Framework Applied: Distributed Fertilizer Production

The Problem

The global fertilizer system is built on a hidden subsidy: cheap diesel makes centralised production and long-distance transport appear economically rational. Nitrogen fertilizer is made from air — a free, ubiquitous input — yet it is manufactured in massive centralised plants, shipped across the world under pressure as a toxic gas, distributed through long markup chains, and applied to fields where roughly 50% is lost to atmosphere or waterways.

When transport costs are honestly priced — including carbon, infrastructure, and strategic risk — the calculus inverts. Local production from local inputs becomes competitive or superior across a wide range of contexts.

This document applies the framework to that opportunity.

Objects

Resources

  • Nitrogen (N₂): 78% of atmosphere, free, unlimited, requires energy to fix
  • Straw / crop residue: agricultural waste, currently burned or ploughed in
  • Methane: produced from anaerobic digestion of organic waste
  • Hydrogen: produced from methane (SMR) or water (electrolysis)
  • Ammonia (NH₃): fixed nitrogen, the target output, toxic and pressurised
  • Digestate: nutrient-rich slurry, co-product of anaerobic digestion
  • Electricity: input and potential output
  • Carbon credits: tradeable value from methane capture and emissions reduction
  • Water: input to SMR and electrolysis
  • Lime (calcium hydroxide): pretreatment agent, breaks lignin structure in straw
  • Iron catalyst: standard Haber-Bosch catalyst, commodity, known since 1909
  • Nickel catalyst: standard SMR catalyst, commodity

Technologies

  • Anaerobic digester: converts organic waste to methane and digestate
  • Lime pretreatment unit: soaks straw in lime solution to break lignin, improves methane yield by 60-80%
  • Steam methane reformer (SMR): converts methane + water to hydrogen + CO₂
  • Pressure swing adsorption (PSA) unit: separates nitrogen from air
  • Haber-Bosch reactor: combines hydrogen and nitrogen under pressure and heat with iron catalyst to produce ammonia
  • Ammonia storage: pressurised vessels for ammonia holding and distribution
  • Gas genset: converts methane to electricity
  • Electrolyser: alternative hydrogen source, splits water using electricity (not recommended as primary path — see dragons)

Actors

  • Farmers: feedstock suppliers, fertilizer consumers, capital contributors
  • Farming cooperative: governance, capital aggregation, logistics coordination
  • Government: food security motivation, potential emergency funder, regulatory authority
  • Carbon market: pays for methane capture, emissions reduction
  • Grid operator: potential buyer of surplus electricity
  • Fertilizer importers / distributors: incumbent whose margin disappears under this model
  • Technology vendors: SMR suppliers, Haber-Bosch unit suppliers, digester suppliers

Institutions

  • Global fertilizer supply chain: incumbent system, optimised for cheap transport
  • Carbon credit verification bodies: gatekeepers to carbon revenue
  • Regulatory bodies: safety, environmental, planning approvals

Properties

Straw

  • Yield per hectare: 3-4 tonnes (confidence: high)
  • Lignin content: high — resists anaerobic digestion without pretreatment (confidence: high)
  • Collection cost (baling + transport): $25-45/tonne (confidence: medium — highly location dependent)
  • Farmer willingness to collect year one: 20-50% of available (confidence: low — behavioural unknown, dragon)
  • Seasonality: highly seasonal, requires storage consideration (confidence: high)

Methane yield from straw

  • Without pretreatment: 60-100 m³/tonne (confidence: medium)
  • With lime pretreatment: 120-200 m³/tonne (confidence: medium — lab proven, field scale uncertain, dragon)
  • Real world discount vs theoretical: 20-40% (confidence: medium — digesters consistently underperform theory)

Anaerobic Digester

  • Capex: f(straw_volume, site_conditions, civil_works, local_labour_cost)
    • Estimated range: $800k-2.5m (confidence: medium)
    • Skew: pessimistic — civil works highly variable
    • Dragon: civil works cost in remote locations
  • Opex: f(maintenance_rate, labour_cost, lime_cost, throughput_utilisation)
    • Estimated range: $150-400k/year (confidence: low)
  • Lifetime: 20+ years with maintenance (confidence: high)
  • Failure modes: biological upset, gas handling failure, digestate overflow — all recoverable

Steam Methane Reformer

  • Capex: f(hydrogen_volume_required, purity_specification, local_installation_cost)
    • Estimated range: $500k-1.5m (confidence: low)
    • Dragon: hydrogen purity requirement for Haber-Bosch may require additional cleanup
  • Opex: f(methane_input_cost, catalyst_replacement_frequency, labour_cost)
  • Catalyst: nickel, commodity, replacement every 3-5 years
  • Efficiency: 70-80% methane to hydrogen conversion (confidence: medium)

PSA Nitrogen Separation

  • Capex: $150-400k (confidence: medium-high — commodity technology)
  • Opex: electricity dominated, low
  • Output purity: >99.5% nitrogen (confidence: high)

Haber-Bosch Reactor

  • Capex: f(ammonia_volume_required, pressure_specification, local_installation_cost)
    • Estimated range: $500k-3m+ (confidence: very low — this is the central dragon)
    • Skew: strongly pessimistic — small scale units not yet commodity
    • Dragon: single most important unknown in the entire system
  • Opex: f(energy_input, catalyst_replacement, labour, maintenance)
  • Catalyst: iron, commodity, known since 1909, replacement every 5-10 years
  • Energy requirement: ~10 MWh per tonne ammonia (confidence: high — well established)
  • Operating conditions: 150-300 bar, 400-500°C (confidence: high)
  • Small scale viability: physically proven, commercially uncertain (dragon)

Nitrogen Fertilizer

  • Current global commodity price: $300-400/tonne nitrogen (confidence: high)
  • Farm gate price after last mile (remote): $800-1,500/tonne nitrogen (confidence: medium)
  • Emergency price (supply disruption): $2,000-4,000+/tonne (confidence: low — scenario dependent)
  • Application efficiency: ~50% — half is lost to atmosphere or waterways (confidence: medium)

Electrolyser (alternative hydrogen path)

  • Capex: falling rapidly but not yet commodity at relevant scale (confidence: medium)
  • Stack lifetime: 60,000-100,000 hours claimed, real world uncertain (dragon)
  • Vendor concentration: high — not yet open market (dragon)
  • Dependency: requires reliable cheap electricity — problematic in remote locations
  • Recommendation: not primary path, monitor for future integration

Carbon Credits

  • Value: f(carbon_price, methane_volume_captured, verification_cost, political_stability)
  • Current price range: $20-50/tonne CO₂e (confidence: medium)
  • Dragon: politically volatile, cannot be relied upon as primary revenue stream
  • Methane warming equivalence: 25x CO₂ — captured methane generates significant credits

Relationships

Physical Flows

  • Straw → [collection, transport] → Lime pretreatment unit
    • Rate: f(farmer_uptake, collection_cost, seasonality)
    • Cost: $25-45/tonne (confidence: medium)
    • Dragon: uptake rate year one
  • Lime pretreatment unit → [treated straw] → Digester
    • Lime dosing: ~5-10% of straw weight
    • Yield improvement: 60-80% more methane (confidence: medium, lab proven)
  • Digester → [methane, digestate]
    • Methane yield: 120-200 m³/tonne treated straw (confidence: medium)
    • Digestate: high nutrient content, returned to farms
  • Methane → [6-12% of total] → SMR → Hydrogen
    • Conversion: 2.5 kg hydrogen per kg methane
    • Remainder: electricity generation

  • Methane → [88-94% of total] → Genset → Electricity

  • Hydrogen + Nitrogen (from PSA) → Haber-Bosch → Ammonia
    • Ratio: 3:1 hydrogen to nitrogen by mass
    • Energy: ~10 MWh/tonne ammonia
  • Ammonia → [storage, distribution] → Farms
    • Distribution cost: f(distance, diesel_price, infrastructure)
  • Digestate → [processing, transport] → Farms
    • Value: replaces lime and micronutrient purchases
    • Estimated value: $80-150/tonne (confidence: low)

Economic Flows

  • Fertilizer saving → Cooperative → Farmers
    • Value: f(ammonia_volume, current_fertilizer_price, distribution_cost)
  • Electricity → Grid → Revenue
    • Value: f(electricity_price, grid_access, surplus_volume)
    • Dragon: grid access in remote locations may be unavailable or prohibitively expensive
  • Methane capture → Carbon market → Revenue
    • Value: f(carbon_price, methane_volume, verification_cost)
    • Dragon: politically volatile
  • Straw delivery → Farmers → Fertilizer credit
    • Farmers paid in fertilizer credit not cash
    • Net cash cost to system: near zero if credit value ≥ collection cost

Actor Relationships

  • Farmer trust → Cooperative → Straw uptake rate
    • Trust builds over time, non-linear
    • Early adopters reduce risk perception for later adopters
    • Dragon: initial trust level unknown
  • Cooperative → Capital → System build
    • Governance already exists in many contexts
    • Capital appetite unknown without specific engagement (dragon)
  • Government → Emergency funding → Capital availability
    • Activated by emergency state
    • Can compress payback period to near zero
  • Open source installations → Model improvement → Reduced uncertainty
    • Each installation resolves dragons
    • Loot shared with all subsequent actors

Calculations

All calculations take distributions as inputs and produce distributions as outputs. Point estimates below are central estimates only.

Methane Available

methane_available = farm_hectares × straw_yield_per_ha × farmer_uptake_rate 
                    × lime_pretreatment_yield_improvement × digester_efficiency
                    × methane_content_of_biogas

Confidence: medium. Skew: pessimistic — multiple uncertain terms multiply.

Hydrogen Required for Full Fertilizer Replacement

hydrogen_required = nitrogen_demand_tonnes × 220 kg_H2_per_tonne_N

Confidence: high — well established chemistry.

Methane Required for Hydrogen (via SMR)

methane_for_SMR = hydrogen_required / (2.5 × SMR_efficiency)

Confidence: medium.

Surplus Methane for Electricity

surplus_methane = methane_available - methane_for_SMR
electricity_generated = surplus_methane × energy_content × genset_efficiency

Confidence: medium — depends on methane_available calculation.

System Capex

system_capex = digester_capex(straw_volume, site) 
             + pretreatment_capex(straw_volume)
             + SMR_capex(hydrogen_volume)
             + PSA_capex(nitrogen_volume)
             + haber_bosch_capex(ammonia_volume)    ← central dragon
             + storage_capex(ammonia_volume)
             + genset_capex(surplus_methane)
             + grid_connection_capex                 ← location dragon
             + civil_and_installation(site_conditions, remoteness)

Confidence: low overall — Haber-Bosch and grid connection dominate uncertainty.

System Opex

system_opex = lime_cost(straw_volume, lime_price)
            + labour_cost(automation_level, local_labour_rate)
            + maintenance_cost(capex, maintenance_rate)
            + catalyst_replacement(volume, replacement_frequency)
            + electricity_cost(self_powered_fraction)

Confidence: low — labour cost in remote locations particularly uncertain.

Annual Value Generated

annual_value = fertilizer_saving(ammonia_produced, fertilizer_price)
             + electricity_revenue(surplus_electricity, grid_price, grid_access)
             + carbon_credits(methane_captured, carbon_price)
             + digestate_value(digestate_volume, nutrient_content)
             - system_opex

Confidence: low-medium. Skew: optimistic risk — electricity revenue and carbon credits are dragons.

Payback Period

payback = system_capex / annual_value

Output is a distribution, not a point estimate. Range: 1-15 years depending on scenario state.

Pathways

Path A: Digestate Only

Description: Optimise capture and application of nitrogen already in waste stream. No synthesis.

Objects: Straw → Digester → Digestate → Farms (+ Methane → Genset → Electricity)

Properties:

  • Nitrogen replacement: 20-30% of synthetic fertilizer need (confidence: medium)
  • Capex: $800k-1.5m (confidence: medium-high)
  • Build time: 6-12 months
  • Dragon count: low

Viability: High across all scenarios. Viable today.

Option value: High — infrastructure built here is reused by Paths C and D. Does not foreclose any future path.

Forced moves after choosing: Straw collection logistics must be solved. Digestate application system required.

Path B: Digestate + Bulk Ammonia Hub

Description: Local system handles organic nitrogen. Remaining synthetic fertilizer purchased in bulk to town, not per farm.

Objects: Path A + Bulk ammonia storage + Cooperative purchasing

Properties:

  • Cost reduction vs current: 20-40% on purchased fertilizer (transport markup eliminated)
  • Additional capex: $200-400k for bulk storage (confidence: medium)
  • Dragon count: very low

Viability: High across all scenarios. Viable today. No novel technology.

Option value: High — preserves all future options, adds bulk storage infrastructure Path C needs.

Path C: Full Synthesis — SMR + Haber-Bosch

Description: Full nitrogen independence from local feedstock via methane → hydrogen → ammonia pathway.

Objects: Path A + Lime pretreatment + SMR + PSA + Haber-Bosch + Ammonia storage

Properties:

  • Nitrogen replacement: up to 100% (confidence: medium — depends on straw volume)
  • Capex: $4-12m (confidence: low — Haber-Bosch is central dragon)
  • Opex: $1.5-4m/year (confidence: low)
  • Dragon count: high — SMR purity, Haber-Bosch capex, labour, grid connection
  • All-commodity components: yes — iron catalyst, nickel catalyst, standard pressure vessels

Viability:

  • Normal market: marginal to viable depending on Haber-Bosch capex resolution
  • High fuel price: viable
  • Emergency: strongly compelling

Option value: Medium — forecloses electrolyser path if SMR infrastructure dominates site

Forced moves after choosing: Methane purity specification forces SMR design. SMR forces hydrogen cleanup specification. Haber-Bosch forces ammonia storage safety compliance.

Key dragon: Small scale Haber-Bosch unit cost. Everything else can be estimated with reasonable confidence. This cannot. First dragon to slay.

Path D: Electrolyser + Haber-Bosch

Description: Green hydrogen via electrolysis rather than SMR.

Objects: Path A (electricity only) + Electrolyser + PSA + Haber-Bosch + Ammonia storage

Properties:

  • Capex: higher than Path C currently (confidence: medium)
  • Opex: electricity price dominated
  • Dragon count: very high — electrolyser stack lifetime, electricity price, vendor concentration
  • Carbon accounting: cleaner than Path C

Viability:

  • Normal market: not viable currently
  • If electricity price falls and electrolyser commoditises: viable in 3-5 years
  • Emergency: unlikely — electrolyser supply chains are concentrated and fragile

Recommendation: Monitor. Design Path C to accept electrolyser as future substitution for SMR when economics improve. Do not depend on it.

Path E: Open Source Design Package

Description: Not a physical pathway. A meta-pathway that improves all others.

Objects: All of the above + Knowledge commons + Peer review network + Installation data feedback

Properties:

  • Reduces Haber-Bosch capex uncertainty as installations accumulate
  • Eliminates dragons progressively
  • Changes actor relationships — cooperative becomes knowledge node
  • Accelerates adoption across all contexts

Viability: Always viable. Zero marginal cost to share knowledge.

Relationship to other paths: Does not compete with Paths A-D. Amplifies all of them. Each installation is a team walking territory and returning loot.

Constraints

(Ambient, conditional, cross-cutting — not owned by any object)

Hard constraints (physical, invariant):

  • Nitrogen fixation requires energy input regardless of process — thermodynamic floor
  • Haber-Bosch requires elevated temperature and pressure — cannot be designed away
  • Straw is seasonal — storage or process scheduling required
  • Geography — distance between farms and facility is fixed

Soft constraints (regulatory, financial, behavioural — may dissolve):

  • Ammonia storage safety regulations — significant compliance cost, location dependent
  • Grid connection approval — Western Power or equivalent may refuse or price prohibitively
  • ACCU verification requirements — administrative burden, lag between action and credit
  • Cooperative governance rules — capital decisions require member agreement
  • Planning approval for industrial facility in agricultural zone

Conditional constraints (activated or dissolved by state change):

  • Emergency state dissolves: planning approval timelines, capital availability constraints, regulatory timelines
  • Emergency state activates: government funding, strategic priority status, fast-track approval
  • Farmer trust below threshold: straw uptake too low to sustain system — minimum viable cooperative size required
  • Carbon price below threshold: carbon revenue insufficient to contribute meaningfully to payback

Unknown constraints (dragons — may exist, not yet discovered):

  • Western Power grid capacity at specific Wheatbelt locations
  • Australian safety compliance cost for ammonia at this scale
  • CBH governance rules around novel capital investments
  • Water availability and quality requirements for SMR at specific sites

Phase Space

Current Position

Centralised fertilizer system dominant. Transport cost subsidy intact but eroding. Emergency risk real and growing. Local production technically viable but not yet demonstrated at relevant scale.

Symmetry Points (genuine open choices right now)

  1. Technology choice: SMR vs electrolyser for hydrogen. Genuine choice — path dependent, not reversible cheaply once infrastructure built
  2. Scale: Individual farm vs cooperative vs town. Cooperative is almost certainly right but not yet proven in this context
  3. Ownership model: Farmer-owned cooperative vs external operator vs open source commons
  4. First market: Which context has the right combination of need, ability to pay, and cooperative structure to be first?

Breaking Points (where system snaps)

  1. Haber-Bosch capex resolution: If a vendor quotes $500k-1m installed at relevant scale, Path C becomes compelling in normal market. If quote is $3m+, Path C is emergency-only.
  2. First cooperative commits: Changes actor behaviour in neighbouring cooperatives — demonstration effect
  3. Emergency trigger: Fertilizer price spike changes all calculations simultaneously, activates government funding
  4. Open source first release: Changes the nature of the problem from commercial to collaborative

Forced Moves

  • Choose SMR → forced to solve hydrogen purity for Haber-Bosch
  • Choose cooperative ownership → forced to engage CBH governance process
  • Build Path A infrastructure → forced to solve digestate application logistics (but this is a good forcing)
  • Choose Path C before Path A → forced to solve farmer trust and straw logistics simultaneously with synthesis — high risk

Dragons (priority order)

  1. Haber-Bosch unit cost at relevant scale: Central dragon. Slay first. Get vendor quotes.
  2. Real world methane yield from lime-pretreated straw at field scale: Lab proven, field uncertain
  3. Grid connection cost and feasibility at specific locations: May eliminate electricity revenue entirely
  4. Farmer straw uptake rate year one: Behavioural, unknown without pilot
  5. CBH cooperative capital appetite: Institutional, unknown without direct engagement
  6. Australian ammonia safety compliance cost: Regulatory, location specific
  7. SMR hydrogen purity output vs Haber-Bosch requirement: Technical, resolvable with engineering study
  8. Electrolyser stack lifetime at continuous operation in remote conditions: Monitor, not urgent