pfas removal is usually the practical choice over GAC. Studies show strong-base anion exchange resins outperform granular activated carbon for many short-chain PFAS; see the comparative review by Appleman et al. and the technology survey by Ross et al. (Appleman 2014, Ross 2018).

Strengths and tradeoffs: Ion exchange provides high ionic selectivity, compact vessel sizing, and an option for chemical regeneration, but it is matrix-sensitive. High DOC, sulfate, nitrate and chloride reduce dynamic capacity and accelerate breakthrough; conversely, low-DOC groundwater and AFFF-impacted plumes are where IX delivers the most value. The real tradeoff is operational complexity: pretreatment and brine handling add cost, and disposable-resin strategies trade higher media cost for simpler O&M.

Typical use cases where IX is the better choice

• Groundwater remediation: Long runs at controlled flowrates where a small footprint and modular trains are required for plume containment and pumping-and-treatment.
• AFFF-impacted sites: High ionic PFAS loads with limited organics favor strong-base anion exchange for rapid capacity use and easy swap-out of exhausted media.
• Finished water polishing: When utilities need reliable low-pptr effluent to meet drinking water goals and cannot rely on GAC for short-chain compounds.

Concrete example: A medium-size municipal plant replacing an underperforming GAC polishing train piloted a strong-base polystyrene anion resin on a 10 gpm column. With pretreatment to drop TOC by 40 percent and an EBCT of 5 minutes, the IX pilot achieved regulatory target concentrations for the short-chain suite and ran three times longer before breakthrough than the GAC column on the same water. The plant elected staged IX lead-lag vessels and a vendor-managed regeneration contract to avoid on-site brine handling.

What people get wrong: Designers often assume lab isotherms scale linearly. In practice, pilot-derived dynamic capacity is the only reliable basis for sizing because competing anions and hydrodynamics commonly reduce usable capacity by tens of percent. Rely on site-specific pilot data and incorporate conservative scale-up multipliers in procurement language.

Next consideration: After deciding IX is appropriate, move immediately to resin chemistry comparisons and pilot design to avoid costly under-sizing or unexpected early breakthrough.

Resin chemistry and selection criteria

Key point: Select the resin that matches the water matrix, not the lab brochure. For PFAS water treatment the dominant drivers are the resin functional group (strong-base vs weak-base), the polymer matrix (polystyrene-divinylbenzene vs polyacrylic), and the physical structure (gel vs macroporous) because these three factors control selectivity, kinetics, fouling risk, and regenerability.

Functional groups and their practical consequences

Functional-group rule of thumb: Strong-base anion resins (quaternary ammonium) are the standard choice for removing anionic PFAS at environmental pH because they provide consistent capacity across pH and faster kinetics. Weak-base resins can look attractive on paper but are highly pH-dependent and often lose capacity in alkaline water or under variable pH conditions.

Tradeoff to watch: Strong-base resins give reliable uptake but generate concentrated PFAS brine on regeneration and are more sensitive to competing anions such as sulfate. If your feed has high sulfate or nitrate, expect meaningful capacity loss; ask suppliers for dynamic column tests run on your site water rather than relying on distilled-water isotherms.

Polymer matrix and structure: what actually matters in the field

Polystyrene vs polyacrylic: Polystyrene-divinylbenzene resins are more hydrophobic and therefore often capture short-chain PFAS better because uptake benefits from combined ionic and hydrophobic interactions. Polyacrylic matrices are typically more hydrophilic and resist fouling in high-DOC waters, but they may sacrifice some capacity on short-chain species.

Gel vs macroporous and bead size: Macroporous beads and coarser particle sizes improve hydraulic resilience and fouling tolerance but can slightly lower volumetric capacity. Smaller beads speed kinetics (shorter EBCT possible) at the cost of higher headloss. Choose bead size based on your available backwash capability and allowable pressure drop.

• Ask suppliers for these deliverables: dynamic column breakthrough on site water, regeneration recipe (brine % and contact time), pressure-drop curves for proposed bead size, and documented pilot runs with competing-anion concentrations similar to your feed.
• Verify oxidant tolerance: many anion resins are sensitive to chlorine or permanganate; if upstream disinfection or intermittent chlorination is used, specify oxidant-resistant materials or include dechlorination ahead of the IX train.

Practical insight: Don't chase the highest nominal capacity number. A resin that lists a high static capacity measured in DI water can underperform by 30–70 percent in real feed water loaded with DOC and sulfate. Design around dynamic capacity from column tests and plan a conservative safety factor when sizing.

Concrete example: A remediation project treating an AFFF-impacted groundwater plume switched from a generic acrylic strong-base resin to a polystyrene strong-base resin after a side-by-side pilot. The polystyrene column gave faster breakthrough for long-chain PFAS but maintained regulatory-level effluent for short-chain PFAS at lower EBCT. The team accepted more frequent mechanical cleaning and tighter pretreatment because the polystyrene resin met the effluent targets where the acrylic did not.

Judgment call: In practice, polystyrene strong-base resins win when short-chain PFAS dominate and pretreatment can control DOC; acrylics or sacrificial strategies make sense when fouling is the limiting factor or when you need simpler, lower-risk operation. Anchor your spec to site-specific column tests and a clear regeneration or disposal path before final procurement.

Pretreatment to protect resin and extend service life

Direct point: Pretreatment is not optional for reliable ion exchange pfas removal; it is the difference between predictable run times and constant, expensive surprises. Fouling, iron/manganese coating, and high biodegradable organic carbon are the usual killers — each requires a different, measurable countermeasure.

Where projects go wrong: Teams often treat IX like a drop-in polishing step and skip upstream process controls. That saves capex up-front but forces frequent resin swaps, unplanned regeneration, or premature disposal. Plan pretreatment to protect resin kinetics first, absolute capacity second.

Practical pretreatment sequence and why it matters

1. Particle control: Install cartridge or multimedia filtration targeted at the site silt load to prevent mechanical clogging of the resin bed and porting around nozzles.
2. Iron/manganese removal: Use oxidation plus filtration or a greensand/chemical filter where soluble metals exceed trace levels; metal coatings irreversibly reduce accessible exchange sites.
3. DOC control: When DOC is the dominant foulant, coagulation/flocculation followed by dual-media filtration or a short GAC prefilter reduces organics that otherwise consume capacity.
4. Oxidant management: If you disinfect upstream, remove residual oxidant before the IX train. Oxidants damage many anion resins; the safe approach is controlled dechlorination or bypass during chlorination events.

Trade-off to weigh: Using GAC as a prefilter lowers influent DOC but also adsorbs PFAS and creates a parallel waste stream that needs its own handling. In practice, GAC prefilters can mask true IX capacity during pilots and cause sudden PFAS pulses when the GAC breaks through. Only choose GAC prefiltration when you have a clear disposal or regeneration path for the spent carbon.

Hydraulic hygiene matters as much as chemistry: Design the IX vessel and controls for routine backwash sequences sized to your bead size and headloss tolerance. Backwash intensity and frequency should be driven by measured differential pressure and turbidity at the vessel outlet, not by a calendar. Proper expansion during backwash removes trapped solids and reduces channeling; under-backwashing is a common and costly mistake.

Concrete example: At a former industrial site, soluble iron and heavy silt caused uninterrupted headloss and rapid resin fouling. The team installed an oxidation contactor followed by a pressure sand filter and 5-micron cartridge staging ahead of the IX skid. Resin run time increased from weeks to more than a year, and the project avoided frequent resin handling and off-site disposal costs.

Operational judgement: Target pretreatment to extend usable dynamic capacity rather than to maximize theoretical resin life. A modest DOC reduction that doubles dynamic capacity is usually a better investment than chasing marginal static capacity gains with an expensive resin. Pilot pretreatment configurations as part of your pilot testing scope and lock performance guarantees to pilot results.

Key takeaway: Protect the ion exchange train by controlling solids, metals, and biodegradable organics upstream; pick pretreatment based on what fouls your resin, not on what is cheapest to install.

Engineering design and sizing: formulas, EBCT, bed depth and a worked example

Direct design rule: Size the IX train from pilot-derived dynamic capacity and site flow, not from static lab numbers. Use a mass-balance for PFAS loading, then reconcile that required resin volume with hydraulic sizing driven by EBCT and allowable superficial velocity.

Key formulas and definitions

Essential formulas: EBCT (min) = Bed volume (L) / Flowrate (L/min). Bed volume (L) = Bed area (m2) × Bed depth (m) × 1000. Superficial velocity v = Q (m3/h) / Area (m2) (convert units to gpm/ft2 or m/h as preferred).

Mass-balance for resin sizing (practical form):
1) Daily PFAS mass = Q (m3/d) × (Cin - Cout) (µg/L) × 1000 L/m3 → µg/d → convert to mg/d by /1000.
2) Resin capacity (dynamic) = Cd (mg PFAS per L resin) from pilot.
3) Required resin volume (L) = (Daily PFAS mass (mg/d) × desired runtime (d)) / Cd (mg/L).
4) Resin mass (kg) ≈ Resin volume (L) × bulk density (kg/L) (use supplier bulk density).

• Units matter: Keep Cin/Cout in µg/L and capacities in mg/L (1 mg/L = 1000 µg/L) so conversions are consistent.
• EBCT tradeoff: Longer EBCT improves kinetics and often increases usable dynamic capacity, but it inflates vessel volume and footprint.
• Hydraulics check: If EBCT-derived bed volume exceeds the resin volume from the mass-balance, you must add parallel vessels, increase bed depth (within headloss limits), or accept a shorter runtime.

Practical insight: BDST and Thomas models are useful for turning pilot breakthrough curves into expected service time, but they require pilot data at the design EBCT; do not substitute short-EBCT microcolumn tests for full-scale kinetics. Use the models to estimate conservative scale-up multipliers rather than a single point estimate.

Worked numeric example (illustrative)

Concrete example: Design flow = 500 m3/d (≈ 347 L/min). Influent total PFAS = 0.20 µg/L, target effluent = 0.01 µg/L → delta = 0.19 µg/L. Daily PFAS mass = 500 × 0.19 × 1000 = 95,000 µg/d = 95 mg/d.

Assume pilot dynamic capacity Cd = 10 mg PFAS/L resin (site-derived) and target runtime = 90 d. Required resin volume = (95 mg/d × 90 d) / 10 mg/L = 855 L resin. With bulk density 0.8 kg/L, resin mass ≈ 684 kg.

Pick EBCT = 8 min (design choice driven by pilot kinetics). Bed volume needed = 8 min × 347 L/min = 2,776 L. That bed volume is larger than the 855 L resin required by PFAS mass, so practical layouts are: (a) build one vessel sized for EBCT and fill to the EBCT volume (use extra resin to maintain bed porosity), or (b) use multiple parallel vessels sized for EBCT with total resin equal to the 855 L requirement and accept slightly lower EBCT, then verify with pilot/BDST.

Trade-off judgment: If you reduce EBCT to save footprint, expect lower dynamic capacity and shorter runtime; the net result can be more frequent regeneration or resin change-outs that raise lifecycle cost. When in doubt, lean conservative on EBCT and verify with a pilot at that EBCT.

Operational caveat: Hydraulic limits — headloss, backwash expansion, and available backwash flow — often constrain maximum bed depth in practice. Confirm nozzle design and backwash capability before specifying tall beds.

Takeaway: run the math from pilot capacity to hydraulic EBCT and then reconcile the two — mismatches are routine and require decisions on footprint, vessel count, or acceptable runtime. If you skip this reconciliation you will under- or over-spec the system and pay for it in operational headaches.

Pilot testing protocol and breakthrough data interpretation

Direct rule: Run a pilot that replicates the operational hydraulics and pretreatment you intend to use in full scale — anything less will understate early breakthrough and overstate resin life. Use your pilot to produce a defensible time-to-failure curve, not a static capacity number.

Pilot setup and sampling plan

1. Match EBCT: size pilot columns so their EBCT equals the design EBCT or the EBCT you plan to test. Do not infer full-scale kinetics from microcolumns unless you explicitly correct for wall effects and mass-transfer limitations.
2. Replicates and modes: run at least two parallel columns: one to steady breakthrough and one as a control or guard bed. Include a lead-lag run to assess staged operation performance.
3. Pretreatment mimic: feed the IX pilot with water that has passed the same coagulation, filtration, or GAC prefilters scheduled for the plant. If you cannot run on fully treated water, spike the pilot feed to reproduce DOC, iron, and turbidity levels seen downstream of your pretreatment.
4. Sampling ports and schedule: install ports at influent, effluent and mid-bed heights. Early sampling cadence should be high: daily for the first week, then every 2–3 days until a clear rising trend, then weekly as appropriate. Use field blanks and duplicates and follow chain-of-custody.
5. Instrumentation: log flow, pressure differential, conductivity and turbidity continuously. Use pressure trends and conductivity as operational proxies but confirm with laboratory PFAS data using EPA methods such as EPA 537.1 or EPA 533.

Practical tradeoff: Larger-diameter pilot columns reduce wall effects and give cleaner kinetics but require more resin and water. If site constraints force small columns, extend run time and be conservative when translating results to full scale.

How to produce and read breakthrough data

Plotting choice matters: plot normalized concentration (C/C0) vs bed volumes and vs time. Bed-volume plots expose mass-transfer limitations; time plots expose practical runtime. Define your breakthrough threshold ahead of time — either a fraction of influent (for kinetics) or the regulatory effluent target (for compliance planning).

Modeling approach: fit both BDST and Thomas models to the rising portion of the curve to extract dynamic uptake and rate constants. BDST gives a quick estimate of required bed depth for a target runtime; Thomas is better for scale-up when column kinetics approximate plug flow. Be aware each model has assumptions—use them as guidance, not gospel.

Conservative scaling: in real projects translate pilot dynamic uptake to full scale with a safety factor. In practice a scale-up factor between 0.5 and 0.75 of pilot-measured capacity is reasonable when pretreatment and hydraulics are comparable; tighter factors require matched pilot/full-scale hydraulics and repeated pilots.

Common pilot failure modes: inadequate pretreatment (masks fouling), intermittent oxidant exposure (destroys exchange sites), and using GAC ahead of the pilot (hides PFAS mass that will hit IX in real operation). If you see erratic spikes, check flow splits, by-pass valves, and sampler representativeness before reinterpreting chemistry.

Concrete example: A remediation team ran two 50 mm-ID columns at the design EBCT of 6 minutes using groundwater post-filtration. Daily samples showed a slow rise in short-chain PFAS; BDST fits produced a usable dynamic capacity of 6.5 mg PFAS/L resin. The design team applied a 0.6 scale factor and sized full-scale vessels accordingly, then specified vendor acceptance testing to validate after commissioning.

Takeaway: A pilot is only useful if it reproduces your plant hydraulics and pretreatment. Fit models, but size conservatively — and lock acceptance testing to on-site pilot outcomes before awarding full-scale contracts.

Regeneration, disposal, and end-of-life management

Direct point: the practical cost-effectiveness of ion exchange pfas removal hinges on your chosen path for exhausted resin and spent regenerant long before you buy vessels. Regeneration reduces media purchases but creates a concentrated PFAS waste stream; disposal or destruction avoids brine handling but transfers cost and permitting burdens to transport and thermal treatment.

Practical options and the trade-offs that matter

• On-site chemical regeneration: Use a 3–10% NaCl brine (site-specific) with contact times and displacement steps per supplier. Pros: extends media life and lowers media spend. Cons: produces a high-PFAS brine that needs containment, sampling, and likely off-site treatment or permitted disposal.
• Off-site thermal regeneration/destruction: Spent resin or concentrated brine shipped to a certified incinerator. Pros: reliable PFAS destruction and reduced liability. Cons: high unit cost, transport logistics, air-permit considerations, and potential classification as hazardous waste by some states.
• Disposable-resin strategy: Replace and send spent resin directly for permitted incineration or landfill (where allowed). Pros: simplifies O&M and avoids brine handling. Cons: significantly higher recurring media cost and transport/disposal fees.
• Vendor-managed closed-loop services: Supplier accepts spent resin or brine and handles regeneration plus destruction. Pros: shifts regulatory and logistics burden. Cons: usually carries premium pricing and requires strict contractual SLAs on residual PFAS and turnaround time.

Limitation to plan for: chemical regeneration seldom restores full capacity for PFAS to pre-service levels. Expect partial recovery that declines with each cycle, especially for short-chain and highly ionic PFAS. Design your lifecycle model assuming diminishing returns after several regenerations and include a budget for eventual resin replacement.

Concrete example: A municipal utility serving 30,000 people piloted an on-site brine regeneration program but hit a regulatory snag: no local discharge route and state rules treating the brine as hazardous. They switched to a vendor-managed thermal destruction contract. Upfront service costs rose, but operational disruptions fell and the utility avoided installing large brine tanks and gaining complex waste permits.

Operational and regulatory checklist

1. Confirm waste classification early: Engage state regulators and check federal guidance at EPA PFAS and local agencies documented on regulatory updates.
2. Permit provisions: plan secondary containment, spill response, and storage limits for on-site brine; include manifest and transport requirements for off-site shipments.
3. Analytical gating: require the supplier or hauler to provide PFAS concentration data on spent brine/resin and set acceptable shipping thresholds in the contract.
4. Contract language: specify responsibility for sampling, liabilities for noncompliance, turnaround times, and unit prices for regeneration, destruction, and disposal.
5. Operational controls: provide secure staging, trained personnel for handling hazardous loads, and routine sampling of regenerant to detect PFAS spikes or incomplete desorption before shipment.

Judgment for design teams: compare total lifecycle cost curves — media, regeneration, transport, disposal, and regulatory compliance — not just initial capital. Many small-to-midsize utilities find vendor-managed destruction easier and more predictable despite higher per-cycle cost; larger systems with predictable PFAS loads can justify brine infrastructure and controlled regeneration.

Next consideration: lock your regeneration/disposal route into procurement and acceptance testing so the chosen lifecycle option drives resin selection, pilot scope, and the O&M contract rather than being an afterthought.

Monitoring, instrumentation, operational controls and troubleshooting checklist

Key control point: Effective long-term ion exchange pfas removal depends on marrying targeted laboratory PFAS monitoring with continuous operational telemetry so you detect capacity loss before compliance samples arrive. Relying on lab results alone guarantees surprise; choose a layered detection strategy that uses real-time proxies, defined lab triggers, and automated control actions.

Instrumentation and sampling strategy

Instrumentation aim: Install sensors that detect the mechanical and chemical precursors to breakthrough rather than trying to measure PFAS in-line. Use lab PFAS analysis to confirm and quantify, but use continuous signals to drive alarms and automated responses.

Operational controls and alarm logic

Control principle: Automate conservative actions for early indicators and require human review for lab-confirmed PFAS excursions. Automation reduces operator load but poor setpoints create nuisance trips or missed events.

1. Tiered alarm approach: low-level proxy alarm (e.g., conductivity or TOC trend), pre-emptive action (start guard bed or switch lead/lag), lab sample request; high-level alarm (lab PFAS confirmation or sustained proxy trend) initiates regeneration/change-out.
2. SCADA interlocks: force flow-diversion to a standby IX train when dP or proxy exceedances persist beyond a defined hysteresis window.
3. Data retention and QA: store raw sensor data and chain-of-custody lab results together for post-event root-cause analysis and regulatory recordkeeping.

Troubleshooting checklist

• If you see early PFAS lab rise: confirm sampler representativeness first; check flow splits, valves and sample line integrity. Then review recent conductivity/TOC trends and upstream pretreatment events before assuming resin failure.
• If dP climbs rapidly: inspect prefilters and backwash sequence. If backwash expansion is inadequate, increase backwash flow per supplier guidance and inspect for nozzle plugging.
• If conductivity spikes without dP change: suspect ionic load change (sulfate/nitrate) or brine carryover. Pull a mid-bed sample and check regeneration valve timing and rinse volumes.
• Intermittent spikes in short-chain PFAS: look for GAC prefilter breakthrough, upset in coagulation, or sampler cross-contamination rather than immediate resin loss.
• Biological growth signs (odd odors, slow headloss recovery): check residual oxidant history; confirm dechlorination is working and consider periodic biocide-compatible cleanings or scheduled out-of-service disinfection.

Operational insight: Operators often overestimate the fidelity of single proxies. Conductivity and TOC are useful but non-specific; they should be treated as early warning indicators that trigger targeted sampling and protective actions, not as definitive evidence of PFAS breakthrough.

Real-world example: At a small municipal plant treating mixed PFAS in groundwater, a sudden upstream well rehabilitation changed ionic strength and conductivity rose by 15 percent. The control system automatically switched flow to the lag vessel and flagged a lab sample. Lab confirmation showed a small PFAS uptick that was contained by the lead-lag swap — avoiding an accidental regulatory exceedance and giving time to schedule a controlled regeneration.

Next consideration: After setting instruments and alarms, run drills that simulate sensor failure, sampler contamination, and a sudden ionic load change so your team learns which automated controls to trust and which situations require immediate operator intervention.

Cost drivers, lifecycle assessment and procurement considerations

Bottom line: The procurement choice you make on resin lifecycle handling and vendor responsibilities usually determines whether ion exchange pfas removal is a manageable recurring cost or an open-ended liability. Treat lifecycle planning as the primary design driver rather than an afterthought.

Major cost drivers and where money is actually spent

Primary cost buckets: Capital for pressure vessels and skid, capital for pretreatment to protect resin, initial resin purchase, routine O&M (energy, backwash water, operator time), analytical program costs, and end-of-life handling (regeneration brine management, transport, thermal destruction or landfill). Hidden drivers that commonly surprise teams are frequent resin change-outs driven by fouling, and regulatory-driven increases in disposal cost or monitoring frequency.

• Resin and media lifecycle — purchase price plus expected number of regenerations or replacements; dependent on feedwater matrix.
• Pretreatment scope — coagulation, sand/GAC prefilters, or cartridge filtration to protect IX; increases initial capex but often multiplies usable dynamic capacity.
• Waste handling — brine storage, sampling, transport manifests, and thermal destruction fees; can dominate annual O&M if on-site regeneration is used.
• Analytics and compliance — routine EPA-method lab runs, quality control, and incident sampling; often underestimated in bid comparisons.

Simple lifecycle-cost framework and an illustrative example

How to model it: Annualize capital (use your municipal discount rate), add annual O&M, add recurring disposal/regeneration cost, and divide by annual treated volume to get levelized cost per cubic meter. In procurement, present both a base case (on-site regeneration) and an alternative (vendor-managed destruction) so bidders price both architectures.

Illustrative example: A small system treating ~200 m3/d assumes total capex (vessels + pretreatment + install) of $900,000, initial resin $100,000, and recurring O&M/disposal of $100,000/yr. Over a 15-year horizon that totals about $2.4M lifecycle cost, giving a levelized cost near $2.20/m3. This is illustrative only — changing resin life, regeneration frequency, or disposal fees swings costs dramatically.

Judgment: Maintenance and disposal assumptions move the needle far more than small differences in resin unit price. If you see two vendor bids with similar capex but wildly different O&M profiles, follow the O&M math — the cheaper capital option is often more expensive over a few years.

Procurement must-haves and contract language to insist on

• Pilot-linked guarantees: require performance guarantees expressed as dynamic capacity at the design EBCT on your site water, with documented pilotQA and acceptance tests (see pilot testing guidelines).
• Clear waste responsibility: define who samples, transports, and pays for brine/resin disposal or thermal destruction; include not-to-exceed unit prices or indexed escalators.
• Acceptance testing: mandate on-site proof-of-performance at full flow and EBCT before final payment; include remedial steps and rework clauses if targets are missed.
• Service-level metrics: specify regeneration turnaround time, permitted PFAS levels in returned resin (after regeneration), and KPI penalties for missed deadlines.

Practical negotiation tip: Ask bidders to provide a 10-year cash-flow table showing assumed resin life, number of regenerations, brine disposal cost per cycle, and annual lab costs. Compare total cost-of-ownership, not just equipment list prices.

Concrete example: A regional utility negotiated a vendor-managed destruction contract with a fixed per-cycle freight-and-destruction fee. Their per-cycle fee was higher than on-site brine handling, but the utility reduced staff time and avoided building brine tanks and secondary containment. The predictable OPEX made budgeting and regulatory reporting simpler.

Action to take next: run a rapid lifecycle sensitivity using your expected influent PFAS loading and two disposal scenarios; use that output to decide whether to pursue on-site brine infrastructure or a vendor-managed destruction path, and then write the pilot scope and contract around that choice. For regulatory context consult EPA PFAS.