CO2 Injection Cooling: Use Enthalpy, Not Sublimation Alone

The mixer misses its target temperature or consumes more liquid CO₂ than a sublimation-only estimate predicts. Calculate the requirement from the CO₂ inlet-to-outlet enthalpy change, then commission the estimate against measured batch performance.

Discard the quick fixes that double-count cooling

Do not calculate separate Q expansion and Q sublimation terms and add them without defining thermodynamic states. Across an approximately adiabatic injection valve with no shaft work, throttling is modeled as an isenthalpic process. The pressure drop changes the phase distribution, but it does not create a separate enthalpy credit to add again.

Do not assume all injected liquid immediately becomes dry-ice snow. The downstream pressure and inlet enthalpy determine whether the discharge contains gas, solid, liquid, or a mixture. Even when snow forms, sublimation is only one part of the full path from liquid in the supply line to CO₂ leaving the mixer.

Quick fix	Why it fails	Use instead
Use only 246 Btu/lb	The proposed value represents sublimation alone and omits other inlet-to-outlet enthalpy changes.	Read both endpoint enthalpies from one consistent property source.
Add expansion heat to sublimation heat	A throttling pressure drop is not an independent heat-removal term.	Calculate h_out - h_in.
Assume outlet gas equals product temperature	The leaving gas can be colder or warmer than the chilled mix.	Measure the CO2 exhaust temperature.
Size from flour mass alone	Sugar and every other batch component also carry sensible heat.	Sum the load for all ingredients.

Check: Before proceeding, confirm that the worksheet contains one CO₂ enthalpy difference rather than independent expansion and sublimation credits.

Calculate the complete batch heat load

For a batch with no product phase change, calculate the sensible load for every ingredient:

Q_product = sum[m_i × Cp_i × (T_i,in - T_target)]

Use mass and heat-capacity units that produce the required heat unit. For example, matching mass in pounds with heat capacity in Btu/(lb·temperature) produces Btu when the temperature interval uses the corresponding degree increment. Do not mix SI and customary units inside one term.

If heat capacity changes materially across the cooling interval, replace the constant-Cp term with:

Q_i = m_i × integral from T_target to T_i,in of Cp_i(T) dT

Flour and sugar can enter at different temperatures and have different heat capacities. Calculate each contribution separately. Include other ingredients rather than folding them into the flour term. If the batch experiences moisture freezing, evaporation, crystallization, or another phase change, add that latent load from verified material-property data.

The mixer shell, paddles, shaft, and trapped heel can also exchange heat with the batch. Mechanical agitation and warm surroundings can add heat during injection. Handle these effects either as explicit load terms or through a measured utilization factor during commissioning; applying both methods to the same loss double-counts it.

1. Record the charged mass and inlet temperature of each ingredient.
2. Obtain each ingredient heat capacity over the actual temperature interval.
3. Calculate each sensible-load term and sum them.
4. Add only measured or otherwise defensible non-product loads.

Check: Recalculate the weighted batch temperature from the ingredient data. Investigate any ingredient with missing mass, temperature, or heat-capacity information before sizing CO₂.

Define the liquid CO2 inlet state

Measure the CO₂ temperature at the injection supply line as close to the control valve or nozzle as practical. Also record line pressure and verify whether the fluid is liquid at that point. Temperature alone identifies saturated-liquid enthalpy only when the line is actually at saturation; subcooled or two-phase flow requires the appropriate pressure-and-temperature state.

A useful preliminary approximation is saturated liquid at the measured line temperature. Mark that assumption plainly on the calculation. Replace it with a pressure-and-temperature property lookup when the supply condition is uncertain, pressure loss causes flashing before the nozzle, or the calculated consumption does not match production.

Read h_in from a single authoritative CO₂ property dataset or an established chemical-engineering property reference. Record the unit and reference state used by that dataset. Absolute enthalpy values from different tables may use different zeros, so subtract values drawn from the same source.

1. Stabilize CO₂ flow under the normal operating condition.
2. Record supply-line temperature and pressure.
3. Identify the inlet phase from the selected property data.
4. Read and document h_in for that state.

Check: Stop here if the measured pressure and temperature do not describe the assumed liquid state. Correct the state definition before calculating mass.

Define the CO2 outlet state

The useful refrigerating effect ends at the state in which CO₂ leaves the mixer, not automatically at the final product temperature. Mixing by paddles and gas motion improves contact, but it does not force thermal equilibrium between the exhaust and product.

Measure exhaust temperature in a representative gas stream while avoiding direct impingement from liquid droplets or dry-ice particles. Use the mixer or exhaust pressure corresponding to that location. If the discharge is superheated gas, obtain h_out at the measured outlet temperature and pressure from the same property source used for h_in.

If the exhaust still carries solid CO₂, the assumed superheated-gas endpoint is wrong. Escaping solid represents refrigerating capacity that was not transferred to the product. Observe the discharge, inspect for deposits where permitted by the operating procedure, and use an outlet model that accounts for each exiting phase or correct the contact problem experimentally.

Observation	Likely modeling error or process condition	Next check
Exhaust colder than product	Gas left before reaching product temperature.	Use measured exhaust temperature for h_out.
Exhaust warmer than product	Gas gained heat from warmer zones, hardware, or surroundings.	Check probe location and use the representative discharge state.
Dry ice leaves the mixer	Not all available latent capacity reached the product.	Check distribution, residence time, and phase-specific outlet enthalpy.
Temperature varies across samples	CO2 distribution or mechanical mixing is uneven.	Sample multiple locations before accepting the endpoint.

Check: Confirm that the selected h_out represents what crosses the mixer boundary, including any solid carryover, rather than an assumed equilibrium with the product.

Calculate the theoretical CO2 mass

With both states on the same enthalpy basis, calculate the refrigerating effect per unit mass:

q_CO2 = h_out - h_in

Then calculate the ideal CO₂ requirement:

m_CO2,theoretical = Q_product / q_CO2

When explicit vessel, mixer, or environmental loads are available, use:

m_CO2,theoretical = (Q_product + Q_additional) / (h_out - h_in)

The numerator is heat that must leave the process; the denominator is heat absorbed by each unit mass of CO₂. Both must use compatible units. A positive cooling interval T_i,in - T_target and a positive enthalpy gain h_out - h_in produce a positive mass.

This endpoint method automatically includes flashing, any solid formation and sublimation represented by the path, and warming of the resulting gas. Adding the proposed 246 Btu/lb sublimation value to the endpoint difference would count part of the same state change twice.

Check: Perform a dimensional audit. Dividing total heat by heat per unit mass must return mass; any other result signals mismatched units or an incorrect property entry.

Commission a practical injection quantity

The theoretical mass assumes that the modeled enthalpy gain becomes useful product cooling. Real mixers can vent cold gas early, discharge solid CO₂, cool metal, gain heat from agitation, or produce temperature gradients. Determine the practical correction from controlled batches rather than assigning an unverified efficiency.

1. Start from the theoretical mass while observing all established equipment and process limits.
2. Measure actual CO₂ mass delivered using the installed mass measurement or the difference in verified supply inventory.
3. Record ingredient masses and temperatures, injection start and stop conditions, exhaust temperature, mixer pressure, mixing duration, and final product temperatures.
4. Sample the finished mix at enough locations to detect stratification.
5. Calculate an empirical utilization factor as eta = Q_product / [m_CO2,actual × (h_out - h_in)].
6. For comparable future batches, estimate m_CO2,practical = Q_required / [eta × (h_out - h_in)].

Use the factor only within the operating range that produced it. A nozzle change, altered paddle operation, different fill level, changed vent restriction, or different ingredient loading can alter contact and invalidate the factor.

Check: Repeat a comparable batch. Accept the factor only when delivered CO₂, exhaust state, and final temperature distribution reproduce within the plant's process tolerances.

Verify the complete cooling cycle

Run the calculation and the process as one boundary balance. The batch must reach the target throughout the mix, the measured CO₂ use must agree with the commissioned prediction, and the outlet-state assumption must match the observed exhaust.

1. Confirm all ingredient masses and initial temperatures before injection.
2. Verify the CO₂ inlet phase and record its temperature and pressure.
3. Deliver the calculated practical mass while maintaining normal paddle operation.
4. Measure exhaust temperature and check for liquid or solid carryover.
5. Allow the normal mixing interval, then sample product temperature at representative locations.
6. Compare actual mass and temperature reduction with the predicted energy balance.

If the target is missed uniformly, check delivered mass, property states, heat-capacity data, and unmodeled heat loads. If only part of the batch is warm, treat the problem as distribution or mixing rather than adding CO₂ blindly. If consumption rises while the endpoint stays unchanged, inspect for premature flashing, cold exhaust, solid loss, supply-state changes, or measurement drift.

Check: Release the setup only after the product temperature distribution, CO₂ consumption, and inlet and outlet states all pass their defined production limits.

FAQ