GCI TECH NOTES ©

---

Volume 10, Number 1

A Gossman Consulting, Inc. Publication

March 2005

---

The Impact of Chlorine Input on Cement Kiln Metal SREs

by

David Gossman, Gossman Consulting, Inc.

EPA research on chlorine’s impact on metal emissions from incinerators led to the inclusion of a provision in the  Hazardous Waste Combustor Maximum Achievable Control Technology (“HWC MACT”) regulations that controls chlorine input to the feedrate during stack tests for metal emissions, 40CFR 63.1209(n)(4).  Recognizing that not all control parameters would be appropriate for all hazardous waste combustion devices, EPA, in 40CFR 63.1209(g)(1)(ii), provides an option to “waive an operating parameter limit” when technically appropriate and supported with information from the owner/operator.

In March 1995, at the AWMA Specialty Conference on Waste Combustion in Boilers and Industrial Furnaces, a paper was presented entitled “The Effect of Process Differences on System Removal Efficiencies (SREs) and the Fate of Metals in Cement Kilns”.  This paper analyzed data from 33 different cement kilns generated as a result of the 1992 Boiler and Industrial Furnace Certification of Compliance testing.  Metal system removal efficiencies (“SREs”) were specifically compared with chlorine input rates, HCl emissions and Cl₂ emissions. A multivariant regression analysis including these parameters as well as particulate emissions and APCD inlet temperatures revealed no statistically significant correlations.  In particular, the study showed that no direct relationship appeared to exist between chlorine input, normalized for gas flow rates, and cement kiln SREs for any metal.

In December 1997, at the Rock Products Conference, Gossman Consulting, Inc. presented a paper entitled “An Evaluation of a Cement Kiln’s Emissions While Under Worst Case Operating Conditions”.  It provides a detailed comparison of emissions during the Continental Cement, Hannibal, MO, 1996 Trial Burn while operating under both normal and worst-case operating conditions.  A comparison of metal SREs under both conditions, which included an 8-fold increase in chlorine input, revealed no impact on metal SREs. 

These studies demonstrate that the rate of chlorine input does not impact metal emissions from cement kilns.  The original basis for EPA’s concerns in this area is based on the well-known, relatively high volatility of many metal chloride compounds such as those of lead, beryllium, cadmium and mercury.  The reason this does not impact cement kilns (as opposed to incinerators) is easy to understand.   Cement kilns have large quantities of raw materials that interact with combustion byproducts.  These raw materials contain large quantities of sodium, potassium and calcium.   Chloride salts of sodium, potassium and calcium have much higher bond strengths than chloride salts of the HWC MACT metals.  Therefore, in the presence of sodium, potassium and calcium, the formation of any other metal chloride salt is extremely limited and insignificant.  The fundamental principles of inorganic chemistry and thermodynamics easily explain why the rate of chlorine input into a cement kiln does not impact HWC MACT metal SREs.

Table 1 provides a summary of bond strengths in diatomic molecules.  The values in the second column of this table represent the amount of energy required to break the bond, referred as the dissociation energy.  In the high temperature combustion zone of a cement kiln, these bonds are continuously broken and reformed.

As the gasses cool and move up the kiln, the molecules that form first and are not immediately broken will be those that require higher dissociation energies, e.g., BeO, KCl, CrO, NaCl, CaO and CaCl.  On the cation side of the formation, the sodium and potassium, and to a lesser extent the calcium, lock up the chlorine as NaCl, KCl and CaCl₂ preventing the chloride from becoming available to react with other metals.   Oxygen also locks up selected metals such as chromium, where these bond energies are higher than the corresponding chlorine bonds.  It is easy to see why, from this table, Be, Cd, Cr, Hg, Pb and Sb tend to form oxides or more complex oxide derived compounds while Na and K, and to a lesser extent Ca, form chlorides to the extent that chlorides are available.

Incinerators, with their fast gradient from high to low temperatures, tend to prevent this sequence of events from occurring and most incinerators lack the relatively high levels of Na, K and Ca that are found in cement kilns.

The scientific understanding of the process and the empirical data demonstrate that the chlorine input controls used for incinerators are technically inappropriate to use for cement kilns.  It is therefore technically appropriate to waive the chlorine control operating parameter for cement kilns relative to metal emissions.

Table 1

BOND STRENGTHS-DIATOMIC MOLECULES

CRC Handbook of Chemistry & Physics, 70^th Edition.