Concerns on information within SIMS Application for a City of Chicago Large Recycling Facility (LRF) Permit

Prepared by Donald Wink,1

Summary: In November, 2021, Metal Management Midwest, Inc., also known as “SIMS Metal Management,” filed an application to the City of Chicago to operate a Large Recycling Facility permit within the rules that the city issued in June, 2020. The comment period for this permit application goes until February 28, 2022. However, it is not possible to provide meaningful comments at this point, because the application, especially its Modeling Analysis, lacks clarity or information that can be interpreted in a meaningful way. Two problems are prominent. First, there is inadequate documentation of the results of the crucial Air Dispersion Modeling Analysis. Second, the data used for the Modeling Analysis is certainly based on incorrect information that also compromises any Modeling Analysis no matter how it is reported. Indeed, members of the ILEPA have indicated that they will not do their own, independent, modeling of the emissions from SIMS using this data. These concerns mean that the application may not be something that should be accepted for review in its current form.

I. Introduction

SIMS Metal Management operates a recycling facility on Paulina Street just south of Blue Island Avenue and west of Ashland Avenue. A key part of the operations, and certainly the largest source of hazardous pollutants, is the shredding of vehicles by a Hammerhill Shredder that, among other operations, processes “end-of-life” vehicles. This location is near to homes in the Pilsen community located north of Blue Island and across the South Branch of the Chicago River from homes in Bridgeport. There are two Chicago Public Schools facilities nearby: Whittier Elementary School (450 meters, 1500 feet) and Benito Juarez Community Academy (700 meters, 2300 feet).

The SIMS facility is subject to review and approval at three different levels: the Federal government (led by the United States Environmental Protection Agency, USEPA), the State of Illinois (led by the Bureau of Air within the Illinois Environmental Protection Agency, ILEPA), and the City of Chicago, under rules issued by the Chicago Department of Public Health (CDPH). At this time, there are various reviews going on. The CDPH is considering a permit for a Large Recycling Facility. The ILEPA, specifically the Bureau of Air, is involved in reviewing both an application for a Federally Enforceable State Operating Permit and an application for a construction permit required as part of an action by the Illinois Attorney General. The USEPA is awaiting the issuance of the FESOP as part of an Administrative Consent Order agreement after violations of the Clean Air Act that were revealed in 2019.

The most immediate review in terms of time will be the review by the CDPH. This includes a public comment period for the application that SIMS has filed with CDPH. This white paper concerns material in that LRF application and, briefly, some issues with a pending construction permit for a new emissions system as part of the Illinois Attorney General’s action of October, 2021.

II. Background information

II.a. Recap of major events with respect to SIMS permitting and emissions data

There have been multiple issues with pollution from SIMS, dating back more than 5 years. In December, 2018 the United States Environmental Protection Agency issued an Administrative Consent Order and entered into a Consent Agreement with SIMS, imposing a $225,000 settlement and laying out a process whereby SIMS was required to obtain a Federally Enforceable State Operating Permit (FESOP) from the State of Illinois. An application for this Permit was filed in January 2019 and amended in January 2020. The draft permit has not been issued.

One reason for the delay in issuing a draft permit was the requirement that SIMS obtain accurate information on emissions from the shredder. The original FESOP application included emissions data from a different SIMS facility, in Rhode Island. As part of the Administrative Consent Order, SIMS agreed to have testing done by an independent group. This was done by Mostardi Platt in September 2019.2 That data, which showed emissions fully 10 times that suggested based on the Rhode Island data, was then put into a revised FESOP application in 2020. According to the ILEPA, the next step would be to issue a draft permit for public comment and, since the area around SIMS is considered an Environmental Justice community by the State of Illinois, public hearings about the permit. There has been no action taken on this application as of January, 2022.

One reason why the draft permit has not been issued came to light in Fall, 2021. At that time, it was revealed that testing was done in May, 2021 that showed major problems with the ability to capture, let alone measure, emissions. Specifically, the test involved adding calibrated amounts of an inert tracer gas (sulfur hexafluoride) into the shredder’s air stream. The test showed “it was likewise confirmed that the capture efficiency of the system was estimated to be less than 50%."3

Although the Spring, 2021 testing did not measure the emissions of any pollutants, a direct conclusion is that any measurements made of emissions from the shredder would be inaccurate, likely by a factor of at least 2.

As a result of the discovery that emissions could not be captured properly, it was determined that SIMS was likely emitting more than 25 tons per year of volatile organic material (VOM), in violation of the Air Pollution Regulations of the State. This was the basis of a suit filed by the Illinois Attorney General preliminary injunction order against SIMS. The order, which was agreed to by SIMS, included a requirement that SIMS immediately develop a plan to properly capture emissions and, at the same time, to implement pollution controls on the shredder.

As this is being worked out (the application for the new controls was just filed with the state on December 20), the facility continues to operate with existing technology and controls. These are the technology and emission controls that are referred to in the LRF permit application to the CDPH.

b. CDPH Modeling Analysis Requirements

As mentioned, the SIMS facility is currently under scrutiny by the USEPA, the ILEPA, and the CDPH. The most immediate issue is with the LRF application to the CDPH. This application was filed by SIMS in November, 2021 and the City has given the community until the end of February, 2022, to give comments. This paper relates to whether such comments can be meaningfully provided, based on the data and Modeling Analysis in the LRF application.

The LRF application needs to adhere to rules issued by the Chicago Department of Public Health in 2020.4 These rules have been criticized in other venues but for the purpose of this comment the rules will be treated as is.

Among the rationale for the rules is a clear concern that “these facilities can be significant sources of dust, contaminated storm and process water discharges, metal-containing particulate or vapor, and possible radiation with the potential to harm human health and the environment, and cause a public nuisance or adversely impact the surrounding area or surrounding users.” As a consequence, the rules require that permits include an Air Quality Impact Assessment (Section 3.9.21). This requires a study that “shall evaluate PM10 emissions that may be generated at the Facility,” where PM10 refers to particulate matter that is 10 microns or greater in size. The rules for the Modeling Study also require that “In addition to PM10, Facilities that receive scrap metal or metallic Recyclables shall evaluate the following HAPs in the modeling study: antimony, arsenic, beryllium, cadmium, chromium, cobalt, lead, manganese, nickel, and selenium compounds.”

This Modeling Analysis is then the basis of a required dust monitoring plan to be implemented on an ongoing basis at the perimeter of the Facility and “each location of the Facility or Property determined in the air-dispersion modeling study to potentially exceed EPA’s 24-hour standard for PM10 or relevant acute or chronic health screening limits or standards for the modeled HAPs.”

Therefore, a central part of the CDPH rules is the availability of an air modeling analysis for both PM10 and specific HAPs that can be interpreted in a meaningful way and that can allow for comment. Absent a meaningful Modeling Analysis, the community (and presumably the CDPH) cannot understand the potential impact of the proposed LRF operation and cannot make effective comments.

III. Examination of Modeling Analysis in the SIMS application

The Modeling Analysis in the SIMS application is provided in Appendix R, also prepared by Trinity Consultants. This uses the required USEPA AERMOD software, which is publicly available from the USEPA. The Modeling Analysis notes that the AERMOD is “incorporated within Trinity’s BREEZE™ AERMOD Pro software”, which is, apparently, proprietary and therefore not available for review.

III.a. Meteorological and geographic information.

The Analysis details how important information on the influence of wind and weather is input into the software. Though using wind data from Midway airport is not ideal, it is likely the best possible highly accurate input. The Analysis also presents a detailed, and meaningful, discussion of how the surrounding land use was considered. On the other hand, there is no information about how the particular geography of Pilsen, including especially the presence of the Chicago River and its impact on wind patterns, was considered. Regardless, a reasonable interpretation is that the Modeling Analysis was prepared following standard practice and, though data from closer to Pilsen would be better, it is difficult to improve on these considerations.

III.b. Modeling of PM10.

More important are considerations of how to model the parameters that are specific to SIMS and its operations. Within the report, there is an indication of what input data for different emissions sources (including but not limited to the shredder) are used as input data for the software. The report indicates “PM10 emission rates are based on existing permit limits, AP-42 emission factors, and stack test results” (emphasis added). This is the first indication of a severe problem interpreting the results. At this critical point, information is provided that cannot be interpreted by anyone wishing to comment on the permit application. For example:

  • In general terms, there is no definition of the units or the time span for the rates.

  • The discussion of the inputs for the Modeling Analysis is incomplete, for example regarding why it would use permit limits and, if so, which permit limits are used.

  • Terminology is not explained, for example the meaning or use of an AP-42 emission factor.

  • There is no indication of which stack test results are used. Even given that there are no valid stack tests to work from (see later), the reader at least needs to know whether what was used here were the Rhode Island data, the revised data from Mostardi Platt, or some correction given the May 2021 capture test results.

The Modeling Analysis does provide Attachment Tables with values for 24 hour PM10 emission rates from different locations in the facility.5 Presumably these are some form of average emission rates over a 24 hour period. The major source is the shredder (Attachment Table B-1, pdf page 385), with a value of 0.7507 g/s. This would equate to 64860 grams per day (143.2 pounds) and 2.369 × 10^7^ grams per year (52,300 pounds, 26.15 tons). How the value of 0.7507 g/s was determined is not described, though this does seem in line with the data in the revised FESOP permit, which has a total potential to emit value for all PM emissions of 29.38 tons per year (Revised Table 1).

The analysis then presents what are deemed to be the “Criteria Pollutant Modeling Results” (Section 2.10 of Appendix R). This is done in two ways. One, Table 2-2, “presents the predicted highest 6th high 24-hour average concentration over five (5) years from 2012 to 2016.” This table is presented in the figure below, a screen shot from the report (taken from p. 378 of the full pdf).

Table Description automaticallygenerated

There are several reasons why this Table may be inadequate for the requirements of the CDPH rules and is certainly inadequate for the ability of the community to offer substantial comment on the application. Among these are the following:

  • There is no discussion of why 2012-2016 data are used. These are certainly out of date for any understanding of what the facility might be emitting today.

  • This offers only information on the “highest 6th 24-hour average.” Besides not explaining what that means, this seems to embed an assumption that peak and not total PM10 emissions is the basis of health concerns.

  • The location of the coordinates is not explained. There are multiple areas of concern, not just the one select for reporting.

The second representation of the results is given in a figure, shown here (it is on p. 378 of the full pdf). It is easy to understand that the SIMS facility covers the space that is presented at the center of the figure in a “clear” and not color-coded region site at the center of the figure. The meaning of the red, orange, yellow, green, and blue coding is not provided. The description of this Figure suggests that it is “outlining the 24-hour PM10 concentrations surrounding the Paulina Street Facility.” But it is not explained what the colors mean or what “24-hour PM10 concentrations” mean.

Map Description automaticallygenerated

Problems with this representation include the following:

  • All of the comments presented for Table 2-2, above, including the dates of analysis and just what 24-hour results mean.

  • A lack of information on the color coding, both in absolute terms and in comparison to available standards and references.

  • Information on the impact outside of the Figure—presumably extending further north, east, and west in Pilsen and south / southeast into the Bridgeport community.

Hence, the information provided about the PM10 modeling is not in a form or with explanation to allow meaningful interpretation and comment by the community.

III.b. Modeling of hazardous air pollutants

The next section of the Modeling Analysis covers the requirement in the rules for *"*antimony, arsenic, beryllium, cadmium, chromium, cobalt, lead, manganese, nickel, and selenium compounds.” This does reference a specific stack test: the Mostardi Platt results from September 2019. The Analysis also presents the specific standards (taken from different sources, presented in Table 3-1). The presentation of the table is somewhat confusing at first glance, especially given that in some cases there are multiple references for some HAPs. But, it does provide the expected benchmarks. It would be better if the table noted which of these are incorporated in regulation (lead, for example) and which are guidelines that have not been validated (manganese). There may also be reasons to think that the standards are not ideal, especially for example with manganese and lead, based on the latest scientific information. However, they are the ones that are currently applicable.

The methodology used for HAP modeling is not described in a meaningful way, making the results unreliable. Specifically, the Modeling Analysis includes an indication that “metal HAP emission rates obtained during sampling were used to scale the PM10 emission rates for all emission units to their respective metal HAP emission rates.” Why this was done and, importantly, the actual values of inputs used for the scaling are not presented. Further, the known problems with the September 2019 data again mean that the input data are unreliable as an indication of emission rates from the shredder.

The actual results in this case are presented in terms of a single output value of a “Predicted average” for different time scales, as is probably appropriate given the different standards that are used. Screenshots for two cases, lead and manganese (excerpted from Table 3-3) are presented here.

Table Description automaticallygenerated

Table Description automaticallygenerated{width="6.5in” height="3.2263888888888888in”}

There are again several problems with this presentation of these results that prevent interpretation for the purposes of preparing comments to the permit. Among these are:

  • No indication of where, geographically, these results are determined and certainly no presentation of a map showing how the levels might vary depending on location or by time of year and weather patterns.

  • Unexplained presentation of results from several years ago and, for the case when the averaging period is less than one full year, which averaging period was selected.

  • In the case of lead, presentation of a “maximum rolling average,” which is actually the simple average of results from three different periods.

  • For the HAPs other than lead, presentation of a “maximum predicted impact” for the HAPs other than lead that seems to be just a maximum from a list of unspecified dates within a particular year.

  • No information on what modeling might be for the current situation (i.e, December 2021) or even at the last time any kind of actual emissions data were actually collected (September, 2019).

As with the PM10 data, the Modeling Analysis presented for the HAPs lacks essential information and explanation to let the community comment on this application at this point.

IV. Issues of emissions

A key question for the new modeling analysis is associated with the uncertain nature of what SIMS has been emitting and its impact on the community. Still, it is helpful to note some of what has been reported and to raise issues that the Modeling Analysis, properly done, should address.

IV.a. Questions raised by SIMS Emission Testing, September 2019

In September 2019, an independent company (Mostardi Platt) did emissions testing at the SIMS site in Pilsen. This included manganese along with lead (and other metals). Excerpts from the reports are given below.

Table Description automaticallygenerated
Summary Test Results for Particulates and Metals from Mostardi Platt report, September 2019

Table Description automaticallygenerated
Summary Test Results for Organic Emissions from Mostardi Platt report, September 2019

These and other data in the September 2019 test were used in a revised table in the FESOP application.

Table Description automaticallygenerated
Potential Emissions from FESOP Application January 2020

In general, these results showed emission rates more than 10 times higher than originally suggested (based on the Rhode Island data), including 15 pounds per year for lead. The results also noted a potential to emit 11 pounds per year for manganese, which was not part of the Rhode Island data. These results can be compared to a similar test done at the shredder at the General Iron location in Lincoln Park, which was conducted in Fall, 2018. Emissions measured at SIMS for these metals are 8-9 times higher than those that had been measured at General Iron.

There is one other important point to consider about these data. As noted, the May, 2021 capture efficiency testing that was required by the ILEPA showed that less than half of the emissions were likely being captured at the measurement point. As a result, the actual emissions may be much higher than are being used in the Modeling Analysis. This is never mentioned in the LRF application.

With this in mind, it is clear that the data used in the Modeling Analysis in the LRF application is not reliable at this point. *Members of the ILEPA Bureau of Air have indicated to PERRO members that they will not do their own, independent, modeling of the emissions from SIMS using this data. *

IV.b. Airborne lead and other metals in the Pilsen neighborhood: Perez Elementary School data

One of the important parts of the environment in Pilsen is the continuous monitoring of the air adjacent to the H. Kramer site at 21^st^ and Loomis for many metals, including lead. This includes a station at Perez Elementary School that is close to H. Kramer but also about 1300 meters (4300 feet, 0.80 miles) from SIMS. This shows a dramatic decrease in lead in the air in the period since 2010, along with the almost complete elimination of lead releases as reported by H. Kramer through the EPA Toxic Release Inventory, to just 5 pounds in the 2018 and 2019 reporting year. Data provided to PERRO by the ILEPA and graphed in the figure below shows that lead in the air has increased since 2018 (as have manganese, chromium, and cadmium).6

IV.c. Implications of SIMS and Pilsen monitoring data

The values of lead in the air at Perez are, it should be noted, below the current levels that require action by the USEPA (0.15 μg/m^3^), though in 2020 they were close to 10% of the USEPA action level. Still, given that the lead measured at Perez has not gone away and has in fact shown a recent increase, it is highly likely there is another source of airborne lead in the community. We know from the Fall 2019 data that, at a minimum, SIMS is operating a shredder capable of putting more than 15 pounds of lead in the air per year, and may well be emitting more than twice that. That, and these data, suggest that SIMS may be the source of the continuing, and growing, measurements at Perez, 1800 meters from SIMS. It is also important to note that the Perez monitor is by no means “close” to SIMS. By comparison, Whittier Elementary is only 450 meters away.

IV.d. Implications of December, 2021 SIMS proposal for additional VOM controls

As mentioned, an important change may be coming for the operation of the shredder at SIMS, as outlined in a recent submission of a plan for additional controls based on the outcome of the Illinois Attorney General’s injunctive action in October, 2021. The proposal for this change centers on the introduction of a regenerative thermal oxidizer (RTO) to process emissions prior to release in the atmosphere. The RTO will likely have a significant impact on the rate of emissions of volatile organic material per ton of processed material, since it introduces an additional combustion step applied to the gas emissions, turning that organic matter into carbon dioxide and water. There is no information or expectation that that process will have any impact on particulate matter, which is often mineralized. And it will certainly have no impact on emissions of HAPs, since all additional oxidation will do is, perhaps, to turn metal dust into metal oxide dust.

  1. This comment is prepared by Donald J. Wink, PhD, Professor of Chemistry at the University of Illinois Chicago. This is part of his work to provide scientific advice and analysis to PERRO. This comes as a continuation of work initiated with the NSF-Funded UIC INCLUDES project (Award 1649298) and its support of community-identified science, technology, engineering, and mathematics issues. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation, the University of Illinois System, or the University of Illinois at Chicago. ↩︎

  2. Metal Shredder Emissions Report, October 18, 2019, Mostardi Platt, Inc., Elmhurst, IL. ↩︎

  3. Memo “Metals Management Midwest (031600FFO) Proof-of-Concept Test for Alternative Capture Technique (May 13 & 14, 2021)” from Kevin Mattison, Illinois Environmental Protection Agency, dated July 14, 2021. ↩︎

  4. Rules for Large Recycling Facilities, City of Chicago, Effective June 5, 2020 ↩︎

  5. The Modeling Analysis does not include a discussion of the emissions that will result from the trucks that are used to bring materials to SIMS. It is noted that these emissions are not required for the LRF permit, though it is important to note that the impact of the facility on the community certainly includes those truck emissions. This, as is noted elsewhere, may be a problem with the CDPH permitting process and, therefore, I am not including them in this paper. ↩︎

  6. The graph does show an increase in lead in 2017, something that can be explained based on data on emissions for H. Kramer. What cannot be explained in that way is the increase since 2018. ↩︎