Ohio Lumex Advances EPA-Funded HAP Sorbent Trap Method Toward Final Approval

Background

During this Phase IIb SBIR project, Ohio Lumex continued development of a sorbent trap method for continuous monitoring of metal hazardous air pollutant (HAP) emissions. Metal HAPs from stationary sources are usually estimated using emissions factors based on periodic stack testing, feed stream data, and plant operating parameters. These estimates often carry large uncertainties, especially at facilities where feed composition varies significantly. Continuous measurement provides more accurate data, but current technologies are limited and often too costly for routine use. The sorbent trap method addresses this challenge by allowing repeated in-stack sampling and periodic analysis of time-integrated samples collected over several days. This approach is modeled after EPA Performance Specification 12B (PS 12B) for mercury.

Beyond continuous monitoring, the traps are also being developed as an alternative to Method 29 for intermittent sampling events such as relative accuracy test audits (RATAs). To qualify as a substitute for Method 29, at least one Method 301 validation study must be successfully completed and approved by EPA. The requirements for this application differ from those for continuous monitoring: sampling times are limited to four hours, different performance criteria are used, traps contain only two sections, and spikes are applied to the filter rather than section 3. The alternative test method version of the sorbent trap was the primary focus of Phase IIb.

These two distinct applications have produced two slightly different trap designs, as illustrated below.

Figure 1 – Sorbent Traps for Metal HAPs

The sorbent trap methods developed in this project were primarily designed for coal-fired power plants and waste incinerators, but they are also suitable for cement kilns, metal smelting operations, iron and steel production, and secondary smelting facilities. A modified version was also created to address the measurement needs of the biogas and renewable natural gas (RNG) industries.

Phase IIb focused on several key objectives: field testing to validate the methods (including a Method 301 study and direct comparisons with Method 29), reduction of background metal concentrations in sorbent materials, improvements to digestion procedures and trap design, and preparation of a draft alternative test method.

The metals evaluated during Phase II included Be, Cr, Mn, Co, Ni, As, Se, Cd, Sb, Pb, and Hg. Samples were digested using microwave-assisted digestion in accordance with EPA Method 3052 and analyzed by inductively coupled plasma mass spectrometry (ICP-MS) following EPA Method 6020B.

Phase IIb Objectives

Building on prior SBIR achievements, the Phase IIb effort concentrated on four goals:

1. Reduce background metal concentrations in sorbent media to improve detection limits and blank-subtraction reliability.
2. Conduct side-by-side field comparisons with EPA Method 29 to quantify accuracy and precision under real-world conditions.
3. Complete a formal EPA Method 301 validation, a prerequisite for Alternative Test Method approval.
4. Refine laboratory digestion procedures, including the use of an autoclave-style microwave

At the outset of Phase IIb (October 2024), the method already exhibited low breakthrough, acceptable spike recoveries, and good pair agreement, but background variability and incomplete digestion limited accuracy at low concentrations for a small number of metals.

Key Accomplishments

1. Reduction of Background Metals

A new sorbent material production process was implemented and resulted in substantially lower and more consistent background levels, particularly for nickel, chromium, and manganese.

2. Field Comparison 1 – Dry-Stack Test (December 2024)

A comprehensive comparison of sorbent-trap and Method 29 sampling was performed at a 570-MW coal-fired boiler equipped with an electrostatic precipitator and dry stack. Both methods were run simultaneously under isokinetic conditions for nine three-hour tests at three points in the stack.

Results

• Overall bias between methods: -7.7 % (sorbent traps measured slightly lower).
• Element-specific findings:
  • Chromium: -10.8 % average bias (-6.7 % excluding an outlier run)
  • Manganese: -5.0 %
  • Cobalt: -8.8 %
  • Nickel: -22 %, attributed to variable sorbent background in older batches.
  • Arsenic: initially +37 %, corrected to -6 % after adjusting calibration to a 2% nitric-acid matrix, revealing incomplete digestion as the cause of bias.
  • Selenium: -7 %
  • Lead: +0.7 %

Interpretation

The small overall negative bias and strong agreement between methods demonstrate equivalency of the sorbent-trap method to Method 29 for dry flue-gas conditions.

3. Field Comparison 2 – Wet-Stack Test (January 2025)

A second EPRI-sponsored trial was conducted at a coal-fired unit with wet flue-gas desulfurization, using the same side-by-side sampling protocol.

Sorbent-Trap Performance

• Average breakthrough: 0.9 % of total metals.
• Pair agreement: 1.5 % RPD.
• Spike recoveries: mean 92.5 %, all within 70–130 % acceptance.
• Outliers limited to one contaminated antimony sample and one poor spike-delivery event

Method 29 Performance

Significant reagent-blank contamination and probable glassware carryover were identified:

• Chromium blank levels equaled 35–80 % of measured sample mass.
• The final impinger in the sample train often contained anomalously high metals, up to ten times higher than the filter and first impinger.

Overall Agreement

Despite these Method 29 complications, the two methods exhibited a total bias of 1.3 %. This was aided by the fact that selenium agreement was excellent and the stack gas metal composition was dominated by selenium (≈90 % of total metals). Metals affected by reagent contamination showed the largest discrepancy, confirming that disagreement stemmed mainly from Method 29 complications.

4. Laboratory Process Improvements

Installation of an autoclave-style microwave digestion system replaced the earlier rotor-based unit for digestion of sorbent material. The new system:

• Enables complete digestion of entire sorbent sections without splitting.
• Reduces the minimum dilution required to eliminate matrix effects.
• Improves spike recoveries
• Reduces digestion time by ≈50 % and requires no cleaning cycles between runs

5. Draft Alternative Test Method and Performance Specification

Ohio Lumex completed draft versions of both an Alternative Test Method and a Performance Specification for continuous monitoring applications. These documents incorporate the validated sorbent trap designs, analytical procedures, QA/QC criteria, and acceptance limits derived from the Phase IIb studies. Both drafts have been submitted to EPA and are under review.

6. Method 301 Validation

A formal EPA Method 301 analyte-spiking validation was conducted in October 2025 under EPRI sponsorship. At the time of final report submittal, sample analysis had not yet been completed. The completed dataset will be transmitted to EPA to support formal ATM approval.

Discussion of Results

The data collected during Phase IIb demonstrated that the sorbent trap method is accurate, precise, and provides major practical advantages:

• Simplified field operation: traps can be inserted and recovered with minimal reagents or fragile components.
• Lower cost: reduced labor compared to Method 29.
• Improved safety: elimination of acids and oxidizers.
• Long-duration sampling: sorbent traps can be sampled for several days at a time, producing time-integrated emission trends impossible with discrete Method 29 tests

Contamination studies further highlight a systematic weakness in Method 29, where reagent purity and filter blanks can bias low-level metal data. The sorbent trap method avoids these pitfalls.

Broader Impact and Commercialization Potential

Accurate measurement of metallic HAPs is critical for regulatory programs and for compliance demonstrations by utilities, waste combustors, and other source categories. The Phase IIb results establish that sorbent traps can meet these needs at a fraction of the cost and complexity of traditional systems.

Commercial advantages include:

• Compatibility with existing PS 12B mercury sorbent-trap sampling hardware (with the addition of isokinetic sampling capabilities), enabling multi-pollutant monitoring with minimal additional equipment.
• Use in RATA testing where Method 29 is impractical or cost-prohibitive.
• Extension to biogas and RNG facilities, which face similar challenges measuring trace metals in gas streams

These attributes position the technology for rapid adoption by source testers, emissions monitoring system vendors, and plant operators seeking cost-effective compliance options.

Conclusions

Phase IIb of this SBIR project achieved all major technical milestones and demonstrated the readiness of the sorbent trap method for field deployment. Key conclusions are summarized below:

1. Material Improvement – Implementation of new production procedure resulted in reductions in background metals, improving measurement accuracy and blank subtraction at low levels.
2. Analytical Performance – Field comparisons with EPA Method 29 at dry and wet stacks yielded total-metal biases of -7.7 % and +1.3 %, respectively. Discrepancies observed in nickel and arsenic were traced to external factors and have since been mitigated.
3. Digestion and QA/QC Enhancements – Installation of an autoclave-style microwave digestion system reduced matrix interferences, reduced runtime by half, and improved spike recoveries.
4. Regulatory Readiness – Draft Alternative Test Method and Performance Specification documents have been submitted to EPA, supported by the completed Method 301 study.
5. Commercial Readiness – The technology is now positioned for transfer to market as both a short-term compliance tool and a platform for continuous metal-HAP monitoring.

Once approved, the sorbent-trap method will offer industry a reliable, lower-cost, and safer alternative to Method 29 and a foundation for continuous metals monitoring analogous to mercury PS 12B systems. Its adoption is expected to enhance data quality, reduce compliance uncertainty, and provide regulators with more accurate emission inventories for metals HAPs.