Doc #98 — Glass Production from NZ Materials

---

RECOMMENDED ACTIONS

First week (Phase 1)

1. Classify O-I Glass Penrose workforce as essential personnel. The approximately 200–350 workers at the plant — furnace operators, batch engineers, forming machine operators, quality control staff, maintenance personnel — hold knowledge that cannot be replaced.⁴ Prevent redeployment.
2. Verify raw material inventories at the Penrose plant — tonnes of silica sand, soda ash, limestone, cullet (recycled glass), and minor batch materials (feldspar, dolomite, colouring/decolouring agents) on site and in NZ supplier stocks. This determines how long the plant can continue without intervention.
3. Verify soda ash stocks nationally — O-I plant stocks, any NZ distributor stocks, industrial chemical suppliers (soda ash is used in water treatment, pulp and paper, and various industrial processes). Total NZ soda ash inventory is a critical strategic number.
4. Secure the Penrose plant site — prevent unauthorised removal of materials or equipment.
5. Assess furnace condition — determine remaining campaign life of the glass furnace(s). A glass furnace operates continuously for years; forced shutdown and restart is possible but damages the refractory lining and shortens furnace life substantially.⁵

First month (Phase 1)

6. Implement reduced production rate at Penrose, informed by soda ash and other consumable inventories. Prioritise production of food preservation jars and bottles (Doc #78) over other glass products.
7. Begin national glass collection and recycling program. Cullet (broken or waste glass) reduces the need for virgin raw materials and lowers melting energy by approximately 2.5–3% for every 10% of cullet in the batch — a batch running entirely on cullet would use roughly 25–30% less energy than a virgin-batch furnace.⁶ NZ has an existing kerbside glass recycling system; expand this to total collection of all waste glass, including building demolition glass.
8. Assess Parengarenga sand supply chain — verify mining operation status, transport logistics (the deposit is remote — northern tip of the North Island, approximately 450 km from Auckland by road), and current sand stocks at any NZ locations.⁷
9. Begin knowledge capture from O-I Glass technical staff — document furnace operation, batch formulation, forming techniques, quality control procedures, and maintenance practices specific to the Penrose plant.
10. Inventory all laboratory glassware nationally — through the skills and asset census (Doc #8). Every piece of borosilicate glass (Pyrex-type) in NZ universities, hospitals, industrial laboratories, and school science departments is a strategic asset.

First 3 months (Phase 1)

11. Assess NZ soda ash production pathways — commission an engineering assessment of: (a) Solvay process feasibility using NZ salt and limestone; (b) seaweed ash (kelp ash) as a soda source; (c) wood ash as a potash source (for potash glass as a substitute); (d) any NZ trona or natron deposits. This assessment determines the long-term viability of NZ glass production.
12. Begin experimental small-scale glassblowing using existing NZ materials and skills. NZ has a small but real community of art glass workers and scientific glassblowers — identify them through the census and begin trials of laboratory glassware production from available materials.
13. Assess flat glass stocks — window glass in NZ distributor warehouses, at glaziers, and in building supply stores. Establish a rationing framework for flat glass allocation to essential uses (hospital windows, greenhouse construction for food production under nuclear winter).
14. Coordinate with Doc #97 (Cement) — limestone supply for glass production shares sources with cement production; coordinate allocation.

First year (Phase 1, entering Phase 2)

15. Achieve pilot-scale soda ash production if the Solvay process or an alternative is assessed as feasible. Even small quantities extend the glass production timeline.
16. Establish a standard bottle and jar pattern library — prioritise production of standard-size food preservation jars with reusable seal surfaces, and standard bottles for pharmaceutical and chemical use.
17. Begin training glassblowing apprentices — both machine-assisted (for the Penrose plant production line) and hand glassblowing (for laboratory glassware and specialty items). Hand glassblowing requires years of practice to reach competence.⁸
18. Trial production of basic flat glass — either at the Penrose plant (if adaptable) or using a smaller furnace and hand-drawing or cylinder-blowing methods.
19. Assess NZ feldspar and dolomite sources — secondary batch ingredients that improve glass durability and workability.

Phase 2–3 (Years 1–7)

20. Scale up soda ash production to support continued glass manufacturing as imported soda ash stocks deplete.
21. Establish laboratory glassware production workshop — dedicated facility with trained scientific glassblowers producing distillation apparatus, beakers, flasks, condensers, and other items critical for Doc #113 (sulfuric acid), Doc #119 (pharmaceutical production), and Doc #135 (computer construction).
22. Develop flat glass capability for window production — the float glass process requires infrastructure NZ does not have (see Section 7); alternatives include crown glass, cylinder glass, or drawn sheet glass methods.
23. Begin optical glass trials — producing lens blanks suitable for grinding into corrective lenses (Doc #110) and basic scientific instrument optics.
24. If trade with Australia develops, prioritise import of soda ash, boron compounds (for borosilicate glass), and specialty glass raw materials.

Phase 4+ (Years 7+)

25. Full domestic glass supply chain operational — NZ sand, NZ-produced soda ash, NZ limestone, domestic furnace construction and relining capability.
26. Expanded flat glass production for housing stock window replacement as nuclear winter eases and construction resumes.
27. Optical glass grinding workshop(s) producing corrective lenses at scale (Doc #110).
28. Fiberglass production trials — glass fiber for insulation and composite materials, if NZ can develop continuous fiber drawing capability.

---

ECONOMIC JUSTIFICATION

The cost of not producing glass

Glass is an enabling material — its value lies not in the glass itself but in what it makes possible. Without glass:

• Food preservation collapses to salt-and-smoke methods only. Glass jars with hermetic seals are the primary reusable container for home canning and food preservation under indefinite reuse. Metal lids corrode and rubber seals degrade, but the glass jar itself is effectively permanent. Every jar lost to breakage without replacement is a permanent reduction in preservation capacity (Doc #78). If NZ’s stock of approximately 5–15 million preserving jars degrades over decades through breakage without replacement, the food preservation system progressively loses a critical capability.⁹
• Chemical and pharmaceutical production stalls. Sulfuric acid (Doc #113), pharmaceutical intermediates (Doc #119), and dozens of other chemical processes require glass reaction vessels, distillation columns, and storage containers. Metal and ceramic alternatives exist for some applications but glass is the only material that is transparent (allowing visual monitoring of reactions), chemically inert to most acids and bases, and formable into complex shapes (condensers, graduated cylinders, separating funnels). Without laboratory glassware, NZ’s ability to develop domestic chemical industry is severely constrained.
• Vision correction becomes impossible to replace. Approximately 1.8–2.2 million New Zealanders use corrective lenses.¹⁰ Existing eyeglasses break, scratch, and require replacement. Without glass lens production, a progressively larger fraction of the population loses effective vision correction — with consequences for productivity, safety, and quality of life that compound over decades (Doc #110).
• Window replacement stops. NZ’s approximately 1.8 million dwellings contain an estimated 20–40 million individual window panes.¹¹ Under nuclear winter conditions, window glass breakage from storms, accidents, or thermal stress cannot be replaced without glass production. Every broken window that is boarded up reduces natural lighting (already diminished under nuclear winter cloud cover) and — if poorly sealed — reduces thermal insulation, increasing heating demand.

Labour cost of glass production

Operating the Penrose plant at reduced capacity requires approximately 150–300 direct workers plus supply chain labour (sand mining, transport, soda ash production) — perhaps 200–500 total person-equivalents depending on production rate and the extent of domestic raw material processing required.¹²

Breakeven: Glass production, like cement production (Doc #97), pays for itself immediately because the plant already exists and the workforce already has the skills. The investment is in operational continuity, not construction. The additional investment required — developing soda ash production, training new glassblowers, building flat glass capability — represents perhaps 10–30 person-years of development effort over Phase 2–3, paid back many times over by the products enabled.

The real question is not whether glass production is worth the labour, but how to allocate limited soda ash stocks and furnace capacity across competing glass products. A jar that preserves 1 kg of food for a year versus a window pane that reduces heating demand versus a laboratory flask that enables pharmaceutical production — these are the allocation decisions that matter, not whether to produce glass at all.

---

1. NZ’S EXISTING GLASS INDUSTRY

1.1 O-I Glass New Zealand — Penrose, Auckland

O-I Glass (formerly Owens-Illinois, trading in NZ as O-I Glass New Zealand, and historically as ACI Glass Packaging NZ) operates a glass container manufacturing plant at Penrose, an industrial suburb of Auckland.¹³ This is NZ’s only significant glass manufacturing facility.

Products: The Penrose plant produces glass bottles and jars — primarily for the NZ food and beverage industry (beer bottles, wine bottles, spirit bottles, food jars, sauce bottles). The plant does not produce flat glass (windows), laboratory glassware, or optical glass.¹⁴

Process: Container glass manufacturing at Penrose follows the standard industry process:

1. Batch preparation: Raw materials (silica sand, soda ash, limestone, cullet, and minor ingredients) are weighed and mixed to a precise formulation.
2. Melting: The batch is fed into a continuous glass furnace — a large refractory-lined tank — where it is melted at approximately 1,500–1,600°C. The furnace operates continuously, 24 hours a day, 365 days a year. A glass furnace campaign (continuous operation before the refractory lining must be rebuilt) typically lasts 10–15 years.¹⁵
3. Conditioning: Molten glass flows from the melting tank through a forehearth (a channel) where it is cooled to a uniform working temperature (approximately 1,050–1,200°C depending on the product).
4. Forming: Gobs of molten glass are cut and dropped into moulds on forming machines (IS machines — Individual Section machines) that shape the glass into bottles or jars using a combination of blowing and pressing. This is a highly automated process producing bottles at rates of 100–400 per minute per machine.¹⁶
5. Annealing: Formed containers pass through an annealing lehr — a long, temperature-controlled oven that slowly cools the glass from forming temperature to room temperature over approximately 30–60 minutes. Annealing relieves internal stresses that would otherwise make the glass fragile.
6. Inspection and packing: Finished containers are inspected (automated inspection machines check for cracks, dimensional defects, and visual flaws) and packed for distribution.

Capacity: The Penrose plant’s production capacity is approximately 150,000–200,000 tonnes of glass containers per year — a rough estimate based on typical container plant sizes and NZ market data.¹⁷ NZ’s total glass container consumption is approximately 200,000–250,000 tonnes per year, with the balance imported (primarily from Australia and Asia).¹⁸

Workforce: The plant employs approximately 200–350 people, including operators, engineers, technicians, and support staff.¹⁹ Key specialist roles include: furnace operators and hot-end engineers (who manage the melting process), batch plant operators, IS machine operators and setup technicians, mould technicians, quality control laboratory staff, and maintenance tradespeople (electrical, mechanical, refractory).

Furnace energy: Glass furnaces use a combination of gas or oil firing and electric boosting. The Penrose furnace is believed to use natural gas as the primary fuel with electric boosting — resistive heating elements submerged in the molten glass that supplement the gas firing.²⁰ NZ has domestic natural gas from the Taranaki Basin (see Doc #89, Section 8.3). The electric boosting component draws from the NZ grid.

Conversion to all-electric: All-electric glass melting furnaces exist and are increasingly used globally. NZ’s renewable grid makes all-electric glass melting a realistic long-term option, eliminating the natural gas dependency. All-electric furnaces are smaller (limited by electrode current capacity) and typically used for specialty glass, but technology for larger all-electric furnaces is advancing. Converting or replacing the Penrose furnace with all-electric melting is a Phase 3–4 possibility that would remove the fossil fuel dependency entirely.²¹

1.2 Other NZ glass capability

Beyond O-I Penrose, NZ has:

• Art glass studios and craft glassblowers: A small community of glass artists and craft workers in Auckland, Wellington, Christchurch, and other centres. These individuals work with furnaces, glory holes (reheating chambers), and hand tools at a small scale. Their skills in hand glassblowing are directly relevant to laboratory glassware production.
• Scientific glassblowers: A very small number of specialist scientific glassblowers working at universities (particularly University of Auckland, University of Canterbury, Victoria University of Wellington) and in industrial laboratories. These individuals can fabricate custom borosilicate (Pyrex-type) glassware — distillation apparatus, custom reaction vessels, repairs to broken equipment. This is one of the rarest and most valuable skills in recovery NZ.²²
• Glaziers and window installers: NZ has a substantial glazing trade — people who cut, fit, and install flat glass. These tradespeople do not manufacture glass but they understand its properties and handling. They are the workforce for window glass distribution and installation.
• Flat glass processing: Some NZ companies cut, temper, laminate, and insulate flat glass — but from imported sheet. NZ does not produce flat glass from raw materials.²³

1.3 What NZ does NOT have

NZ has no domestic capability for:

• Float glass production — the process that produces modern flat glass (windows, mirrors, architectural glass). Float glass plants are enormous capital investments.
• Borosilicate glass production — the heat-resistant glass (Pyrex, Schott Duran) used for laboratory equipment and ovenware. This requires boron compounds (borax, boric acid) that NZ does not produce.
• Optical glass production — high-purity, precisely formulated glass for lenses, prisms, and optical instruments.
• Glass fiber production — continuous glass filaments for insulation and composite materials.
• Soda ash production — the key flux for all common glass types.

All of these represent gaps that must be addressed through development (Sections 5–9) or through trade (if it materialises).

---

2. RAW MATERIALS

2.1 Silica sand

Silica sand (SiO₂) is the primary ingredient in all glass — typically 70–75% of the batch by weight for soda-lime glass.²⁴ Glass-grade silica sand must be high purity (>95% SiO₂, preferably >98%), low in iron oxide (Fe₂O₃ < 0.05% for clear glass — iron imparts a green tint), and of appropriate grain size (typically 0.1–0.5 mm).

Parengarenga Harbour: NZ’s premier silica sand deposit is at Parengarenga Harbour, near Cape Reinga at the northern tip of the North Island. This deposit has been mined for decades, primarily for export to Australian and Asian glass manufacturers.²⁵ The Parengarenga sand is world-class quality — approximately 96–99% SiO₂ with very low iron content, making it suitable for clear glass production. The deposit is large — estimated at tens of millions of tonnes of recoverable sand, sufficient for centuries of NZ glass production at any plausible rate.²⁶

Mining and transport: Parengarenga sand has been mined by dredging from the harbour and from adjacent beach and dune deposits. The mining operation involves dredging equipment (or dragline excavation from beach/dune faces), a washing plant to remove salt and clay contamination, drying or drainage infrastructure, stockpile areas, and loading facilities for vessel or truck transfer — each of these represents a capital item that must be assessed and maintained under recovery conditions. The challenge is distance: Parengarenga is approximately 450 km by road from Auckland (the Penrose plant location), via roads that are partly unsealed and winding in the Far North.²⁷ Pre-event transport has been by coastal barge or truck. Under recovery conditions, coastal barging from Parengarenga to Auckland is the most efficient transport method — small coastal vessels can carry hundreds of tonnes per trip. Road transport is feasible but fuel-intensive for bulk materials.

Alternative NZ sand sources: Other silica sand deposits exist in NZ, though generally of lower purity than Parengarenga:²⁸

• Pakiri and Mangawhai (Northland coast): Beach sand deposits with moderate silica content. May be suitable for non-optical glass (green or amber bottles) where iron content is less critical.
• West coast North Island (ironsand): These deposits are predominantly titanomagnetite (Doc #89) and are unsuitable for glass — far too much iron. Not a viable glass sand source.
• Waikato region: Some inland sand deposits with variable silica content.
• Various quartz outcrops: NZ has quartz-bearing rocks in several locations. These could be crushed and processed to produce high-purity silica, but at significantly higher cost and effort than mining natural sand.

Assessment: Parengarenga sand is the obvious primary source for NZ glass production. The deposit is large, high quality, and already established for mining. The transport logistics are a challenge but not a barrier — this is a supply chain management problem, not a materials availability problem. For the foundry at Doc #93, NZ’s silica sand deposits were noted for moulding sand; the glass industry’s requirements are more stringent (higher purity) but the Parengarenga deposit meets them.

2.2 Soda ash (sodium carbonate)

Soda ash (Na₂CO₃) is the critical gap in NZ’s glass raw material chain. Soda ash is the principal flux in soda-lime glass, reducing the melting temperature of silica from approximately 1,700°C (impractical for sustained production) to approximately 1,500–1,600°C in the furnace. Standard soda-lime glass contains approximately 12–15% Na₂O by weight, derived from soda ash in the batch.²⁹

NZ’s soda ash supply: NZ does not produce soda ash. All soda ash used in NZ is imported — primarily from the USA (which has enormous natural trona deposits in Wyoming) and from China.³⁰ NZ’s total soda ash consumption (for all industrial uses — glass, water treatment, detergents, pulp and paper, chemical processes) is probably in the range of 30,000–60,000 tonnes per year, with the glass industry being the largest consumer.³¹ In-country stocks at the time of the event depend on import timing and distributor inventories — perhaps a few months’ supply if stocks are managed well.

This is the single most important raw material constraint on NZ glass production. Without soda ash, the Penrose plant cannot produce soda-lime glass. Section 5 addresses domestic soda ash production pathways in detail.

2.3 Limestone (calcium carbonate)

Limestone (CaCO₃) provides the calcium oxide (CaO) component of soda-lime glass — typically 8–12% CaO by weight. Calcium oxide improves the chemical durability of glass (without it, soda-lime glass would be water-soluble — early medieval glass made with soda and sand but insufficient lime deteriorated rapidly).³²

NZ has abundant limestone. The same deposits that serve the cement industry (Doc #97) — Golden Bay/Portland in Northland, Waikato deposits, and others — are suitable for glass production. Limestone for glass must be relatively low in iron (to avoid green tint in clear glass) but the purity requirements are less stringent than for optical glass. Limestone availability is not a constraint.

2.4 Minor batch ingredients

Standard soda-lime glass batches typically include small amounts of other materials:

• Feldspar (or nepheline syenite): Provides alumina (Al₂O₃, typically 1–3% in the glass), which improves chemical durability and working properties. NZ has feldspar deposits in several locations (Fiordland, West Coast, Coromandel) though none have been commercially mined at large scale recently. Feldspar requirements for glass are modest in tonnage.³³
• Dolomite: Provides both CaO and MgO (magnesium oxide, typically 1–4% in glass). MgO improves chemical resistance and reduces devitrification tendency. NZ has dolomite deposits (Doc #89, footnote 38).
• Iron oxide (deliberate addition): For amber and green bottle glass, iron oxide is actually desired — it provides the colour and UV protection. NZ has iron-bearing sands in abundance.
• Decolourising agents: Selenium, cobalt oxide, or manganese dioxide are used in small quantities to neutralise the green tint from trace iron in clear glass. These are specialty chemicals that NZ imports. Without them, “clear” glass produced from anything but the purest Parengarenga sand will have a slight green tint — functional for most container and window uses but a real limitation for optical glass (where even trace colouration distorts colour perception through lenses) and for pharmaceutical glass where visual inspection of the contents is required. The performance gap is minor for food jars and windows, more significant for laboratory and optical applications. Note: amber and green bottle glass deliberately uses iron and chromium colourants — these NZ can source from ironsand and other minerals, so the pharmaceutical packaging priority should shift toward amber glass for light-sensitive medications rather than clear glass where possible.
• Cullet (recycled glass): The most important “ingredient” after the primary raw materials. Cullet substitutes directly for virgin batch materials (reducing the need for sand, soda ash, and limestone proportionally) and melts at a lower temperature, reducing energy consumption by approximately 2.5–3% for every 10% of cullet in the batch.³⁴ NZ’s existing glass recycling infrastructure collects a significant fraction of post-consumer glass — approximately 65,000–80,000 tonnes per year.³⁵ Under recovery conditions, all waste glass should be collected and recycled.

2.5 Boron compounds (for borosilicate glass)

Borosilicate glass — known commercially as Pyrex (Corning), Duran (Schott), or Kimax (Kimble) — contains approximately 12–15% boron trioxide (B₂O₃), which gives it extremely low thermal expansion (making it resistant to thermal shock — critical for laboratory use where glassware is heated and cooled repeatedly).³⁶

NZ has no known commercial boron deposits. Boron minerals (borax, colemanite, ulexite) are found primarily in Turkey, the USA (California), and South America. Without boron, NZ cannot produce borosilicate glass.

Consequence: NZ’s existing stock of borosilicate laboratory glassware is finite and irreplaceable domestically. Every broken Pyrex flask or condenser is a permanent loss unless trade develops. Conservation measures — careful handling, designated glassware storage, repair of cracked items where possible — are essential from Day 1. Laboratory glassware produced domestically will be soda-lime glass, which is less resistant to thermal shock and chemical attack than borosilicate. This performance gap is significant for laboratory use (see Section 8).

Australia has some boron deposits (in Western Australia) and, if Tasman trade develops (Doc #150), boron compounds should be a priority import item — small in volume, high in strategic value.

---

3. THE SODA ASH PROBLEM

This section addresses NZ’s most critical glass production challenge in detail, because without soda ash (or a substitute flux), glass production cannot continue.

3.1 The Solvay process

The Solvay process (also called the ammonia soda process) is the standard industrial method for producing soda ash from salt (sodium chloride) and limestone (calcium carbonate). It was developed in Belgium in the 1860s and remains the dominant production method worldwide outside the USA (where natural trona deposits are more economical).³⁷

Chemistry overview:

The overall reaction is: 2 NaCl + CaCO₃ → Na₂CO₃ + CaCl₂

In practice, this is achieved through a series of intermediate reactions involving ammonia as a recyclable intermediary:

1. Brine (NaCl solution) is saturated with ammonia (NH₃).
2. Carbon dioxide (CO₂) from limestone calcination is passed through the ammoniated brine, precipitating sodium bicarbonate (NaHCO₃).
3. The sodium bicarbonate is filtered and heated (calcined) to produce soda ash (Na₂CO₃), releasing CO₂ and water.
4. The ammonia is recovered from the ammonium chloride (NH₄Cl) byproduct by reacting it with lime (CaO from limestone calcination), regenerating ammonia for reuse.

NZ feasibility:

• Salt: NZ has sea salt production capability — solar evaporation of seawater or direct evaporation using heat. NZ also has inland salt springs and saline aquifers in some areas. The salt requirement for a Solvay plant is substantial — approximately 1.5 tonnes of salt per tonne of soda ash.³⁸ Producing this quantity of salt domestically from seawater requires extensive shallow evaporation ponds (typically hectares of surface area per tonne of daily output), earthworks or pond lining, harvesting equipment, washing and drying facilities, and a workforce to operate the system through seasonal production cycles. NZ’s climate in drier regions (the Marlborough Sounds coast, Hawke’s Bay, and parts of Canterbury) supports solar evaporation, but yields per unit area are lower than in tropical climates, and salt production is seasonal — meaning storage infrastructure is also required to buffer supply across the year.³⁹
• Limestone: Abundant in NZ (Section 2.3 and Doc #97).
• Ammonia: This is the real constraint. The Solvay process requires a reliable supply of ammonia. NZ does not currently produce ammonia domestically — it was formerly produced at the Kapuni ammonia-urea plant in Taranaki, which has operated at various times but may not be running at the time of the event.⁴⁰ Ammonia synthesis (the Haber-Bosch process) requires hydrogen, nitrogen, high pressure, high temperature, and a catalyst — it is one of the most demanding industrial chemical processes and is classified as a multi-decade development project in this library (Doc #114). However, the Solvay process recycles its ammonia — losses are modest (a few percent per cycle), so even a small ammonia supply allows sustained operation. If any NZ ammonia stocks exist (in fertiliser stocks, industrial chemical inventories, or refrigeration systems), they could prime a Solvay plant that then operates with minimal ongoing ammonia input.
• Equipment: A Solvay plant requires carbonation towers, filters, calciners, ammonia recovery stills, pumps, and piping — all of which must handle corrosive solutions (brine, ammonium chloride, carbonate solutions). Stainless steel and glass-lined equipment would be ideal but NZ’s ability to fabricate these is limited. Carbon steel with protective linings, combined with glass equipment where available, would be the practical approach.

Assessment: The Solvay process is technically feasible for NZ in principle — all the fundamental materials exist — but establishing the capability represents a major industrial chemistry project requiring 2–5 years of development, significant engineering effort, and solution of the ammonia supply question. This is a Phase 2–3 project, not a quick fix. Feasibility: [B–C] — the chemistry is known and NZ has the inputs, but the engineering and ammonia supply chain are difficult.

3.2 Seaweed ash (kelp ash)

Before the industrial Solvay process, soda ash was produced by burning seaweed (kelp) and extracting sodium carbonate from the ash. This was a major industry in Scotland, Ireland, and Scandinavia from the 17th to 19th centuries.⁴¹ Kelp ash typically contains 3–5% sodium carbonate along with potassium carbonate, sodium chloride, and other salts. The sodium carbonate is extracted by dissolving the ash in water, filtering, and evaporating to crystallise.

NZ feasibility: NZ has extensive coastline with significant kelp and seaweed populations. Bull kelp (Durvillaea species) is abundant around southern NZ coasts. However, there are important limitations:

• Yield is low: Producing 1 tonne of soda ash from kelp ash requires approximately 20–30 tonnes of dry kelp, which in turn requires approximately 80–150 tonnes of wet kelp.⁴² NZ glass production at even a modest rate of 50,000 tonnes per year would require thousands of tonnes of soda ash per year — which would require harvesting hundreds of thousands of tonnes of wet kelp. This is an enormous logistical effort.
• Environmental impact: Intensive kelp harvesting could damage coastal ecosystems. Sustainable harvest rates would likely limit soda production to a fraction of glass industry needs.
• Quality: Kelp ash soda is impure and requires further processing to achieve glass-grade quality.

Assessment: Kelp ash is not a scalable solution for industrial glass production, but it could supplement soda ash supply for small-scale specialty glass production (laboratory glassware, optical glass blanks) where volumes are small and quality can be managed through additional purification. It may also serve as an interim source while Solvay process capability is developed.

3.3 Wood ash (potash)

Wood ash is rich in potassium carbonate (potash, K₂CO₃) rather than sodium carbonate. Potash can substitute for soda ash as a flux in glass — the resulting “potash glass” or “forest glass” was the standard glass of continental Europe for centuries.⁴³ Potash glass has slightly different properties from soda-lime glass: higher viscosity at working temperatures (making it harder to blow quickly but easier to work for longer — advantageous for hand glassblowing), slightly higher refractive index, and different chemical durability profile.

NZ feasibility: NZ has abundant wood resources (radiata pine plantation forests, native timber). Wood ash from NZ timber species contains variable but meaningful potash content — typically 5–15% K₂CO₃ in hardwood ash, lower in softwood ash (radiata pine).⁴⁴ The potash must be extracted by leaching the ash with water, filtering, and evaporating — the traditional “potash” process that gave the chemical its name.

Yield: Like kelp ash, the yield is low. Producing 1 tonne of potash requires burning approximately 100–200 tonnes of wood (depending on species) and leaching the ash.⁴⁵ This is labour-intensive but uses a resource (wood) that NZ has in great abundance. NZ’s forestry industry generates significant wood waste — sawdust, bark, offcuts — that could serve as potash feedstock without diverting useful timber.

Assessment: Potash from wood ash is a viable flux source for small-to-medium-scale glass production. It does not scale to replace soda ash entirely at industrial volumes, but it can support hand glassblowing workshops producing laboratory glassware, specialty bottles, and other products where production rates are measured in tonnes per year rather than hundreds of thousands of tonnes. Combined with the Solvay process for bulk soda ash, potash glass offers a complementary pathway that can begin sooner (the technology is simpler) and at smaller scale.

3.4 Sodium sulfate (Glauber’s salt)

Sodium sulfate (Na₂SO₄) can partially substitute for soda ash in glass batches — it provides sodium but also introduces sulfur into the glass (used as a refining agent to remove bubbles, but excessive sulfur causes amber colouration). Glass batches historically and currently include small amounts of sodium sulfate (salt cake) as a refining agent.⁴⁶

NZ may have small natural sources of sodium sulfate (in evaporite deposits or geothermal brine) and can produce it from sulfuric acid (Doc #113) reacted with salt. This is a supplementary source rather than a primary replacement for soda ash.

3.5 Summary of soda ash pathways

Pathway	Volume potential	Timeline	Feasibility
Solvay process	Industrial (tens of thousands of tonnes/year)	2–5 years to develop	[B–C] — ammonia supply is the key constraint
Kelp ash	Small (tens of tonnes/year of soda ash)	Months to begin	[B] — labour-intensive but proven technology
Wood ash (potash)	Small to moderate (hundreds of tonnes/year)	Months to begin	[A–B] — NZ has abundant wood; proven historical process
Sodium sulfate	Supplementary only	Depends on sulfuric acid availability	[C] — secondary pathway
Trade with Australia	Potentially large	Depends on maritime trade development	Unknown

Recommendation: Pursue all pathways in parallel. Begin potash and kelp ash production immediately (Phase 1–2) to support small-scale specialty glass. Develop Solvay process capability as a Phase 2–3 major project for industrial-scale soda ash. Coordinate with Doc #113 (sulfuric acid) and Doc #114 (ammonia) on the chemical industry prerequisites.

---

4. FURNACE TECHNOLOGY AND ENERGY

4.1 The glass tank furnace

The Penrose plant’s furnace is a continuous tank furnace — a large rectangular refractory-lined basin in which glass is melted, refined (bubbles rise out of the melt), and conditioned (brought to uniform working temperature). Raw batch materials are fed in at one end; molten glass is drawn out at the other. The furnace operates 24 hours a day, typically for 10–15 years before the refractory lining is too deteriorated to continue and the furnace must be rebuilt.⁴⁷

Capacity: A medium-sized container glass furnace like the one at Penrose may produce 200–500 tonnes of glass per day (roughly 70,000–180,000 tonnes per year). At reduced recovery production rates, throughput would be lower.⁴⁸

Energy: Glass melting is energy-intensive. The theoretical minimum energy to melt glass from batch is approximately 2.5–2.8 GJ per tonne (varying with batch composition and cullet ratio). Practical furnace energy consumption ranges from approximately 4–8 GJ per tonne for gas-fired furnaces and 3–5 GJ per tonne for electric furnaces. A furnace producing 300 tonnes per day at 5 GJ/tonne requires approximately 1,500 GJ of thermal energy per day — equivalent to approximately 60,000 m³ of natural gas or 17 MW of continuous electrical power.⁴⁹

NZ energy options:

• Natural gas: From the Taranaki Basin (Doc #89, Section 8.3). Gas supply to Auckland via the North Island pipeline is assumed operational under baseline conditions. Gas is the current primary fuel for the Penrose furnace.
• Electricity: NZ’s renewable grid can supply glass melting. All-electric glass furnaces use molybdenum or tin oxide electrodes submerged in the glass melt, passing current directly through the molten glass (which is electrically conductive at high temperatures). Electric melting is technically mature and well-suited to NZ’s energy situation. At 17 MW continuous for a 300 tonne/day furnace, annual electricity consumption would be approximately 150 GWh — roughly 0.35% of NZ’s total generation. This is a significant but manageable load.⁵⁰
• Hybrid (gas + electric): The existing Penrose furnace likely uses electric boosting (10–20% of energy from submerged electrodes, the rest from gas firing). This hybrid approach could transition gradually toward higher electric fraction as gas supply becomes uncertain.

Assessment: Energy for glass melting is available from NZ sources — either natural gas or electricity. The grid is the long-term solution. Converting to all-electric melting requires new electrode systems and potentially a new furnace design, but is technically feasible and eliminates fossil fuel dependency.

4.2 Small-scale pot furnaces

For specialty glass production — laboratory glassware, optical glass, art glass — small pot furnaces are appropriate. A pot furnace holds one or more refractory crucibles (pots), each containing a batch of molten glass. The furnace is heated externally by gas, electric elements, or (historically) wood or coal.

Capacity: A single pot typically holds 50–500 kg of glass — enough for a day’s production by a hand glassblowing team. Multiple pots in one furnace allow different glass compositions to be melted simultaneously.⁵¹

NZ construction: Pot furnaces are within NZ’s construction capability. The refractory structure can be built from firebrick (NZ fireclay — see Doc #89, Section 5 and Doc #93) and the pots from high-alumina refractory (more difficult — imported material, or potentially producible from NZ alumina-bearing clays with development). Heating by electric resistance elements connected to the grid is the most practical NZ approach.

For laboratory glassware production: A small pot furnace facility with 2–3 workstations, each manned by a trained glassblower, could produce significant quantities of laboratory glassware — beakers, flasks, condensers, distillation apparatus, graduated cylinders, storage bottles — from soda-lime or potash glass. This is a Phase 2 development priority.

4.3 Furnace refractory

Glass furnaces are lined with specialised refractory materials that must resist not only high temperatures but also chemical attack from molten glass (which is highly corrosive to most refractory materials at melting temperatures):⁵²

• Crown (roof): Silica brick — NZ has silica sand (from Parengarenga or other sources) that can be processed into refractory-grade silica brick.
• Sidewalls (above glass line): AZS (alumina-zirconia-silica) refractory — the standard modern glass furnace wall material. AZS is an imported specialty product. NZ cannot produce it. When existing stocks are exhausted, alternatives include dense alumina brick or high-alumina castable, both of which have shorter service life in contact with molten glass.
• Tank bottom and lower walls (in contact with glass): AZS or dense alumina/zirconia refractories. The most demanding application — these refractories must resist continuous corrosion by molten glass for years.
• Regenerator or recuperator: Refractory brick for heat recovery systems. Less demanding than tank refractories — basic fireclay is often adequate.

Assessment: Furnace refractory is a serious long-term constraint — the same refractory supply challenge faced by NZ Steel (Doc #89) and the cement industry (Doc #97). AZS refractories are a specialty import. NZ’s domestic refractory capability (fireclay, silica, dolomite, potentially seawater magnesia) can serve lower-duty applications but not the most demanding glass-contact zones. As AZS stocks deplete, furnace campaigns will shorten (more frequent rebuilds) and glass quality may decline (refractory contamination of the melt). This is manageable — shorter campaigns with domestic refractories are better than no furnace at all — but it is an honest performance degradation.

---

5. CONTAINER GLASS — BOTTLES AND JARS

5.1 Why container glass matters for recovery

Glass containers are the only fully reusable, hermetically sealable, chemically inert food preservation vessel that NZ can produce. Metal cans require tin plate (imported — NZ does not produce tin),⁵³ rubber seals degrade, and plastic containers cannot be produced domestically. A glass jar with a properly designed closure can be reused hundreds of times for canning, pickling, and preserving food (Doc #78).

Under the baseline scenario, NZ’s existing stock of glass jars and bottles is large — probably tens of millions of individual containers in homes, commercial kitchens, and warehouses.⁵⁴ But glass breaks. At even a modest breakage rate of 2–5% per year, the stock declines continuously. Without replacement production, NZ’s glass container stock halves within 15–30 years. Given the importance of glass jars for food preservation, this slow attrition is a strategic concern.

5.2 Production priorities

Under recovery conditions, the Penrose plant should prioritise:

1. Standard preserving jars — wide-mouth jars (500 mL and 1 litre sizes) suitable for home canning with metal or glass lids. These serve the food preservation system (Doc #78) directly.
2. Standard bottles for pharmaceutical and chemical use — amber glass bottles (to protect light-sensitive contents) in standard sizes for medication storage and distribution (Doc #116, Doc #119).
3. Standard beverage bottles — reusable beer, wine, and spirit bottles. Under recovery conditions, beverage glass is lower priority than food preservation glass, but maintaining reusable bottle systems for locally produced beverages is practical. Reuse rates for standardised bottles can exceed 20 cycles.⁵⁵
4. Specialty containers — laboratory reagent bottles, carboys for bulk chemical storage, demijohns for fermentation — as capacity allows.

5.3 Closures and seals

A glass container is only useful if it can be sealed. Closures are a dependency:

• Metal lids (twist-off, crown caps): NZ can produce steel from Glenbrook (Doc #89), and stamping thin steel sheet into caps is within NZ’s manufacturing capability (Doc #91). However, the rubber or PVC gasket liner that provides the hermetic seal inside a metal lid is an imported material. Without gasket material, metal lids do not seal reliably.
• Rubber rings: NZ has no rubber production (Doc #33). Existing rubber seal stocks deplete.
• Glass lids with wire-bail closures: The traditional preserving jar (Kilner/Mason type) uses a glass lid held in place by a wire bail, with a rubber gasket providing the seal. The glass lid and wire are producible in NZ; the rubber gasket is the weak link.
• Ground-glass stoppers: For laboratory use, ground-glass joints and stoppers provide hermetic seals without rubber or gasket materials. The glass-to-glass contact surface is precision-ground so that the two surfaces mate intimately. This is a standard laboratory technique and is producible with NZ-made glass and grinding capability.⁵⁶
• Wax seals: Beeswax or rendered tallow can seal glass containers for non-pressure applications (storage, not pressure canning). This is an inferior but functional seal for many preservation methods.
• Cork: NZ does not produce cork (which requires the bark of the cork oak, Quercus suber, a Mediterranean species not grown commercially in NZ).⁵⁷ Some cork oaks may exist in NZ botanical gardens and private estates — propagation is theoretically possible but would take years to produce useful cork.

Assessment: The closure problem is real but manageable through a combination of NZ-produced metal caps (with development of a suitable gasket substitute — potentially tallow-impregnated cotton or woven harakeke fiber as a temporary gasket material; these substitutes are not proven at scale and require experimental validation before reliance),⁵⁸ ground-glass stoppers for laboratory and chemical use, and wax seals for lower-demand applications. This is an area requiring specific development work.

---

6. FLAT GLASS — WINDOWS

6.1 NZ’s window glass situation

NZ’s approximately 1.8 million dwellings and commercial buildings contain a very large stock of existing window glass.⁵⁹ Under nuclear winter conditions, window glass becomes more important than normal — reduced daylight means every window must admit light efficiently, and thermal insulation (including glazing integrity) is critical for heating efficiency (Doc #162).

NZ does not produce flat glass. All window glass is imported — primarily float glass from Australia and Asia.⁶⁰ When imports cease, the existing stock of installed window glass plus warehouse stocks at glaziers and building suppliers is all that is available.

Breakage rate: Flat glass breakage rates in buildings are low under normal conditions — perhaps 0.5–1% of total panes per year from storms, accidents, vandalism, and thermal stress.⁶¹ Under nuclear winter conditions with potentially more severe weather, this rate may increase. At even 1% per year, NZ loses 200,000–400,000 window panes per decade without replacement. The effect is cumulative: boarded-up windows progressively darken buildings and degrade insulation.

6.2 Can the Penrose plant make flat glass?

Container glass plants and flat glass plants use different forming equipment. The Penrose plant’s IS machines are designed to make bottles, not flat sheets. However, the furnace itself produces the same soda-lime glass that flat glass is made from — the difference is in the forming step, not the melting step.

Options for flat glass from a container glass furnace:

• Hand-blown cylinder glass: The oldest industrial flat glass method. A glassblower blows a large cylinder (up to approximately 600 mm diameter and 1,000 mm long), then cuts the cylinder open along its length and flattens it in an annealing oven. This produces a flat sheet with some surface distortion (not optically perfect but functional for windows). This method was used commercially until the early 20th century and produces sheets of approximately 0.5–1.0 m² per cylinder.⁶² Labour-intensive but requires no equipment NZ cannot build.
• Drawn sheet glass (Fourcault or Pittsburgh process): A continuous process where a flat sheet of glass is drawn vertically upward from a pool of molten glass through a slot or over a shaped block (debiteuse). The sheet is gripped by rollers as it cools and hardens. This produces flat glass of reasonable quality (some surface waviness but acceptable for windows) at rates of 1–3 m² per minute. The equipment — a drawing machine, annealing lehr, and cutting table — could be built in NZ, though it represents a significant engineering project.⁶³
• Cast (rolled) glass: Molten glass is poured onto a flat table and rolled to a uniform thickness with a steel roller. This produces glass with one patterned surface (from the table texture) — commonly used for obscure or patterned glass. Simpler equipment than drawn glass but lower optical quality.

Float glass: The modern float glass process (glass floating on a bath of molten tin to produce a perfectly flat, fire-polished surface) requires approximately 100–150 tonnes of molten tin in a carefully controlled atmosphere chamber — a major industrial installation that NZ cannot build in the near term. Float glass is not feasible for NZ. The alternatives listed above produce functional window glass, but it will not match the optical clarity of modern float glass.⁶⁴

Assessment: Flat glass production is a Phase 2–3 development. Hand-blown cylinder glass can begin as soon as skilled glassblowers and a suitable furnace are available, producing limited quantities for highest-priority applications (hospital windows, greenhouse glass). Drawn sheet glass is a more significant engineering project but could produce window glass at useful rates by Phase 3–4.

6.3 Prioritising window glass allocation

Until domestic flat glass production is established, existing window glass stocks must be rationed:

1. Hospitals, clinics, laboratories — functional need for natural light and controlled environments
2. Greenhouses for food production — glass-covered growing space extends growing seasons under nuclear winter (see Greenhouse Construction document — cross-reference number to be confirmed)⁶⁵
3. Schools and community buildings — maintaining function of shared public spaces
4. Residential repair — replacing broken window panes in occupied dwellings
5. New construction — lowest priority; new buildings should minimise glass use or use alternatives (translucent corrugated roofing, oiled paper or cloth panels for non-critical windows)

---

7. LABORATORY GLASSWARE

7.1 Why this is critical

Laboratory glassware is the single most strategically important glass product for NZ’s recovery. Without glass laboratory equipment, NZ cannot:

• Produce sulfuric acid beyond crude methods (Doc #113) — distillation requires glass or equivalent chemically resistant equipment
• Purify pharmaceutical compounds (Doc #119) — distillation, crystallisation, and filtration all depend on glass apparatus
• Produce the chemical precursors for computer construction (Doc #135) — germanium purification requires fractional distillation in glass apparatus
• Perform quality control on food, water, and pharmaceutical products
• Conduct the chemical analyses needed to develop new industrial processes from NZ materials

Laboratory glassware is the enabling infrastructure for NZ’s entire chemical industry development path.

7.2 The borosilicate problem

Standard laboratory glassware is made from borosilicate glass (Pyrex, Duran, Kimax) because of its low thermal expansion coefficient — approximately 3.3 x 10⁻⁶/°C, compared to approximately 8.5 x 10⁻⁶/°C for soda-lime glass.⁶⁶ This means borosilicate glass can withstand rapid temperature changes (heating a flask on a burner, quenching in water) without cracking, while soda-lime glass in the same situation would shatter.

NZ cannot produce borosilicate glass without boron compounds (Section 2.5). This is a genuine constraint, not a solvable problem with NZ materials alone.

Soda-lime glass as a laboratory substitute: Soda-lime glass can be used for many laboratory applications if thermal shock is avoided:

• Storage bottles, graduated cylinders, volumetric flasks: These are used at or near room temperature. Soda-lime glass works well.
• Beakers and flasks for room-temperature mixing: Functional.
• Distillation apparatus: Here the limitation bites. Distillation involves sustained heating and temperature gradients. Soda-lime glass distillation apparatus must be heated slowly and uniformly (using a water bath or oil bath rather than a direct flame), and cooled gradually. The glassblower must anneal the joints and bends carefully. Soda-lime glass distillation apparatus is less convenient and more fragile than borosilicate, but it works — chemists used soda-lime glass for centuries before borosilicate was invented (1893).⁶⁷
• Condensers: Soda-lime glass condensers are feasible if designed with gradual temperature transitions (avoiding cold water jacketing directly on a hot glass surface).
• Thermometers: Soda-lime glass thermometers are standard — most non-precision thermometers are already soda-lime glass.

Performance gap: Soda-lime glass laboratory equipment breaks more easily, requires more careful handling, limits the temperature range of operations, and cannot be directly flame-heated. This is a real degradation — laboratory operations will be slower, more cautious, and will lose more glassware to breakage. But it does not make laboratory chemistry impossible. It makes it harder.

7.3 Production of laboratory glassware

Laboratory glassware is produced by hand glassblowing — a skilled craft that uses glass tubing and rod, worked with a gas-oxygen torch (lampworking) or gathered from a furnace (off-hand blowing).

Skills: NZ has a very small number of scientific glassblowers — perhaps 5–15 people nationally with professional-level competence.⁶⁸ This is a critically small number. Training additional scientific glassblowers should be a high priority, but glassblowing competence takes years of practice — this is not a skill that can be taught in a short course. Expect 2–3 years of apprenticeship before a trainee can reliably produce functional laboratory apparatus, and 5+ years for high-quality complex items.

Equipment needed:

• A glassblowing torch (gas-oxygen) — NZ can produce natural gas and oxygen (from air separation, as at NZ Steel — Doc #89, Section 3.4)
• Glass tubing and rod — must be drawn from molten glass, which requires a glass furnace and tube-drawing equipment. Tube drawing is a specialised process: molten glass is gathered on a blowpipe, inflated, and drawn out into a long tube while maintaining air pressure to keep the bore open. This is feasible by hand at small scale but requires skill.⁶⁹
• Annealing oven — an electrically heated kiln for stress-relieving finished glassware
• Grinding and polishing equipment — for ground-glass joints

Priority items for domestic production:

Item	Urgency	Difficulty
Distillation flasks (round-bottom, various sizes)	Phase 2 — essential for chemical production	Moderate — standard glassblowing
Condensers (Liebig, Graham, Allihn types)	Phase 2 — essential for distillation	High — requires concentric tube construction
Separating funnels	Phase 2 — liquid-liquid extraction	Moderate
Graduated cylinders	Phase 2 — volumetric measurement	Moderate — requires calibration
Beakers and Erlenmeyer flasks	Phase 2 — general laboratory use	Low — straightforward shapes
Reagent bottles with ground-glass stoppers	Phase 2 — chemical storage	Moderate
Thermometers	Phase 2 — temperature measurement	High — requires calibration and capillary tube
Pipettes and burettes	Phase 3 — quantitative analysis	High — precision calibration required
Vacuum apparatus (desiccators, vacuum flasks)	Phase 3 — advanced chemical processing	High

---

8. OPTICAL GLASS — LENSES

8.1 The vision correction need

Approximately 1.8–2.2 million New Zealanders use corrective lenses — roughly 36–44% of the population.⁷⁰ Eyeglasses break, lenses scratch, and prescriptions change (particularly for children, whose vision develops, and for adults over 40, who develop presbyopia). Without new lens production, the fraction of the population with adequate vision correction declines steadily over time.

Doc #110 addresses the broader eyecare problem, including frames, optometry, and lens grinding. This section covers the glass production aspect: producing lens blanks of suitable optical quality that can be ground and polished to prescription.

8.2 Requirements for optical glass

Optical glass for corrective lenses must meet standards that are more demanding than container or window glass:

• Homogeneity: The glass must be uniform in composition and free of striae (streaks of different refractive index caused by incomplete mixing of the melt). Inhomogeneity causes optical distortion.
• Refractive index: Standard ophthalmic crown glass has a refractive index of approximately 1.523 (the d-line value). Different refractive indices allow thinner lenses for stronger prescriptions (higher-index glass). The basic crown glass composition — a soda-lime-silica glass with minor additions — is achievable from NZ materials.⁷¹
• Freedom from bubbles and stones: Bubbles in the glass cause visual defects in lenses. Stones (unmelted batch particles or refractory fragments) cause stress concentrations that lead to breakage during grinding. The glass must be well-refined (held at high temperature long enough for bubbles to rise out of the melt).
• Colour: Ophthalmic glass should be essentially colourless — requiring low iron content in the batch (high-purity Parengarenga sand) and adequate decolourising treatment.
• Annealing: Optical glass must be very carefully annealed (cooled from melting temperature at a controlled rate) to eliminate residual stresses that cause birefringence (double refraction), which degrades optical performance.

8.3 NZ production feasibility

Producing basic crown glass lens blanks is within NZ’s capability using:

• High-purity Parengarenga silica sand
• Soda ash (from the pathways described in Section 3) or potash
• Limestone and possibly minor additions (borax if available via trade, zinc oxide, barium carbonate for higher-index glass — these are specialty materials that NZ does not produce)
• A small pot furnace with careful temperature control
• Extended refining time (holding the melt at high temperature to eliminate bubbles)
• Very careful annealing (slow cooling over 24–48 hours in a temperature-controlled kiln)
• Skilled personnel who understand optical glass quality requirements

The process: Small batches of glass (50–200 kg) are melted in a refractory pot, refined for hours to remove bubbles, stirred (or the pot is rotated slowly) to improve homogeneity, then poured into moulds to form lens blanks — flat or slightly curved discs approximately 60–80 mm diameter and 10–15 mm thick. After careful annealing, blanks are inspected for defects and those that pass inspection are sent to a lens grinding workshop (Doc #110) for surfacing to prescription.⁷²

Quality assessment: Domestically produced optical glass will not match the quality of pre-event commercial ophthalmic glass (which benefits from centuries of industrial refinement). Specific limitations:

• Some striae and minor inhomogeneity — producing slight optical distortion that may be noticeable but not disabling for most wearers
• Limited to standard crown glass refractive index (~1.52) — meaning stronger prescriptions require thicker, heavier lenses than modern high-index alternatives
• Possibly a slight colour tint (green from residual iron)

These are real performance gaps. A person wearing NZ-produced lenses will notice the difference from their pre-event glasses. But functional vision correction from imperfect glass is vastly preferable to no correction at all.

Timeline: Lens blank production is a Phase 3 capability — it requires glass production infrastructure, reliable raw material supply, and skilled personnel, all of which take time to develop. In the interim, NZ’s existing stock of pre-event eyeglasses and lens blanks provides a bridge (Doc #110).

---

9. GLASSBLOWING SKILLS AND TRAINING

9.1 The skills challenge

Glassblowing is a manual craft that requires years of practice to master. The key skills are:

• Gathering: Collecting a precisely sized ball of molten glass from the furnace on the end of a blowpipe (off-hand blowing) or manipulating a piece of glass tubing in a torch flame (lampworking). Temperature judgment, timing, and hand coordination are essential.
• Blowing: Inflating the glass bubble with breath through the blowpipe, controlling wall thickness and shape simultaneously.
• Shaping: Using hand tools (jacks, paddles, tweezers, shears) and centrifugal force (spinning the blowpipe) to shape the glass while it is hot and workable.
• Annealing judgment: Knowing when a piece must be placed in the annealing oven to avoid cracking from thermal stress.
• Lampworking (for laboratory glass): Heating glass tubing in a flame to soften it, then bending, stretching, joining, and blowing to form complex apparatus. Lampworking is more precise than off-hand blowing and is the primary technique for laboratory glassware.

9.2 Training pathway

Phase 1 (identification and preservation): - Identify all NZ glass workers — art glass, scientific glass, craft glass — through the census (Doc #8) - Classify experienced scientific glassblowers as critical-skills personnel - Begin knowledge capture: film techniques, document procedures, pair experienced workers with apprentices

Phase 2 (apprenticeship): - Establish formal glassblowing apprenticeships — target 10–20 apprentices in the first cohort - Curriculum: glass chemistry basics (3 months), basic gathering and blowing (6 months), intermediate forming (6 months), lampworking introduction (6 months), laboratory glassware specialisation (12+ months ongoing) - Training location: wherever a furnace and torch are available — likely Auckland (near Penrose plant and existing art glass community) and possibly Wellington or Christchurch if equipment exists

Phase 3–4 (expanding workforce): - Target: 30–50 trained glassblowers nationally, including 10–15 with scientific glassblowing competence - Distributed across regions to serve local laboratory and production needs - Ongoing training from the experienced cohort

---

10. PHASED DEVELOPMENT STRATEGY

Phase 1 (Months 0–12): Preserve and extend existing capability

• Keep the Penrose plant running as long as soda ash stocks allow
• Maximise cullet (recycled glass) collection to stretch raw material supply
• Prioritise food preservation containers and pharmaceutical bottles
• Secure and inventory all laboratory glassware nationally
• Begin experimental small-scale production (potash glass from wood ash, lampworking trials)
• Assess soda ash production pathways

Phase 2 (Years 1–3): Develop domestic raw material chain

• Pilot soda ash production (Solvay process if ammonia is available; potash from wood ash as fallback)
• Establish 1–2 small pot furnace workshops for specialty glass (laboratory, optical)
• Begin scientific glassblowing training program
• Maintain Penrose container production at whatever rate raw materials support
• Develop flat glass capability (hand-blown cylinder method initially)

Phase 3 (Years 3–7): Expand and diversify

• Scale soda ash production toward industrial volumes
• Penrose plant operating on substantially domestic raw materials
• Flat glass production at useful rates (drawn sheet method)
• Laboratory glassware workshop producing for chemical and pharmaceutical programs
• Optical lens blank production beginning
• First generation of NZ-trained glassblowers reaching competence

Phase 4 (Years 7–15): Mature domestic industry

• Full domestic glass supply chain operational
• Multiple production facilities (container, flat, specialty)
• Consider all-electric furnace conversion at Penrose (eliminating gas dependency)
• Fiberglass trials (insulation, composite materials)
• Export of glass products to trade partners if surplus exists

Phase 5+ (Years 15+): Industrial glass economy

• Expanded product range including technical and specialty glass
• Self-sustaining workforce training pipeline
• Glass as a mature NZ industry with established domestic supply chains

---

CRITICAL UNCERTAINTIES

Uncertainty	Impact if Wrong	Resolution Method
Soda ash stocks at Penrose and in NZ distribution	Determines how long existing container production continues. If less than estimated, production stops sooner.	Direct verification with O-I Glass management — first week
Penrose furnace remaining campaign life	If the furnace is near end of campaign, a forced rebuild consumes refractory stocks and causes months of downtime.	Furnace condition assessment — first week
Solvay process feasibility with NZ materials	If ammonia supply cannot be established, industrial-scale soda ash production is blocked. The entire long-term glass production plan depends on this.	Engineering assessment — first 3 months; pilot trials Phase 2
NZ ammonia availability	Critical input for Solvay process. If no ammonia stocks exist and synthesis is decades away (Doc #114), soda ash must come from lower-yield sources.	Inventory ammonia stocks nationally — first month
Parengarenga sand transport logistics	If coastal barging is disrupted and road transport proves too fuel-intensive, sand supply to Auckland is constrained.	Assess transport options — first month
Number and skill of NZ scientific glassblowers	If fewer than estimated (~5–15), laboratory glassware production capacity is extremely limited.	Census identification — first 3 months
Potash yield from NZ timber species	If NZ wood ash produces less potash than estimated, the wood ash pathway yields less flux.	Testing — begin Phase 1
Refractory availability for furnace rebuild	If AZS refractory stocks are insufficient for a furnace rebuild, the Penrose furnace cannot be restarted after end of campaign. Alternative refractories result in shorter campaigns.	Inventory of glass-grade refractories in NZ — first month
Quality of soda-lime laboratory glassware	If thermal shock breakage rates are unacceptably high for laboratory use, the productivity of NZ chemical programs is reduced.	Pilot production and user trials — Phase 2
Flat glass production feasibility	If hand-blown cylinder glass and drawn sheet glass prove too labour-intensive or too poor in quality for NZ needs, window replacement remains a long-term constraint.	Pilot trials — Phase 2–3

---

CROSS-REFERENCES

• Doc #1 — National Emergency Stockpile Strategy (glass container inventory, soda ash stocks, laboratory glassware as strategic assets)
• Doc #8 — National Skills and Asset Census (scientific glassblowers, art glassblowers, glass factory workforce, laboratory glassware inventory)
• Doc #135 — Computer Construction (germanium purification requires glass distillation apparatus)
• Doc #33 — Tires (transport of Parengarenga sand)
• Doc #78 — Food Preservation (glass jars as primary reusable preservation containers)
• Doc #89 — Greenhouse Construction (glass for greenhouse glazing under nuclear winter)
• Doc #89 — NZ Steel Glenbrook (steel for furnace construction; wire for glass container moulds; oxygen supply for glassblowing torches; shared refractory challenge)
• Doc #91 — Machine Shop Operations (mould fabrication for container glass; grinding and polishing equipment for lens blanks)
• Doc #93 — Foundry Operations (casting furnace components and glass production tooling)
• Doc #97 — Cement Production (shared limestone sources; shared refractory challenge)
• Doc #100 — Harakeke Fiber (potential gasket material for glass container closures)
• Doc #110 — Eyeglass Lens Grinding (downstream consumer of optical glass lens blanks)
• Doc #113 — Sulfuric Acid (requires glass laboratory equipment; sulfuric acid needed for some glass chemistry processes)
• Doc #114 — Ammonia Synthesis (ammonia supply for Solvay process soda ash production)
• Doc #116 — Pharmaceutical Rationing (glass bottles for medication storage and distribution)
• Doc #119 — Local Pharmaceutical Production (requires glass laboratory and production equipment)
• Doc #138 — Sailing Vessel Design (coastal transport of Parengarenga sand)
• Doc #151 — Trans-Tasman Relations (boron compounds and soda ash as priority trade items)
• Doc #157 — Trade Training Priorities (glassblowing as a training priority trade)
• Doc #160 — Heritage Skills Preservation and Transmission (hand glassblowing as a heritage craft skill)
• Doc #163 — Housing Insulation (window integrity for thermal performance)

---

---

1. O-I Glass (Owens-Illinois) operates a glass container manufacturing plant at Penrose, Auckland — NZ’s only significant glass manufacturing facility. The plant has operated under various ownerships since the mid-20th century. O-I is the world’s largest glass container manufacturer. See: O-I Glass company information. https://www.o-i.com/ — NZ operations are described in company reports and NZ business directory listings.↩︎

2. Parengarenga Harbour silica sand: The Parengarenga deposit in the Far North of the North Island contains high-purity silica sand (96–99% SiO₂) that has been mined and exported for glass manufacture for decades. The deposit is one of the highest-quality glass sand sources in the Australasian region. See: GNS Science mineral resource data; Crown Minerals NZ. https://www.nzpam.govt.nz/ — Also: NZ Geological Survey publications on industrial minerals.↩︎

3. Soda-lime glass composition: The standard composition for container and flat glass is approximately 70–75% SiO₂, 12–15% Na₂O (from soda ash), 8–12% CaO (from limestone), with minor additions of Al₂O₃, MgO, and other oxides. The melting point of pure silica is approximately 1,713°C; the addition of soda ash (the flux) reduces the working temperature of the batch significantly. See: Shelby, J.E., “Introduction to Glass Science and Technology,” Royal Society of Chemistry, 2nd ed., 2005; Varshneya, A.K., “Fundamentals of Inorganic Glasses,” Academic Press, 2nd ed., 2006.↩︎

4. O-I Glass NZ workforce: Estimated at approximately 200–350 employees based on typical staffing levels for a glass container plant of the Penrose facility’s scale. Exact current staffing should be verified directly with O-I Glass NZ management. Specialist roles include furnace operators, batch engineers, IS machine operators, mould technicians, quality control chemists, and maintenance trades.↩︎

5. Glass furnace campaign life: Continuous tank furnaces for glass production typically operate for 10–15 years between major rebuilds (campaign life). During a campaign, the furnace operates 24 hours a day, 365 days a year — shutdown and restart damages the refractory lining through thermal cycling and is avoided unless absolutely necessary. A planned rebuild takes several months and requires substantial refractory inventory. See: Shelby (note 3); Tooley, F.V. (ed.), “The Handbook of Glass Manufacture,” 3rd ed., Ashlee Publishing, 1984.↩︎

6. Cullet (recycled glass) in glass production: Each 10% of cullet substituted for virgin batch reduces furnace energy consumption by approximately 2.5–3% because cullet melts at a lower temperature than raw batch materials (it is already in the glassy state). Cullet also reduces batch dust and emissions. The typical cullet content in container glass production ranges from 20–90% depending on availability and colour requirements. See: Glass Packaging Institute; International Commission on Glass publications on glass recycling.↩︎

7. Parengarenga location and transport: Parengarenga Harbour is located at the far northern tip of the North Island, approximately 450 km by road from Auckland. The last section of road (north of Kaitaia) includes unsealed portions. Pre-event sand transport has been by coastal barge and road truck. See: Far North District Council road information; NZ Transport Agency.↩︎

8. Glassblowing training timelines: Professional competence in hand glassblowing typically requires 2–5 years of regular practice. Scientific glassblowing, which demands higher precision and knowledge of glass properties, requires additional specialist training. The American Scientific Glassblowers Society and the British Society of Scientific Glassblowers provide professional standards and training frameworks. See: ASGS training guidelines. https://www.asgs-glass.org/↩︎

9. NZ glass container stock: The estimate of tens of millions of glass jars and bottles in NZ is based on: NZ’s population of ~5.1 million, average household ownership of preserving jars, commercial jar stocks, and the total container glass market of ~200,000–250,000 tonnes per year with typical container weight of 200–500 g (implying roughly 400 million–1 billion containers produced or imported per year, with a significant fraction accumulating in the economy as reusable items). The actual stock of preserving-suitable jars (wide-mouth, with intact sealing surfaces) is a subset of total glass containers and is uncertain.↩︎

10. NZ corrective lens use: The figure of approximately 2 million NZ residents using corrective lenses is estimated from NZ Health Survey data indicating that approximately 36–40% of adults report using corrective lenses, extrapolated to the total population including children. See: Ministry of Health NZ, NZ Health Survey. https://www.health.govt.nz/ — Also cited in Doc #110.↩︎

11. NZ window glass stock: NZ has approximately 1.8–1.9 million dwellings (Stats NZ). The estimate of 20–40 million individual window panes is based on typical NZ dwelling window counts (10–25 panes per dwelling for the mix of houses, apartments, and commercial buildings). This is a rough estimate — actual total depends on building size and type.↩︎

12. Glass production workforce: The estimate of 200–500 total person-equivalents for operating the Penrose plant plus supply chain labour is based on the plant’s direct workforce (200–350) plus additional labour for Parengarenga sand mining and transport, soda ash production (if domestic), limestone supply, cullet collection and processing, and distribution. Under recovery conditions with reduced mechanisation, labour requirements may be higher.↩︎

13. O-I Glass (Owens-Illinois) operates a glass container manufacturing plant at Penrose, Auckland — NZ’s only significant glass manufacturing facility. The plant has operated under various ownerships since the mid-20th century. O-I is the world’s largest glass container manufacturer. See: O-I Glass company information. https://www.o-i.com/ — NZ operations are described in company reports and NZ business directory listings.↩︎

14. O-I Glass NZ products: The Penrose plant produces glass containers (bottles and jars) for the food and beverage industry. It does not produce flat glass, laboratory glassware, fiberglass, or specialty glass. See: O-I Glass company information; NZ glass packaging industry publications.↩︎

15. Glass furnace campaign life: Continuous tank furnaces for glass production typically operate for 10–15 years between major rebuilds (campaign life). During a campaign, the furnace operates 24 hours a day, 365 days a year — shutdown and restart damages the refractory lining through thermal cycling and is avoided unless absolutely necessary. A planned rebuild takes several months and requires substantial refractory inventory. See: Shelby (note 3); Tooley, F.V. (ed.), “The Handbook of Glass Manufacture,” 3rd ed., Ashlee Publishing, 1984.↩︎

16. IS (Individual Section) machines: The standard forming machine for glass containers, invented by Hartford-Empire in the 1920s. Each “section” of the machine independently forms one or more containers per cycle — gob feeding, blank mould pressing or blowing, transfer to blow mould, final blowing, and takeout. Production rates depend on the number of sections and containers per section. A typical modern IS machine has 8–12 sections, each producing 1–4 containers per cycle at 6–12 cycles per minute. See: Tooley (note 5); O-I Glass technical publications.↩︎

17. Penrose plant capacity: Estimated at approximately 150,000–200,000 tonnes per year based on typical single-furnace container plant capacities and NZ market size. O-I does not publicly report individual plant capacities. This estimate requires verification.↩︎

18. NZ glass container market: Total NZ glass container consumption is estimated at approximately 200,000–250,000 tonnes per year based on industry data and per-capita consumption estimates. NZ production (Penrose) supplies approximately 60–80% of domestic demand; the balance is imported. See: Glass Packaging Forum NZ; Packaging NZ; Stats NZ trade data.↩︎

19. O-I Glass NZ workforce: Estimated at approximately 200–350 employees based on typical staffing levels for a glass container plant of the Penrose facility’s scale. Exact current staffing should be verified directly with O-I Glass NZ management. Specialist roles include furnace operators, batch engineers, IS machine operators, mould technicians, quality control chemists, and maintenance trades.↩︎

20. Penrose furnace fuel: Glass container furnaces in NZ and Australia typically use natural gas as the primary fuel, often with electric boosting (submerged electrodes that supplement gas firing with direct electrical heating of the glass melt). The exact fuel configuration of the Penrose furnace should be verified with O-I management. NZ natural gas is from the Taranaki Basin, delivered by pipeline.↩︎

21. All-electric glass melting: All-electric (cold-top) glass furnaces use submerged electrodes (typically molybdenum or tin oxide) to heat the glass melt directly. They are commercially proven for specialty glass and smaller container furnaces. Larger all-electric furnaces are under development globally as part of the glass industry’s decarbonisation effort. NZ’s renewable electricity makes all-electric melting a natural long-term option. See: Glass industry publications; International Commission on Glass; British Glass technical guidance.↩︎

22. NZ scientific glassblowers: A very small number of specialist scientific glassblowers work at NZ universities and industrial laboratories. The exact number is uncertain — probably 5–15 with professional-level competence in fabricating custom laboratory apparatus from borosilicate glass tubing. This is a rare and highly valuable skill set. See: NZ university laboratory services; compare with overseas professional bodies such as ASGS (American Scientific Glassblowers Society) and BSSG (British Society of Scientific Glassblowers) for context on workforce size.↩︎

23. NZ flat glass: NZ does not manufacture flat glass. All window glass, mirror glass, and architectural glass consumed in NZ is imported, primarily from Australia (where Viridian/CSR and others operate float glass plants) and from Asian manufacturers. NZ glass processing companies cut, temper, laminate, and fabricate insulating glass units from imported flat glass sheet. See: NZ glass industry publications; Window & Glass Association NZ.↩︎

24. Soda-lime glass composition: The standard composition for container and flat glass is approximately 70–75% SiO₂, 12–15% Na₂O (from soda ash), 8–12% CaO (from limestone), with minor additions of Al₂O₃, MgO, and other oxides. The melting point of pure silica is approximately 1,713°C; the addition of soda ash (the flux) reduces the working temperature of the batch significantly. See: Shelby, J.E., “Introduction to Glass Science and Technology,” Royal Society of Chemistry, 2nd ed., 2005; Varshneya, A.K., “Fundamentals of Inorganic Glasses,” Academic Press, 2nd ed., 2006.↩︎

25. Parengarenga Harbour silica sand: The Parengarenga deposit in the Far North of the North Island contains high-purity silica sand (96–99% SiO₂) that has been mined and exported for glass manufacture for decades. The deposit is one of the highest-quality glass sand sources in the Australasian region. See: GNS Science mineral resource data; Crown Minerals NZ. https://www.nzpam.govt.nz/ — Also: NZ Geological Survey publications on industrial minerals.↩︎

26. Parengarenga sand reserves: The Parengarenga silica sand deposit is estimated to contain tens of millions of tonnes of recoverable high-purity silica sand. Exact reserve figures depend on the mining method, the boundaries of the deposit, and environmental constraints on extraction. At any plausible NZ glass production rate (even 200,000+ tonnes of glass per year requires roughly 150,000+ tonnes of sand), the deposit provides centuries of supply. See: GNS Science; Crown Minerals; mining company resource assessments.↩︎

27. Parengarenga location and transport: Parengarenga Harbour is located at the far northern tip of the North Island, approximately 450 km by road from Auckland. The last section of road (north of Kaitaia) includes unsealed portions. Pre-event sand transport has been by coastal barge and road truck. See: Far North District Council road information; NZ Transport Agency.↩︎

28. NZ silica sand deposits (other than Parengarenga): Various sand deposits exist along the Northland coast and elsewhere in NZ. These generally have higher iron content and lower purity than Parengarenga sand, making them suitable for coloured glass (amber, green) but not for clear glass without additional processing. See: GNS Science NZ Mineral Occurrence Database; Crown Minerals.↩︎

29. Soda-lime glass composition: The standard composition for container and flat glass is approximately 70–75% SiO₂, 12–15% Na₂O (from soda ash), 8–12% CaO (from limestone), with minor additions of Al₂O₃, MgO, and other oxides. The melting point of pure silica is approximately 1,713°C; the addition of soda ash (the flux) reduces the working temperature of the batch significantly. See: Shelby, J.E., “Introduction to Glass Science and Technology,” Royal Society of Chemistry, 2nd ed., 2005; Varshneya, A.K., “Fundamentals of Inorganic Glasses,” Academic Press, 2nd ed., 2006.↩︎

30. Global soda ash production: The world’s largest soda ash producer is the USA, where vast natural trona (sodium sesquicarbonate) deposits in Wyoming are mined and refined. China, Turkey, and other countries produce synthetic soda ash via the Solvay process. No soda ash production exists in NZ or Australia. See: USGS Mineral Commodity Summaries — Soda Ash; general industrial chemistry references.↩︎

31. NZ soda ash consumption: Estimated at approximately 30,000–60,000 tonnes per year based on NZ glass production volumes and other industrial uses (water treatment, detergent manufacture, pulp and paper). The glass industry is typically the largest consumer of soda ash in any economy. Exact NZ consumption should be verified from import statistics (Stats NZ trade data).↩︎

32. Role of calcium oxide in glass: CaO (from limestone) is essential for making glass chemically durable. Soda-silica glass (without lime) is water-soluble — “water glass” or sodium silicate. The addition of ~10% CaO produces a stable, durable glass suitable for containers and windows. This was discovered empirically by ancient glassmakers. See: Shelby (note 3); any glass chemistry reference.↩︎

33. NZ feldspar deposits: Feldspar-bearing rocks occur in several NZ locations, including Fiordland, the West Coast, and the Coromandel Peninsula. NZ has not had significant commercial feldspar mining, but deposits have been identified by geological surveys. See: GNS Science; Crown Minerals NZ mineral occurrence data.↩︎

34. Cullet (recycled glass) in glass production: Each 10% of cullet substituted for virgin batch reduces furnace energy consumption by approximately 2.5–3% because cullet melts at a lower temperature than raw batch materials (it is already in the glassy state). Cullet also reduces batch dust and emissions. The typical cullet content in container glass production ranges from 20–90% depending on availability and colour requirements. See: Glass Packaging Institute; International Commission on Glass publications on glass recycling.↩︎

35. NZ glass recycling: NZ’s glass recycling system collects approximately 65,000–80,000 tonnes of post-consumer glass per year through kerbside collection and commercial recycling. The Glass Packaging Forum and local councils administer the collection. See: Glass Packaging Forum NZ. https://www.glassforum.org.nz/ — Ministry for the Environment waste data.↩︎

36. Borosilicate glass properties: Borosilicate glass (approximately 80% SiO₂, 12–15% B₂O₃, 4% Na₂O, 2% Al₂O₃) has a thermal expansion coefficient of approximately 3.3 x 10⁻⁶/°C — roughly one-third that of soda-lime glass (~8.5 x 10⁻⁶/°C). This gives it excellent thermal shock resistance. Invented by Otto Schott in 1893 (Jena, Germany). Corning Glass Works commercialised it as “Pyrex” in 1915. See: Shelby (note 3); Schott AG technical publications.↩︎

37. Solvay process: Developed by Ernest Solvay in Belgium in 1861 and commercialised from the 1860s. The process produces sodium carbonate (soda ash) from sodium chloride (salt), limestone, and ammonia (which is recycled). It replaced the older Leblanc process as the dominant soda ash production method. See: Any industrial chemistry textbook; Hou, T.P., “Manufacture of Soda,” Reinhold, 1942 (historical but comprehensive).↩︎

38. Solvay process material requirements: Approximately 1.5 tonnes of NaCl and 1.1 tonnes of limestone per tonne of soda ash produced, with ammonia recycled (losses of a few percent per cycle). The process also produces calcium chloride (CaCl₂) as a byproduct — this has limited industrial use but is not a disposal problem. See: Industrial chemistry references on the Solvay process.↩︎

39. Solar salt production requirements: Commercial solar salt production requires shallow evaporation ponds (typically 0.5–1.5 m deep) covering large areas — in Australia’s Lake Eyre basin, ponds cover thousands of hectares to produce hundreds of thousands of tonnes per year. NZ-scale production at the quantities required for a Solvay plant (tens of thousands of tonnes of salt per year) would require hundreds of hectares of evaporation surface in a suitable climate zone, plus concentration ponds, crystallisation ponds, washing and harvesting equipment, and storage. Construction lead time for a salt field is typically 2–4 years. See: Salt production engineering references; Ramsay, P.J. and Cooper, J.A.G., “Late Quaternary Sea-Level Change in South Africa” for context on tidal flat characteristics; general solar salt production literature.↩︎

40. NZ ammonia production: The Kapuni ammonia-urea plant in Taranaki was NZ’s only domestic ammonia production facility. Its operational status at any given time depends on commercial conditions and natural gas feedstock availability. The plant produces ammonia from natural gas via steam methane reforming and the Haber-Bosch process. See: MBIE energy and chemical industry data; Ballance Agri-Nutrients (operator). https://www.ballance.co.nz/↩︎

41. Kelp ash for soda production: Burning seaweed to produce soda ash (barilla) was a major coastal industry in Scotland, Ireland, Scandinavia, and other regions from the 17th to 19th centuries. The industry declined after the development of the Leblanc and Solvay processes for synthetic soda ash. Kelp ash contains approximately 3–5% sodium carbonate along with potassium compounds, chlorides, and other salts. See: Clow, A. and Clow, N.L., “The Chemical Revolution,” Batchworth Press, 1952; general histories of the alkali industry.↩︎

42. Yield estimates for kelp and wood ash: These are approximate. Kelp ash yield depends on species, preparation, and burning conditions. Wood ash potash yield depends on species (hardwoods produce more potash than softwoods), burning temperature, and leaching efficiency. Historical sources provide ranges of yields. See: Allen, R.C., “The British Industrial Revolution in Global Perspective,” Cambridge University Press, 2009; historical chemistry references.↩︎

43. Potash glass (forest glass): European glass from the medieval period through the early modern era was predominantly made with potash flux from wood ash (producing “Waldglas” or forest glass), in contrast to Mediterranean glass which used natron or plant ash soda. Potash glass has a slightly different working character and optical properties than soda-lime glass. See: Shelby (note 3); Charleston, R.J., “English Glass and the Glass Used in England,” Allen & Unwin, 1984; general glass history references.↩︎

44. Potash content of wood ash: The potassium carbonate content of wood ash varies significantly by species. Hardwoods generally produce ash with 5–15% K₂CO₃; softwoods produce less (3–8%). NZ’s dominant commercial timber (radiata pine) is a softwood with relatively lower potash content. Native NZ hardwoods (beech, rata) may produce higher-potash ash but these timber sources are more limited and ecologically sensitive. See: General wood chemistry references; Ritter, G.J. and Fleck, L.C., “Chemistry of Wood,” various USDA Forest Products Laboratory publications.↩︎

45. Yield estimates for kelp and wood ash: These are approximate. Kelp ash yield depends on species, preparation, and burning conditions. Wood ash potash yield depends on species (hardwoods produce more potash than softwoods), burning temperature, and leaching efficiency. Historical sources provide ranges of yields. See: Allen, R.C., “The British Industrial Revolution in Global Perspective,” Cambridge University Press, 2009; historical chemistry references.↩︎

46. Sodium sulfate in glass: Sodium sulfate (Na₂SO₄, historically called “salt cake”) is added to glass batches as a refining agent — it decomposes at glass melting temperatures, releasing SO₂ gas bubbles that rise through the melt, sweeping up smaller bubbles and improving glass clarity. Typical addition is 0.3–1.0% of the batch. Excessive sodium sulfate causes amber colouration. See: Tooley (note 5); Shelby (note 3).↩︎

47. Glass furnace campaign life: Continuous tank furnaces for glass production typically operate for 10–15 years between major rebuilds (campaign life). During a campaign, the furnace operates 24 hours a day, 365 days a year — shutdown and restart damages the refractory lining through thermal cycling and is avoided unless absolutely necessary. A planned rebuild takes several months and requires substantial refractory inventory. See: Shelby (note 3); Tooley, F.V. (ed.), “The Handbook of Glass Manufacture,” 3rd ed., Ashlee Publishing, 1984.↩︎

48. Glass furnace output: Container glass furnaces vary widely in size and output. A medium furnace of the type likely at Penrose may produce 200–500 tonnes per day. This estimate should be verified with O-I. See: Glass industry engineering references; O-I Glass technical data.↩︎

49. Glass melting energy: The theoretical minimum energy to convert batch to glass is approximately 2.68 GJ per tonne (endothermic reactions plus sensible heat). Practical furnace energy consumption is 3–8 GJ per tonne depending on furnace type, size, cullet ratio, and fuel source. All-electric furnaces are more thermally efficient (3–4.5 GJ/tonne) because they heat the glass directly without hot flue gas losses. See: Shelby (note 3); International Commission on Glass energy studies; European Container Glass Federation (FEVE) technical data.↩︎

50. All-electric glass melting: All-electric (cold-top) glass furnaces use submerged electrodes (typically molybdenum or tin oxide) to heat the glass melt directly. They are commercially proven for specialty glass and smaller container furnaces. Larger all-electric furnaces are under development globally as part of the glass industry’s decarbonisation effort. NZ’s renewable electricity makes all-electric melting a natural long-term option. See: Glass industry publications; International Commission on Glass; British Glass technical guidance.↩︎

51. Pot furnaces for specialty glass: Small pot furnaces holding 50–500 kg of glass per pot are used for art glass, specialty technical glass, and optical glass production. The pot (a refractory crucible) sits inside a heated chamber. Multiple pots allow different glass compositions to be melted simultaneously. See: Tooley (note 5); art glass and scientific glass production references.↩︎

52. Glass furnace refractories: The primary glass-contact refractory for modern glass tank furnaces is AZS (alumina-zirconia-silica) — a fused-cast refractory with excellent resistance to glass corrosion. The crown (roof) is typically silica brick. These are specialty products manufactured by companies such as Saint-Gobain (SEFPRO), RHI Magnesita, and Monofrax. NZ does not produce glass-grade refractories. See: Tooley (note 5); glass furnace refractory engineering literature.↩︎

53. Tin plate for metal cans: NZ does not produce tin. Tin plate (tin-coated steel sheet used for food cans) depends on imported tin metal. NZ Steel (Glenbrook) can produce steel sheet, but coating it with tin requires an electrolytic tinning line and tin metal, neither of which NZ has domestically. See: Doc #89 (NZ Steel Glenbrook); USGS Mineral Commodity Summaries — Tin.↩︎

54. NZ glass container stock: The estimate of tens of millions of glass jars and bottles in NZ is based on: NZ’s population of ~5.1 million, average household ownership of preserving jars, commercial jar stocks, and the total container glass market of ~200,000–250,000 tonnes per year with typical container weight of 200–500 g (implying roughly 400 million–1 billion containers produced or imported per year, with a significant fraction accumulating in the economy as reusable items). The actual stock of preserving-suitable jars (wide-mouth, with intact sealing surfaces) is a subset of total glass containers and is uncertain.↩︎

55. Glass bottle reuse: Reusable glass bottles (with standardised dimensions and closure systems) can be washed, inspected, and refilled many times. Industry experience shows 20–50+ reuse cycles for well-managed bottle systems (typical of European beer and soft drink bottle pools). Refill systems require standardisation of bottle sizes and shapes — which should be designed into NZ recovery glass production from the start. See: WRAP (Waste & Resources Action Programme) studies on glass reuse; European bottle pool data.↩︎

56. Ground-glass joints: Standard taper ground-glass joints (e.g., 14/23, 19/26, 24/29, 29/32 — denoting cone diameter/length in mm) provide hermetic, reusable connections between laboratory glassware components without rubber or other gaskets. The glass surfaces are precision-ground to mate with a thin film of vacuum grease or silicone providing the seal. These are standard in laboratory chemistry. Producing ground-glass joints requires the glass piece to be formed first, then ground with an abrasive slurry on a lathe or by hand. See: Any laboratory chemistry equipment reference.↩︎

57. Cork oak in NZ: The cork oak (Quercus suber) is native to the Mediterranean basin (southern Europe, North Africa). It is grown as an ornamental tree in some NZ locations but has not been cultivated commercially for cork production. Cork requires bark stripped from trees aged at least 25 years, with first commercial harvests typically at 40–50 years. Even if propagation began immediately, useful cork production from NZ-grown trees would be decades away. See: MBIE plant biosecurity data; general cork production references.↩︎

58. Gasket substitutes for glass closures: The use of tallow-impregnated cotton or harakeke fiber as a gasket material for glass jar seals is a historical technique (similar to early canning methods before purpose-made rubber gaskets). Performance — particularly seal integrity during pressure canning (which generates internal steam pressure of approximately 100 kPa above atmospheric) — has not been validated for NZ materials. These substitutes are likely adequate for low-pressure applications (fermentation crocks, dry storage) but seal integrity for pressure canning would need experimental confirmation. See: Historical food preservation references; Mason jar development history.↩︎

59. NZ window glass stock: NZ has approximately 1.8–1.9 million dwellings (Stats NZ). The estimate of 20–40 million individual window panes is based on typical NZ dwelling window counts (10–25 panes per dwelling for the mix of houses, apartments, and commercial buildings). This is a rough estimate — actual total depends on building size and type.↩︎

60. NZ flat glass: NZ does not manufacture flat glass. All window glass, mirror glass, and architectural glass consumed in NZ is imported, primarily from Australia (where Viridian/CSR and others operate float glass plants) and from Asian manufacturers. NZ glass processing companies cut, temper, laminate, and fabricate insulating glass units from imported flat glass sheet. See: NZ glass industry publications; Window & Glass Association NZ.↩︎

61. Window glass breakage rates: Industry data on glass breakage in buildings is sparse. The figure of 0.5–1% per year is an estimate based on insurance claims data and building maintenance experience — it has not been verified against published NZ data. Actual rates vary significantly by building type, age, location (coastal vs. inland, wind exposure), and maintenance regime. The actual NZ rate should be estimated from insurance industry data or building maintenance records. This figure requires verification.↩︎

62. Cylinder glass (hand-blown flat glass): The cylinder process for flat glass production was the standard method before machine-drawn sheet glass (early 1900s) and float glass (1950s). A skilled glassblower gathers a large amount of glass on a blowpipe, blows and swings it to form a long cylinder, cuts the ends, slits the cylinder lengthwise, and reheats it in a flattening oven where it opens out into a flat sheet. Sheet size is limited by the glassblower’s lung capacity and physical strength — typically 0.5–1.0 m² per sheet. Surface quality is good but not perfectly flat. See: Glass history references; McGrath, R. and Frost, A.C., “Glass in Architecture and Decoration,” Architectural Press (historical reference).↩︎

63. Drawn sheet glass processes: The Fourcault process (1904, Belgium) draws a continuous ribbon of glass vertically upward from a tank of molten glass through a shaped clay block (debiteuse) with a slot. The Pittsburgh process (1926, USA) is similar but draws glass over a drawbar. Both produce flat glass with some surface waviness (not as perfect as float glass) at rates of approximately 1–3 m² per minute per drawing machine. The equipment — drawing machine, annealing lehr, and cutting table — is substantial but within NZ’s fabrication capability with time and engineering effort. See: Tooley (note 5); Shelby (note 3); flat glass technology references.↩︎

64. Float glass process: Invented by Alastair Pilkington at Pilkington Brothers (UK) in the 1950s and commercialised from 1959. Molten glass is poured onto a bath of molten tin in a controlled atmosphere (nitrogen-hydrogen) chamber. The glass floats on the tin, spreading into a flat sheet of uniform thickness with fire-polished surfaces. A float glass line is a major industrial installation — approximately 300 m long — requiring approximately 120 tonnes of molten tin and producing continuous glass ribbon at approximately 6,000 tonnes per week. There are approximately 260 float glass plants worldwide. NZ does not have one and cannot build one in the foreseeable future. See: Pilkington, L.A.B., “Review Lecture. The Float Glass Process,” Proceedings of the Royal Society A, 1969; float glass industry references.↩︎

65. Greenhouse construction cross-reference: The original draft contained a self-reference to Doc #98 (this document) in the context of greenhouse glazing — an error introduced during drafting. The correct cross-reference should be to the Greenhouse Construction document in the library; however, the catalog number for that document could not be confirmed at time of editing and should be resolved during the next catalog reconciliation. See also Doc #89 reference in the Cross-References section, which appears to assign “Greenhouse Construction” incorrectly to the NZ Steel document number — both references need catalog reconciliation.↩︎

66. Borosilicate glass properties: Borosilicate glass (approximately 80% SiO₂, 12–15% B₂O₃, 4% Na₂O, 2% Al₂O₃) has a thermal expansion coefficient of approximately 3.3 x 10⁻⁶/°C — roughly one-third that of soda-lime glass (~8.5 x 10⁻⁶/°C). This gives it excellent thermal shock resistance. Invented by Otto Schott in 1893 (Jena, Germany). Corning Glass Works commercialised it as “Pyrex” in 1915. See: Shelby (note 3); Schott AG technical publications.↩︎

67. History of laboratory glassware: Glassware has been used for chemical experimentation since the medieval period. The alchemists used soda-lime and potash glass apparatus extensively. Modern laboratory glassware transitioned to borosilicate glass after its invention in 1893, but soda-lime glass continued in use for many applications. The key limitation of soda-lime glass is thermal shock — it requires more gradual heating and cooling than borosilicate. See: History of chemistry and laboratory practice references.↩︎

68. NZ scientific glassblowers: A very small number of specialist scientific glassblowers work at NZ universities and industrial laboratories. The exact number is uncertain — probably 5–15 with professional-level competence in fabricating custom laboratory apparatus from borosilicate glass tubing. This is a rare and highly valuable skill set. See: NZ university laboratory services; compare with overseas professional bodies such as ASGS (American Scientific Glassblowers Society) and BSSG (British Society of Scientific Glassblowers) for context on workforce size.↩︎

69. Glass tube drawing: Glass tubing is traditionally produced by the Danner process (1912) — molten glass flows over a rotating hollow mandrel, is drawn into a tube, and is blown with air to maintain the bore. Smaller-scale tube production can be done by hand — a glassblower gathers glass, blows a bubble, and an assistant draws the glass out while the blower maintains air pressure. The resulting tube is irregular in diameter but functional for laboratory lampworking. See: Tooley (note 5); scientific glassblowing references.↩︎

70. NZ corrective lens use: The figure of approximately 2 million NZ residents using corrective lenses is estimated from NZ Health Survey data indicating that approximately 36–40% of adults report using corrective lenses, extrapolated to the total population including children. See: Ministry of Health NZ, NZ Health Survey. https://www.health.govt.nz/ — Also cited in Doc #110.↩︎

71. Ophthalmic glass: Standard ophthalmic crown glass has a composition similar to optical crown glass — a soda-lime-silica glass with additions to achieve the desired refractive index (typically nd = 1.523) and Abbe number (approximately 58–59). Higher-index ophthalmic glasses require additions of barium, titanium, or lead oxides — materials NZ may not have access to. See: Ophthalmological optics references; Schott Glass optical glass catalogue.↩︎

72. Optical glass production: Small-scale optical glass production involves melting batch in refractory pots, extended refining to eliminate bubbles, stirring or rotation for homogeneity, pouring into moulds to form blanks, and very careful annealing (slow cooling over 24–48 hours or longer for larger pieces). Quality assessment includes visual inspection for bubbles, striae (using transmitted light), and birefringence testing (using crossed polarisers). See: Izumitani, T., “Optical Glass,” AIP Translation Series, 1986; Schott AG, “Schott Guide to Glass.”↩︎