Continuous Emissions Monitors (CEMs)

Field Studies of Dioxin/Furan CEMs

Studies on CEM testing at Belgian, Japanese and Italian incinerators, make it clear that continuous monitors are desirable because: Semi-continuous sampling is more common in Europe and is now also used at the newest trash incinerator in Canada, but is not required anywhere in the U.S. except by some local governments, like the City of Baltimore, Maryland.

As early as 2004, European scientists were calling for semi-continuous testing of dioxin from industrial dioxin sources:
"The case studies thus confirmed the conclusions of the Stage I report and highlight the necessity to establisha more comprehensive picture about the European emissions from sintering plants and hospital waste incinerators. To achieve this, the sinter plants not yet investigated should be subjected to emission measurements; with respect to their overall importance and inorder to initialise abatement measures an obligation to monitor dioxin emissions frequently, for instance by applying semi-continuous sampling, should be established."
Source: "The European Dioxin Air Emission Inventory Project––Final Results" (Chemosphere, Volume 54, Issue 9, March 2004, Pages 1319-1327).

Dioxin emissions many times higher during startup/shutdown periods, when dioxin emissions aren't monitored

The following was written by Alan Watson, CEng, for the Global Alliance for Incinerator Alternatives in relation to revisions to the Eurpean Union Directive on integrated pollution prevention and control:

We know that high dioxin emissions can be generated and released particularly during start-up periods (Gass, Lder et al. 2002; Nordsieck, Neuer-Etscheidt et al. 2003; Environment Agency 2006; Neuer-Etscheidt, Nordsieck et al. 2006; Tejima, Nishigaki et al. 2007; Wang, Hwang et al. 2007; Wang, Hsi et al. 2007; Chen, Lin et al. 2008). It is notable that these start-up periods are never monitored with spot dioxin tests although they are clearly a time when it would be anticipated that dioxins could be formed within incinerators from previously deposited carbonaceous deposits.

Tejima tested a modern Japanese incinerator and found that even a single incinerator start-up released more dioxins to air than operating the incinerator in steady state conditions non-stop for over 2 months. Contamination levels of ash were also increased. If an incinerator was started more than four times a year the majority of the dioxin emissions are likely to come from the unregulated start-up emissions.

The results from Wang are more worrying. To verify the PCDD/F characteristics of incinerators during start-up a continuous MSW incinerator was investigated for two years. The elevated PCDD/F emissions of the MSWI during start-up could reach 96.9 ng I-TEQ N m3 (nearly 1,000 times the EU limit of 0.1 ng/m3) and still maintained a high PCDD/F emission (40 times higher than the Taiwan emission limit) even 18 h after the injection of activated carbon.

From the four MSW incinerators they studied the estimated annual PCDD/F emission from normal operational conditions was 0.112 g I-TEQ. However they calculated that one start-up procedure can generate 60% of the PCDD/F emissions for one whole year of normal operations. Furthermore the PCDD/F emission from the start-ups of some incinerators was at least two times larger than that of a whole years normal operations. This was even without consideration for the PCDD/F emission contributed by the long lasting memory effect.

Dioxin emissions are, not surprisingly (as they are products of incomplete combustion) reported as being much higher during combustion upsets (De Fre and Rymen 1989; Halonen, Tuppurainen et al. 1997; Hart 2001; Zimmermann, Blumenstock et al. 2001; Gullett, Touati et al. 2006; Neuer-Etscheidt, Orasche et al. 2007; Neuer-Etscheidt, Orasche et al. 2007).

MSW = Municipal Solid Waste

MSWI = MSW incinerator

PCDD/F = polychlorinated dibenzo dioxins/furans

I-TEQ = a standard international unit of dioxin measurement

AMESA = Adsorption Method for Sampling of Dioxins and Furans
Technical Briefing Paper: Quasi-Continuous Sampling of PCDD/F in Incinerator Stack Gas

By Pat Costner, senior scientist for Greenpeace, 29 February 2003

Continuous emissions monitoring for PCDD/F content offers the most accurate and reliable means for estimating PCDD/F releases and evaluating potential impacts on public health and the environment. One of the more important earlier studies of the AMESA system for monitoring PCDD/F releases in stack gases is that by De Fre and Wevers (1998). These scientists found that, in comparison to the AMESA system, the standard 6-8 hour sampling time "underestimated the average [dioxin] emission by a factor 30 to 50."

Subsequently, the AMESA system has not only been validated by the German EPA but is now in widespread use in Europe. For example, all municipal waste incinerators in the Flanders portion of Belgium are now required to use the AMESA method and, through such use, have demonstrated substantial reductions in dioxin releases to air. Reinmann et al. (2001) summarize as follows:

“In the last three years the continuous dioxin and furan monitoring system AMESA came to a standard for the continuous control of the dioxin emissions in waste incinerators in Belgium. Due to these results and the operators feedback, some new features were developed and results showed that the continuous emission control of dioxins and furans is necessary. … Actual results of the Walloon region of Belgium, where the continuous control of the dioxin emission is obliged since the 1st January of 2001 shows once again, that plants which were still controlled by short time dioxin measurements can have too high dioxin emission when they were controlled continuously.”

In a still more recent report, Reinmann (2002) concluded as follows:

“The actual results of the Walloon region of Belgium show the different advantages for all, the public, the environmental and the operator. If defects in the plant happen, which lead to higher dioxin emissions, these defects are recognised earlier and help to reduce the dioxin emissions. On the other side, if the values are constantly low, the public acceptance is higher and help to increase or to keep high the capacity of the plant. More and more countries start to think about a continuous control of the dioxin emissions of the waste incinerators. Therefore in different countries verification projects were started. If such regulations will be introduced global the dioxin emissions can be reduced world-wide in a strong way.”

With respect to the on-going application of the AMESA system at the 11 municipal waste incinerators in Wallonia in Belgium, Idczak et al. (2003) noted as follows:

“Evolutions of results since the launching of the programme show the benefits brought. Continuous monitoring with fast availability and publicity of results triggered efforts of operators. They managed their process and prevented breakdowns of their abatement system so that remaining problems are very limited. … Actually samples for 14 days periods are unquestionably more representative than sampling durations of 6 or 8h as usual with the manual method. Continuous sampling throughout the year is comprehensive.

Results of the quasi-continuous monitoring of dioxin releases in the stack gases of the 11 incinerators in Wallonia are available 2 weeks after sample collection and are posted on Environment Directorate’s web site at the authorities’ request.” [See]

The AMESA system has also been used for hazardous waste incinerators. For example, four years ago, Mayer et al. (1999) described in detail the use of the AMESA system, as required by German authorities, in a study of high- and low-temperature operation of a hazardous waste incinerator.

A detailed technical description of the AMESA system can be found on the website of its developer, Becker-Messtechnik. (See

In summary, the AMESA system for the quasi-continuous monitoring of dioxin releases in incinerator stack gases is well proven and has accumulated a substantial record of successful application with solid documentation of the benefits of such quasi-continuous monitoring in comparison to the standard occasional 6-8 hour sampling period. While the development of other methods of continuous or quasi-continuous stack gas sampling and analysis is to be encouraged, there is no sound rationale for waiting for their development rather than using the AMESA system, which is already well-developed, widely used and well-proven.

  • De Fre, R., Wevers, M. 1998. Underestimation in dioxin emission inventories. Organohalogen Cpds. 36:17-20.
  • Reinmann, J., Rentschler, W., Becker, E., 2001. New Results and features of the continuous dioxin/furan monitoring system AMESA. Presented at Dioxin 2001 -- 21st International Symposium on Halogenated Environmental Organic Pollutants & POPs, Kyoung ju, Korea, Sept. 9-14, 2001.
  • Reinmann, J., 2002. Results of one year continuous monitoring of the PCDD/PCDF emissions of waste incinerators in the Walloon Region of Belgium with AMESA. Organohalogen Cpds. 59: 77-80.
  • Idczak, F., Petitjean, S., Duchateau, P., Bertrand, L., 2003. Validation and optimization of continuous sampling to monitor PCDDs, and PCDFFs emissions of waste incinerators. Organohalogen Cpds. 60: 537-540.
  • Mayer, J., Rentschler, W., Sczech, J., 1999. Long-term monitoring of dioxin emissions of a hazardous waste incinerator during lowered incineration temperature. In: Proceedings of Dioxin 99 Conference, Venice, Italy, September 12-17, 1999.

Below are many of these studies. They're arranged with the most recent at the top.

Reinmann: Validation Tests for PCDD/PCDF Long-Term Monitoring Systems: Short Comings of Short Term Sampling and Other Lessons Learned
2008 - Germany - Examination and recommendations on how to standardize AMESA reporting.
"Additionally, the detailed evaluation of the results demonstrates again the insufficient monitoring capability by manual short term sampling for evaluation of the real dioxin emissions of an incinerator when not considering start-up and shut down phases."

Bergmans: Monitoring of PCDDs, and PCDFs Emissions of Waste Incinerators in Wallonia: Uncertainty Evaluation
2007 - Belgium - 11 incinerators in Wolloon region are being continuously monitored. Calculated the uncertainty for dioxin estimates to be 30-40%.

Chachignon: Identification of Key Surrogates for Real-Time Dioxins Monitoring
2007 - France - Automatic continuous sampling devices for dioxin have been installed over the last ten years, but since they don't provide real-time reporting (results take about 2 weeks), they are lookng for something that is simpler to test for that would be a good indicator of dioxin levels - either a surrogate (one type of dioxin) or a precursor polychlorinated benzene/phenol. Found TEQ correlation coefficients >.92 for 2,3,4,7,8-PeCDF, 1,2,3,4,7,8-HxCDF, 1,2,3,6,7,8-HxCDD, 1,2,3,7,8,9-HxCDD and Cl6Bz. Suggests use of these monitoring devices: Resonance Enhanced Multi-Photon Ionization coupled with a Mass Spectrometer Time of Flight, gas chromatographs with Electron Captures Detector, on-line ion trap mass spectrometer.

Hattori: Effect of De Novo Dioxin Synthesis from Boiler Dust on Dioxins Emission from Irregular Operation of MSW Incinerator
2007 - Japan - Found that dust on the superheater, even though lower in dioxin content, produced more dioxin during startup than economizer dust, because superheater dust was subject to temps over 250 C.

Horie: Comparison of the Performance of Long-Term Automated Sampling Method of AMESA and those of JIS-Type I and Type III Manual Sampling Methods for Dioxins in Flue Gas
2007 - Japan - Compared AMESA with JIS-Type I sampling method and found high correlation during short-term testing in 2 new facilities.

Kawamoto: Dioxin Surrogate Study Under Startup Conditions of Municipal Waste Incinerator
2007 - Japan - Testing correlation of surrogates (organic halogenated compounds, chlorophenols and chlorobenzenes with dioxins, in effort to identify cheaper & faster testing methods. Found good correlation during steady state operation.

Paoli: Micropollutants Production in RDF-Coal Co-Combustion in the Fusina-Venice Power Plant
2007 - Italy - Results of refuse-derived fuel (RDF) co-fired with coal.

UK Environment Agency: Use of Continuous Isokinetic Samplers for the Measurement of Dioxins and Furans in Emissions to Atmosphere (April 2006)

Gullett: Transient PCDD and PCDF Concentrations in an MWC
2006 - USA - Tested dioxin precursors during different phases of boiler operation. "The transient, upset conditions tend to promote PCDD formation. 52% of the TEQ value is due to PCDDs during startups, versus 39% during steady state operation."
"Large differences exist in PCDD/F levels in identical, parallel boilers... these differences suggest that even steady state conditions are amenable to process monitoring and control in order to minimize conditions that lead to PCDD/F formation."
"The rapidity of the response of PCDDs, PCDFs, and other co-pollutants to transients and the distinctions between similar boilers suggest that rapid on-line monitoring will be necessary to effect changes in operating conditions that will reduce or prevent favorable PCDD/F formation conditions. However, variation in the response time of pollutants to these transients and facility-specific differences suggest that such measures will require a timelagged correlation between PCDD/F and monitorable surrogate compounds that is statistically established for each particular facility.".

Neuer-Etscheidt: PCDD/F and Other Micropollutants in MSWI Crude Gas and Ashes during Plant Start-Up and Shut-Down Processes
2006 - Germany -

Reinmann: Continuous Monitoring of Unintentionally Formed POPs Listed Under the Stockholm Convention (PCDDs/PCDFs, PCBs, HCB) Using AMESA® Long Term Sampling System
2006 - Germany - "Because of the fact, that manual sampling can give only insufficient information of the total dioxin emissions due to a spot measurement of several hours during one year, the general wish is to have an on-line monitoring, which is however not yet possible. Therefore long term sampling with AMESA® is the practical choice for continuous monitoring of PCDD/PCDF emission from waste incinerators and other industrial emission sources."
"After a network of 12 AMESA monitors was installed in this region, the total dioxin emissions of 4 waste incinerators were reduced by a factor of 10, whereby up to now the quantity of the total burned waste was increased by almost 50 % (fig. 4)."
"The results of the present sampling tests show that the AMESA® sampling system can be used for long term monitoring of the full range of U-POPs (PCDD/PCDF, PCBs, HCB) listed under the Stockholm Convention. Furthermore the test revealed that the AMESA® system is capable to adsorb and hence monitor even molecules with higher volatility (pentachlorobenzene, tri– to pentachlorophenol). Since all POPs listed in Stockholm Convention are less volatile (Aldrine, Chlordane, DDT, Dieldrine, Endrine, Heptachlor, Mirex, Toxaphen) the AMESA® system has the potential to sample the full range of POPs. This indicates that the AMESA® system can be applied for monitoring and supervision of air emissions of POPs remediation projects and POPs destruction processes."

Tejima: Characteristics of dioxin emissions at startup and shutdown of MSW incinerators
2006 - Japan -

Nakui: Online Measurement of Organic Halogenated Compounds (OHC) for Monitoring of Dioxins at Waste Incinerator
2005 - Japan - At stoker-type incinerator, Organic Halogenated Compunds. "The OHC increased when the gas temperature fell or rose. The peaks of OHC were observed when the gas temperature was in the range of 300 to 500 deg. Celsius."

Paoli: Correlation between quantization methods for POPs emission: Dioxin Monitoring System vs. manual sampling in the MSWI of Venice
2005 - Italy - Results of CEM near Venice, compared with traditional stack testing.

Reinmann: Cost Analysis of Automated Long-Term Sampling in Comparison to Existing Application Modes of Manual Short-Term Sampling
2004 - Germany - Cost comparison of manual stack test vs. AMESA CEM.
"Both examples from Taiwan and from Europe show that by higher sampling frequencies an automated sampling system is the more cost-efficient solution. Additional such an instrument is a really good tool to detect really the total yearly dioxin and furan emissions of a plant. Therefore quite more accurate dioxin emission inventories can be given by relatively low costs. Additionally such an instrument is suitable to optimize the operation of a plant with the effect, that the dioxin emissions could be reduced effectively, like it was shown several times over the last few years."

Lamparsk: A Simplified Sample Train for Collection of Chlorinated Dibenzo-p-dioxins and Dibenzofurans from Incinerator Gaseous Emissions
2004 - USA - Description of DOW Chemical's alterations to EPA standard stack test equipment to reduce glassware, leaks, and sample time. Altered eqipment gives higher readings than standard stack test.

Idczak: Validation and Optimization of Continuous Sampling to Monitor PCDDs, and PDDFs Emissions of Waste Incinerators
2003 - Belgium - Reported results from stack test and the newly-installed AMESA continuous monitoring systems.
"Evolutions of results since the launching of the programme show the benefits brought. Continuous monitoring with fast availability and publicity of results triggered efforts of operators. They managed their process and prevented breakdowns of their abatement system so that remaining problems are very limited."

Ludwig: First Long-Term Field Test of the PCDD/F-Analyzer DioxinCop
2003 - Germany - Hazardous Waste incinerator trial using DioxinCop CEM.

Reinmann: Results of One Year Continuous Monitoring of the PCDD/PCDF Emissions of Waste Incinerators in the Walloon Region of Belgium with AMESA®
2002 - Germany - Analysis of Belgian CEM studies.

Riley: Long Term Semi-Continuous D/F Monitoring Systems
2002 - USA - Power Point on AMESA and DMS methods

Pranghofer: Destruction of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans in Fabric Filters
2001 - Germany/USA - Belgium field results with REMEDIA Catalytic Filter System.

Reinmann: New Results and Features of the Continuous Dioxin-/Furan Monitoring System AMESA
2001 - Germany - AMESA is good to use until an on-line method is available.

McKone: Managing the Health Impacts of Waste Incineration
2000 - USA - discussion of current health impact knowledge

DeFre: Underestimation in Dioxin Emission Inventories
1998 - Belgium - There's discrepancy between dioxin levels in the environment and what is known to be emitted. Study found the discrepancy could be accounted for when considering the difference between stack-test estimates of dioxin emissions and two-week continuous emission tests. Continuous testing showed dioxin emissions 30 to 50 times higher than stack tests indicated.