Dioxins and Furans
in Rillito, Arizona

in consultation with the Arizona Department of Health Services
January 2006

Dioxins and Furans in Rillito, Arizona

Arizona Department of Environmental Quality,
in consultation with the Arizona Department of Health Services

January 2006

Publication Number OFR 06-01

1

Executive Summary
Concerns of Rillito citizens about emissions of dioxins and furans from the Arizona
Portland Cement Plant (APCC) and stack testing at APCC led to an ambient air
monitoring program conducted in 2004 - 2005. Ambient concentrations of dioxins and
furans proved to be within health-based guidelines. These concentrations were somewhat
higher than those measured at background sites but considerably lower than at urban
sites. The distribution of the 17 toxic dioxins and furans in the ambient air of Rillito
closely resembled the mobile source profile but was quite different from the APCC
profile. The dioxin and furan concentrations measured during the one-year study
suggested that mobile source emissions had a much greater impact at Rillito than the
cement plant emissions.
Introduction to Dioxins and Furans
“Dioxins and furans” is a term used to describe a group of environmentally persistent
chemicals that are metabolized slowly and therefore tend to bioaccumulate, especially in
the fat and in the liver. These compounds are the unintended by-products of virtually all
combustion. Although numbering 210 different compounds, the environmental
community has focused on the 17 found to have adverse health effects in humans or
animals. Several international organizations have classified one of the two most toxic of
these compounds -- 2,3,7,8-tetrachlorodibenzo-para-dioxin (2,3,7,8-TCDD)—as a human
carcinogen.
To assess the cumulative toxicity of a mixture of these compounds – also called
“congeners” – the World Health Organization has assigned a Toxic Equivalency Factor
(TEF) to each compound relative to the toxicity of 2,3,7,8-TCDD. These TEFs are the
results of scientific judgment of a panel of experts using all of the available data and are
selected to account for uncertainties in the available data and to avoid underestimating
risk. Thus, they are “public health conservative” values. When this TEF is multiplied by
its congener concentration, and all of these products are summed, the result is the Toxic
Equivalency (TEQ) of the mixture. Table 1 gives the names, abbreviations, and toxicity
factors for the 17 toxic dioxins and furans.
Typical units of measurement for ambient dioxins and furans are picograms per cubic
meter (pg/m3) or femtograms per cubic meter (fg/m3). A picogram is one trillionth of a
gram, or 0.000000000001 of a gram, or, in scientific notation,
1x10-12 gram. Femtograms are 1000 times smaller than picograms: one quadrillonth of a
gram, or 0.000000000000001 of a gram, or 1x10-15 gram. Results of stack tests are
reported in picograms per dry standard cubic meter (pg/dscm). Concentrations expressed
in these units, both ambient and stack, can be reported directly or can be converted into
the relative toxicity or TEQ. In this paper most results will be reported as their TEQ, the
ambient in femtograms; the stack, in picograms.
Dioxins and furans belong to that class of compounds which are found in both the
gaseous and particulate phase. Depending on the individual compound, temperature, and

2

general particulate concentration, a particular congener can be either gaseous or
particulate. The term “semi-volatile organic compounds” is used to describe these
compounds that can partition between the two phases. In this paper concentrations of
both the particulate phase and gaseous phase dioxins and furans will be presented.
Table 1. Dioxins and Furans
Name
2,3,7,8-tetrachlorodibenzo-p-dioxin
1,2,3,7,8-pentachlorodibenzo-p-dioxin
1,2,3,4,7,8-hexachlorodibenzo-p-dioxin
1,2,3,6,7,8-hexachlorodibenzo-p-dioxin
1,2,3,7,8,9-hexachlorodibenzo-p-dioxin
1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin
1,2,3,4,6,7,8,9-octachlorodibenzo-p-dioxin
2,3,7,8-tetrachlorodibenzo-p-furan
1,2,3,7,8-pentachlorodibenzo-p-furan
2,3,4,7,8-pentachlorodibenzo-p-furan
1,2,3,4,7,8-hexachlorodibenzo-p-furan
1,2,3,6,7,8-hexachlorodibenzo-p-furan
1,2,3,7,8,9-hexachlorodibenzo-p-furan
2,3,4,6,7,8-hexachlorodibenzo-p-furan
1,2,3,4,6,7,8-heptachlorodibenzo-p-furan
1,2,3,4,7,8,9-heptachlorodibenzo-p-furan
1,2,3,4,6,7,8,9-octachlorodibenzo-p-furan

Abbreviation
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF

TEF*
1.0
1.0
0.1
0.1
0.1
0.01
0.0001
0.1
0.05
0.5
0.1
0.1
0.1
0.1
0.01
0.01
0.0001

*TEF: toxic equivalency factor: 2,3,7,8-TCDD and 1,2,3,7,8-PeCDD are the two most
toxic congeners and have a TEF set to 1.0. The toxicity of the other 15 is expressed as a
decimal fraction of the most toxic two. Their toxicity ranges from 50% (0.5) to 0.01%
(0.0001) of 2,3,7,8-TCDD.
Environmental scientists for the last two decades have determined that different types of
combustion produce different patterns, or “profiles”, of dioxins and furans congeners.
Profiles from specific types of combustion can then be compared with those profiles in
ambient air measurements. These comparisons shed light on which sources are most
influential in the dioxin and furan composition of the ambient air.
Background of the Present Study
The kilns at APCC are subject to the dioxin and furan standard in the Portland Cement
Maximum Achievable Control Technology Standard (40 CFR 63 Subpart LLL). The
MACT Standard was promulgated on June 14, 1999. APCC was required to conduct a
test to demonstrate compliance with the dioxin and furan emission limits by December
14, 2002. The permit (MP190310P1-00) requires that subsequent tests be conducted on
an annual basis.
3

During tests conducted in December 2002, APCC failed to demonstrate compliance with
the standard in four out of five operational configurations. APCC conducted follow-up
tests in January 2003, and failed to demonstrate compliance in one out of five operational
configurations. It was determined that the earlier exceedances were due to Alumina
Cement Additive (ACA) in the feedstock and subsequently ACA was removed from the
feedstock. An additional set of tests was conducted in late February 2003. The results
from the February tests demonstrated that the emissions during the February tests were
below the MACT standard. In June 2004, APCC conducted a dioxin furan test on Kilns
1, 2, and 3 to determine if they could operate at a higher baghouse inlet temperature and
still meet the dioxin/furan (D/F) standard. APCC failed this performance test and was
again required to show compliance at their normal operating temperature range.
Subsequent tests for Kilns 1, 2, and 3 conducted in July 2004 and March 2005 showed
that the standard was again being met.
Tests which demonstrated noncompliance with the dioxin and furan standard raised
concerns in the community about potential exposure levels and the associated health
impacts. In response to these concerns, ADEQ agreed to conduct an ambient air study to
assess Dioxin and Furan ambient concentrations in the Rillito community. ADEQ staff
enlisted the help of EPA Region 9 (San Francisco) and Region 7 (Kansas City) to obtain
the necessary technical advice and laboratory assistance to carry out a dioxin and furan
ambient air study. Conducted from February 2004 through January 2005 by ADEQ
staff, this monitoring work at Rillito has provided the dioxin and furan concentrations
necessary to assess human health risk. These concentrations will be discussed in the
remainder of this paper, in the context of dioxin and furan congener profiles from APCC
and from mobile sources.
Sources
Dioxins and furans are produced by combustion, so only combustion emission sources
figure directly into their airborne concentrations. For example, mobile source emissions
in this comparison are limited to exhaust, and cement plant emissions are limited to
combustion emissions from stacks. In this section the air pollution sources that affect
Rillito are described. Where available, dioxin and furan emissions will be presented, as
well as the conventional pollutants. The intention here is to provide the relative strengths
of these sources. Considering only three sources – the APCC, Interstate 10, and the
Union Pacific Railroad – the former has much greater emissions of all pollutants except
volatile organic compounds (VOC) than the mobile sources (Table 2).
The total range of sources contributing to the dioxins and furans in this study needs to be
explained. Dioxins and furans, like virtually all other air pollutants, are deposited on the
ground and foliage surfaces through both wet and dry deposition. This deposition leads
to appreciable concentrations of dioxins and furans on their surfaces. Windblown and
vehicle-generated dusts that have been subject to dioxins and furans deposition can
contribute to ambient concentrations at such sites as Rillito. Quantifying this source was

4

beyond the scope of this study, so the focus remains on the combustion sources
themselves.
Table 2.

1
2

Annual Air Pollution Emissions for Three Major
Sources in Rillito

Units
APCC1
Total Mobile
I-10
Railroad
Ratio

PM10

CO

NOx

SOx

VOC

Tons/yr
439.0
1.7
1
0.7
258.2

Tons/yr
3956.2
281.6
279
2.8
14

Tons/yr
3460.2
69.1
50
19.5
50.5

Tons/yr
11.6
2.5
0.4
2.1
4.6

Tons/yr
5.7
33.8
33
1.0
0.2

DF
grams/yr
TEQ
0.5 2
0.0006122
0.0006100
0.0000022
817

Average of emission rates from 2001, 2002, and 2003 Emission Inventory Reports
This emission estimate is based on the average of test results from February 2003 tests and July and
April 2004 test results.

PM10
CO
NOx
SOx
VOC
DF
I-10
Railroad
Ratio

Particles 10 microns and smaller
Carbon monoxide
Nitrogen oxides
Sulfur oxides
Volatile organic compounds
Dioxins and furans, with the units of grams per year
Based on 52,000 vehicles per day on one mile of roadway, 10% diesel
Based on one mile of line haul emissions, 20 trains per day
APCC to total mobile

Description of Measurements
APCC Stack Testing
Five stack tests for Dioxins and Furans were conducted in December 2002. An
additional five stack tests were conducted in January 2003. Four stack tests were also
conducted on Kiln 4 in April 2004 and one stack test was conducted on Kilns 1-3 in June
2004. One stack test was also conducted in July 2004. The most recent five stack tests
were conducted in March 2005. The required test method is EPA Method 23. Initial
tests are required by the Portland Cement MACT Standard (40 CFR 63 Subpart LLL).
Kiln 4 is required to be tested in two different configurations, raw mill on, and raw mill
off. The subsequent tests only need be performed in the configuration that resulted in the
highest emission rate. Tests were done on the following kilns and conditions (Table 3):

5

Table 3a - g. Stack Test Conditions and Results at APCC
Table 3a
December 9-14, 2002
Kiln

Raw Mill
Configuration

1-3

N/A

4

Off/Low
Temp.
Off/High
Temp.
On/Low
Temp.
On/High
Temp.

4
4
4

Average
Baghouse
Inlet
Temperature
(º F)
635

Results
ng/dscm
TEQ

Allowable
ng/dscm
TEQ

Pass/Fail

0.62

0.20

Fail

519

0.29

0.20

Fail

694

4.7

0.20

Fail

414

0.05

0.20

Pass

687

0.6

0.20

Fail

Average
Baghouse
Inlet
Temperature
(º F)
607

Results
ng/dscm
TEQ

Allowable
ng/dscm
TEQ

Pass/Fail

0.02

0.20

Pass

484

0.03

0.20

Pass

690

0.76

0.20

Fail

459

0.03

0.20

Pass

678

0.06

0.20

Pass

Table 3b
January 14-18, 2003
Kiln

Raw Mill
Configuration

1-3

N/A

4

Off/Low
Temp.
Off/High
Temp.
On/Low
Temp.
On/High
Temp.

4
4
4

6

Table 3c
February 24-28, 2003
Kiln

Raw Mill
Configuration

1-3

N/A

4

Off/Low
Temp.
Off/High
Temp.
On

4
4

Average
Baghouse
Inlet
Temperature
(º F)
K1(718)
K2(619)
K3(646)
570

Results
ng/dscm
TEQ

Allowable
ng/dscm
TEQ

Pass/Fail

0.05

0.20

Pass

0.04

0.20

Pass

600

0.10

0.20

Pass

700

0.04

0.20

Pass

Results
ng/dscm
TEQ

Allowable
ng/dscm
TEQ

Pass/Fail

0.06

0.20

Pass

Table 3d
April 13-29, 2004
Kiln

Raw Mill
Configuration

4

Off

Average
Baghouse
Inlet
Temperature
(º F)
625

4

Off

600

0.05

0.20

Pass

4

Off-Oil Temp.

600

0.06

0.20

Pass

4

On

692

0.04

0.20

Pass

7

Table 3e
June 2-3, 2004
Kiln

Raw Mill
Configuration

1-3

N/A

Average
Baghouse
Inlet
Temperature
(º F)
705(Kiln 1)
720 (Kiln 2)
706 (Kiln 3)

Results
ng/dscm
TEQ

Allowable
ng/dscm
TEQ

Pass/Fail

0.22

0.20

Fail

Average
Baghouse
Inlet
Temperature
(º F)
653 (Kiln 1)
652 (Kiln 2)
671 (Kiln 3)

Results
ng/dscm
TEQ

Allowable
ng/dscm
TEQ

Pass/Fail

0.04

0.20

Pass

Results
ng/dscm
TEQ

Allowable
ng/dscm
TEQ

Pass/Fail

0.13

0.20

Pass

0.11

0.20

Pass

0.13

0.20

Pass

Table 3f
July 26-27, 2004
Kiln

Raw Mill
Configuration

1-3

N/A

Table 3g
March 4-5, 8, 2005
Kiln

Raw Mill
Configuration

1-3

N/A

1-3

N/A

4

Off

Average
Baghouse
Inlet
Temperature
(º F)
688 (Kiln 1)
687 (Kiln 2)
686 (Kiln 3)
649 (Kiln 1)
648 (Kiln 2)
670 (Kiln 3)
693

4

Off

620

0.19

0.20

Pass

4

On

693

0.18

0.20

Pass

These stack test data above are summarized in Figure 1

8

1

Figure 1 - Summary of Dioxin/Furan Tests

0.9

Tested value was 4.7 ng/ dscm TEQ

0.8

K4: Kiln 4

0.7

K123: Kilns 1,2,3
K12: Kiln 1,2
RM: Raw Mill
T: Temper at ure

0.6

TEQ: Toxic Equivalent
ng/ dscm: (nanogr ams)/ (dr y st andard cubic met er)

0.5

0.4

0.3

0.2

M ACT STANDARD

FAIL
PASS

0.1
0.2

0
K4 RM On/High K4 RM Off/High K4 RM On/Low
T
T
T

December 2002 Testing
April 2004 Testing
March 2005 Testing

K4 RM
Off/Low T

January 2003 Testing
June 2004 Testing

K4 RM Off w ith
Oil

K123

February 2003 Testing
July 2004 Testing

9

To produce a congener profile, several of these tests have been averaged for their
congener concentrations (Table 4)
Table 4.

1

Rillito Dioxins and Furans: Stack Tests from APCC, with All
Concentrations Expressed as Toxic Equivalents
Name
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
2,3,7,8-TCDD (TEQ)

TEF
1.0
1.0
0.1
0.1
0.1
0.01
0.0001
0.1
0.05
0.5
0.1
0.1
0.1
0.1
0.01
0.01
0.0001
---

pg TEQ/dscm1
6.533
7.576
0.292
0.351
0.257
0.078
0.001
16.055
2.591
33.821
1.601
1.699
1.483
0.392
0.117
0.018
0.000
72.864

Average of stack tests conducted in April 2004 on Kiln 4 (RM On, RM Off with Oil, RM Off at 625 °F, RM Off at
600 °F), June 2004 test conducted for Kilns 1-3,and July 2004 test conducted for Kilns 1-3.

Ambient Air Measurements
Detailed descriptions of the sampling and analysis methods for dioxins and furans can be
found in “Quality Assurance Plan: Monitoring for Polychlorinated Dibenzo-p-Dioxins
and polychlorinated Dibenzofurans in Rillito, Arizona”, Arizona Department of
Environmental Quality, February 6, 2004 and “Determination of Polychlorinated,
Polybrominated and Brominated/Chlorinated Dibenzo-p-dioxins and Dibenzofurans in
Ambient Air”, EPA, 1999. Dioxins and furans can be found in two phases: as particles,
and as gases, or “semi-volatile organic compounds”. This dual phase phenomenon
requires that two sampling media be employed: a quartz fiber filter for airborne
particulates; followed by a polyurethane foam cartridge for the gaseous and semi-volatile
organic compounds. In the Rillito work, the long-term air pollution monitoring site at the
Gremmler residence was used.
Conducted from February 2004 through January 2005, the sampling schedule followed
that of the National Dioxin Air Monitoring Network (NDAMN). The sampling periods
were selected to coincide with the quarterly NDAMN sampling. Each sampling period
10

consisted of 26 (or 27) days, with 20 days actually being sampled. These 20 sampling
days were divided into four-, and five-day periods, each period being separated by two
days. This schedule allows the operator to change the filter and foam substrate in the
daytime, rather than at midnight, the hour that the sampling begins. Because of a contract
lapse in the national program, the month of May was not sampled at any NDAMN sites
throughout the United States, nor was it sampled at Rillito. Dates of the sampling periods
are given in Table 5. All sampling media were shipped to the EPA Region 7 laboratory in
Kansas City for analysis.
Table 5. Dioxin and Furan Sampling Dates in Rillito
Number
Start Date
End Date
of Days
2/12/2004
3/9/2004
26
3/11/2004
4/6/2004
26
4/8/2004
5/4/2004
26
5/27/2004
6/22/2004
26
6/24/2004
7/20/2004
26
8/5/2004
8/31/2004
26
9/2/2004
9/28/2004
26
9/30/2004 10/26/2004
26
11/3/2004 11/30/2004
27
12/2/2004 12/28/2004
26
1/6/2005
2/1/2005
26
2/12/2004
2/1/2005
287

Actual
Sampling
Days
20
20
20
20
20
20
20
20
20
20
20
220

The results of the ambient air sampling program are summarized in this section. These
concentrations should be interpreted with the following insights:
1. Each value is expressed as the toxic equivalent (TEQ) of the concentration in
femtograms per cubic meter (fg/m3).
2. The annual average is calculated in the following way. First, the average of the
11 semi-volatile organic compound concentrations is determined. Second, the
average of the 11 particulate concentrations is determined. Third, the two
averages are added. Fourth, this annual average concentration is converted to the
TEQ.
3. In laboratory analyses zero concentrations are never reported. The laboratory can
determine two lower limits. The lower of these two is called the minimum
detection limit. The higher of the two is called the minimum reporting limit. In
this study the concentrations at the lower end of laboratory sensitivity will be
based on the more conservative values (i.e. minimum reporting limits).
11

4. Values below the minimum reporting limit have been treated in two ways. In the
first method, they have been assigned a concentration of one half of the reporting
limit. In a separate analysis, they have been set equal to zero.
5. When most of the laboratory analyses are below the reporting limit, as Table 6
shows, the treatment of sub-reporting-limit values has important consequences in
calculating the annual averages.
Table 6.

Dioxins and Furans in Rillito: Frequency of Values below the
Reporting Limit
Statistic
Number of analyses
Number of Non-reported
Percentage of Non-reported

SemiParticulates volatiles Total
187
187
374
75
175
250
40%
94% 67%

6. This importance can be seen in Table 7, in which the TEQs have been calculated
first with the sub reporting limit values as zero and second with the sub reporting
limit values as one half of the reporting limit. The latter method produces a TEQ
3.74 times greater than the former.
7. Various statistical methods can be applied to such data sets to reduce the
influence of the nondetectable values (see, for example, “Guidance for Data
Quality Assessment”, EPA QA/G-9, EPA/600/R-96/084, January 1998). Cohen’s
method and the method of proportions are two such methods. Their application is
restricted to those compounds whose nondetectable frequency is within the
method’s limits. For Cohen’s method, only three compounds are suitable and
they comprise only 14% of the TEQ. The method of proportions could be applied
to five compounds (15% of the TEQ), but this method does not yield a specific
numerical value for the average. Of the total TEQ, 74% of the toxicity resides in
compounds that were either not detected at all or only one to three times out of the
eleven. No statistical method can be applied to these compounds. In summary,
the available statistical treatments for these dioxins and furans data either don’t
apply or don’t change the answer appreciably.

12

Table 7.

Annual Average Dioxin and Furan Concentrations in Rillito with Two
Methods of Treating Sub Reporting Limit Values (TEQ fg/m3)
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
TEQ

Non-reported Values Treated As
Zero
½ Reporting Limit
0.000
1.000
0.000
7.000
0.155
0.791
0.618
1.127
0.800
1.309
1.245
1.280
0.046
0.046
0.527
0.559
0.000
0.350
0.409
3.750
0.491
1.095
0.255
0.859
0.073
0.741
0.682
1.223
0.425
0.457
0.074
0.134
0.003
0.004
5.802
21.726

Given the insights just discussed, the ambient concentrations are now presented as annual
averages, with the particulate and semi-volatile phases shown separately, in Tables 8 and
9. In Table 8 the averages have been calculated with those values below the reporting
limit set to zero. In Table 9 the averages are based on setting those values below the
reporting limit to one half the value of that limit. The logical interpretation of these
tables is to view their respective averages as a range. The low end of the range is 5.8
fg/m3 and the upper end is 21.7 fg/m3. Dioxins and furans in Rillito, averaged for the
year, lie somewhere in this range, not below and not above it. Because of the generally
low levels of dioxins and furans that led to multiple nondetectable values, and because of
the uncertainty of those concentrations below the minimum reporting limit (they can be
anywhere from zero up to the limit itself), it is impossible to pinpoint an exact
concentration. All that can be said for certain is that the annual average of dioxins and
furans in Rillito, expressed as the toxic equivalent in femtograms per cubic meter,
is 5.8 – 21.7.

13

Table 8.

Rillito Dioxins and Furans: Annual averages, with all concentrations
expressed as Toxic Equivalents in fg/m3, and with values below the
Reporting Limit set to zero
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
TEQ

TEF
1.0
1.0
0.1
0.1
0.1
0.01
0.0001
0.1
0.05
0.5
0.1
0.1
0.1
0.1
0.01
0.01
0.0001

SemiParticulate volatiles Ambient
0.000
0.000
0.000
0.000
0.000
0.000
0.155
0.000
0.155
0.618
0.000
0.618
0.800
0.000
0.800
1.245
0.000
1.245
0.046
0.000
0.046
0.100
0.427
0.527
0.000
0.000
0.000
0.409
0.000
0.409
0.491
0.000
0.491
0.255
0.000
0.255
0.073
0.000
0.073
0.682
0.000
0.682
0.415
0.011
0.425
0.074
0.000
0.074
0.003
0.000
0.003
5.364
0.438
5.802

Notes:
All particulate and semi-volatile concentrations reported below the reporting limit were
assigned a value of zero.
“Ambient” concentrations are the sum of the particulate and semi-volatiles
concentrations.
TEF: toxic equivalency factor
TEQ: toxic equivalency

14

Table 9.

Rillito Dioxins and Furans: Annual averages, with all concentrations
expressed as Toxic Equivalents in fg/m3, and with values below the
Reporting Limit set to one half the reporting limit
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
TEQ

TEF
1.0
1.0
0.1
0.1
0.1
0.01
0.0001
0.1
0.05
0.5
0.1
0.1
0.1
0.1
0.01
0.01
0.0001

Particulate
0.500
3.500
0.441
0.777
0.959
1.245
0.046
0.127
0.175
2.000
0.745
0.509
0.391
0.873
0.415
0.099
0.003
12.805

Semivolatiles Ambient
0.500
1.000
3.500
7.000
0.350
0.791
0.350
1.127
0.350
1.309
0.035
1.280
0.001
0.046
0.432
0.559
0.175
0.350
1.750
3.750
0.350
1.095
0.350
0.859
0.350
0.741
0.350
1.223
0.043
0.457
0.035
0.134
0.001
0.004
8.921
21.726

Notes:
All particulate and semi-volatile concentrations reported below the reporting limit were
assigned a value one half the limit.
“Ambient” concentrations are the sum of the particulate and semi-volatiles
concentrations.
TEF: toxic equivalency factor
TEQ: toxic equivalency

15

Discussion of Ambient Measurements
Comparison with AAAQG
The Arizona Ambient Air Quality Guidelines (AAAQG) have recommended
concentrations not to be exceeded for about 300 toxic compounds, including 2,3,7,8tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD), one of the two most toxic of the 17 dioxins
and furans. With the 1999 value of 23 fg/m3 as a guideline, the dioxin and furan
concentrations, which average 5.8 – 21.7 fg/m3, are within this guideline.
The ambient concentration determined in Rillito is the sum of the 17 congeners,
expressed as the toxic equivalency of the mixture relative to the single congener 2,3,7,8TCDD. Since the TEQ range of the mixture is less than the guideline value of the
congener, the overall toxicity of the Rillito dioxins and furans is within the health-based
guideline.
Comparison with Urban and Background Sites
Ambient sampling programs for dioxins and furans, conducted by the U. S. EPA, the
California Air Resources Board, and other investigators around the world, have
accumulated a wealth of data in a wide variety of settings. In the EPA program in 2001,
the annual average dioxin and furan concentrations at 22 rural sites ranged from 2 to 28
fg/m3 TEQ. At eight remote sites the range was 0.3 to 2.8 fg/m3 TEQ. Included in these
remote sites are two Arizona sites: the Grand Canyon (1.4 fg/m3 TEQ) and Chiricahua
National Monument (0.4 fg/m3 TEQ). The California program has sampled at five sites in
the San Francisco area and at five in the Los Angeles area: concentrations have ranged
from 20 to 40 fg/m3 TEQ. Figures 2 and 3, which display these and other published data,
place the Rillito concentration of 5.8 – 21.7 fg/m3 TEQ (average of 13.7) above the
remote sites, within the range of the rural sites, and somewhat lower than most of the
urban sites. Of considerable interest are the highest annual average concentrations
reported: from 160 to 480 fg/m3 TEQ.

16

Figure 2.

Annual Average Dioxin and Furan Concentrations from
Sites with Lower Concentrations
Chiracahua
Grand Canyon

Crockett, CA (SF)
Rillito
Richmond, CA (SF)
Australia-Background
Boyle Heights, CA (LA)
Wilmington, CA (LA)
Oakland
San Francisco
Rubidoux, CA (LA)
0

5

10

15

20

25

30

35

fg/m3 (TEQ)

Figure 3.

Annual Average Dioxin and Furan Concentrations from
Urban Sites with Higher Concentrations

San Jose, CA (SF)
Reseda, CA (LA)
Livermore, CA (SF)
Syndey, Australia - 1
Catalonia, Spain
Japan
Syndey, Australia - 2
0

100

200

300

400

500

600

fg/m3 (TEQ)

17

Seasonal Variation
Concentrations of nearly all air pollutants (the photochemically produced ozone is an
exception) tend to be higher in winter than summer, with spring and fall midway between
the two. This seasonal pattern is a result of the surface nocturnal temperature inversions
lasting three to four hours longer in the winter. This pattern also reflects the generally
higher daytime wind speeds in the summer. Because winter conditions have less
horizontal and vertical dispersion, air pollutant concentrations tend to be higher. Dioxin
and furan concentrations in Rillito are no exception, as Figures 4 and 5 illustrate.
Figure 4. Monthly Variation of Dioxin and Furan Concentrations
40
35

TEQ(fg/m3)

30
25
20
15
10
5
0
Jan

Feb

Mar

April

May

June

July

Aug

Sept

Oct

Nov

Dec

Figure 5. Seasonal Variation of Dioxin and Furan Concentrations
30

25

TEQ(fg/m3)

20

15

10

5

0
Dec,Jan,Feb

Mar, April

June, July, Aug

Sep, Oct, Nov

18

Comparison of Congener Profiles
In this section three profiles of the 17 dioxins and furans are examined:
1. the profile determined in the ambient monitoring at Rillito;
2. the profile of the APCC kiln emissions determined by the stack testing; and
3. the profile for mobile sources, virtually the same for diesel and gasoline fuels,
taken from the literature (D. Cleverly et al, “The Congener Profiles of
Anthropogenic Sources of Chlorinated Dibenzo-p-Dioxins and Chlorinated
Dibenzaofurans in the United States”, Organohalogen Compounds, Vol 32: 430435, 1997).
Each profile is simply a bar chart that shows the percentage contribution of each of the 17
congeners to their total concentration, unweighted by the toxic equivalency. For the
ambient profile, the percentages come from the annual averages calculated by setting the
nondetectable values to one half the reporting limit. When the profile of an emissions
source is similar to an ambient profile, then the measured ambient concentrations can be
viewed as resulting from these emissions. Conversely, when the source profile is
substantially different from the ambient profile, its emissions are not a major contributor
to the ambient concentrations. The following Figures 6-7 depict these three profiles.
In the figures and following discussion, the 17 congeners are assigned a letter name, as
shown in Table 10.
Table 10. Dioxins and Furans: Letter Names
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF

A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q

19

Figure 6.

Dioxin and Furan Congener Profiles, Expressed as
Percentages of their Total Concentrations, Unweighted by
Toxic Equivalency, for (a) Rillito Ambient Air, (b) Cement Kiln
Emissions, and (c) Mobile Sources

70
60
50

Percent

(a) Ambient
40
30
20
10
0
A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

70
60

Percent

50
40
(b) Kilns
30
20
10
0
A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

70
60
(c) Mobile

Percent

50
40
30
20
10
0
A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

20

Figure 7.

Dioxin and Furan Congener Profiles, Expressed as
Percentages of their Total Concentrations, Unweighted by
Toxic Equivalency, for (a) Ambient Air and Cement Kiln
Emissions, and (b) Ambient Air and Mobile Emissions

70
60

Percent

50
40
Ambient
Kilns

30
20
10
0
A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

70
60

Percent

50
40
Ambient
Mobile

30
20
10
0
A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

21

These bar graphs clearly show that both sources, the cement plant and the mobile,
contribute to the ambient concentrations, although the mobile source profile resembles
the ambient profile considerably more than does the cement plant’s. From Figure 7a the
three most prominent congeners in the kiln profile – H, I, and J – are nearly absent from
the ambient profile. In contrast, Figure 7b reveals that the four most prominent
congeners of the mobile profile – F, G, O, and Q – coincide with the four most prominent
congeners of the ambient profile. Taken together, these profiles strongly suggest that
dioxins and furans from the interstate and railroad have a much greater effect on the
ambient concentrations in Rillito than do the emissions from the cement plant. The better
match of the mobile profile with the ambient can also be seen in the percentages given in
Table 11.
Table 11.

Dioxin and Furan Profiles, the Percentage Contribution to the
Total Concentration, Unweighted by Toxic Equivalency (TEQ),
for Rillito Ambient Air, APCC Kiln Emissions, and Mobile Sources

Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF

Letter
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q

Rillito
APCC
Mobile
Ambient
Kilns
Sources
0.13
1.68
0.55
0.89
1.95
0.22
1.00
0.75
0.11
1.43
0.90
0.22
1.66
0.66
0.11
16.27
2.00
5.59
58.95
2.33
27.41
0.71
41.37
1.64
0.89
13.35
1.10
0.95
17.43
1.10
1.39
4.12
1.32
1.09
4.38
1.43
0.94
3.82
0.77
1.55
1.01
1.10
5.81
3.03
6.58
1.70
0.46
0.77
4.62
0.74
50.00

Ambient & mobile similarities
Kiln and ambient dissimilarities
Kiln and ambient similarity

22

To summarize the congener profile analysis, the data are highly suggestive that the
ambient dioxins and furans, while influenced to a small degree by the cement plant, are
much more closely related to the mobile source emissions. What the data also suggest is
that other sources are contributing, as well. For example, congeners F and G of the
ambient profile far outweigh the same congeners of the mobile profile, while the mobile
congener Q is ten times greater than its ambient counterpart. These ambiguities may
stem from deposition of dioxins and furans onto soil, followed by chemical alteration,
with their subsequent resuspension, or from other combustion sources in the vicinity.
Health Significance of Ambient Concentrations
Dioxins and furans are produced as trace byproducts of many combustion processes
including burning of trash and landfills, forest fires, industrial combustion processes
(such as Portland cement kilns), and combustion of gasoline and diesel in vehicles. They
accumulate in the food chain, and, as a result, people are exposed to small amounts in the
food supply and have low concentrations of dioxins and furans in their fat tissue.
In this report, evaluation of human health risks associated with exposure to emissions of
dioxins and furans is based on an assessment of long-term exposure levels and potential
health outcomes. In the case of exposures to dioxins and furans, the health outcome of
concern is increased cancer risk. The annual AAAQG for 2,3,7,8-TCDD was set at a
level that corresponds to a lifetime cancer risk of one-in-one million (0.000001). That is,
the concentration of 2,3,7,8-TCDD in air that would result in a theoretical increase in
cancer risk for an individual of one-in-one million after a lifetime of exposure at
that level.
As discussed above, people are exposed daily to dioxins and furans through their diet.
The USEPA has estimated that more than 90 percent of dioxin and furan exposure in the
United States occurs though food, and current typical daily intake of dioxins and furans
in food are in the range of 0.000001 microgram TEQ per kilogram body weight per day.
The daily dose of dioxins and furans received through inhalation of air at concentrations
equal to the annual AAAQG is small compared to current daily dietary intake,
comprising less than 0.5% of the dose received daily from food.
Conclusions
1. Dioxin and furan concentrations in Rillito are lower than the Arizona Ambient Air
Quality Guidelines.
2. Emissions from the cement plant contribute marginally to the ambient dioxins and
furans at Rillito, with mobile source emissions being the dominant source.
Because the ambient and mobile profiles do not match perfectly, however, other
sources such as resuspended soil dusts onto which dioxins and furans have been
deposited may be a factor.

23