For glass wool, the right silica sand is not simply the one with the highest SiO₂ number. It is the material whose chemistry, particle-size distribution, moisture and consistency fit the complete batch balance - including cullet - and support a sufficiently homogeneous melt for fiberization. Process references show sand being used alongside recycled glass, carbonates, borates and other composition-adjusting materials [3].
In Morocco, qualification should therefore start with the glass to be produced and the actual process, then work backward to measurable raw-material requirements. Moroccan standards NM EN 13162 and NM EN 14303 cover manufactured mineral-wool products; they do not provide a universal recipe for incoming silica sand [1–2].
1. What role does silica sand actually play in a glass-wool batch?
Glass wool is produced by melting a glass formulation and then converting the melt into fibres. Industrial and academic process descriptions show that the batch may combine silica sand, limestone or dolomite, soda ash, borates and a substantial share of recycled glass or process returns [3].
Sand primarily supplies silica to the glass network. But buyers should not reason as if sand were the only oxide source: cullet already brings a complete glass composition. Its proportion and chemistry therefore change the amount and profile of virgin raw materials required.
Procurement implication: a sand specification isolated from the batch recipe can look chemically impressive while being poorly matched to the industrial process.
Some publications describe melting around 1350 °C, while others report ranges around 1400–1500 °C [3]. These values describe particular processes and formulations; they are not universal operating temperatures.
2. Purity: why SiO₂ alone is not enough
A certificate showing high SiO₂ answers one question: what share of the reported chemical composition is expressed as silica? By itself, it does not answer the industrial question: will this lot move the batch toward the target composition with acceptable variability?
Qualification therefore needs the oxide profile relevant to the formulation: Fe₂O₃, Al₂O₃, CaO, MgO, alkalis and any other constituents controlled by the plant. Their importance depends on the target glass, cullet and other raw materials.
| Parameter | Why monitor it | Mistake to avoid |
|---|---|---|
| SiO₂ | Silica balance in the batch | Turning it into a universal “quality” threshold |
| Fe₂O₃ | Part of the impurity balance and potentially relevant to optical/redox behaviour | Copying a clear-glass limit without glass-wool validation |
| Al₂O₃, CaO, MgO | Contribute to the total oxide balance | Treating them as inherently undesirable |
| Alkalis | Can affect the balance with soda ash and cullet | Ignoring them because the material is sold as “sand” |
| Moisture / loss on drying | Changes the dry solids actually dosed | Confusing a wet tonne with a tonne of dry material |
Accordingly, this guide does not present any SiO₂ minimum or Fe₂O₃ maximum as a Moroccan standard or universal glass-wool threshold. Contract values should come from formulation and process validation.
3. Particle size: a melting parameter, not just a sieve result
Particle-size distribution affects how quartz grains are incorporated into molten glass. PNNL work shows that quartz dissolution depends on initial particle size: fine fractions may be consumed earlier, whereas coarser grains persist longer and dissolve within a continuous glass phase [4].
Other work shows that an unsuitable distribution can favour either strong foaming behaviour for some fine fractions or persistence and clustering of coarser grains [5]. The important procurement lesson is simple: “finer” is not automatically “better.”
These studies used experimental batches different from a commercial glass-wool formulation. Their particle sizes and temperatures are therefore mechanistic observations, not a ready-made glass-wool specification.
What a useful particle-size specification should state
- the test method and sample preparation;
- the sieves or metrics actually monitored;
- the acceptable window, including fines and oversize where process-sensitive;
- test frequency;
- the lot-acceptance rule;
- how lot-to-lot drift is handled, not only a single result.
The distribution should be selected together with the furnace, formulation, batch mixing and observed melting behaviour.
4. Cullet changes how silica sand should be qualified
Glass-wool manufacturing descriptions show that recycled glass can represent an important part of the feed [3]. An Isover environmental declaration documents, within its own products and scope, formulations that can contain a high share of recycled glass together with sand, carbonates and borates [6]. That percentage is an industrial example, not a general technical requirement.
For supplier qualification, this changes two things.
First, cullet chemistry reduces or shifts the need for virgin raw materials. Second, variability in cullet can alter the correction required from sand and other components. The sand specification therefore needs to fit the overall formulation system.
A plant that increases cullet without revisiting oxide balances can keep receiving sand that is “compliant” on paper while shifting the batch composition.
5. Moisture and dosing: work on a dry-mass basis
Even when it does not change intrinsic mineral chemistry, moisture changes the mass of solids actually delivered by one tonne of material. If batch dosing is mass-based, uncorrected moisture variation can create an apparent dosing variation.
The objective is not to impose a universal maximum, but to define:
- a reproducible measurement method;
- the calculation basis used for purchasing and dosing;
- test frequency;
- mass correction where required by the process;
- storage and handling conditions that limit drift.
For more detail, see our guide to silica-sand moisture measurement and lot control.
6. Lot consistency can matter more than one exceptional lot
An industrial formulation is tuned around known raw materials. An exceptionally pure lot that differs sharply from previous deliveries can require more adjustment than a less spectacular but stable and predictable material.
Supplier control should therefore track both average values and dispersion for critical parameters. A series of consistent CoAs tied to incoming checks is more informative than a single certificate.
| Incoming control | Industrial question |
|---|---|
| Identity / traceability | Does the received material match the certified lot? |
| Chemistry | Does the oxide balance remain within the formulation window? |
| Particle size | Is expected mixing and melting behaviour stable? |
| Moisture | Is the dry mass actually dosed under control? |
| Appearance / contamination | Is there evidence of a handling or storage deviation? |
| Lot-to-lot trend | Is the supplier drifting before a formal limit is exceeded? |
7. The cited Moroccan standards do not replace the sand specification
IMANOR lists NM EN 13162 for manufactured mineral-wool products used in buildings and NM EN 14303 for manufactured mineral-wool products used in building equipment and industrial installations [1–2].
These references are relevant for placing mineral wool within the Moroccan product framework. They should not, however, be presented as incoming silica-sand specifications.
The raw-material specification remains a process document: it should translate glass composition, batch design, furnace behaviour, fiberization, quality controls and producer-validated tolerances.
8. A pragmatic supplier-qualification sequence
Step 1 - Define the reference glass and batch
Document target composition, batch materials, cullet type and proportion, and process-sensitive parameters.
Step 2 - Translate the balance into raw-material characteristics
Identify the oxides to control, useful particle-size distribution, moisture and any secondary contaminants or parameters that genuinely matter.
Step 3 - Define methods before limits
A limit without a test method is ambiguous. The contract should state the method, sampling, dry or wet basis, and rounding/interpretation rule.
Step 4 - Compare several lots, not a showcase sample
Check lot-to-lot variability and compare supplier CoAs with incoming measurements. For representativeness, see our guide to silica-sand sampling and lot control.
Step 5 - Validate on the real process
Observe batch preparation, melting, homogeneity and relevant fiberization indicators. Laboratory testing cannot fully replace industrial validation.
Step 6 - Lock the contractual window and change-management rules
The final specification should reflect what has actually been validated. It should also require notification of source, processing or material-distribution changes that could affect the process.
Conclusion: buy a material that fits the system, not a number
For glass wool, silica-sand qualification should connect chemistry, particle size, moisture and lot consistency to the complete batch, cullet and melting behaviour. Quartz-dissolution studies explain why grain size matters, but they do not provide a universal PSD to copy. Moroccan mineral-wool standards address the finished product; they do not replace the incoming raw-material specification.
A strong specification therefore starts from the process, defines measurable methods, checks multiple lots and locks tolerances only after industrial validation.
References
[1] IMANOR, NM EN 13162 - Thermal insulation products for buildings - Factory made mineral wool (MW) products - Specification.
[2] IMANOR, NM EN 14303 - Thermal insulation products for building equipment and industrial installations - Factory made mineral wool (MW) products - Specification.
[3] S. Gerbinet, A. Léonard, V. Briard, Modeling of a glass mineral wool process in view of Life Cycle Analysis, University of Liège / SETAC Europe, 2014.
[4] P. Hrma et al., Conversion of batch to molten glass, II: Dissolution of quartz particles, Journal of Non-Crystalline Solids 357 (2011), 820–828.
[5] M. J. Schweiger et al., Cluster formation of silica particles in glass batches during melting, Journal of Non-Crystalline Solids 356 (2010), 1359–1367.
[6] Saint-Gobain Isover, Environmental Product Declaration - mineral wool / glass wool.