How does bottle glass composition affect electrical conductivity?

A furnace can look stable, then one pull change turns boosting wild. Conductivity shifts can be the hidden reason behind that chaos.

Glass-melt conductivity is mostly ionic. More mobile alkalis and higher temperature raise conductivity fast. Stable conductivity keeps boosting power, glass temperature, and quality more predictable.

A simple conductivity playbook for bottle-glass furnaces

Electrical conductivity ¹ in bottle glass is not a lab curiosity. It is a control variable that decides how smoothly an electric boost system behaves and how evenly heat moves through the melt. Conductivity is also one of the easiest properties to “accidentally change” when cullet rises, when alkali sources change, or when forehearth temperatures drift.

The first step is to treat conductivity as an output of three things:

– Composition (mainly total alkali and which alkali is used)

– Temperature (conductivity rises very fast with temperature)

– Structure and redox (usually smaller than alkali and temperature for common soda-lime ² melts, but still relevant in some cases)

A bottle plant rarely measures conductivity directly every minute. Still, the plant feels it every minute through booster current, voltage, electrode load, temperature stability, and sometimes defect rates. So the practical job is to control the inputs that move conductivity the most, then use fast signals to detect drift before it becomes a furnace problem.

What moves conductivity most in container glass

Driver	What changes in the melt	Conductivity trend	What operators notice first
Temperature (melter/forehearth)	Ion mobility rises	Up fast	Booster current rises, power balance shifts
Total alkali (Na₂O + K₂O)	More mobile charge carriers	Up	Easier current flow, different boost efficiency
Alkali type mix	Mobility and “pathways” change	Can be non-linear	Small recipe changes cause surprises
Cullet percent and cullet chemistry	Alkali and impurities shift	Often up, sometimes unstable	Batch-to-batch electrical behavior changes
Water/volatiles in batch	Local melt chemistry shifts	Often transient	Short-term swings and hotspots

A buyer spec should not demand “one conductivity number” without context. A better request is a stability plan: a composition window, a temperature window, and a method to detect conductivity drift early.

Now the key questions.

This next section explains what actually carries current in glass melts, because that mechanism guides every control decision.

What mechanisms govern ionic conduction in glass melts?

Small changes in alkali or temperature can look like small chemistry changes. In conductivity terms, they can behave like big electrical changes.

Ionic conduction in molten soda-lime glass is dominated by mobile cations, mainly Na⁺ and K⁺. Conductivity rises with temperature in an Arrhenius-like way because ion motion needs activation energy.

The core mechanism: ions hopping through a disordered network

Molten and high-temperature glass is an ionic conductor. The silica-based network provides a disordered structure. Alkali ions sit in that structure and move when thermal energy is high enough. When voltage is applied (like in electric boosting), the electric field biases that motion and a net current flows.

Two ideas explain most real plant behavior:

1) Charge carrier concentration

More alkali oxide usually means more mobile ions available. That raises conductivity.

2) Ion mobility

Mobility rises sharply with temperature. It also depends on how “open” the network is. A more modified network often lets ions move more easily.

Many melts show an Arrhenius ³-type temperature dependence: conductivity increases exponentially with temperature. In practical terms, a small temperature rise can cause a noticeable conductivity rise, which then changes current flow in boosting circuits.

Mixed-alkali behavior: why Na₂O and K₂O can surprise

Container glass often uses mostly Na₂O. Some recipes add K₂O or have K₂O variation from raw materials. Mixed-alkali systems can show non-linear transport behavior. That means a “small substitution” can create a conductivity minimum or an unexpected slope change. This matters when a plant changes soda ash source, feldspar source, or cullet blend and unintentionally shifts the Na/K ratio.

Why redox matters less (usually) than alkalis and temperature

For typical bottle soda-lime compositions, alkali content and temperature dominate conductivity. Redox can matter more in specialty systems with electronic conduction contributions or with multivalent ions that change charge state. In normal container operations, redox often shows up indirectly by changing temperature profiles, fining behavior, or foam, which then shifts conductivity through temperature and composition mixing.

A mechanism table that ties to control

Mechanism piece	What it depends on	How it shows up	What to control
Carrier count	Na₂O + K₂O level, cullet chemistry	“Base” conductivity level	Alkali window, cullet grade
Mobility	Temperature, network openness	Sensitivity to temperature drift	Tight temperature control
Non-linear mixing	Na/K ratio, local structure	Unexpected drift during recipe changes	Manage substitutions and blends

Once the mechanism is clear, the furnace reason becomes obvious: conductivity is not only a property. It is a stability lever.

Why does conductivity control matter for furnace stability?

A furnace can run smooth for months, then a few weeks of instability burn electrodes faster and increase defects. Conductivity drift is one common root cause.

Conductivity control matters because electric boosting depends on current flowing through the melt. If conductivity rises or falls, the same boost setting can deliver different heating, shift thermal gradients, and create runaway behavior unless power control is tight.

Boosting is a “glass-as-a-resistor” system

Electric boosting heats glass by passing current through the melt. The melt behaves like a resistor. Heat is generated by Joule heating ⁴. So, if resistivity ⁵ drops (conductivity rises), current can rise at a given voltage. That changes:

– electrode loading,

– local heating near electrodes,

– convection patterns,

– temperature uniformity.

A stable furnace needs stable heat distribution. If conductivity drifts, heat distribution drifts too.

The hidden feedback loop that causes trouble

A positive feedback loop can appear:

1) A hot zone forms.

2) Hotter glass becomes more conductive.

3) More current flows through that zone.

4) That zone heats even more.

Modern power systems reduce this risk with controlled power electronics and careful electrode design, but the physics still exists. A recipe that becomes “too conductive” at operating temperature can make control more sensitive. A recipe that becomes “not conductive enough” can limit boost capability and reduce flexibility during pull changes.

Why container plants feel this more now

Many bottle plants push:

– higher cullet percent,

– higher pull rate flexibility,

– tighter energy targets.

All three increase the value of stable boosting and stable melt behavior. Cullet ⁶ helps energy but adds chemistry variability. Pull changes stress heat balance. Energy optimization reduces the safety margin. Conductivity control ties these together.

What stability looks like in operational terms

Stability goal	Conductivity-related signal	Risk if uncontrolled	Practical safeguard
Stable glass temperature	Stable boost current/voltage patterns	temperature waves, cords	closed-loop boost power control
Stable electrode life	Lower current spikes and balanced load	electrode wear, downtime	alarm limits and ramp rules
Stable quality	Fewer thermal gradients	seeds, striae, color drift	smooth pull-change strategy
Stable energy	predictable kWh/ton with boosting	energy spikes	model-based setpoints

Conductivity control is not only about “getting more heat.” It is about keeping heat predictable across time, batch, and cullet changes.

Next is the buyer-facing part: how to tune alkalis and temperature without breaking durability or formability.

How to tune alkalis and temperature to manage conductivity?

It is tempting to fix conductivity problems by adding alkali or increasing temperature. That approach can create new issues fast.

Conductivity is managed best by setting an alkali window, controlling Na/K mix, and holding a tight temperature window. Small changes should be validated with a property model or a controlled trial because conductivity responds sharply to temperature and Na₂O level.

Alkalis: the strongest composition knob

In soda-lime container glass, Na₂O is the main driver. Raising Na₂O tends to raise conductivity and lower resistivity. That can help boosting and melting, but it can also:

– increase thermal expansion ⁷,

– reduce chemical durability ⁸,

– increase volatility and carryover,

– shift redox needs.

K₂O can behave differently in mobility and can create mixed-alkali effects when combined with Na₂O. So it is safer to control total alkali and also control the ratio, not only one number.

A practical approach is to define three ranges:

– Na₂O target band

– K₂O maximum drift band

– total alkali band (Na₂O + K₂O)

Then the plant links these to cullet chemistry control. Many conductivity surprises come from cullet shifts, not from deliberate batch changes.

Temperature: the fastest lever, but also the most sensitive

Conductivity rises strongly with temperature. Temperature changes happen every day:

– pull rate shifts,

– burner tuning,

– batch blanket changes,

– electrode adjustments.

So temperature control should be viewed as a conductivity control tool. The best plants use smooth ramps and avoid sharp changes that create local hotspots and local conductivity spikes.

How to tune without hurting formability

A bottle furnace needs a stable viscosity ⁹ window at the working end. Changes that help conductivity can hurt viscosity. So, every tuning step should check:

– viscosity curve shift,

– liquidus/devitrification margin,

– forming temperature stability,

– defect trend.

A tuning decision table that stays safe

Goal	Primary lever	Secondary lever	What must be watched
Boosting becomes “weak”	Slightly higher total alkali	Slightly higher operating temp	durability, expansion, volatility
Boosting becomes “too aggressive”	Reduce alkali drift from cullet	Reduce hotspots and smooth temp	electrode load, cords
Conductivity swings batch-to-batch	Tighten cullet grade and blend	tighten raw material substitution rules	chemistry trend charts
Need more output with stable quality	increase clean cullet + stable boost	model-based setpoints	stability during pull changes

The safest path is to make changes small, then verify with measured signals. A big recipe change to chase conductivity often costs more than it saves.

Now the final question: can sensors and models predict conductivity shifts in real time?

Are inline sensors predicting conductivity shifts in real time?

Many plants want a simple “conductivity meter” for molten glass. The environment is harsh, but real progress exists.

Real-time prediction is usually done with indirect signals (electrode voltage/current, temperature maps, batch chemistry trends) and a model that estimates resistivity. Direct in-melt probes exist in research and specialty melters, and electrical resistance tomography and impedance methods are proving that online monitoring is possible when hardware can survive.

What is realistic today in container operations

In most bottle furnaces with boosting, the best “inline conductivity signal” is already present:

– each electrode pair has measured current and voltage,

– power controllers and SCADA ¹⁰/PLC systems trend these values,

– temperature sensors trend melter and forehearth behavior.

From these, the plant can estimate effective resistivity in the electrode zone. This is not a perfect “whole furnace conductivity,” but it is extremely useful because it connects directly to the heating system.

A simple predictive setup uses:

– composition data (batch + cullet targets),

– temperature profile,

– historical electrode I/V data,

– a resistivity model to forecast how the next shift will behave.

This approach often catches drift early, before quality problems appear.

Direct sensors: possible, but still selective

Direct conductivity probes in molten glass exist in research and in some specialized melter environments. These probes typically use noble-metal electrodes and protective shafts. They can measure conductivity trends and detect events like crystal buildup or local property shifts. The barrier in container plants is durability, maintenance access, and the need to keep probes from becoming defect sources.

Electrical resistance tomography and multi-frequency resistance approaches are also being explored to infer melt state. These methods can provide richer information than a single point measurement. They also need robust installation and careful interpretation.

The smartest control pattern: hybrid sensing

A hybrid approach works best:

– Use electrode electrical data as the continuous signal.

– Use temperature imaging and thermocouples to detect hotspots.

– Use batch/cullet chemistry data to predict baseline conductivity.

– Use periodic glass composition checks to keep the model honest.

A practical “sensor stack” buyers can ask about

Measurement layer	What it predicts	Update speed	Why it matters
Electrode current + voltage	effective resistivity near electrodes	seconds	protects electrodes and stabilizes boost
Furnace temperature profile	mobility and hotspot risk	seconds to minutes	prevents runaway zones
Batch + cullet chemistry tracking	carrier count trends	hours to days	avoids slow drift surprises
Periodic glass composition checks	model correction	daily/weekly	keeps predictions reliable

Inline sensing is improving. The strongest gains come when the plant stops treating conductivity as a mystery and starts treating it as a modeled property tied to real electrical signals.

Conclusion

Bottle-glass conductivity is driven by mobile ions and temperature. Stable alkalis, stable temperature, and smart electrical trending keep boosting stable and quality predictable.

Footnotes