Why an Isfahan data center needs its own 2026 water brief
A 100 MW data center can demand up to 2 million litres of water per day, the same envelope that anchors the Najaf data center water brief (IDE 2026, via HydropureWater 2026) and that Iranian EPC firms should treat as the design floor for any 5–50 MW site in Isfahan Province. Isfahan is not Najaf and not Riyadh: the city sits on a critically low-flow Zayandeh-Rud basin, with a Karaj-feed groundwater chemistry that is hard (Ca²⁺ + SO₄²⁻), silica-bearing, and prone to CaSO₄ scaling, and with Bandar Abbas clearance plus roughly 1,300 km of overland haul before equipment reaches the Isfahan industrial parks. The 2026 hyperscale water reuse math is unforgiving here, and an off-the-shelf Middle East design brief will not pass a DOE Isfahan Province review.
Three site-specific realities drive the design. First, Zayandeh-Rud scarcity: the river has been in critically low-flow status for multiple recent years, which means there is no realistic path for high-volume surface discharge of CTBD or treated effluent; freshwater withdrawal permits are also tightening in lockstep. Second, summer ambient runs 38–42°C, so biological oxygen transfer is derated similarly to the 45–50°C Najaf baseline (HydropureWater 2026); aeration must be oversized 10–15% or specified with disc diffusers carrying a guaranteed SOTE above 6.5 kg O₂/kWh at design temperature. Third, the Iranian industrial grid in Isfahan parks averages 2–6 hours of planned outage per week plus unscheduled faults, which forces a 7-day chemical autonomy buffer and a 2-year consumables plus critical spares kit to be written into the WWTP CAPEX line, not into a later change order. Logistics alone shifts the FOB-to-commissioned-ready lead time from the 10–16 weeks the Najaf line assumes to 12–20 weeks for Isfahan, because the overland corridor from Bandar Abbas is roughly 2.6× longer than Umm Qasr to Najaf.
Isfahan data center wastewater: the four streams and their chemistry
An Isfahan data center site splits into four distinct streams, each with a different chemistry and a different downstream train. Segregating them at the head of the plant is the single most important cost lever, because it prevents the high-volume, low-strength humidification stream from being over-treated and the high-TDS CTBD stream from contaminating the rest of the reuse train.
| Stream | Typical flow share | TDS (mg/L) | SS (mg/L) | Key contaminants | Default destination |
|---|---|---|---|---|---|
| Data-hall humidification + process water | 30–40% | 500–1,500 | 10–50 | Silica 5–30 mg/L, corrosion products, biofilm fragments | DAF → MBR → RO → CT make-up |
| Cooling-tower blowdown (CTBD) | 60–70% | 1,200–6,000 | 10–50 | CaCO₃, CaSO₄, silica, biocides, inhibitors | Side-stream filtration → UF → RO → NF/MVC |
| RO reject + UPW-loop blowdown | 5–10% | 200–800 | <5 | Low-TDS, high-purity; NH₃-N 50–500 mg/L if standby gensets scrubbed | Boiler feed or process wash; NH₃ side-stream stripped |
| Sanitary + cafeteria | 5–10% | 300–800 | 100–250 | BOD, pathogens, surfactants | Buried A/O package plant, not the industrial train |
The per-MW scaling math comes from Genesis Water Technologies (2026): at 4 cycles of concentration a typical data center loses 25–30% of its make-up water to blowdown, so a 10 MW Isfahan site at 10 million gallons per month of make-up discharges roughly 2.5–3 million gallons per month of CTBD alone. Pushing the tower to 6–8 cycles cuts that stream by half but lifts CTBD TDS into a CaSO₄ + silica scaling regime that conventional brackish RO cannot always handle (Genesis 2026). The data-hall humidification drain is the cleanest stream on the site, and it is the worst stream to send to the sewer, because it is too clean to waste and too large to ignore on a permit.
2026 compliance envelope for Isfahan: DOE, FEPC, and Zayandeh-Rud
Iran does not operate a single codified federal industrial discharge schedule comparable to U.S. EPA categorical effluent guidelines. Enforcement sits with the Department of Environment (DOE) at the provincial level, with the Forest, Range and Watershed Management Organization (FEPC-aligned) reviewing reuse applications, and 2023–2025 enforcement records show administrative penalties plus operational suspension as the default response to non-compliance (HydropureWater 2026, regional analogue applied to Iran). The 2026 Isfahan envelope, pending a province-specific ministerial decree, is best cross-walked to the regional benchmark used in the Najaf brief: BOD₅ ≤50 mg/L, COD ≤200 mg/L, TSS ≤50 mg/L, free Cl <1 mg/L, total Cr ≤0.5 mg/L, verified by 24-hour composite sampling (HydropureWater 2026).
Two Isfahan-specific risks sit on top of the numeric envelope. First, Zayandeh-Rud liability: even where the limits are met on paper, large-volume surface discharge to a critically low-flow river is a permit liability and a public-consent issue that the DOE renewal audit will revisit. Second, the cost of non-compliance is asymmetric. On a USD 5M/yr revenue line at the typical 8–12% industrial margin, a two-week operational stoppage exceeds USD 75,000 in lost contribution margin; the DOE's first response to a failed composite sample is suspension, not a warning letter (HydropureWater 2026, via S2). Translated to 2026 Iran industrial tariffs, the same stoppage window sits in the 30–50 billion rial band depending on the site's toman-denominated power and colocation contracts, which is the number finance teams will want to see alongside the WWTP CAPEX line.
Data-hall train: DAF → MBR → multi-media → RO
The data-hall stream is the lower-strength side of the site, and the train is sized to the humidification plus process-water load, not to the fab load. The four-stage layout is the 2026 default and it is defensible stage by stage in a DOE design review. Stage 1 is a DAF sized for 4–300 m³/h, 13 models in the ZSQ range, which removes the 10–50 mg/L suspended-solids load of corrosion products and biofilm fragments and buffers hydraulic surges from humidification cycles. Stage 2 is a containerized MBR operating at MLSS 8,000–12,000 mg/L with HRT 8–14 h, delivering BOD <10 mg/L, COD <60 mg/L, and TSS <5 mg/L in roughly 60% smaller footprint than conventional activated sludge (HydropureWater 2026). On the Isfahan summer envelope, the aeration train is oversized 10–15% or specified with high-efficiency disc diffusers carrying a guaranteed SOTE >6.5 kg O₂/kWh at design temperature.
Stage 3 is a multi-media filter to hold SDI below 5 and protect the downstream RO from the biocides and corrosion inhibitors that concentrate in the circulating water; without it the RO membranes foul on the 1.0–2.0 SDI feed that humidification drain chemistry typically produces. Stage 4 is brackish RO polishing at 65–75% recovery, returning polished flow to cooling-tower make-up or UPW make-up. The 2026 design envelope target for the data-hall stream is 85–90% overall site recovery once CTBD is folded in (HydropureWater 2026, via S2). Decision rule that holds for Iranian sites as well as Iraqi ones: if total project flow is below 200 m³/day and reuse economics are weakest, drop the RO and discharge DAF + MBR + ClO₂ to sewer; most Isfahan data centers above 5 MW IT load will exceed this threshold on flow alone.
CTBD reuse ladder: side-stream filtration → UF → RO → NF → MVC
CTBD is the larger stream at 60–70% of site volume and the harder one to treat, because it carries concentrated CaCO₃, CaSO₄, silica, biocides, and corrosion inhibitors. The 2026 technology choice is a stepped ladder, not a single membrane shortcut, and each step has a defensible cost and recovery band. Step 1 is side-stream filtration at 1–5% of circulation flow using 10–25 µm self-cleaning spiral screens, which cuts suspended solids to levels the downstream membranes can tolerate; CAPEX is $50K–$200K for a typical data-center installation (Genesis Water Technologies 2026). Step 2 is UF pretreatment ahead of the blowdown RO at 0.01–0.1 µm, 10–30 psi, 90–95% recovery, with chemical cleaning every 1–3 months (Genesis 2026).
| Step | Technology | Operating pressure | Recovery | Permeate / output quality | CAPEX band | OPEX band |
|---|---|---|---|---|---|---|
| 1 | Side-stream filtration, 10–25 µm | 5–15 psi | Continuous bleed 1–5% of circ. | SS cut to RO-tolerable | $50K–$200K | Solids disposal only |
| 2 | UF, 0.01–0.1 µm | 10–30 psi | 90–95% | SDI <3, no biocides | Site-specific | CIP every 1–3 months |
| 3 | Brackish RO | 150–400 psi | 50–85% | 10–50 mg/L TDS, 95–99% salt rejection | $250K–$500K per 50,000 GPD | $1.50–$3.00/kgal |
| 4 | NF (partial softening alt.) | 75–150 psi | 70–85% | 30–50% of feed TDS | Site-specific | Site-specific |
| 5 | MVC evaporator (ZLD finish) | n/a (thermal) | 95–98% | Distillate <10 mg/L TDS | $1–3M per 10,000–30,000 GPD | 15–25 kWh/1,000 USG |
| 6 | Full ZLD with crystallizer | Thermal + mechanical | 95–99% overall | Solid salt cake, <1% liquid waste | $3–8M total | $5–$15/kgal |
Step 3 is the brackish RO polishing at 50–85% recovery with permeate at 10–50 mg/L TDS; a 50,000 GPD unit is $250K–$500K installed CAPEX and $1.50–$3.00/kgal OPEX (Genesis Water Technologies 2026). Step 4 is an NF alternative at 70–85% recovery and 75–150 psi where partial softening, not full demineralization, is the goal; permeate sits at 30–50% of feed TDS (Genesis 2026). Step 5 is an MVC evaporator for the ZLD finish at 95–98% recovery and distillate below 10 mg/L TDS, drawing 15–25 kWh per 1,000 USG; full ZLD with crystallizer runs $3–8M CAPEX and $5–$15/kgal OPEX (Genesis 2026). Automatic antiscalant and biocide dosing skids are mandatory on Isfahan Karaj-feed chemistry: without them the RO concentrate scales within hours, not days. A softener train ahead of the RO can take the hardness load off the membrane and push recovery toward 85% on Karaj feed.
Final disinfection and reuse polishing for the Isfahan train
The binding discharge constraint is free chlorine <1 mg/L, which makes an on-site ClO₂ generator the 2026 default; chlorine gas or NaOCl would trip the THM formation risk on the long humidification residence times typical of an Isfahan data hall. UV polish is optional for RO permeate returning to humidification make-up, not a substitute for ClO₂ on the discharge line, and the UV sterilizer in the ZS line is sized to the permeate flow plus a 1.5× redundancy factor.
Reuse priority order for an Isfahan site, ranked by water quality and economic return: cooling-tower make-up first, then UPW make-up, then boiler feed, then landscape or equipment washdown. The data-hall humidification drain is the cleanest stream on the site, and it is the most damaging stream to send to the sewer, because it is too clean to waste and too large to ignore on a permit. Ammonia wet-scrubber blowdown from standby generators, when present, carries 50–500 mg/L NH₃-N (HydropureWater 2026, via S2) and must be nitrified or air-stripped separately before it joins the main biological train.
Decision framework: reuse vs discharge vs ZLD for an Isfahan site
The 2026 decision rule for an Isfahan data center collapses to three triggers. Choose reuse when site flow exceeds 200 m³/day or the DOE flags non-revenue water on the renewal audit; choose discharge only when project flow is below 200 m³/day and on-site operators are limited (HydropureWater 2026). Choose ZLD finish when Zayandeh-Rud discharge is non-permittable, when DOE requires it, or when the hyperscale site is in a closed basin; 95–99% overall recovery is achievable but at $3–8M CAPEX and $5–$15/kgal OPEX (Genesis Water Technologies 2026).
Push cooling-tower cycles from 4 to 6–8 to cut make-up water demand 30–50%; that single lever, paired with automatic antiscalant dosing, is the largest water and CAPEX offset an Isfahan project has. The data-hall humidification drain, low-strength and high-volume, is the worst candidate to send to sewer, and the best candidate for direct RO reuse back to cooling-tower make-up. Most Isfahan data centers above 5 MW IT load will trip the reuse threshold on flow alone, and any hyperscale site above 20 MW should be designed for ZLD finish from day one, because the Zayandeh-Rud permit liability does not scale down.
2026 CAPEX, OPEX, and delivery timeline for an Isfahan data center WWTP
Budgeting a 2026 Isfahan data center WWTP requires three layered cost lines, and each layer has an Iran-specific uplift over the Najaf regional benchmark. Energy: an MBR-equipped plant runs 0.8–1.6 kWh/m³, a DAF-only scope 0.4–0.9 kWh/m³, and Iran industrial tariffs run 1.5–3× the unit cost seen in lower-tariff markets, so energy alone is the largest variable OPEX line (HydropureWater 2026, directional). Sludge dewatering with a plate-and-frame filter press reaches 22–28% dry solids, and landfill disposal follows regional Iran industrial rates, with the press maintenance protocol extending element life by 40% on Karaj-feed abrasion.
| Tier | Scope | Indicative CAPEX envelope | OPEX driver | Best fit |
|---|---|---|---|---|
| Small | DAF + MBR + ClO₂ (sewer discharge) | Lowest (single-digit USD M for <200 m³/day) | Energy 0.4–0.9 kWh/m³, Iran tariff | <5 MW IT, no reuse economics |
| Mid | DAF + MBR + RO + ClO₂ | USD 250K–$500K per 50,000 GPD RO alone, plus MBR and civils | Energy 0.8–1.6 kWh/m³, $1.50–$3.00/kgal RO | 5–20 MW IT, 60–85% site recovery |
| Hyperscale | Above + NF/MVC or full ZLD with crystallizer | $1–3M MVC, $3–8M full ZLD | 15–25 kWh/1,000 USG, $5–$15/kgal ZLD | 20–50 MW IT, 85–99% recovery, Zayandeh-Rud closed |
Logistics is the second Iran-specific cost layer. Containerized WWTP skids clear Bandar Abbas in 7–14 days for standard ISO containers, then require 7–10 days for the roughly 1,300 km overland haul to the Isfahan industrial parks; total FOB-to-commissioned-ready lead time is 12–20 weeks, roughly 4–6 weeks longer than the Najaf line because of the longer corridor and additional customs handling (uplift from HydropureWater 2026, Najaf baseline). A rotary mechanical bar screen at the head of the train is mandatory to protect the membranes during the overland-transport debris window, and membrane spares should ship in the same logistics window to avoid a 6–10 week replacement-part wait from China or Europe. Specify a 7-day chemical autonomy buffer and a 2-year consumables plus critical spares kit to absorb 2–6 hours per week of planned and unscheduled grid outages; the buffer should include NaOCl backup, antiscalant, and CIP chemicals in sealed totes.
Frequently Asked Questions
What is the 2026 reuse-vs-discharge decision rule for an Isfahan data center?
Choose reuse when site flow exceeds 200 m³/day or DOE flags non-revenue water; choose discharge only below 200 m³/day with limited operators. Most Isfahan data centers above 5 MW IT load trip the reuse threshold on flow alone (HydropureWater 2026).
What is the 85–90% overall site recovery target, and how is it reached?
The 2026 design envelope is 85–90% overall site recovery once CTBD is folded in, reached by pairing the four-stage data-hall train with a CTBD side-stream filtration → UF → RO ladder and pushing cooling-tower cycles from 4 to 6–8 (HydropureWater 2026; Genesis Water Technologies 2026). For hyperscale sites, MVC or full ZLD pushes recovery to 95–99% at $3–8M CAPEX.
What is the 2026 DOE Isfahan Province discharge envelope?
BOD₅ ≤50 mg/L, COD ≤200 mg/L, TSS ≤50 mg/L, free Cl <1 mg/L, total Cr ≤0.5 mg/L, verified by 24-hour composite sampling (HydropureWater 2026, regional analogue). Zayandeh-Rud scarcity adds a permit-liability layer on top of the numeric envelope.
What is the indicative CAPEX and OPEX for a 50,000 GPD RO polishing CTBD on an Isfahan site?
The directional benchmark is $250K–$500K installed CAPEX and $1.50–$3.00/kgal OPEX (Genesis Water Technologies 2026), uplifted for Iran logistics, ambient derating, and the 1.5–3× energy tariff multiplier (HydropureWater 2026).
What is the Bandar Abbas to Isfahan logistics lead time for a containerized WWTP?
Bandar Abbas ISO container clearance 7–14 days plus roughly 1,300 km overland haul to Isfahan at 7–10 days; total FOB-to-commissioned-ready is 12–20 weeks, with a 2-year consumables and critical spares kit shipped in the same window (HydropureWater 2026, adapted from the Najaf baseline).
