The acidity coefficient Mk, defined for the mineral-wool industry as the mass ratio (SiO2 + Al2O3) / (CaO + MgO), is the spec window cupola operators actually have to hold in the taphole stream, with 1.05 giving a melilite-dominated phase field and 1.20 shifting precipitation to anorthite under continuous cooling at DSC conditions [S1]. North American iron-foundry cupolas instead use the metallurgical basicity index CaO/SiO2 and target 0.5-0.6 in routine operation [S4]. The two numbers are not interchangeable, and a mineral wool line that borrows a foundry slag-control recipe without re-deriving it will misfire on fiberization within the first shift.
The practical cupola for mineral wool runs with about 5-6 parts mineral charge to 1 part coke, melts at 1300-1650 °C, and drops the melt onto a spinner or Powell-process rotor [S2]. That same furnace, if its lining is acid, gets eaten by any attempt to push basicity toward the gray-iron range; acid linings and basic slags do not coexist [S4]. This is why the controlling spec in a wool plant is the acidity coefficient, not the metallurgical index, even though both travel under the loose label of "slag basicity."
Why the index differs from iron-foundry basicity
Mineral wool is a glassy fiber, so the melt must resist crystallization during attenuation. Modified blast-furnace slag samples with Mk 1.05 show melilite as the primary crystallization phase and a critical precipitation rate of 50 °C/s for melilite and 20 °C/s for anorthite, while Mk 1.20 samples precipitate anorthite with critical rates of 20 °C/s and 15 °C/s respectively [S1]. The activation energy of crystallization rises from 698.14 kJ/mol at Mk 1.05 to 1292.50 kJ/mol at Mk 1.20, which is the kinetic reason a higher Mk gives a more fiberization-friendly melt [S1]. Foundry basicity, by contrast, is tuned for desulfurization and refractory life, not fiber spinnability [S4].
The acidity coefficient is the right knob for a wool line because the viscosity-temperature curve and the crystal nucleation path both move with it. A BF-slag/coal-ash system titrated for slag-wool production with integrated waste-heat recovery is governed by a liquidus and viscosity window that enables fiberization by a high-speed air injection method, rather than the viscosity range suited to iron pouring [S3]. Mineral-wool producers feeding a cupola furnace therefore treat slag chemistry as a fiberization parameter first, and a refractory-life parameter second.
Charge geometry, coke rate, and the basicity drift
EPA's AP-42 chapter on mineral wool manufacturing documents the canonical US charge as alternating layers of slag/rock and coke at 5-6:1 mineral-to-coke by weight, with melt at 1300-1650 °C and combustion air introduced through tuyeres near the cupola bottom [S2]. Coke rates quoted in the Chinese slag-wool literature run 350-800 kg of standard coal per ton of wool, which is roughly an order of magnitude worse than the integrated waste-heat route, where a 70% energy cut has been reported at industrial scale [S3]. Any of these charge changes that pulls FeO, SiO2, or CaO in the melt will move the acidity coefficient, and the operator sees it first as a fiber-quality drift, not as a slag-color shift.
Mineral charge composition is a primary control, not a secondary one. A cupola patent for rock-wool production makes the discipline explicit: a strict furnace operating sequence keeps the outflow's flow rate, temperature, and acidity coefficient stable [S8]. In practice this means the raw-material weigh feeder, not the tuyere, is the basicity actuator. Adding diabase to a blast-furnace-slag blend for stone-wool primary fiber layers changes both the Mk and the resulting fiber mechanics, so any swap in the rock fraction must be re-validated against the wool spec sheet, not just the melt rate [S5].
Acidic vs basic lining: which index the plant can hold
All North American foundry cupolas are acid-lined, and acid linings cannot tolerate basic slags because the slag neutralizes the refractory and accelerates wear [S4]. A mineral-wool cupola that wants to push toward higher basicity has to either accept short lining life or move to a basic lining, and the operating economics of a wool line typically do not support the latter. Plants that buy the BF-slag-plus-rock-wool route effectively lock themselves into the acidic side of the phase diagram, where the natural basicity of the slag is held in check by added siliceous rock [S6].
Slag-wool facilities processing molten BF slag, where the ironmaking slag is itself basic, handle the contradiction by adding coal ash or other siliceous modifiers to bring Mk back into the wool window. The BF-slag/coal-ash study from the Chinese literature is built around exactly this lever, with the mBF/m(CA) ratio selected to land both liquidus and viscosity in a fiber-friendly band [S3]. The same lever is what makes a rock wool cupola survivable on basic BF slag feed without changing the furnace.
Comparison: which "basicity" to use for which decision
Three different numbers are in circulation, and a process engineer who mixes them up will mis-spec a charge. The table below lines them up against the decisions a mineral-wool plant actually has to make. [S4]
Acidity coefficient Mk = (SiO2 + Al2O3) / (CaO + MgO): target 1.05-1.20 for BF-slag-modified mineral wool, governs crystal phase (melilite at 1.05, anorthite at 1.20) and critical cooling rate, validated by DSC work [S1]. Foundry basicity CaO/SiO2: target 0.5-0.6 for North American iron cupolas, governs desulfurization and refractory compatibility with acid linings [S4]. Slag-to-coke mass ratio: 5-6:1 for the US wool industry, 350-800 kg standard coal per ton of wool in the Chinese cupola route, 70% energy reduction reported for the integrated waste-heat slag-wool route at industrial scale [S2][S3]. The decision rule is simple: use Mk to set the rock/slag blend before the melt, use CaO/SiO2 to monitor lining life, use the fuel ratio to size the coke budget.
Operating limits and failure modes
The single most common failure mode on a wool cupola is devitrification on the spinner, and it is a basicity symptom. Move Mk above ~1.20 and the melt's critical cooling rate drops into a range where ambient attenuation can no longer quench the fiber glassy; the rotor builds up crystalline nodules and fibers break short [S1]. Move Mk below ~1.05 and viscosity at 1300-1400 °C climbs into a band that the Powell or Downey rotor cannot attenuate into the 4-7 micron diameter range that the wool market accepts [S2]. Both end states are first observed as a slag chemistry drift on the taphole sample, which is why the basicity number belongs on the operator's shift log, not just the monthly melt audit.
Refractory life is the second failure mode and is dominated by the lining's acid-base match, not the wool quality target [S4]. Plants that try to drift the basicity index to chase iron-furny economics tend to discover the lining budget is the constraint, not the slag assay. For operators in the BF-slag-feed segment, the corollary is that waste-heat recovery and integrated slag-wool production are less a green-marketing story and more a refractory-and-energy story, with the 70% energy and 90% emission cuts coming from skipping the cupola remelt altogether [S3].
Standards, sourcing, and what to verify on the data sheet
There is no single international standard that pins the mineral-wool acidity coefficient, which is part of why the value drifts from plant to plant. Industry practice is documented in three places: the EPA AP-42 mineral-wool chapter for charge geometry and melt temperature [S2], the Chinese academic work on BF-slag modification for Mk and crystallization kinetics [S1][S3], and the Modern Casting cupola-operations guidance for the acid-lining constraint and the 0.5-0.6 CaO/SiO2 reference target [S4]. The blast-furnace-slag-replacement logic for rock-wool is captured in the patent literature as a closed-loop control of Mk, melt temperature, and flow at the taphole [S8], and the broader process description is available in the Natural England mineral-wool background note [S6]. Graphalloy's slag-wool field report is a useful secondary reference on the extreme service conditions inside a coke-fired wool cupola [S9].
For an engineer writing a spec, the verifiable items are: the target Mk band, with a documented DSC critical cooling rate inside that band [S1]; the lining material and its acid/base class [S4]; the coke-to-mineral mass ratio and the melt temperature window [S2]; and whether the BF-slag feed will be modified on-site with a siliceous additive to hold Mk inside the wool window [S3]. A wool spec that does not list all four is missing the levers that actually move basicity on the cupola. For a closer look at how the charge-side refractories are managed on a cupola furnace line, the encyclopedia entry covers the tuyeres, water-cooled trough, and the lining wear patterns the basicity index has to live with.
Detailed specification references: access control.
Background reading: Counterbalance Forklift vs Reach Truck: Picking the Right Truck for a Narrow-Aisle.