DSHW-1999-005230 - 0901a06880158023

lr
TRC JUL 0 2 1899
Division of Solid & Hazardous Waste
Utah Department of Environmental Quality
•^mmsmmmimmmmiimmm:mfmi iiiiiPi«illll$l6lliiillllfli|^
iMIiiiRiilOIWillillEilililiER
•^;^^
w$&.
W^:i^'$
W^^^§M
5|f'^^B^-
::$H;^-M(^^
•-^^^-^i-r;:;^ :T^.:•~'••;^.^•;'"^•-;;••y•v:v;K.:.•-••. , •-:y-'.'----i:;,^'•,.r~''^.'\!r.-'^v-i^ ••-• "^^^ .-•••••?'-'-•'- - - -• .-.- '.. '
'.iJMMXBIm ••:/•• i:.: ^.- im^0^'WMmi$ifm0( ^ • • • '-yyr: yi
y^':'i^0^'0}^^$i-l^^ ••••. '^''§^:-4-Syf:-0-
''J^'rM^!0W^^: 1 ••^ilvT'i'o'^^o' '^Prpor^ipn^'Utah;i>^^^ '••; •::^'-•iv>#-^'^>;i!;;:i;
• •• '•^i'-S'SSSft;\li:>^^^^ AFB, Dhio>.V-
'•'• ':i;''^'K??''';*^i^^;'i;!S\Uriii^ •TecRholoqiesVGorploratlon^^Ca ifbrhia •.;•••?; 'i.-': •;^:;i;-'^':--:- •
• •••: -'•^-;>s;^M:^^-s-;;:-r;^i ••;• •; • -^t-y .:••••
A^l;'
••'•.••..'>;T-
TABLE OF CONTENTS
I
}
PAGE NO.
1.0 INTRODUCTION AND BACKGROUND 1
2.0 SAMPLING AND ANALYSIS 1
2.1 Sample Collection 1
2.2 Analysis 2
2.2.1 lon-Chromatography 2
2.2.2 Triphenylstibonium Titration 2
3.0 DISCUSSION 3
TABLES
APPENDIX A: AVAILABLE UPON REQUEST
ATTACHMENT 1: DETERMINATION OF PERCHLORATE BY
lON-CHROMATOGRAPHY
Ail TRC paper is recyclable and made from recycled paper. TRC
1.0 INTRODUCTION AND BACKGROUND
The presence of naturally-occurring perchlorate has been identified in sodium- and
potassium-nitrate mined from caliche deposits in Chile (i.e., in Chilean saltpeter). Chilean
saltpeter is used as a nitrate fertilizer throughout the worid. Schilt (1979) describes
observations of perchlorate in Chilean saltpeter as far back as 1896, with concentrations
ranging from trace amounts to just under 7 percent. Levels of perchlorate of 1 to 1.5 percent
in refined and crude Chilean saltpeter, respectively, measured in 1914 were also described
(Schilt, 1979). A study in 1972 attributed poor soybean growth to elevated concentrations of
perchlorate in Chilean nitrate fertilizer. The relatively broad range of perchlorate
concentrations was attributed to varying degrees ofthe liquid recycling during the production
of the nitrate salts, and the possible nonuniform distribution of naturally-occurring perchlorate
in the caliche. Review of available literature and limited testing of selected samples of chemical
fertilizer indicate that some chemical fertilizers contain perchlorate.
Based on the above-referenced literature, data and the results ofthe limited fertilizer sampling
and analysis of Califomia fertilizers for perchlorate conducted in August of 1998, additional
studies to determine if perchlorate is typically present in nitrate type fertilizers was performed.
In order to confirm the above results and also to obtain more information on the presence of
perchlorate associated with the use of fertilizers, the following activities were completed, and
are presented in this report:
• Confirmatory sampling and analysis of previously tested products.
• Collection and analysis of fertilizers nationwide.
• Collection and analysis of fertilizer raw materials.
2.0 SAMPLING AND ANALYSIS
2.1 SAMPLE COLLECTION
1. To confirm the results of the previous perchlorate data indicated above, samples were obtained
from different production lots of the previously tested Califomia fertilizers. These samples
were purchased from various commercial suppliers and retail sources.
2. To obtain information on the potential presence of perchlorate associated with the use of
chemical fertilizers, additional samples were collected from the Kansas City, Missouri and
Long Island, New York areas. Samples of the fertilizers shown in Table 1 were collected from
various commercial suppliers and retail sources.
6/18/99 - Rev. 1 TRC
3. As part of this study, samples of two raw materials u.sed in chemical fertilizer production were
also collected. Caliche (Chilean nitrate) was obtained from the Aconcagua Nitrate Mine in
Calama, Chile by TRC. A Langbeinite (a potassium/magnesium sulfate mineral) ore sample
(crashed) was provided by IMC Minerals in Albuquerque, New Mexico.
4. The fertilizers and raw material samples were collected under chain-of-custody procedures and
forwarded to the primary laboratory for analysis.
5. The primary laboratory prepared extracts of each sample, which were then shipped to six
outside laboratories for confirmatory analysis.
2.2 ANALYSIS
2.2.1 lON-CHROMATOGRAPHY
1. The primary laboratory (Del Mar Analytical, Irvine, Califomia) analyzed the fertilizers using
the California Department of Health Services Method of lon-Chromatography for Perchlorates
(Attachment 1). Table 1 provides a summary ofthe lon-Chromatography perchlorate results.
2. As discussed above, six outside laboratories analyzed the perchlorate extracts, for confirmalion
purposes. However, these laboratories did not all perform the analyses using the Califomia
DHS Method for perchlorates. Table 2 provides a list of the methods used by each laboratory
and the differences between the method used and the Califomia PHS method. Table 1
provides a summary of the results of the analyses conducted by the outside laboratories
discussed above. As noted in Table 1, the outside laboratories analyzed a 1:10 extract. The
results in Table 1 for the outside laboratories were adjusted by a factor often to allow
comparison with the primary laboratory data. Copies of the laboratory reports are provided in
Appendix A.
2.2.2 TRIPHENYLSTIBONIUM TITRATION
1. As part of this study, one set of the fertilizer extracts, and various control samples prepared by
TRC were analyzed using the triphenylstibonium titration method after purification of the
extracts using ion exchange resin adsorption. The purpose of performing the titration study
was to chemically confirm the presence of perchlorate in the extracts. The triphenylstibonium
titration method has been shown to be highly specific for perchlorate, and therefore can be
J
I • • .
1 .6/18/99 - Rev. 1 2 ilY^
used to confirm the lon-Chromatography results, which arc highly accurate, but may be less
specific in detecting perchlorate. Table 1 also incliulcs a summary ofthe triphenylstibonium
titration results.
3.0 DISCUSSION
1. The results of the sampling and analysis of the fertilizers indicates that perchlorate was
present at varying levels in approximately 29 out of the ^^ samples tested (i.e., 90 percent).
The detectable perchlorate levels in these fertilizers ranged from approximately 2 ppm to over
11,000 ppm with most being greater than 500 ppm. T^ho perchlorate concentration averaged
roughly one-half of one percent (0.5 percent or 5,000 pp,,,), as measured in the 29 fertilizers
that were positive for perchlorate. These results indicate that perchlorate levels can vary
significantly between fertilizer brands and formulation types. Further, the results ofthe analysis
of differem production lots ofthe previously tested fonili/ers showed significant varialion.s
between the different lots, indicating cither variations in pit^duction practices or perchlorate
levels in the raw materials. Both raw materials tested, c\ilichc and Langbeinite contained
elevated levels of perchlorate; 30,000 ppm and 13,000 pp,„ (i.e., 1.3 to 1.5 percent), respectively.
2. A comparison of the perchlorate results from the primary laboratory and the outside laboralories
shows an excellent correlation. As indicated in Tabic 1, the standard deviations are very low,
indicating a high level of precision. Table 1 also indicates the various duplicate samples and
their results. The duplicate samples were found to goneially agree within normal laboratory
limils (i.e., ±50 percent), with the exception of one diiplie;,(e sample PS-17/PS-24 which
showed an unexplained significant difference. This dil leience is most likely due to an error in
the extract preparation, sinceall of the laboratories had similar results for these samples.
3. The results of the triphenylstibonium titration, appear to generally agree with the
ion-chromatography results, although the method appears K, slightly over estimate the
perchlorate levels. Quality Control (QC) samples submiueij to the laboratory containing
known levels of perchlorate, chlorate, chlorite, ammonim,, nitrate and thiosulfate showed no
interference with the perchlorate analysis. Additional duplicate samples submitted also showed
good correlation, within the expected limits.
6/18/99 - Rev. 1 TRC
used to confirm the lon-Chromatography results, which arc highly accurate, but may be less
specific in detecting perchlorate. Table 1 also includes a summary ofthe triphenylstibonium
titration results.
3.0 DISCUSSION
1. The results of the sampling and analysis of the fertili/ois indicates that perchlorate was
present at varying levels in approximately 29 out of the ^^ samples tested (i.e., 90 percent).
The detectable perchlorate levels in these fertilizers ranged from approximately 2 ppm to over
11,000 ppm with most being greater than 500 ppm. The perchlorate concentration averaged
roughly one-half of one percent (0.5 percent or 5,000 pp,„), as measured in the 29 fertilizers
that were positive for perchlorate. These results indicate that perchlorate levels can vary
significantly between fertilizer brands and formulation types. Further, the results ofthe analysis
of different production lots ofthe previously tested fonili/ers showed significant variations
between the different lots, indicating cither variations in production practices or perchlorate
levels in lhe raw malerials. Both raw materials tested. c\,lichc and Langbeinite contained
elevaied levels of perchlorate; 30,000 ppm and 13,000 pp,,, (i.e., 1.3 to 1.5 percent), respectively.
2. A comparison of the perchlorate results from the primary laboratory and the outside laboralories
shows an excellent correlation. As indicated in Table 1, il,e standard deviations are very low,
indicating a high level of precision. Table 1 also indicates the various duplicate samples and
their results. The duplicate samples were found to generally agree within normal laboratory
limits (i.e., ±50 percent), wilh the exception of one duplicate sample PS-17/PS-24 which
showed an unexplained significant difference. This ditfeience is most likely due lo an error in
the extract preparation, since all ofthe laboratories had similar results for these samples.
3. The results of the triphenylstibonium titration, appear to generally agree wilh the
ion-chromatography results, although the method appears to slightly over estimate the
perchlorate levels. Quality Control (QC) samples submitted to the laboratory containing
known levels of perchlorate, chlorate, chlorite, ammonium nitrate and thiosulfate showed no
interference with the perchlorate analysis. Addilional iliiplicate samples submitted also showed
good correlation, within the expected limits.
6/18/99 - Rev. 1 TRC
TABLE 1
FERTILIZER PERCHLORATE LIQUID CHROMATOGRAPHY TESTING PRELIMINARY RESULTS
)
•
MANUFACTURER BRAND NAME
pbuined 1/27/99 Kanssis Ciiy, Missouri
PS-II
PS-13
PS-IO
PS-()7
PS-()5
PS-14
PS-Ol
PS-()8
IPS-12""
|PS-02
ps.m">
PS-()3
PS-04
Scon's
Shullc
Acme
Sudbui>
Peters
lobe's
Fchilome
Osmocole
Osmocote
Ringer
'Ringer
Peiers
HIYield
Mirocle-Gro Lawn Food
Rose Plus
Stump Remover
Potash
All Purpose Planl Food
Plant Food Spikes
Start-N-Grow Planl Food
Vegetable und Bedding
Plant Food
Vegciuble and Bedding
Plant Food
Supreme Gardens
'Supreme Gardens
Lawn Food
Nilrale of Soda
Obliiined 1/28/99 Long Island, New York
PS-15
PS-21'"
PS-20
PS-16
PS 19
PS-18
PS-23
PS-17
PS-24"'
PS-22
Ringer
Ringer
Frank's
Peiers
Osmocote
Osmocote
Vigom
Jonathan Green
Jonathan Green
Scons
Lawn Restorer
Lawn Restorer
Gmw
All Purpose Plant Food
Vegetable and Bedding
Plant Food
Outdoor and Indoor Flam
Food
Tomato and Vegetable
Plant Food
Fall Fertilizer
Fall Feililizer
Miracle-Gro Lawn
Pertllizer
Ohiiiiiiiid frnm Califomia Source
l'S-31
PS-.33
PS-28
PS-25
PS-26
PS-27
PS-29
PS-32
PS-()6
CNC?"
ncsl'"
Grow More Corp'"
Best'"
Bandini'"
Plant Marvel
Dexol
K Power
Peiers
Champion Polassium
NilRlf
K-Power Polassium Nilrale
Umw More
Triple Sixteen
Sui Po Mag
Natriculture
Slump Remover
Mini Prills
All Purpose
Sample of Raw Malerials and Conlrol Sample 2/8/99
PS-34
'S-30
LCS-I"
LCS^'"
Blank
MC Group
-
--
Caliche (Chilean NiUTite 1
So||rc()
Langbeinite Ore
-
"
•-
N-P-K'"
RATIO
36-6-6
19-24-24
NA
0-0-44
20-20-20
16-2-6
18-6-12
14-14-14
14-14-14
7/7A)7
7/7/07
38-4-4
16-0-0
l(V2/0f>
IQ/2A)6
I5-.30-15
20-20-20
14-14-14
18-6-12
I(V8/I4
1(1/18/20
l(V18/20
31-3-9
13.5-0--15
13.75-0-46
—'.V30.3*;-^
16-16-16
(M)-22
12/31/14
Unknown
I3.75-(MA
20-20-20
Unknown
Unknown
Unknown
Unknown
Unknown
PRIMARY
CONSTITUENT
TRIPHENYL-
STIBONIUM
TITRATION
(mg/kg)'"
1 ION-
CHROMATO-
GRAPHY DEL
MAR
ANALYTICAL
RESULTS
1 (mg/kg) LABI
SPLIT SAMPLE ANALYSIS RESULTS (mgflig)'"
LAB 2
Potash derived from
^assiyin-tfjttale
Potassium Nitrale
Polassium Nitrate
Potassium Chioridc
Polu,ssiuin Nitrate
Pola,sstum Nitrate
Polassium Nitrate
Ammonium Nitrate
Ammonium Nitrate
Nitrate of Soda
Nilrale of Soda
Potassium Nilrale
Nitrale of Soda
8.6(X
•^^
• • <5f
5.44(
7,I5(
10,351
. 5,9(X
l.25(
75(
2,951
5,2(X
9.9(X1
6,8(X
7.4(X
13
<0.4
4,8(X
6,.3(X
8.I(X
4,7(X
94(
54(
3,2(X
3,9(X
8,6(X
7.4(X
8.I6(
<6.(
<6,(
4,73(
• 6,7(X
9,481
4,681
\.m
63(
3.49(
1 4,3«
8.95(
7,62(
LAB 3 LAB 4 USAF
6.26(J 8,8(xJ 8.29(
<4I <0,8 9,(
i . <4( <:0.4 <3.(
i 8.83(M S,tXX 4,921
4,760 6,2(Xl' 6,.1(X
8,6411 9.IXX< 9.921
5.1511 5,.3IX1 4.711
968 970 1,011
597, 5.10 581
2..I60' 3.11X1 3.24(
5.12(1 4„1(«ll 4,38(
7.73(1 9,4(X) 8,86(
6,3(XVI 1,840"'" • «,<XX) 7.66(
USEPA
7.0681 12.061
1 "
i 4,36'
6,89)
6.89(
• 4.87;
80.1
52f
3,071
4.045
7.678
6.908
Nitrate of Soda
Nilrale nf Soda
Murialc of Pniush
Polassium Nilrale
Ammonium Niirate
Ammonium Nilrale
Muriate of Potash,
Sui Po Mag
Muriate of Potash
Muriate of Potash
Murialc of Potash
6,140
3,4(X
6.25(
6,9R(
3.I(X1
740
450
<5n
2,850
l,7(XI
5,.3(X
2.4IX
5.4IX
7,3(X
2.S(X)
6(XI
360
<().4
2,400
l,3(XI
6,IXiO
2,96(
5,3K(
7,511
2.650
733
464
<6.0
2,590
1.549
5.191
2.5f.O
KUXXI
5.641
3.271
832fl93"°
738
<4(
2.84(
2,646
6,IIX
2,9(X)
4,8IX
7.4(X
2.7(X
750
.141
2,2
2,21X1
l.4(XJ
6.080
3,(XX)
5,560
7,211
2.660
m
4.11
. •cl.f
2,480
1,531
4,673
2.376
7,098
6,194
2,053
904
514
(11
2,530
1.995
9,571X14)
1
6,61(
9,721
13.3.11
5.721
1,381
5.11
4.65(
6,23(
I2.67(
l().34(
STATISTICAL
EVALUATION
MEAN"-'
PERCENTAGE 01
PERCHLORATE
STANDARD 1
[DEVIATION" 1
8.29 l.86(
9 f
i <3.« )
5,609 l„S9;
6,705 1.49"
9.338 2,()t(
5,019 .19:
1.023 17c
563, 4.
3,302 " 685
4,623' 807
9,127 1.686
8,153 1,175
8,84(
3.4 K
4.24(
I2.55(
3.84(1
160
97(1
(1
3,97(1
1,68(1
6.035
2,801
6,068
7,686
2.810
678
545
4
2,716
1,729
I.-I53
375
1.943
2.257
578
255
229
8
586
462
^.,..,,.^.,.,^-,i_
~
-
Potassium Nilrale,
Unknown
Chileans Nitrate
Potassium Nitrate
Sodium Nitrale
Magnesium. Sulfur,
Potassium
"
-
lA.AlXI
8.25(1
--^—-nriiSaa
5,250
3.55C
6,31X1
3,8(K
S.2(X
.1,400
I2.(XX)(7_S(X))"
7,tO()(27,5(X))"
=2-;J(M>;i;4,«<y)';'
4,200 (18.(XW)"
3,000 (|5,(XX))"
" ' 3,.30(1
" 4,9(X)
6.M
.3.l9(l|
13.290
7,563
,_.^2,.5'.»
4,22.1
3.073
6.680
5,189
ft..1«
3.220
14,380
2.6(X)
3,680
4,810
7.120
5,29(i
6..1(X1
3,680
I3,(XXI
7,4(X
^^-^2,2{i<:)
4,(XX
3,2(X
5.5(X
5.4(X
6.11X1
3,IXXl|
32,8(XI
15,600
90
105
<50
30.(XX)
I3,(XX1|
-
--
26.310
I4,3(X)
96
104
<1).(X)3
32,980
20.090|
99
94
<0.(X)3
27,(XX)
I4.(XXI
IIXI
99
<(I,IX)4
15,KX)
7.871
-iii.~2,4S0
4,3 id
3.160
6,450
5,420
6,56(1
3,090
11.749
7.201
-::-ud
3,727
2,834
6,246
4,57f
5.862
.1.128
30,9(X)
I6,4(X)
941
97
<().(X)3
26,120
13.649
99
98'
21.14(1
10,14(1
7.0511
4,511
17.440
6,73(1
8,5.1(
4.74(1
41.35(1
21.160
(1
14,1)58
8,582
.. 3..5()?
5,441
4.3691
7.819
5.358
6.576
3,43.5]
5.947^
4.455
2.101
2.754
2,18(1
4.290
676
889
615
30.666
16.086
98
98
<0.(X)l
.5..17I
3,288
2
4
l).(XX)7
0.8
1 o.ix'tf^B
<,(XX)
l).5(
•: 0.6"
0,9:
0,
1. "•
l),(K
0.3?
0.4(
0.91
0.81
i
0.28
().(
0,77
0,28
0,07
0,05
().(XX)4
0,27
0,17
1.41
0.85
_ .,0,35
0.4;
. ' • • 0,78
0.53
0.65
0,14
3.06
l.l
().(X)9
• 0,(X)9
O.IXXXX))
[
. • ' "
"
.
i
^^-^^
'" Laboratoiy exinici results (mg/L) were converted to mg/kg unils for comparalive purposes, by mulliplyiiig Ihe result by 10 to accounl for the 1:10 dilulion used to prepare Uie extracts.
'" Nitrogen, Phosphorus and Polassium Levels, _^
'" Repeat of Prior Testing.
'" Prior Testing Results, •'
'*' Duplicale for PS-()2 1,::.,.^ .v.-';'..
'*'Duplicale for PS-()8 . '•
'" Duplicale for PS-17 '
'" Duplicale for PS-15
'•" Laboraiory Conlrol Standard of 100 ppb Qualiiy Conlrol Samples "«"'
'"" Inicrim duplicale sample analysis.
"'' Due to interferences, no value was reported.
"" Mean values calculated from results obtained by Del Mar Analytical, Labs I through 4 as noted, the United Stales Air Force and the United
Environmenial Protection Agency. For samples with duplicate analyses, Ihe duplicale results were averaged before calculating the overall mean and standard deviation,
"" Standard deviation values calculaled from results obtained by Del Mar Analytical, Labs I through 4 as noted, Ihc United Slates Air Force and
United Slates Environmental Protection Agency, For samples with duplicale analy,ses, the duplicale results were averaged before calculaling the overall mean and siandard deviation,
"" Considered an outlier - nol used for statistical evaluation.
6/18/99-Rev. I