NCDOT*s Experience Using Ultraviolet - Trb

Report
SESSION: MANAGING CONTAMINATION
North Carolina DOT’s Experience with UV Fluorescence for Measuring Petroleum Contamination in Soil
Cyrus Parker, LG P.E., NCDOT
The North Carolina Department of Transportation (NCDOT) has started using Ultra Violet Fluorescence (UVF) to analyze soil
samples for petroleum compounds instead of the traditional USEPA Method 8015. This presentation will discuss the 3 ways
NCDOT has used the UVF test, the test results and potential cost and time savings associated with using the UVF test instead of
Method 8015.
Cyrus is the GeoEnvironmental Supervisor for the North Carolina Department of Transportation. He has over 18 years of
experience with geoenvironmental engineering as both a consultant and with the Department of Transportation. Mr. Parker has a
bachelor’s degree from North Carolina State University and licenses to practice Geology and Engineering in the state of North
Carolina.
Ultra Violet Fluorescence Soil Analysis
Cyrus Parker, LG, PE
North Carolina Department of Transportation
The Green Box
Overview
Ultra Violet Fluorescence (UVF)
•
•
•
•
Introduction to UVF
Data Comparison with EPA Method 8015
User Feedback
Cost Comparison with EPA Method 8015
Weigh 10 Grams Soil
Add Methanol to soil. Shake. Wait.
Extract Sample and Transfer to Cuvette
Place Cuvette into analyzer, Enter
Sample ID and Click Analyze
Diesel Fuel
Heavy Fuel Oil
Regulatory Approval
Suspected Contaminant
1a. Low Boiling Point Fuels:
gasoline, aviation gasoline, etc. a
1b. Ethanol-Gasoline Blends
2. Medium/High Boiling Point Fuels:
jet fuels, kerosene, diesel, fuel oil #2,
biodiesel (containing diesel), etc.
Varsol, mineral spirits, naphtha.
3. Heavy Fuels: #4, #5, #6 fuel oils,
motor oil, hydraulic fluid, etc.
Mineral oilc
4. Used / Waste Oil
Analytical Methods for
Preliminary Investigation
Samples
EPA 8015C for TPH-GRO
(or UVF for TPH)b
EPA 8015C for TPH-GRO
(or UVF for TPH)b and EPA
8260B
EPA 8015C for TPH-GRO
and EPA 8015C for TPHDRO (or UVF for TPH)b
Analytical Methods for Samples
Collected after Preliminary
Investigation
EPA 8260B and MADEP VPH
EPA 8015C for TPH-DRO
(or UVF for TPH)b
EPA 8270D and MADEP EPH
EPA 8260B, EPA 8270D,
MADEP VPH, MADEP
EPH, (or UVF for TPH and
PAH)b and EPA 3050B or
3051A Prep: Total Metals
(Cr and Pb), EPA 8081B
(pesticides),and EPA 8082A
(PCBs)d
EPA 8260B, EPA 8270D, MADEP VPH,
MADEP EPH, EPA 3050B or 3051A
Prep: Total Metals (Cr and Pb), EPA
8081B (pesticides), and EPA 8082A
(PCBs)e
EPA 8260B, EPA 8270D, MADEP VPH,
and MADEP EPH
Key Notes:
UVF and EPA Method 8015
Data Comparision
NOTE:
This summary represents a review of soil samples collected as grab
samples on-site, analyzed immediately in the field, followed by
submitting a separate container to the laboratory for 8015 analysis.
Some samples noted were submitted for UVF testing for 24 or 48
hour TAT.
Thorough Homogenization was not applied. It is important to note
that the Lab and the UVF did not test the same 10g sample or extract
in this correlation study.
1400
Lincoln DRO Samples
1200
D
1000
R
O
(
m
g
/
k
g
800
QED/DRO
8015/DRO
600
)
400
200
0
0
2
4
6
Sample Number
8
10
1) HA-1
2) HA-2
3) HA-3
4) HA-4
5) HA-5
6) HA-7
7) HA-8
8) HA-9
UVF/DRO
503
68.1
1322
0
2.4
2.5
3.3
3.8
8015/DRO
264
46.2
1060
0
0
0
0
0
•In the last 4 samples the
Lab results a Non Detect
and the UVF results in
below the action limit.
The UVF results show that
there are still trace
amounts in the sample.
Note the fingerprint
example on next slide.
•ONSITE RENTAL
UVF Fingerprint Sample HA-5 (F&R, Lincoln)
5) HA-5
UVF/DRO
2.4
8015/DRO
0
UVF fingerprint trace clearly shows the presence of a degraded
petroleum hydrocarbon. This is consistent with all samples where
the Lab reports 0…
5000
1) 4-1
2) 4-3
3) SB-1
4) SB-2
5) SB-7
Rowan DRO Samples
4500
4000
3500
D
R
O 3000
(
m 2500
g
/ 2000
k
g
1500
QED/DRO
8015/DRO
)
1000
UVF/DRO
40.5
13.8
3504.6
2356
579.2
8015/DRO
123
6.6
3440
3220
528
• Values are very close
for the DRO range.
• A 30% error bar is
incorporated into this
graph to show that the
UVF/8015 values fall
within the 30% error
range.
500
• OFFSITE UVF LAB
0
0
1
2
3
4
Sample Number
5
6
6000
Rowan GRO Samples
5360
5170
4-1
4-3
SB-1
SB-2
SB-7
5230
5000
G
4000
R
O
•
QED/GRO
8015/GRO
)
1000
844.7
6.9
0
1
293.4
0
4.6
0
989.9
0
2
3
4
Sample Number
5
6
8015/GRO
6.9
0
5170
5230
5360
•
Highlighted numbers are very
close to standard calibrator values
for GC 8015.
All other samples correlated
•
OFFSITE LAB
(
m 3000
g
/
k
g 2000
UVF/GRO
4.6
0
844.7
989.9
293.4
SB-1 Lab Result
UVF/GRO
SB-1
844.7
8015/GRO
5170
SB-2 Lab Result
UVF/GRO
SB-2
989.9
8015/GRO
5230
SB-7 Lab Result
UVF/GRO
SB-7
293.4
8015/GRO
5360
Background Organics
SB-1
SB-2
SB-3
SB-4
UVF/GRO
844.7
989.9
0
0
UVF/DRO
3504
2356
0
0
*The top 2 fingerprints represent the presence of the contaminant in high
concentrations.
*The bottom 2 fingerprints represent how a negative on the Rowan site would
appear with background organics.
*Please note the dates in which the samples where collected and
the dates in which they were analyzed at the lab. UVF sample
data was generated within 24 hours.
The UVF fingerprints exhibit the high levels of background
organics in the samples, which may account for the high
recoveries and results in the lab data.
The Matrix Spike and Matrix Spike Duplicate for these samples
show 133% and 155% surrogate recovery. They have a qualifier to
that effect In the QA/QC data.
35000
Guilford DRO Samples
UVF/DRO
8015/DRO
110-3-10
12.9
<7.0
137-7-2
2.1
22.9
155-4-10
30.6
308
66-6-10
31283
26600
137-8-15
31283
30000
26600
66-14-8
190
252
66-19-9
915.3
5460
25000
66-25-12.5
4515.8
4580
D
R
O 20000
66-26-11
7184
0
116-16-10
21.5
148
116-16-12
13.7
17
(
QED/DRO
m
g
/ 15000
k
g
8015/DRO
)
10000
5460
5000
0
0
0.00 22.9
12.9 2.1
2
308
252
30.6
4
190
7184
4580
4515.8
915.3 0
6
8
Sample Number
148 17
21.5 13.7
10
12
*Sample 66-26-11 is a major
discrepancy. A review of
S&ME field notes indicated a
positive PID reading, odor
and discoloration for this
sample.
400
Guilford GRO Samples
380
360
340
320
300
280
G
R 260
O 240
220
(
m
g
/
k
g
200
UVF/GRO
180
176.2
8015/GRO
160
140
)
120
100
83.1
63.8
60
40.1
54
40
12.9
20
0
0
0
0
0
0
2
7.9
0
0
7.6
3.7 4.2
4
6
8
Sample Number
25.4
0
8015/GRO
0
12.9
<7.2
0
696
7.9
<6.7
40.1
0
63.8
120
Sample fingerprints on
the following slides
exhibit background
organics and
explanation of conflicts.
120
80
110-3-10
137-7-2
137-8-15
155-4-10
66-6-10
66-14-8
66-19-9
25-12.5
26-11
116-16-10
116-16-12
UVF/GRO
0
0
3.7
4.2
2432
7.6
83.1
176.2
54
25.4
8.8
8.8
10
12
• ONSITE LAB
UVF/DRO
8015/DRO
137-7-2
2.1
22.9
66-6-10
31283
26600
110-3-10
12.9
<7.0
137-7-2 Exhibits background organics
substantial in the low level sample.
66-6-10 Exhibits an over range sample that was
not diluted or recalculated with the UVF and
may have been closer to Lab result.
110-3-10 Lab reports <7.0 mg/kg for DRO,
UVF reports 12.9 mg/kg DRO. Product is
present in sample.
UVF/
DRO
8015/
DRO
137-8-15
<1.3
<6.1
66-19-9
915.3
5460
UVF/
GRO
3.7
83.1
8015/
GRO
<7.2
<6.7
137-8-15 Normal blank subtraction was not carried out
in this sample. UVF reported 3.7 mg/kg GRO and the
Lab reported <7.2 mg/kg GRO.
66-19-9 High concentrations of degraded diesel would
require a large dilution by GC for the DRO range and it
is possible the GRO range was diluted away in the
process. Lab reports 5,460 mg/kg DRO and < 6.7
GRO respectively. This product is NOT highly
degraded, thus to produce such high levels of DRO
and NO GRO would be atypical for Diesel fuel.
*PLEASE NOTE background and particulate present in the samples from this
site.
600
Pitt DRO Samples
101_SS1
96_SS-7
071_2-2
075_3
079-1
085_1
500
D 400
R
O
391
(
m 300
g
/
k
g 200
QED/DRO
8015/DRO
)
100
37.3
4.8
19.6
0
0
1
2
15.3
12.1
17.1
0
3
4
Sample Number
4.8
1.9
0
5
0
6
7
UVF/DRO
8015/DRO
4.8
37.3
15.3
12.1
4.8
1.9
19.6
391
17.1
0
0
0
• The Fingerprints
clearly show the
presence of
petroleum product.
1
Pitt GRO Samples
UVF/GRO
8015/GRO
0
0
<0.6
<07
<0.6
<0.6
0
0
0
0
0
0
101_SS1
96_SS-7
071_2-2
075_3
079-1
085_1
0.9
0.8
0.7
G
R
O 0.6
(
m 0.5
g
/ 0.4
k
g
0.3
QED/GRO
8015/GRO
)
0.2
0.1
0
0
0
0
0
0
0
2
0
0
0
4
Sample Number
0
0
0
0
6
8
• 100% Correlation
071_2-2
075_3
UVF/DRO
15.3
12.1
8015/DRO
17.1
0
71-2(0-2) Lab reports 17.1 mg/kg DRO and
UVF reports 15.3 mg/kg DRO, which shows
almost a perfect correlation.
75-3(0-2) Lab reports 0 mg/kg for DRO and
UVF reports 12.1 mg/kg DRO. A very clear
fluorescent fingerprint of fuel product.
079-1
085_1
UVF/DRO
4.8
1.9
8015/DRO
0
0
79-1(0-2) Lab reports 0 mg/kg DRO and the
UVF reports 4.8 mg/kg DRO. A very clear
fingerprint of fuel product is shown in the
fingerprint. Perhaps it was detected below the
lab PQL?
85-1(0-2) Lab reports 0 mg/kg DRO and UVF
reports 1.9 mg/kg. Perhaps it was detected below
the lab PQL?
96_SS-7
UVF/DRO
37.3
8015/DRO
391
Batch SS-7 QC Data
SS-7 Chromatogram
96_SS-7
UVF/DRO
37.3
8015/DRO
391
User Feedback
Overall experience using rapid UVF analysis.
9
8
8
7
7
6
5
4
3
2
1
1
0
0
0
Extremely poor
Poor
Satisfactory
Good
Excellent
Would you recommend replacing traditional TPH/GRO/DRO
chemistry with rapid UVF analysis for soil site assessments?
No
Yes
Are You Confident in Rapid UVF Analysis?
No
33%
Yes
67%
Cost Analysis
UVF Options
•
Onsite UVF Equipment Rental for Immediate Results
• ~$800 per day
•
•
•
•
•
Rental
Shipping
Supplies
Additional Labor
Onsite Laboratory for Immediate Results
• ~$1100 per day (all inclusive)
•
Ship to Offsite Laboratory for 24 or 48 hour Results
• 24 hour $55 per sample
• 48 hour $45 per sample
Method 8015 vs UVF Cost Analysis
$100,000.00
$90,000.00
$80,000.00
$70,000.00
$60,000.00
$50,000.00
Laboratory Cost
$40,000.00
$30,000.00
$20,000.00
$10,000.00
$Actual 8015 2 week Turnaround
Price if UVF 48 HR Turnaround
8015 vs UVF Cost Analysis
$350,000.00
$314,874.00
$300,000.00
$250,000.00
$200,000.00
Laboratory Cost
$150,000.00
$100,000.00
$51,682.50
$50,000.00
$Price if 8015 48 HR Turnaround
Price if UVF 48 HR Turnaround
The Real Question
Would the regulatory decision change if UVF was used
instead of Method 8015?
Change recommendations as a result of using UVF for GRO
samples?
Yes
10%
No
90%
Change recommendations as a result of using UVF for DRO
samples?
Yes
20%
No
80%
Change Site Recommendation as a result of using UVF
Based on 49 sites
Yes
16%
No
84%
Conclusions
•
•
•
•
•
Flexibility for onsite or laboratory analysis
Potential cost and time savings
Experienced users like the system
Less Experienced users can ship samples to laboratory
for analysis similar to their current process
Using UVF would not have change our
recommendations, most of the time
More Information?
Felecia Owen
QROS, LLC
[email protected]
919-278-8926
http://qros.us/
Cyrus Parker
NCDOT
[email protected]
919-707-6868

similar documents